DOCUMENT RESUME

ED 315 423 TM 014 423

AUTHOR Facione, Peter A.

TITLE Critical Thinking: A Statement of Expert Consensus
for Purposes of Educational Assessment and
Instruction. Research Findings and
Recommendations.
SPONS AGENCY American Philosophical Association, Newark, Del.
PUB DATE 90
NOTE 112p.; Some pages have broken type.
PUB TYPE Reports - Research/Technical (143)

EMS PRICE MF01/PC05 Plus Postage.

DESCRIPTORS Cognitive Ability; *Critical Thinking; *Curriculum
Development; Cu.riculum Evaluation; *Delphi
Technique; *Educational Assessment; Qualitative
Research
IDENTIFIERS *Experts; *Panel Consensus Technique

ABSTRACT
Using a qualitative research methodology, known as
the Delphi Method, an interactive panel of experts was convened to
work toward a consensus on the role of critical thinking (CT) in
educational assessment and instruction. In Delphi research, experts
participate in several rounds of questions that require thoughtful
and detailed responses. Panelists work toward consensus by sharing
reasoned opinions and reconsidering the opinions with regard to
comments, objections, and arguments offered by other experts. A total
of 46 scholars, educators, and leading figures in CT theory and CT
assessment research were gathered for the panel meetings. About half
of the ranelists were primarily affiliated with philosophy
departments; the others were affiliated with education, social
sciences, or physical sciences. Recommendations resulting from the
discussion rounds address the cognitive skill dimension of CT, the
dispositional dimension of CT, and specific recommendations on CT
instruction and assessment, including development of a CT curriculum.
A discussion of commercially available CT assessment tools, a
bibliography with an emphasis on assessment, and a set of letters
which chronicle the progress of the Delphi research group are
appended. (TJH)

******************************************************Ve***************
Reproductions supplied by EDRS are the best that can be made
from the original document.
********************************************************************ft**
 U.S. DEPARTMENT OF EDUCATION "PERMISSION TO REPRODUCE THIS
Office of Educational Research and Improvement MATERIAL HAS BEEN GRANTED BY
EDUCATIONAL RESOURCES INFORMATION
CENTER (ERIC)
r This document has been reproduced as
received from the person or organizahor
arcx frAcfofve-
originating it
r Minor changes have been made to improve
reproduction (IL:8NY

F oints of view or opinions stated in this cioc TO THE EDUCATIONAL RESOURCES

ment do not necessarily represent official
OERI position or policy INFORMATION CENTER (ERIC)."

CRITICAL THINKING:

A STATEMENT OF EXPERT CONSENSUS

FOR PURPOSES OF

EDUCATIONAL ASSESSMENT AND INSTRUCTION

Research Findings and Recommendations

Prepared for the
Committee on Pre-College Philosophy
of the
American Philosophical Association

by
Peter A. Facione,
California State University, Fullerton

(c) 1990 P. A. Facione

BEST COPY AVAILABLE.

 ONTENTS
I The Critical Thinking Movement and CT Assessment 1

II -- Research Methodology and Purpose 4

III -- The Cognitive Skill Dimension of Critical Thinking 8

IV The Dispositional Dimension of Critical Thinking 20

Procedural, Laudatory and Normative Uses the Term "CT" 21
Dispositions of the Good Critical Thinker 27

V Further Recommendations on CT Instruction and Assessment 27

The CT Goal 28
The CT Curriculum 30
The CT Assessment Tool 30
The CT Instructor 32

VI -- The Delphi Research Panel 33

TABLES
TABLE 1 Consensus Statement Regarding Critical Thinking
and the Ideal Critical Thinker 3

TABLE 2 Project History 7

TABLE 3 Consensus List of CT Cognitive Skills and Sub-Skills....12

TABLE 4 Consensus Descriptions of Core CT Skills and Si.b-Skil1s 13
TABLE 5 Affective Dispositions of Critical Thinking 26

TABLE 6 Consensus Statement on Teaching and Assessing CT Skills 29

TABLE 7 Participating Critical Thinking Experts 35

APPENDICES

Appendix A: Commercial CT Assessment Tools 36

Appendix B: CT Bibliography with Emphasis on Assessment 40

Appendix C: The Delphi Research Letters 51

 CRITICAL THINKING: A STATEMENT OF EXPFRT CONSENSUS
FOR PURPOSES OF EDUCATIONAL ASSESSMENT AND INSTRUCTION

IT Critical Thinking ligtygagni and a Assessment

From New Jersey to California, and from Newfoundland to Florida,

leaders in the critical thinking movement have advocated major
educational reform. They have argued that effective and meaningful
education requires that curricular, pedagogical and assessment strategies
at all levels of education be coordinated so as to foster in students
those cognitive skills and habits of inquiry associated with critical
thinking. They have made the case that educating students to be critical
thinkers is vital for the students themselves and for society in general,
(Ennis, 1962, 1981, 1986; Passmore, 1967; Schievella, 1958; Shefler,
1973; Lipman, 1977; Siegel, 1980, 1988; Gardner, 1983; Arons, 1983;
Beyer, 1985; Costa, 1985; Quellmalz, 1983, 1985; Scriven, 1985;
Sternberg, 1985; Ruggiero, 1938; Paul, 1988 (a) and (b); etc.).
The arguments for critical thinking have been successful.
After decades of relative neglect, the eighties witnessed a growing
accord that the heart of education lies exactly where traditional
advocates of a liberal education always said it was in the processes
of inquiry, learning and thinking rather than in the accumulation of
disjointed skills and senescent information. The critical thinking
movement gained momentum throughout the decade. Conferences and position
papers led to the development of college level critical thinking (CT)
courses. In elementary and secondary schools (K-12) teachers revised
lesson plans to incorporate CT objectives. In the span of a few years
publishing CT textbooks and offering CT staff development programs became
growth industries. The CT movement enjoyed major success when
universities introduced CT requirements into their general education

programs and state departments of education targeted CT in their

curricular frameworks and their standardized testing programs. By the

decade's end CT could no longer be characterized as a cottage industry.

With success come questions: Not new ones necessarily, but, because

of the expectations which have been raised and the investments being

proposed, vexing ones. Intuitively, CT instruction should focus on how

students approach a question and reason about it. CT pedagogy should

develop in students those cognitive skills and affective dispositions

which characterize the good critical thinker. Rather than or in addition

to targeting whether a given answer is correct, CT assessment should

target the quality of the critical thinking the students put into

arriving at that answer. Thus, for all of their successes, CT experts

find they must continue to address some fundamental academic concerns.

What exactly are those skills and dispositions which characterize CT?

What are some effective ways to teach CT? And how can CT, particularly

if it becomes a campus-wide, district-wide or statewide requirement, be

assessed?

When these academic questions are asked by the individual professor

ur teacher seeking to introduce CT into her own classroom, they are

difficult enough. But the questions take on social, fiscal, and

political dimensions when asked by campus curriculum committees, school

district offices, boards of education, and the educational testing and

publishing industries. This is not to say that the experts find these

questions insurmountable. On the contrary, CT experts have worked with

their colleagues in the education community on some remarkable projects.
For example, California and New Jersey have established ways of

introducing CT into their curricular frameworks and statewide testing

7
 programs. The twenty-campus California State University system, which
enrolls hundreds of thousands of students, has established a process for
the approval of CT courses for its general education requirement.
Given the central role played by philosophers in articulating the
value, both individual and social, of CT, in analyzing the concept of CT,
in designing college level academic programs in CT, and in assisting with
efforts to introduce CT into the K-12 curriculum, it is little wonder
that the American Philosophical Association, through its Committee on
Pre-College Philosophy, has taken an interest in the CT movement and its
impact on the profession. In December of 1987 that committee asked this
investigator to make a systematic inquiry into the current state of CT
and CT assessment.

TABLE 1

CONSENSUS STATEMENT REGARDING CRITICAL

THINKING AND THE IDEAL CRITICAL THINKER
We understand critical thinking to be purposeful, self-regulatory judgment
which results in interpretation, analysis, evaluation, and inference, as well as
explanation of the evidential, conceptual, methodological, criteriological, or
contextual considerations upon which that judgment is based. CT is essential as a tool
of inquiry. As such, CT is a liberating force in education and a powerful resource in one's
personal and civic life. While not synonymous with good thinking, CT is a pervasive
and self-rectifying human phenomenon. The ideal critical thinker is habitually
inquisitive, well-informed, trustful of reason, open-minded, flexible, fair-minded in
evaluation, honest in facing personal biases, prudent in making judgments, willing to
reconsider, clear about issues, orderly in complex matters, diligent in seeking relevant
information, reasonable in the selection of criteria, focused in inquiry, and persistent
in seeking results which are as precise as the subject and the circumstances of inquiry
permit. Thus, educating good critical thinkers means working toward this ideal. It
combines developing CT skills with nurturing those dispositions which consistently
yield useful insights and which are the basis of a rational and democratic society.

As Table 1 suggests, a key result of inquiry is the articulation by

a panel of CT experts of a conceptualization of CT it terms of two
dimensions: cognitive skills and affective dispositions. Section II of

3
 this report describes the Delphi research methodology. Section III
address the skill dimension of CT, and Section IV the dispositional
dimension of CT. Research findings are presented throughout the report,
both in the text and in tabular farm. Six recommendations are presented
in Sections III and IV so they can be related most sensibly with their
rationale. Nine additional recommendations which pertain specifically to
CT instruction and assessment are presented in Section V.

-- Research !Yjettmlizlagy and Purpose

The Committee on Pre-College Philosophy suggested several persons

with special expertise in CT and CT whom this investigator might contact
as part of the inquiry into the controversial issues known to lie at the
heart of the profession's concern. This investigator decided to employ
the powerful q:.talitative research methodology known as the Delphi Method.
The Delphi Method requires the formation of an interactive panel of
experts. These persons must be willing to share their expertise and work
toward a consensus resolution of matters of opinion. Using the firs::
group of er,:perts to nominate othors, the Delphi panel soon took shape.
In all forty-six persons, widely recognized by their professional
colleagues to have special experience and expertise in CT instruction,
assessment or theory, made the commitment to parts =spate in this Delphi
project. If it were not for their conscientir:Lts effort, (for which this
investigator is extremely appreciative4 the consensus expressed in this
report could :lot have been ruched.
lit Delphi research experts participate in several rounds of
questions which call for thoughtful and detailed responses. Achieving a
consensus of expert opinion using the Delphi Method is not a matter of

7
 voting or tabulating quantitative data. Rather the expert panelists work
toward consensus by sharing their reasoned opinions and being willing to
reconsider them in the light of the comments, objections and arguments
offered by other experts. In Delphi research, once an expert cixpresses
an opinion, even a dissenting one, it becomes a factor iu the mix and
flow of all subsequent argument and thought. To circumvent undue
influence arising from any given eApert's r..-afessional status, each round
of questions is initiated by the pruiect director and all responses are
coordinated through that pe!-,-son. The project director circulates to the
entire panel direct quotations and synthesized responses, with the names
of their auth l s removed.
The expert panelists themselves, through the thoughtfulness and
ps.rsuasiveness of their written responses, shape the line of inquiry.
The project director endeavors to frame questions which respond to the
direction panel debate is taking and lead the conversation toward
fruitful resolution. As the inquiry proceeds, the project director
assists the panelists with bibliographies and alerts them to other useful
sources of relevant information. As areas of accord or disagreement
emerge these are presented to the panel in the form of drafts of
preliminary findings or crucial follow-up questions. The process
terminates when the project director determines that sufficient accord
has been reached for areas of consens. to be made public. Delphi
findings also include descriptions of residual disagreement and
statements of minority opinion.
A clear and accurate conceptualization of CT is absolutely essential
for the development of valid CT assessment tools and effective CT
instructional programs. qith this in mind, and recognizing that
divergent conceptualizations of CT have hindered curricular and
 assessment efforts, early in the Delphi process the panel decided its
most worthwhile contribution could be the articulation of a clear and
correct conceptualization of CT. The expert panelists devoted their
major effort toward that end. The experts hoped that by coming to
consensus they could offer educators interested in CT assessment or
instruction a conceptualization of CT of sufficient clarity, accuracy and
richness to warrant their serious attention.
To balance the theoretical with the practical, the experts asked
themselves what a generally educated college lower division level
critical thinker should be able to do. However, they did not attempt to
describe the typical college level critical thinker. It soon became
evident that the experts were actually articulating an ideal. It may be
that no person is fully adept at all the skills and sub-skills the
experts found to be central to CT. It may be that no person has fully
cultivated all the affective dispositions which characterize a good
critical thinker. Also humans compartmentalize their lives in ways that
CT is more active and evident in some areas than in others. This gives
no more reason to abandon the effort to infuse CT into the educational
system than that knowing no friendship is perfect gives one reason to
despair of having friends. The experts' purpose in putting the ideal
before the education community is that it should serve as a rich and
worthy goal guiding CT assessment and curriculum development at all
educational levels.

6
 TABLE 2

PROJECT HISTORY

Round 1 (Feb. 11, 1988) and Round 2 (Mar. 14, 1988) initiated the Delphi
process. In both rounds panelists were invited to nominate other CT experts to join
in this research project. The experts reached consensus on the working assumption that
"the concept of CT could be made operational to the extent that important parts of CT
could be assessed validly and reliably." The experts agreed to begin their analysis
of CT by "identifying the core elements of CT which might reasonably be expected
at the freshman and sophomore general education college level." The rationale for this
decision was that the college level theoretical construct of CT could reasonably be used
to guide what might be said about CT at the K-12 level. Also the panelists noted that
most of the participating experts had greater experience at the college level than in K-
12 education.

Round 3 (May 4,1988) was an open-ended invitation for experts to write their own
list of the operations which they conceived of as central to CT. The first synthesis of this
input was presented for expert review in Round 4 (Sept. 23, 1988). This synthesis
focused on the skill dimension of CT. Round 4 invited responses regarding each skill
and sub-skill identified, a proposed [and ultimately rejected] input/output model of CT
operations, a list of closely related cognitive operations which might or might not be
distinguished from CT, a general statement regarding what a skill is and how one is
taught, and is list of caveats and cautions regarding CT instruction and assessment.

Round 5A (Feb. 28, 1989) reviewed the definitions and classification of CT

cognitive skills in the light of expert responses to Round 4. Round 5B (also Feb. 28, 1989)
proposed statements regarding the dispositional dimension of CT and about its possible
normative connotations. Round 5C (Mar. 10, 1989) asked for specific recommendations
regarding CT instruction and assessment, and offered a revision of the general
statement on teaching and assessing a cognitive skill. Round 5 included several
quotations culled from the panelists' earlier responses and invited comments and
reactions.

The experts' comments regarding the various quotations included in each round
added greatly to the project director's understanding of the experts' overall views. From
these and the responses to specific Round 5A, 58 and 5C questions, the project director
assembled a draft report of all Delphi findings, including recommendations. Round 6,
(Sept. 2.5, 1989) circulated that draft and gave the CT experts the opportunity to express
their views or make comments for inclusion in the final report, which went through its last
revisions in Nov. 1989.

7
 III The_CAMitiMrLatill_DiMemaignaGriticAl Thinking
FINDING: As indicated in Table 1, the experts find good critical
thinking to include both a skill dimension and a dispositional
dimension. The experts find CT to include cognitive skills in (1)
interpretation, (2) analysis, (3) evaluation, (4) inference, (5)
explanation and (6) self-regulation. Each of these six is at the core
of CT. Associated with each are criteria by which its execution can be
meaningfully evaluated. However, no attempt is made here to specify
those criteria since ample criteriological discussions exist in the
literature.

Concerned not to generate misunderstandings, the experts offer- many

cautions about the analysis of CT in terms of skills and subskills. The
experts warn that good CT is not rote, mechanical, unreflective,
disconnected execution of sundry cognitive processes. They caution not
to lose sight of the whole while attempting to attend well to its many
parts.

RECOMMENDATION 1: All CT instruction should aim at developing good

critical thinkers -- persons who can integrate successful execution of
various skills in the CT enhanced classroom with the confidence,
inclination and good judgment to use these powerful tools in their
other studies and in their everyday lives. Persons who have
proficiency in CT skills but fail to use them appropriately are most
unlikely to be regarded as good critical thinkers.

RECOMMENDATION 2: Those who seek to infuse CT into the educational

system to be guided by a holistic conceptualization of what it means to
be a pod critical thinker. That some aspects of CT, particularly
features within its skill dimension, are more readily targeted by
existing educational assessment strategies should not distort the
conceptualization of CT nor truncate full-blown CT instruction.

The experts characterize certain cognitive skills as central or core

CT skills. The more one achieves proficiency in these skills, the more
worthy one is of being regarded as adept at CT. The experts are not,
however, sayin that a person must be proficient at every skill to be
perceived gas having CT ability. Considering the panel's purposes and
methodology, trying to analyze CT in terms of necessary and suf Ficient
conditions would have had strong negative utility. Thus, in view of the

8
pi.ecision which the question permits, the panel, early in the Delphi
process, decided to strive for a consensus on the core skills. The panel
was not asked to name skills without which a person is surely not a
critical thinker.

Responses to Rounds 4 and sA reveal the experts to be virtually

unanimous (N>957.) on ir-auding analysis, evaluation, and inference as
cep' rat. to CT. But in response to Round 6 one assessment expert strongly
dissented regarding tt-u inclusion of interpretation, arguing that it was
properly a part of communication, not CT. The same expert noted that
analysis, as defined in this report, overlaps with reading and listening.
These points raise obvious difficulties for CT assessment, particularly
as one attempts to make finer differentiations between CT and
communication or between analysis-in-the-CT-sense and analysis-in-the-
reading-sense. Regarding self-regulation the expert said, "I think this
is where testing must merge with teaching." In response to Round 6
another assessment e:;pert pointed out that, as compared to the others,
self-regulation appears to be a skill of a different kind or level. In
self-regulation one applies the other CT skills to one's own CT, by, for
example, evaluatjny or:e's on inferences. This gives CT an interestingly
Lharaizter. However, as this expert noted, the meta-cognitive
asrect of self -regulation makes it extremely difficult to assess using
the si-.andard kinds of paper and pencil instruments. Nonetheless, strong
conS*-2nSAA (N -87%) exists that interpretation, e:fplanation and self-
regulation are t.'entra 1 to CT. [For detailed results see the response
tables ,_,n it. -Age k)f the Delphi letter for Round 5A in Appendix CO

ti
 FINDING: There is consensus that one might improve one's own CT in
several ways. The experts agree that one could critically c*,amine and
evaluate one's own reasoning processes. One could learn hoe: to think
more objectively and logically. One could expand one's repertoire of
those more specialized procedures and criteria used in di44erent areas
of human thought and inquiry. One could increase one's base of
information and life experience.

It was readily apparent that the experts do not regard CT as a body

of knowledge to be delivered to students as one more school subject along
with others. The panel sees CT, like reading and writing, as having
applications in all areas of life and learning. And, as several pointed
out, CT instruction, like reading and writing, can occur in programs rich
with disciplinespecific content or in programs which rely on the events
in everyday life as the basis for developing one's CT.

FINDING One implication the experts draw from their analysis of CT

skills is this: "while CT skills themselves transcend specific subjects
or disciplines, exercising them successfully in certain contexts
demands domain - specific knowledge, some of which may concern specific
methods and techniques used to make reasonable judgments in those
specific contexts."

Although the identification and analysis of CT skills transcend, in

significant ways, specific subjects or disciplines, learning and applying
these skills in many contexts requires domainspecific knowledge. This
domainspecific knowledge includes understanding methodological
principles and competence to engage in normregulated practices that are
at the core of reasonable judgments in those specificcontexts. The
explicit mention of "evidential, conceptual, methodological,
s:riteriologicel, or contextual" considerations in connection with
explanation reinforces this point. Too much of value is lost if CT is
conceived of simply as a list of logical operations and domainspecific
knowledge is conceived of simply as an aggregation of information.
Inquiry into the nexus of reasonable judgment and actual application can

10

1.3
produce new appreciations of the necessity of robust concepts of both CT
and domain-specific knowledge in education.

RECOMMENDATION 3: Since becoming adept at CT involves learning to use

CT skills effectively in many different contexts, the experts insist
that "one cannot overemphasize the value of a solid liberal education
to supplement the honing of one's CT skills and the cultivating of
one's CT dispositions."

The experts caution that CT skills can usefully be grouped and sub-
classified in a number of legitimate ways. Hence, the sub-classification
which resulted from this Delphi research should not be interpreted as
necessarily excluding all others. Indeed, while declaring themselves to
be in agreement with this sub-classification, various participating
experts have also published their own sub-classifications. There are
areas of overlap in the classification system which emerged from the
Delphi research. However, while characterizing each skill and sub-skill
is important, creating arbitrary differentiations simply to force each
and every sub-skill to become conceptually discrete from all the others
is neither necessary nor useful. In practical contexts the execution of
some skills or sub-skills may presuppose others. Thus, order of the
Delphi listing is not intended to imply the endorsement of any
psychological, logical or epistemological order or skill-sequence, nor as
prescribing any educational taxonomy or skill-hierarchy.
Table 3 lists the skills and sub-skills which the experts identify
as being at the core of CT. No claim is being made that the list
eghaufAs the concept of CT in either breadth or detail. Beyond their
inclusion in CT, many of the skills and sub-skills identified are
valuable, if not vital, for other important activities, such as
communicating effectively. Also CT skills can be applied in concert with
other technical or interpersonal skills to any number of specific

11
concerns such as programming computers, defending clients, developing a
winning sales strategy, managing an office, or helping a friend figure
out what might be wrong with his car. In part this is what the experts
mean by characterizing these CT skills as pervasive and purposeful. It
is also fair to say that a particular skill, such as evaluation, or a
particular sub-skill, such as developing reasons, is essential for
success in a given endeavor, such as properly diagnosing illness. The

experts are not concerned that various skills and sub-skills are widely
used. It is not a problem that the skills might be essential elements in
other endeavors. On the contrary, it would be extremely disconcerting if
they were not, since the case for infusing CT into the educational system
depends un CT's utility across almost all areas of life and learning.
The experts are clear on the point that not every useful cognitive

TABLE 3

CONSENSUS LIST OF CRITICAL THINKING

COGNITIVE SKILLS AND SUB-SKILLS
1. Interpretation Categorization
Decoding Significance
Clarifying Meaning
2. Analysis Examining Ideas
Identifying Arguments
Analyzing Arguments
3. Evaluation Assessing Claims
Assessing Arguments
4. Inference Querying Evidence
Conjecturing Alternatives
Drawing Conclusions

5. Explanation Stating Results

Justifying Procedures
Presenting Arguments
6. Self-Regulation Self-examination
Self-correction
process should be thought of as CT. Not every valuable Lhinking skill is
CT skill. CT is one among a family of closely related forms of higher-
order thinking, along with, for example, problem-:solving, decision
making, and creative thinking. Unfortunately the conceptual overlaps and
complex relationships among all the various forms of higher-order
thinking have yet to be examined satisfactorily. However, that does not
imply that one cannot develop a careful and accurate conceptualization of
the target, CT -- a conceptualization fully adequate to its purpose,
which is to guide CT assessment and instruction.
In addition to accord on the listings in Table 3, the Delphi experts
find remarkable consensus on the descriptions of each of the skills and
sub-kills. These descriptions are presented in Table 4. The examples
as:.,ociated with each sub-skill are intended as clarifications. Some

reader.s ;night see in then suggestions of possible instructional or

assessment strategies. Others might see in them the tools to initiate
staff development conversations about the curricular implications.
Hi3wever, the panel's consensus has to do with the skill and sub-skill
descriptions, and does not necessarily extend to the examples.

TABLE 4

CONSENSUS DESCRIPTIONS OF CORE CT SKILLS AND SUB-SKILLS

1. INTERPRETATION: To comprehend and express the meaning or
significance of a wide variety of experiences, situations, data,
events, judgments, conventions, beliefs, rules, procedures or
criteria.

1.1 CATEGORIZATION:
* to apprehend or appropriately formulate categories,
distinctions, or frameworks for understanding, describing or
characterizing information.
* to describe experiences, situations, beliefs, events,
etc. so that they take on comprehensible meanings in terms of
appropriate categorizations, distinctions, or frameworks.
For eilmple: to recognize ct problem and define its character

13
 without prejudice to inquiry; to determine a useful way of sorting
and sub-classifying information; to make an understandable report
of what one experienced in a given situation; to clalsify data,
findings or opinions using a given classification schema,

1.2 DECODING SIGNIFICANCE:

* to detect, attend to, and describe the informational
content, affective purport, directive functions, intentions,
motives, purposes, social significance, values, views, rules,
procedures, criteria, or inferential relationships expressed in
convention-based communication systems, such as in language,
social behaviors, drawings, numbers, graphs, tables, charts, signs
and symbols.

For example: to detect and describe a person's purposes in asking

a given question; to appreciate the significance of a particular
facial expression or gesture used in a given social situation; to
discern the use of irony or rhetorical questions in debate; to
interpret the data displayed or presented using a particular form
of instrumentation.

1.3 CLARIFYING MEANING:

* to paraphrase or make explicit, through stipulation,
description, analogy or figurative expression, the contextual,
conventional or intended meanings of words, ideas, concepts,
statements, behaviors, drawings, numbers, signs, charts, graphs,
symbols, rules, events or ceremonies.
* to use stipulation, description, analogy or figurative
expression to remove confusing, unintended vagueness or ambiguity,
or to design a reasonable procedure for so doing.
For example: to restate what a person said using different words
or expressions while preserving that person's intended meanings;
to find an example which helps explain something to someone; to
develop a. distinction which makes clear a conceptual difference or
removes a troublesome ambiguity.

2. ANALYSIS: To identify the intended and actual inferential

relationships among statements, questions, concepts, descriptions
or other forms of representation intended to express beliefs,
judgments, experiences, reasons, information, or opinions.
2.1 EXAMINING IDEAS:
* to determine the role various expressions play or are
intended to play in the context of argument, reasoning or
persuasion.
* to define terms.
* to compare or contrast ideas, concepts, or statements.
* to identify issues or problems and determine their
component parts, and also to identify the conceptual relationships
of those parts to each other and to the whole.
For example: to identify a phrase intended to trigger a
sympathetic emotional response which might induce an audience to
agree with an opinion; to examine closely related proposals

14
17
 regarding a given problem an'i to determine their points of
similarity and divergence; given a complicated assignment, to
determine how it might be broken up into smaller, mcwe manageable
tasks; to define an austract concept.

2.2 DETECTING ARGUMENTS:

* given a set of statements, descriptions, questions or
graphic representations, to determine whether or not the set
expresses, or is intended to express, a reason or reasons In
support of or contesting some claim, opinion or point of view.
For example, given a paragraph, determine whether a standard
reading of that paragraph in the context of how and where it is
published, would suggest that it presents a claim as well as a
reason or reasons in support of that claim; given a passage from a
newspaper editorial, determine if the author of that passage
intended it as an expression of reasons for or against a given
claim or opinion; given a commercial announcement, identify any
claims being advanced. along with the reasons presented in their
support.

2.3 ANALYZING ARGUMENTS:

* given the expression of a reason or reasons intended
to support or contest some claim, opinion or point of view, to
identify and differentiate: (a) the intended main conclusion, (b)
the premises and reasons advanced in support of the main
conclusion, (c) further premises and reasons advanced as backup or
support for those premises and reasons intended as supporting the
main conclusion, (d) additional unexpressed elements of that
reasoning, such as intermediary conclusions, unstated assumptions
or presuppositions, (e) the overall structure of the argument or
intended chain of reasoning, and (f) any items contained in the
body of expressions being examined which are not intended to be
taken as part of the reasoning being expressed or its intended
background.

For example: given a brief argument, paragraph-sized argument, or

a position paper on a controversial social issue, to identify the
author's chief claim, the reasons and premises the author advances
on behalf of that claim, the background information used to
support those reasons or premises, and crucial assumptions
implicit in the author's reasoning; given several reasons or
chains of reasons in support of a particular claim, to develop a
graphic representation which usefully characterizes the
inferential flow of that reasoning.

3. EVALUATION: To assess the credibility of statements or other

representations which are accounts or descriptions of a person's
perception, experience, situation, judgment, belief, or opinion;
and to assess the logical strength of the actual or intend
inferential relationships among statements, descriptions,
questions or other forms of representation.

3.1 ASSESSING CLAIMS:

* to recognize the factors relevant to assessing the

1.5

Is
degree of credibility to ascribe to a source of information or
opinion.
* to assess the contextual relevance of questions,
information, principles, rules or procedural directions.
* to assess the acceptability, the level o, confidence
to place in the probability or truth of any given representation
of an experience, situation, judgment, belief or opinion.
For example: to recognize the factors which make a person a
credible witness regarding a given event or credible authority on
a given topic; to determine if a given principle of conduct is
applicable to decidinc what to do in a given situation; to
determine if a given claim is likely to be true or false based on
what one knows or can reasonably find out.

3.2 ASSESSING ARGUMENTS:

* to judge whether the assumed acceptability of the
premises of a given argument justify one's accepting as true
(deductively certain), or very probably true (inductively
justified), the expressed conclusion of that argument.
* to anticipate or to raise questions or objections, and
to assess whether these point to significant weakness in the
argument being evaluated.
* to determine whether an argument relies on false or
doubtful assumptions or presuppositions and then to determine how
crucially these affect its strength.
* to judge between reasonable and fallacious inferences;
* to judge the probative strength of an argument's
premises and assumptions with a view toward determining the
acceptability of the argument.
* to detsrmine and judge the probative strength of an
argument's intended or unintended consequences with a view toward
judging the acceptability of the argument;
* to determine the extent to which possible additional
information might strengthen or weaken an argument.
For example: given an argument to judge if its conclusion follows
either with certainty or with a high level of confidence from its
premises; to chPck for identifiable formal and informal fallaclei,1
given an objection to an argument to evaluate the force of
that objection; tc evaluate tne ..inJ applicability of
analogical arguments; Judge the logical strength of arguments
based or '!oothetical situations or causal rea.Boning; to judge if
a given argument is relevant or applicable or has implirations for
the situation et hand; to determine how possible new data miyht
lead logically to the further confirmation or disconfirmation of a
given :pinic!.1.

4: INFERENCE: To identify and secure elements needed to draw

reasonable conclusions; to form conjectures and hypotheses; to
consider relevant information and to educe the consequences
flowing from data, statements, principles, evidence, judgments,
beliefs, opinions, coneepts, descriptions, questions, or other
forms of representation.

16
I
 4.1 QUERYING EVIDENCE:
* in particular, to recognize premises which require
support and to formulate a strategy for seeking and gathering
information which might supply that support.
* in general, to judge that information relevant to
deciding the acceptability, plausibility or relative merits of a
given alternative, question, issue, theory, hypothesis, or
statement is required, and to determine plausible investigatory
strategies for acquiring that information.

For example: when attempting to develop a persuasive argument in

support of one's opinion, to judge what background information it
would be useful to have and to develop a plan which will yield a
clear answer as to whether or not such information is available;
after judging that certain missing information would be germane in
determining if a given opinion is more or less reasonable than a
competing opinion, to plan a search which will reveal if that
information is available.

4.2 CONJECTURING ALTERNATIVES:

* to formulate multiple alternatives for resolving a
problem, to postulate a series of suppositions regarding a
question, to project alternative hypotheses regarding an event, to
develop a variety of different plans to achieve some goal.
* to draw out presuppositions and project the range of
possible consequences of decisions, positions, policies, theories,
or beliefs.

For example: given a problem with technical, ethical or budgetary

ramifications, to develop a set of options for addressing and
resolving that problem; given a set of priorities with which one
may or may not agree, to project the difficulties and the benefits
which are likely to result if those priorities are adopted in
decision making.

4.3 DRAWING CONCLUSIONS:

* to apply appropriate modes of inference in determining
what position, opinion or point of view one should take on a given
matter or issue.
* given a set of statements, descriptions, questions or
other forms of representation, to educe, with the proper level of
logical strength, their inferential relationships and the
consequences or the presuppositions which they support, warrant,
imply or entail.
* to employ successfully various sub-species of
reasoning, as for example to reason analogically, arithmetically,
dialectically, scientifically, etc.
* to determine which of several possible conclusions is
most strongly warranted or supported by the evidence at hand, or
which should be rejected or regarded as less plausible by the
information given.

For example: to carry out experiments and to apply appropriate

statistical inference techniques in order to confirm or disconfirm
an empirical hypothesis; given a controversial issue to examine

17
 informed opinions, consider various opposing views and the reasons
advanced for them, gather relevant information, and formulate
one's own considered opinion regarding that issue; to deduce a
theorem from axioms using prescribed rules of inference.

5: EXPLANATION: To state the results of one's reasoning; to

justify that reasoning in terms of the evidential, conceptual,
methodological, criteriological and contextual considerations upon
which one's results were based; and to present one's reasoning in
the form of cogent arguments.

5.1 STATING RESULTS;

* to produce accurate statements, descriptions or
representations of the results of one's reasoning activities so as
to analyze, evaluate, infer from, or monitor those results.
For example: to state one's reasons for holding a given view; to
write down for one's own future use one's current thinking about
an important or complex matter; to state one's research findings;
to convey one's analysis and judgment regarding a work of art; to
state one's considered opinion on a matter of practical urgency.

5.2 JUSTIFYING PROCEDURES:

* to present the evidential, conceptual, methodological,
criteriological and contextual considerations which one used in
forming one's interpretations, analyses, evaluation or inferences,
so that one might accurately record, evaluate, describe or justify
those processes to one's self or to others, or so as to remedy
perceived deficiencies in the general way one executes those
processes.

For example: to keep a log of the steps followed in working

through a long or difficult problem or scientific procedure; to
explain one's choice of a particular statistical test for purposes
of data analysis; to state the standards one used in evaluating a
piece of literature; to explain how one understands a key concept
when conceptual clarity is crucial for further progress on a given
problem; to show that the prerequisites for the use of a given
technical methodology have been satisfied; to report the strategy
used in attempting to make a decision in a reasonable way; to
design a graphic display which represents the quantitative or
spatial information used as evidence.

5.3 PRESENTING ARGUMENTS:

* to give reasons for accepting some claim.
* to meet objections to the method, conceptualizations,
evidence, criteria or contextual appropriateness of inferential,
analytical or evaluative judgments.

For example: to write a paper in which one argues for a given

position or policy; to anticipate and to respond to reasonable
criticisms one might expect to be raised against one's political
views; to identify and express evidence and counter-evidence
intended as a dialectical contribution to one's own or another
person's thinking on a matter of deep personal concern.

18
21
6: SELF-REGULATION: Self-consciously to monitor one's cognitive
activities, the elements used in those activities, and the results
educed, particularly by applying skills in analysis and evaluation
to one's own inferential judgments with a view toward questioning,
confirming, validating, or correcting either one's reasoning or
one's results.

6.1 SELF-EXAMINATION:
* to reflect on one's own reasoning and verify both the
results produced and the correct application and execution of the
cognitive skills involved.
* to make an objective and thoughtful meta-cognitive
self-assessment of one's opinions and reasons for holding them.
* to judge the extent to which one's thinking is
influenced by deficiencies in one's knowledge, or by stereotypes,
prejudices, emotions or any other factors which constrain one's
objectivity or rationality.
* to reflect on one's motivations, values, attitudes and
interests with a view toward determining that one has endeavored
to be unbiased, fair-minded, thorough, objective, respectful of
the truth, reasonable, and rational in coming to one's analyses,
interpretations, evaluations, inferences, or expressions.
For example: to examine one's views on a controversial issue with
sensitivity to the possible influences of one's personal bias or
self-interest; to review one's methodology or calculations with a
view to detecting mistaken applications or inadvertent errors; to
reread sources to assure that one has not overlooked important
information; to identify and review the acceptability of the
facts, opinions or assumptions one relied on in coming to a given
point of view; to identify and review one's reasons and reasoning
processes in coming to a given conclusion.

6.2 SELF-CORRECTION:
* where self-examination reveals errors or deficiencies,
to design reasonable procedures to remedy or correct, if possible,
those mistakes and their causes.
For example: given a methodological mistake or factual deficiency
in one's work, to revise that work so as to correct the problem
and then to determine if the revisions warrant changes in any
position, findings, or opinions based thereon.
 IV The Dispositional Dimension of Cxitical Thinking

As is evident, particularly in the descriptions of selfexamination

and selfcorrection, there are dispositional components to critical

thinking. Indeed each cognitive skill, if it is to be exercised

appropriately, can be correlated with the cognitive disposition to do so.

In each case a person who is proficient in a given skill can be said to

have the aptitude to execute that skill, even if at a given moment the

person is not using the skill. But there was a great deal more many
experts wished say in regard to the personal traits, habits of mind,
attitudes or affective dispositions which seem to characterize good

critical thinkers.

FINDING: Although the language here is metaphorical, one would find

the panelists to be in general accord with the view that there is a
critical spirit, a probing inquisitiveness, a keenness of mind, a
zealous dedication to reason, and a hunger or eagerness for reliable
information which good critical thinkers possess but weak critical
thinkers du not seem to have. As water strengthens a thirsty plant,
the affective dispositions are necessary for the CT sLills identified
to take root and to flourish in students.

RECOMMENDATION 4: Modeling that critical spirit, awakening and

nurturing those attitudes in students, exciting those inclinations and
attempting to determine objectively if they have become genuinely
integrated with the high quality execution of CT skills are, for the
majority of panelists, important instructional goals and legitimate
targets for educational assessment. However, the experts harbor no
illusions about the ease of designing apprqpriote instructional
programs or assessment tools.

Procedural, Laudatory and Normative Uses of the Term "CT"

The experts have a consensus regarding the list of affective

dispositions which characterize good critical thinkers. This consensus is

expressed in Table 5. However, whether or not these affective dispositions

are part of the in the way that the cognitive skills are,
meaning of "CT"

was an issue which divided the experts from the first. It became evident

20
that various experts mean different things when they used the term "CT" in
reference to its possible dispositional components.
The deepest division is between the nearly two-thirds majority who
hold that the term "CT" includes in its meaning a reference to certain
affective dispositions and the roughly one-third minority who hold that
"CT" refers only to cognitive skills and dispositions, but not to affective
dispositions. The project director put this issue to the panel in several
different ways, sometimes directly and at other times more obliquely.
Responses, comments and arguments were shared, as were the objections and
counter-arguments which they engendered. In the end the panel remained
divided both numerically and in depth of feeling, with opposing positions
becoming more strident and entrenched as the debate continued.
In Round 513, of those expressing an opinion, the majority (617.)
maintain that the affective dispositions constitute part of the meaning of
"CT." They argue that these dispositions flow from, and are implied by,
the very concept of CT, much as the cognitive dispositions are. These
experts argue that being adept at CT skills but habitually not using them
appropriately disqualifies one from being called a critical thinker at all.
Thus, in addition to using "CT" in its procedural sense, these panelists
als.a use "CT" in its laudatory sense. They find it sensible to say, "This
person is a critical thinker, but this other person is so mentally lazy,
close-minded, unwilling to check the facts and unmoved by reasonable
arguments that we simply cannot call him a critical thinker."
The laudatory use of "CT" can suggest approval of how well a person
applies her CT skills or it can convey praise for the person, because the
person: has the proper affective dispositions. While the two-thirds
majority was eloquent regarding the importance of finding ways to instill
affective dispositions in students, in the final analysis they were unable

21

24
 to persuade the other third of their expert colleagues to view these
dispositions as essential to the concept of CT. The majority was, however,
persuasive in bringing about virtual unanimity regarding using the
affective dispositions to describe the paradigm critical thinker. (See
Table 1.)
In Round 5B a minority (307.) insist on using "CT" in a strict
procedural sense, that is as referring only to a certain judgmental
process. They distinguish sharply between what is true of critical
thinking from what is true of good critical thinkers. Their primary
concern is with the CT skills. They argue that good critical thinkers are
people who have those skills and certain valuable habits as well. If they
are good, critical thinkers, then they use their CT skills appropriately
because good critical thinkers also have some or all of the affective
dispositions listed in Table 5. But those dispositions are not what is
meant by "CT." They argue that one would not want to say a sophist is not
a critical thinker simply because the sophist uses CT skills for deceptive
or self-interested ends. The sophist, they would maintain, is a critical
thinker -- but nut an good one (in an ethical sense). The strict
proceduralists do not -Find it sensible deny that a person is a critical
thinker simply because the person, while skilled in CT, fails to check the
credibilitv of - sources, gives up too soon when asked to work a challenging
prcblem, lacks confidence in using reason to approach everyday problems, or
1.inores painful facts. These experts hold that such a person, because of
his CT skills, should be called a critical thinker but not a good one,
(ill terms of his effective use of those skills).
As suggested above, there are two senses of the term "good" which
might be operating when one uses the phrase "good critical thinker." One

sense applies to the thinker's effectiveness and responds to the question,

22 ,.
"How well is this person using CT?" The second sense applies to the
thinker's morality and responds to the question, "Is this person's use of
CT ethical'?" In order to clarify which sense of "good" the experts wished
to convey, Round 4 asked the panel to respond to a proposal that CT might
have a normative dimension in addition to a skill dimension and a affective
dispositional dimension.

FINDING: The mistaken notion that CT has a normative component is

rejected by the expert panelists. It is an inappropriate use of the
term to deny that someone is engaged in CT on the grounds that one
disapproves ethically of what the person is doing. What "CT" means,
why it is of value, and the ethics of its use are best regarded as
three distinct concerns.

The majority of experts (527.) forcefully reject the proposed

nonnative use of "CT." They hold that it is one thing to say what
something is, and another thing to say how it ought to be used. A
person's skills and attitudes are what they are, even if the person
suffers from certain ethical inadequacies.
Only a small group (177.), argue in favor of using "CT" in a
normative sense. This minority of experts, all of whom also use "CT" in
its commonly understood laudatory sense, hold that the true meaning of
"CT" extends to a certain set of ethical norms and social values. For
example, they would be willing to say that a defense attorney who uses CT
skills to cause a mistrial or win acquittal for a guilty client ought not
be dignified with the title of critical thinker. By the same token, the
prosecutor who uses CT skills to contrive a way to mislead a gullible
jur y into convicting and punishing an innocent person is not a critical
thirAer. Sir rce neither sufficiently value truth and since both appear to
lack the moral fiber to ef4chew deliberate deception in the practice of
their socially important professions, neither attorney should be accorded

2
the moral approval which calling them critical thinkers would imply.
The debate turned out to be instructive in another way as well. The
consensus (74% in support and 47, opposed) was that this report should
express the experts' fullest support and appreciation of the immense
personal and social importance of CT.
The panel shares a solid consensus about the importance of CT as a
tool of inquiry, as a liberating force in education, as a powerful
resource in one's personal life, and as a vital component in a rational
democratic society. It is extremely unlikely that any panelist would
condone using CT for immoral, deceptive, or unjust purposes. However the
personal and civic value of CT and sensitivity to the morality of its use
are not acceptable grounds for building a normative dimension into the
meaning of the term "CT". Some even saw such an effort as misguided and
potentially destructive of the CT movement. Giving "CT" a normative
twist could, they argue, lead to unwarranted limitations on open inquiry
and :JD unjustifiable ideological restrictions on the very concept of
being a "thinking" person. The totalitarian specter this conjures up is
the antithesis of the liberating critical spirit described earlier.

Dispositions of the Good Critical Thinker

FINDING: To the experts, a good critical thinker, the paradigm case,

is habitually disposed to engage in, and to encourage others to engage
in, critical judgment. She is able to make such judgments in a wide
range of contexts and for a wide variety of purposes. Although perhaps
not always uppermost in mind, the rational justification for
cultivating those affective dispositions which characterize the
paradigm critical thinker are soundly grounded in CT's personal and
civic value. CT is known to contribute to the fair-minded analysis and
resolution of questions. CT is a powerful tool in the search for
knowledge. CT can help people overcome the blind, sophistic, or
irrational defense of intellectually defective or biased opinions. CT
promotes rational autonomy, intellectual freedom and the objective,

24
 reasoned and evidence-based investigation of a very wide range of
personal and social issues and concerns.

Thus, in addition to possessing CT skills the good critical thinker

can be characterized by certain affective dispositions or habits of mind.
These dispositions, listed in Table 5 below, flow from two sources:
characteristics which the experts judge to hold true of good critical
thinkers, and the affective dispositions the expert:; judge to be part of
CT in its fullest realization. The majority of the experts (617.) regard
the dispositions listed in Table 5 as part of the conceptualization of
CT. A consensus exists (837.. in favor) that good critical thinkers can be
characterized as exhibiting these dispositions.

TABLE 5
AFFECTIVE DISPOSITIONS OF CRITICAL THINKING
APPROACHES TO LIFE AND LIVING IN GENERAL:
inquisitiveness with regard to a wide range of issues,
concern to become and remain generally well-informed:
alertness to opportunities to use CT,
trust in the processes of reasoned inquiry,
self-confidence in one's own ability to reason,
open-mindedness regarding divergent world views,
flexibility in considering alternatives and opinions,
understanding of the opinions of other people,
fair-mindedness in appraising reasoning,
honesty in facing one's own biases, prejudices,
stereotypes, egocentric or sociocentric tendencies,
prudence in suspending, making or altering judgments,
willingness to reconsider and revise views where honest
reflection suggests that change is warranted.

APPROACHES TO SPECIFIC ISSUES, QUESTIONS OR PROBLEMS:

clarity in stating the question or concern,
orderliness in working with complexity,
diligence in seeking relevant information,
reasonableness in selecting and applying criteria,
care in focusing attention on the concern at hand,
persistence though difficulties are encountered,
precision to the degree permitted by subject and circumstances._
. 25
28
 The experts are not saying that a person whose metaphysical,
epistemological, political, cultural or religious view of the world is
different from one's own is, ipso facto, not a good critical thinker.
Beliefs are not atoms each of which is at any moment subject to being
reconsidered independently. Be lie+s form intricately interconnected
systems of thought. To focus critical attention on any of them,
particularly those more central or fundamental to one's own view of the
world, car, cause reverberations throughout one's entire belief system.
Thus in advocating CT the panelists are not urging ideological
conforMity. Indeed, just as many experts argued that an over-emphasis on
the values of CT could lead to trouble, others warn that an over-emphasis
on the skills dimension of CT to the exclusion of the affective
dispositions might have the unfortunate result of making some students
close-minded, intellectually inflexible and dogmatic.

RECOMMENDATION 5: Just as with the cognitive dimension of CT, when

conceiving of the education or assessment of critical thinkers, it is
important to consider ways of developing materials, pedagogies, and
assessment tools that are effective and equitable in their focus on
these affective dispositions. The cultivation of these dispositions is
particularly important to insure the use of CT skills outside the
narrow instructional setting. Persons who have developed these
affective dispositions are much more likely to apply their CT skills
appropriately in both their personal life and their civic life than are
those who have mastered the skills but are not disposed to use them.

As with the listing of cognitive si -ills earlier, the panel does not

intend that each disposition be considered a necessary condition. The

experts are characterizing the ideal. In setting forth the concept of

the paradigm critical thinker, they intend to express a goal toward which

all might strive. These virtues require a measure of maturity and

personal development not commonly found in college sophomores or twelfth

graders. Yet to dent}, embartting on the practices and disciplines which

2e
will lead to these virtues would be an even more profound mistake.

RECOMMENDATION 6: From early childhood people should be taught, for

example, to reason, to seek relevant facts, to consider options, and to
understand the views of others. It is neither impractical nor
unreasonable to demand that the educational system teach young people
the habits of mind which characterize the good critical thinker,
reinforce those practices, and move students well down the path toward
their attainment.

t_ion ssessment

Several pedagogical and assessment implications follow from the

dispositional dimension of CT, implications which might not be apparent

if educators focused only on the skill dimension of CT. The education of

good critical thinkers is more than training students to execute a set of
cognitive skills. For example, in terms of pedagogy, modeling how to

evaluate critically that information which students would normally accept

uncritically and encouraging them to do the same can do wonders for

developing their confidence in their CT ability. With this confidence

students are much more likely to try thinking for themselves. Just as
instruction should not focus on skills only, assessment which focus on

Skills only may give a misleading or incomplete picture of someone's

strengths as a critical thinker.

The CT Goal

RECOMMENDATION 7: Brcause CT helps students with a wide range of

educational, personal and civic concerns in a rational way, the
acaJemic goal of CT instruction, regardless of the educational level,
should be furthering students in the development of their CT cognitive
skills and affective dispositions.

27
r TABLE 6 "N

CONSENSUS STATEMENT ON TEACHING AND ASSESSING CT SKILLS

A CT skill, like any skill, is the ability to engage in an activity, process or

procedure. In general, having a skill includes being able to do the right thing at the
right time. So, being skilled at CT involves knowing, perhaps implicitly or without the
ability to articulate this knowledge, both a set of procedures and when to apply those
procedures. Being skilled also involves having some degree of proficiency in executing
those procedures and being willing to do so when appropriate. Reflecting on and
improving one's CT skills involves judging when one is or is not performing well, or as
well as possible, and considering ways of improving one's performance. Learning CT
involves acquiring the ability to make such self- reflective judgments.

Skills, particularly CT cognitive skills, can be taught in a variety of ways, such

as by making the procedures explicit, describing how they are to be applied and
executed, explaining and modeling their correct use, and justifying their application.
Teaching cognitive skills also involves exposing learners to situations where there are
good reasons to exercise the desired procedures, judging their performance, and
providing the learners with constructive feedback regarding both their proficiency and
ways to improve it. Instruction might start with situations that are artificially simple,
but should culminate in situations that are realistically complex. Particularly in the case
of CT, the learners must contribute a solid measure of personal effort, attention, practice,
desire, and, as they learn how, self-monitoring. Teaching skills involves motivating
learners to achieve higher levels of proficiency and, particularly in the case of CT,
independence. It also involves coaching learners on how they can achieve those goals.

In theory there are several ways persons can be judged to be more or less
proficient in a given CT skill or at the integrated use of related CT skills. One way
is to observe a person over time performing those activities, processes or procedures
generally regarded as presupposing that skill for proper execution. One then mak...
judgment regarding the degree to which the person possesses the general skill in
question. A second way is to compare the outcomes (if any) that result from executing
a given skill against some set of criteria. A third way is to query persons and receive
their descriptions of the procedures and judgments they are using as they exercise that
skill, would use if they were to perform that skill, or did use when they performed that
skill. A fourth way is to compare the outcomes (if any) that result from performing
another task against some set of criteria, where the performance of that other task
has been shown to correlate strongly with exercising the skill of interest. However,
that such correlations exist between any other task and CT, or any of its sub-skills, has
yet to be established in the research literature.

Each of the four ways of CT assessment has limitations as well as strengths. No

matter which ways are used, it is important to ensure that the assessment conditions
foster an attitude in which the subjects are disposed to use their skills as well as they
can, and are not constrained or inhibited from doing so. In our view it is highly
advantageous to gather evidence regarding CT performance in many situations, using
several assessment methods, so as to compile a composite picture of the subject and to
cross check the results of an one wa of assessment.
 Either to transform CT into one subject field among others, or to
narrow the range of CT applications strictly to domainspecific subject
content, would be to truncate its utility, misapprehend its nature and
diminish its value. Within the overall curriculum the goal of learning
CT can be clearly distinguished from the goal of learning domainspecific
content. And yet: while these two goals can be distinguished, the
experts do not wish to deny one of the best ways to learn CT is within a
subject context.

RECOMMENDATION 8: Direct instruction in CT and assessment of CT should

be an explicit parts of any course granted approval for purposes of
satisfying CT requirements, whether that course is a CT course per se
or a course in a given subject field. The primary academic criterion
in the evaluation of a proposed instructional program for purposes of
achieving the CT goal should be whether the program will further the
development of students' CT skills and dispositions.

The CT Curriculum,
Given that CT has, in many cases, become a college general education
reqvirement, secondary schools can be expected to begin to develop
college preparatory CT programs. However, the value of CT extends well
beyond its importance as a universitylevel inquiry tool. CT is vitally
important in the personal and civic life of all members of society. A
significant percentage of the citizenry will not graduate from high
school, or if they graduate, will not have the benefit of postsecondary
educ:ation.

RECOMMENDATION 9: Thus, CT instruction should not be reserved only for

those who plan to attend college. or should it be deferred until
college, since it is not likely to be effective if it were.

RECOMMENDATION 10: Explicit attention to the fostering of CT skills and

dispositions should be made an instructional goal at all levels of the
K-12 curriculum. The cultivation of CT dispositions and an insistence
on giving and evaluating reasons, should be an integral part of
elementary school education. In middle schools and high schools,

2q
3 -)
 instruction on various aspects and applications of CT should be
integrated into all subject area instruction. Specific courses in CT
and an advanced placement examination program in CT for college bound
students should be developed. Although for good reasons at the post-
secondary level CT programs are generally associated with departments
of philosophy, no academic unit should be restricted in principle from
participating in an institution's CT program, provided that the overall
institutional program in CT equips students to apply CT to a broad
range of educational, personal and civic subjects, issues and problems.

There is growing evidence of the successes, both scientific and

economic, of those industrialized democracies which emphasize demanding
academic assessment and set firm educational standards for career and
professional advancement. Assessment that counts is unquestionably a key
factor in promoting academic achievement.

RECOMMENDATION 11: Thus, minimum CT proficiency expectations should be

set for each educational level, including promotion in grade, high
school graduation, college entrance, and graduate school admission.

The CT Assessment

The development of valid and reliable assessment strategies from

which teachers can draw reasonable inferences about students' CT, in
contrast to their domainspecific knowledge or other academic abilities
(such as reading or writing), is essential. CT assessment strategies,
whether for use in the individual classroom or for broader purposes, must
not simply reward arriving at correct answers. They must, however
recognize achieving correct answers by way of good CT. The challenge of
CT assessment is not to let what is easily measured restrict our sense of
the -Fullness of CT. It would be shameful if those assessment instruments
which fucus only on CT skills drove our CT curricular design and caused
the dispositional components of good CT to be neglected.

RECOMMENDATION 12: In evaluating the acceptability of a CT assessment

strategy or instrument one should consider content validity, construct
validity, reliability, and fairness.

30
33
 (1) Content Validity: The strategy or instrument should be based
on an appropriate conceptualization of CT and a clear understanding of
which aspects of CT the assessment targets. Each task or question should
be evaluated to insure that correctly responding to that item is not a
matter of rote learning or information recall. Whether for the classroom
or for broader educational purposes, CT assessment should include
strategies for targeting CT's dispositional dimension as well as its
cognitive skills dimension.
(2) Construct Validity: In acceptable CT assessment each task or
question should have been evaluated to insure that students who answer
correctly do so on the basis of good CT and that inadequate or wrong
responses are the result of weak or inadequate CT. Entire strategies or
specific items on which good CT leads to wrong answers, or poor CT to
right answers, should not be used.
(3) Reliability: In acceptable CT assessment each task or question
should have been evaluated to insure that good critical thinkers
generally do better on that item than weak critical thinkers. If
different persons are involved in evaluating the results, for example
grading essays or judging presentations, the evaluations of the different
judges should be cross-checked to assure that their findings are
reliable, that is, generally consistent with one another. However, it is
an open question whether the levels of achievement associated with the
different CT sub-skills and affective dispositions are positively
correlated. Empirical research on how the sub-skills correlate with each
other and with various dispositions has yet to be undertaken. Thus, at
this time, due caution should be exercised regarding how to interpret
ti,echnical ftlee'iuriis of test-form reliability in the case of paper and

3J
pencil CT assessment instruments.
(4) Fairness: CT assessment should not unfairly disadvantage or
advantage groups of students on the grounds of reading ability, domain-
specific knowledge (broadly understood as including the evidential,
conceptual, methodological, criteriological, contextual considerations,
or familiarity with technical vocabulary], gender or age related life
experience, ethnicity or socio-economic status, differences in social
norms, or differences in cultural assumptions. CT assessment locates CT
tasks and questions in some assumed context, either subject-specific,
everyday life, or fictional. Thus, guaranteeing that all students,
regardless of their individual backgrounds, will come to the CT
assessment on a perfectly equal basis in terms background knowledge,
reading ability, life experiences, etc. is impossible. However,
examining the assessment strategy or instrument to be sure that these
factors do not unfairly influence the results is prudent and reasonable.
Although one cannot eliminate the influence of these variables, one may
be able to neutralize or control for their affects.
The -Fairness criterion applies both to discipline-neutral and
discipline-specific CT assessment. Within curricular programs
discipline-specific CT assessment is encouraged, since it is possible for
one to be fair in one's presumptions regarding subject-specific criteria,
concepts, methodologies, evidence, information and terminology. The
challenge of such assessment is to factor out the discipline content in
order- to acxess the strength or weakness of the CT. It is worth noting
that discipline-neutral CT assessment also makes similar assumptions
regarding the everyday contexts which form its topic content.

RECOMMENDATION 13: CT assessment should occur frequently, and it should

be used diagnostically as well as summatively. Different kinds of

'41)
 instruments should be employed, depending on which aspect of CT is
being targeted and where students are in their learning -- the
introductory stage, the practice stage, the integration stage or the
generalized transfer stage. Although the veteran CT instructor is able
to assess students continuously, CT assessment should be made explicit
to reinforce its worth in the eyes of the students, their families, and
the public. It should be made explicit to support the goals of
educators seeking to improve the curriculum. And it should be made
explicit to properly inform educational policy formation.

The CT Instructor

RECOMMENDATION 14: Teaching CT is most effective if the instructor

models CT dispositions and the proper use of CT skills in the very
process of instruction. Regardless of the subject area, students
should be encouraged to be curious, to raise objections, ask questions,
point out difficulties in the instructor's position. These objections
and questions should be clarified, interpreted, and examined
objectively. Students should be given reasons for doing things a
certain way, rather than being dogmatically told how to do them.
Instruction should bridge the gap between the subject and the student's
own experience. In the case of CT instruction, the topics of
discussion should not be restricted to factual matters or academic
subjects, but should include issues which have normative, moral,
ethical or public policy dimensions.

The ideal CT instructor will integrate instruction in CT in a

variety of subject areas. She will teach specific CT skills directly
using these subjects as content for the application of those skills. She
will help students elaborate, transfer and generalize these skills to a
variety of contexts. She will create a classroom and school environment
which is supportive of CT. She will model CT in her teaching and her
interactions with colleagues. She will provide her students with
thoughtprovoking subjects to learn about, and projects to undertake.
She will engage students in social activities requiring them to reflect
on, articulate, share and discuss justifications, explanations and
contrasts in how they executed various CT tasks. She will evaluate each
student's progress, achievement or proficiency in CT continuously.

RECOMMENDATION 15: For CT to infuse the K-12 and college curriculum,

teacher "training" should give way to teacher "education." If teachers

33
36
 are to model CT, so must those who have an instructional role in
teacher preparation or staff development. In all instruction, and
particularly in CT instruction, both faculty and leaders of faculty
development should model CT. They should foster the students'
confidence in their own powers of reason, rather than dependency on
rote learning. They should nurture in students open-mindedness,
attention to alternatives, and as much precision of thought as the
subject and circumstances permit.

VI -- The Delphi Rese4rch Pawl

The Delphi research findings reported here result from the

participation of fortysix scholars, educators and leading figures in CT
theory and CT assessment research. Roughly half the panelists are
primarily affiliated academically with Philosophy (52%), the others are
affiliated with Education (22!), the Social Sciences (207.), or the
Physical Sciences (67.).
It would be a mistake to construe participation in this research
project as implying that a person agrees with all the findings. Thus,
where consensus is reported a minority of panelists hold divergent views.
Where near unanimity is reported a some panelists may may not be in full
accord with hew the specifics are expressed. In the end, however, after
reviewing the draft Delphi findings presented in Round 6, only one of the
forty-six explicitly opted to be listed as a participant only, but not as
supporting the document.
These Delphi findings fix an important moment in time. It is a
moment when the efforts of forty-six experts possessing special
experience and knowledge in matters relating to CT converged with a view
toward discovering if some measure of general accord could be found. As
we move from the successes of the eighties into the decade of the
nioe ties, the person' who participated in this project hope that the
findings of expert consensus reported herein will advance critical
!_hintang and help sh..kpe the future of CT instruction and CT assessment.

34
 TABLE 7

PARTICIPATING CRITICAL THINKING EXPERTS

Jonathan Adler Philosophy Brooklyn College

David Annis Philosophy Ball State University
Arnold Arons Physics University of Washington
James Bell Psychology Howard Community College, MD
Barry K. Beyer Education George Mason University
Charles Blatz Philosophy University of Toledo
Rob Brady Philosophy Stetson University
Neil Browne Economics Bowling Green State University
Rez Clemmenson CT Assessment American College Testing (ACT)
Arthur L. Costa Education Sacramento State University
Stan Dundon Philosophy Cal. Polytechnic University, SLO
Robert H. Ennis Education University of Illinois
James B. Freeman Philosophy Hunter College, CUNY
Jack Furlong Freshman Studies Transylvania University
Eugene Garver Critical Thinking Saint John's University
H. Scott Hestevold Philosophy University of Alabama
David Hitchcock Philosophy McMaster University
John Hoag lund Philosophy., Christopher Newport College
Kenneth Howe Education University of Colorado
Ralph H. Johnson Philosophy University of Windsor
Stuart Keeley Psychology Bowling Green State University
Anthony Lawson Zoology Arizona State University
Matthew Lipman Philosophy Montclair State College
David S. Martin Education Gallaudet University
John Martin Philosophy University of Cincinnati
Gary Matthews Philosophy U. Massachusetts, Amherst
Stuart Miller Psychology Towsen State University
Brooke Noel Moore Philosophy CSU Chico
Wayne Neukberger Assmt. and Eval. Oregon Department of Education
Stephen Norris Education Memorial University of Newfoundland
Richard Parker Philosophy CSU, Chico
Richard D. Parry Philosophy Agnes Scott College
Richard Paul Philosophy Sonoma State University
Philip Pecorino Social Sciences Queensborough C. College
William Rapaport Computer Science SUNY Buffalo
Pasqua' Schievella Council of Critical Analysis, Port Jefferson, NY
Zack Seech Behavioral Science Palomar College
Anita Silvers Philosophy San Francisco State University
Richard Stiggins Northwest Regional Educational Laboratory, Portland
Robert J. Swartz Philosophy U. Massachusetts, Boston
Steven Tigner Philosophy University of Toledo
Carol Tucker CT Assessment Educational Testing Service (ETS)
Perry Weddle Philosophy Sacramento State University
Robert Wengert Philosophy University of Illinois
Mark Weinstein Institute for CT Montclair State College
Peter WmoRard Education' University of Kentucky
35
38
 APPENDIX A
4 ;

Prepared for the APA Committee on Pre-College Philosophy

Delphi Research Project on CT Assessment

The Education Testing Service (ETS) Academic Profile Test measures

the academic abilities of CT, reading, writing, and using mathematical
data, all within the context of three major academic areas --
humanities, social sciences and natural sciences. This is a multiple
choice instrument with an optional critical essay (locally scored). It
assumes that students have completed most or all of their general
education -- that is, completed the sophomore college year.

ETS also has sections targeting logical reasoning and analytical

thinking on several of its widely used instruments such as the (a) Law
School Admissions Test, (LSAT) "Logical Reasoning" section, (b) Graduate
Record Examination, (GRE) -- General Test, the Analytical section, (c)
Advanceo Placement Test, subject- matter based CT questions, (d) National
Assessment of Educational Progress, higher order thinking and
laboratory-based questions, (e) Foreign Service Test, in-basket portion,
and (f) Graduate Record Examination -- Advanced Test in Philosophy 1972-
1952. ETS is adding a section of Critical Reasoning Questions to the
rTircluate Management Admissions Test (GMAT). ETS has constructed
branching tests of information-seeking and decision-making, some paper-
and-pencil and some on computers. An example is the clinical practice
test prepared for the National Board of Respiratory Care, Shawnee

36
33
Mission, Kansas.

Stephen Norris and R. King, through the Institute for Educational

Research and Development at Memorial University of Newfoundland, has
developed the Test on Appraising Observations 1983. Using the backdrop
of a common but fictional situation subjects are asked to judge the
relative credibility of the claims made by various characters.

The American College Testing Program (ACT) in 1988 produced the

"Collegiate Assessment of Academic Proficiency" (CAAP). ACT describes
CAAP is designed to measure selected academic skills including reading,
writing, clithematics, CT, and scientific reasoning. Tne CT Test
measures the ability to "clarify, analyze, evaluate, an extend
arguments." The test is composed of passages commonly encountered in a
postsecondary curriculum followed by multiple choice test questions.

Paul M. Ramirez discusses the "Valett Inventory of CT Abilities"

(VICTA) in The Reading Teacher, vol. 41, Dec. 1987, page 348.

THE NINTH MENTAL MEASUREMENTS YEARBOOK, (NMMY), lists commercially

available tests in print along with reviews and research data. Many of
these are also described a d reviewed by Norris and Ennis in their
lAst.i.ful Evaluating CT, Midwest Publications, Pacific erove, CA, 1989.

07/39 "Cornell Critical Thin ing Test, Levels X and Z." :Ennis, Millman,
140), 1961-1987, Midwest Publications, Pacific Grdve, CA.

Revied in Educational and Psychological ileasurements 1983

Vol. 43., pp. 1187-11971 by Modjeski and Michael.

40
 #390 "Ennis-Weir Argumentation Test, Level X: An Essay Test of Rational
Thinking Ability," (Robert Ennis and Eric Weir) 1982, Illinois
Thinking Project, University of Illinois, Urbana, IL.
Reviewed by Herbert Rudman, Michigan State, in NMMY.
#391 "Ennis-Weir Critical Thinking Essay Test: An Instrument for
Testing/Teaching." (Robert Ennis and Eric Weir) 1983, Midwest
Publications, Pacific Grove, CA.
#1347 "Watson-Glaser Critical Thinking Appraisal" 1942-80. Described
and reviewed by two persons in the NMMY many citations of
other research regarding this instrument.
*751 "New Jersey Test of Reasoning Skills," 1983, Virginia Shipman,
Institute for the Advancement of Philosophy for Children.
#1258 "Test of Inquiry Skills" 1979, Australian Council for Educational
Research. For junior high grades, this test purports to
evaluate a range of research, study and critical thinking
skills in the sciences.
#1061 "Ross Test of Higher Cognitive Processes" (John Ross and Catherine
Ross) 1976-79, Academic Therapy Publications. F or grades 4-6,
this Lest includes sub-scores on analogies, deductive
reasoning, missing pretaises, questioning strategies, and
relevance of in-Formation.
#1248 "Test of Cognitive Skills" 1981, McGraw Hill. For grade levels 2-
121 this test includes sub-scores on sequencing, analogies,
rne nary, and verbal reasoning.
#122 "Basic Skills Assessment" 1977-81, McGraw Hill. Included in the
reading p.Act::age is a sub-score on inference and evaluation.
In the writing package is a sub-score on logical evaluation.
#1269 "Test of Problem Solving" 1984, LinguiSystem Inc. For ages 6-12,

3
 this tests a child's thinking and reasoning abilities critical
to events of everyday life. It includes sub-scorem on
explaining inferences, determining causes, negative why
questions, etc.
#272 "Corrective Reading Mastery Test" 1980, Science Research
Associates, Inc. Designed to measure the effectiveness of
corrective reading programs, this test includes sub-scores on
deductions, classifications, analogies, inductions, statement
inference, hypothesis/evidence.
#1302 "Deductive Reasoning Test" (J. M. Verster) 1972-73, National
Institute for Personnel Research, South Africa. Focuses on
syllogistic problems and designed for for candidates for
graduate scientists and higher professions.
#1010 "PSI Basic Skills Test for Business and Industry" 1981-1982,
Psychological Services Inc. Includes sub-scores on problem
solving, decision making, reasoning and classifying.
#106 "Ball Aptitude Battery" the Ball Foundation. Used to tests persons
for occupational placements, this instrument includes sub-
scores on inductive reasoning, analytical reasoning, idea
fluency, and shape assembly.

3
 APPENDIX B
Critical Thinking Bibliography with Emphasis on Assessment

Prepared for the APA Committee on Pre-College Philosophy

Delphi Research on Critical Thinking Assessment

Adler, M., "Why 'CT' Program Won't Work," Education Reek, Sept. 1986.

Annis, David B. and Annis, Linda, "An Empirical Study of the Impact of
Philosophy on Students' CT Ability," Teaching Philosophy, v3, pp. 145-
152, 1980.

Arons, Arnold B. "CT and the Baccalaureate Curriculum," Liberal Education,

v71, n2, Summer 1985.

"Achieving Wider Scientific Literacy," Daedalus, Journal of the

,

American Academy of Arts and Sciences, v112, n2, p91-122, Spr. 1983.

Azima, Kiavach, and Henry, Rebecca, "Teaching Students to Reason: An

Application of Piagetian Psychology to College Teaching No. 76," Learning
and Evaluation Service, Michigan State University, East Lansing, MI, 40p,
1980.

Baker, P. J., "Learning Sociology and Assessing CT," Teaching Sociology, v8,
p325-363, 1981.

Barun, Joan B., and Sternberg, Robert J., Teaching Thinking Skills: Theory and
Practice, W. H. Freeman Publishing, 1987.

Bangert-Drowns, Robert L., et, al., "Individualized Systems of Instruction in

Secondary Schools," RevieN of Educational Research, v53, n2, p143-58,
Summer 1983.

Beck, Ronald A., A Guide to Criterion-Referenced Test Construction, Johns

Hopkins University Press, 1984.

Beyer, Parry K., "Improving Thinking Skills -- Defining the Problem," Phi
Delta Kappan, v65, n7, p4R6-901 March 1984.

1"Improving Thinking Skills -- Practical Approaches," Phi Delta

Kappan, v65, n8, Apr. 1984.

1
"Practical Strategies for the Direct Teaching of Thinking," in
Developing Hinds: A Resource Book for Teaching Thinking, Arthur L. Costa,
ed., Association for Supervision and Curriculum Development, Alexandria,
VA, 1985.

1"A Suggested Format for Testing Thinking Skills," Social Science

Record, v24, nl, p3-5, Spr. 1987.

1
Practical Strategies for the Teaching of Thinking, Boston, Allyn and
Bacon, 1987.

4C
43
Blatt, Charles V., "Contextualism and CT: Programmatic Investigations,"
Educational Theory, v39, n2, 1989.

Block, R. A., and Taylor, S. V., "Cognitive Skills: Enhancement and Assessment
Issues, Presented to the American Psychological Association, Toronto,
Canada, 1984.

Blumberg, Fran, et. al, A Pilot Study of Higher-Order Thinking Skills

Assessment Techniques in Science and Hatheaatics -- Part 1 and Pilot-
Tested Tasks -- Part II, Final Report, National Assessment of Educational
Progress, Princeton, NJ, Nov. 1986.

Bransford, J. D., et a!, "Teaching Thinking: Evaluating Evaluations and

Broadening the Data Base," Educational Leadership, v44, p67 -70, 1986.

Branson, Stimmann Margaret, "CT Skills -- A Continuum for Grades 3-12 in

History/Social Science," Social Studies Review, v25, n2, p24 -32, Winter
1986.

Brandt, Ron, "On Philosophy in the Curriculum: A Conversation with Matthew

Lipman," v46, Educational Leadership, p 38, Sept. 1988.

Braungart-Bloom, Diane S., "Assessing Higher Thinking Skills through Writing,"

10p. Presented at the Annual Meeting of the American Educational Research
Association, San Francisco, CA, April 1986.

Brown, J. L., "On Teaching Thinking Skills in the Elementary and Middle
Schools," Phi Delta Kappan, v64, p709 -714, 1983.

Bryden, David P., "What Do Law Students Learn? A Pilot Study," Journal of
Legal Education, v34, n3, p479 -506, Sept. 1984.

Carlson, E. R., "Implications of Cognitive Theory and Research for Teaching

CT, Presented to the Annual Meeting of the American Psychological
Association, Toronto, Canada, 1984.

Chaffee, John, "Viewing Reading and Writing as Thinking Processes," 9p,

Presented to the Annual Meeting of the American Educational Research
Association, Chicago, IL, Mar-Apr. 1986.

Chance, Paul, Thinking in the Classroom: A Survey of Programs, NY, Teachers

College Press, 1986.

Cierzniak, Susanne Lipetska, The Question of CT: An Annotated Bibliography,

64p. Exit Project, University of Indiana at South Bend, ED 260 069, April
1985. The specific focus of this work is CT in thR secondary schools.

Cornbleth, Catherine, "Assessing Skills and Thinking in Social Studies,"

position paper prepared for Study Group on the national Assessment of
Student Achievement, and cited in Appendix B of their report, "The
Nation's Report Card" (TM 870 049), Journal announcement: RIEJUL87

Costa, Arthur L., ed. Developing Hinds: A Resource Book for Teaching
Thinking?, Association for Supervision and Curriculum Development,
Alexandria, VA, 1985.

41
 "Thinking: How Do We Know Students are Getting Better at It?" ms.,
Dept. of Education, Sacramento State University, CA., 1989.

Costa, Arthur L., and Marzano, Robert J., "Teaching the Language of
Thinking,"
Educational Leadership, v45, p29, Oct. 1987.

D'Angelo, Edward, The Teaching of CT, Amsterdam, B. R. Gruner N. V., 1971.

Deshmukh, M. N., "Teaching the Unteachable: Some Pedagogical Considerations

of Creativity," Psycho-Lingua, v151 n1, p33-40, 1985.

Dewey, John, How We Think, D. C. Heath and Co., Boston, 1933.

Drake, James A., Teaching CT, Danville, IL, Interstate Printers and
Publishers, 1976.

Duck, Lloyd E., "Seven Cardinal Principles for Teaching Higher-Order

Thinking," Social Science Record, v241 nl, p3 -5, Spr. 1987.

Elman, Sandra E., and Lynton, Ernest A., "Assessment in Professional

Education," 24p. Presented to the National Conference on Assessment in
Higher Education, AAHE, Columbia, SC, Oct. 1985.

Ennis, Robert H., 'A Concept of CT," Harvard Educational Review, v32, nl,
1962.

"Operational Definitions," Aperican Educational Research Journal, vl,

n3, p183 -201, May 1964.

"Testing for CT: State of the Art," American Educational Research

1

Assoc., San Francisco, CA, 1968.

"Rational Thinking and Educational Practice," in Philosophy and

1

Education, Jonas Soltis (Ed.), Eighteenth Yearbook of the National

Society for the Study of Education, Chicago, NSEE, p143-183, 1981.

"Goals for a CT/Reasoning Curriculum, Illinois CT Project, U. of

1

Illinois, Champaign, IL, 1984.

"Problems in Testing Informal

,
Logic CT Reasoning Ability," Informal
Logic v6, nl, p3-9. 1984.

1 "A Logical Basis for Measuring CT Skills," Educational Leadership,

v431 p44 -48, 1985.

"A Logical Approach to Measuring CT Skills in the Fourth Grade,"

1

Illinois CT Project, Champaign, IL, draft, April 1985.

, "A Taxonomy at CT Dispositions and Abilities," in Teaching for

Thinking, Joan Baron and Robert Sternberg (Eds.), Freeman, New York, NY,
1987.

1
"A Conception of CT with Some Curriculum Suggestions," Newsletter on
Teaching Philosophy, American Philosophical Assoc., p1-5, Summer 1987.

45 42
 "A Bibliography of Testing CT," CT peas, Center for the Reasoning
5

Arts, CSU Sacramento, v6, nl, Sept.-Oct. 1987.

Ennis, Robert H, Gardiner, William L, et a!, Cornell Class Reasoning Test,

University of Illinois, 1964,

, Cornell Conditional Reasoning Test Illinois CT Project, Champaign,

IL, (date ?)

Ennis, Robert H., and Millman, Jason, Cornell CT Test Level X, Midwest
Publications, Pacific Grove, CA, 1985.

1 Cornell CT Test Level 2, Midwest Pub., Pacific Grove, CA, 1985.

Ennis, Robert H., and Weir, Eric, Ennis -Weir CT Essay Test, Midwest
Publications, Pacific Grove, CA, 1985.

Ennis, Robert H., and Norris, Stephen P., "CT Testing and Other CT Evaluation:
Status, Issues, and Needs," in Issues in Evaluation, Algina, James,
(Ed.), Ablex Press, New York, NY, 1988(?).

Facione, Peter A., "Toward a Theory of CT," Liberal Education, v70, p253 -261,
Fall 1984.

"Testing College Level CT," Liberal Education, v72, p221 -232, Fall
1986.

"Teaching about Fallacies," Teaching Philosophy, v101 pp. 211-217,

Summer 1987.

1 "Some Current Concerns Regarding HOTs Assessment," CT News, v7, n2/31

p3 & 9, 1988.

1 "Assessing Inference Skills," ERIC Clearinghouse on Tests,

Measurevents and Evaluation, TM 012917, Mar. 1989.

1 "Reasons and Premises," CT News v7, n4, p1, 7f, 1989.

1 "A Quick List of CT Tests," CT Ness, v7, n4, p12, 1989.

1 "Some Definitions of CT," CT NeNs, v7, n4, p101 1929.

1 "Strategies for Multiple Choice CT Assessment," in CT at Colleges and

Universities, David Hitchcock, (Ed.), Vale Press, Newport News, VA.,
forthcoming.

Follman, John, "Contemporary CT Bibliography," CT News, Center for the

Reasoning Arts, CSU Sacramento, v61 n2, Nov.-Dec. 1987.

Fraser, Berry, Test of Inquiry Skills, The Australian Council for Educational
Measurement, Hawthorn, Victoria, 1979.

Frederiksen, Norman and Ward, William B., "Measures for the Study of
Creativity in Scientific Problem Solving," Applied Psychological
 Measurement, v2, p1 -24, 1978.

Fritz, Paul A., and Weaver, Aichard L. II, "Teaching CT Skills in the Public
Speaking Course: A Liberal Arts Perspective," 36p, Presented at the
Annual Meeting of the Speech Communication Assoc., Chicago, IL, Nov.
1984.

Gardner, H., Frames of Mind: The Theory of Multiple Intelligences, Basic

Books, New York, 1983.

Glaser, Edward M., An Experiment in the Development of CT, Teachers College,

Columbia University, 1941.

Glaser, R, "Education and Thinking: The Role of Knowledge," American

Psychologist, v39, p93-1041 1984.

6rinols, Ann Bradstreet, (ed.) CT: Reading Across the Curriculum Ithaca, NY,
Cornell University Press, 1984.

Haars, Venant J., and Mason, Emanuel J., "Children's Understanding of Class
Inclusion and Their Ability to Reason with Implication," International
Journal of Behavioral Development, v9, nl, p45 -63, March 1986.

Haney, Walt, "Testing Reasoning and Reasoning about Testing," Review of

Educational Research, v541 n4, p597 -654, Winter 1984.

Harnadek, Anita, CT Troy Mich., Midwest Publications, 1976.

Hey, Ellen, "The CT Movement and Its Implications for the Teaching of Speech
Communication," paper to Speech Communication Assoc., Boston, MA, Nov.
1987.

Hermstein, R. J., et al, "Teaching Thinking Skills," American Psychologist,

v411 p1279-1289, 1986

Hitchcock, David, (Ed.), CT at Colleges and Universities, forthcoming, Vale

Press, Newport News, Virginia.

Howard, George S., and Englehardt, Jean L., "Teaching Rival Hypotheses in
Experimeotal Psychology," Teaching Psychology, v111 nl, p44 -5, Feb. 1984.

Hunter, Jacqueline; Jones, Lester; et. al., "Teaching Cognitive Skills in a

Pre-College Program," Journal of Learning Skills, vl, n2, p24 -26, 1982.

Kean, Michael H., "Testing Future Challenges, Future Responses," Presented at

the Annual Meeting of the National Ccuncil of Measurement in Education,
San Francisco, CA, April 1986.

Kearney, C. Philip, et al, " Assessing Higher Order Thinking Skills," ERIC
Clearinghouse on Tests, Measurement, and Evaluation, Princeton NJ, p86
April 1986

Kneedler, Peter E., "Testing CT at Grade Eight," Social Studies Review, v251
n2, p78-89, Winter 1986.

4' 44
 "History-Social Science Assessment in California," Social Science
,

Record v24 n1 p8 -9, Spr. 1987.

Lawson, Anton E., "The Effects of Causality, Response Alternatives, and

Context Continuity on Hypothesis Testing Reasoning," Journal of Research
in Science Teaching v20, n4, p297-3101 April 1983.

Lee, Graham, and Oakhill, Jane, "The Effects of Externalization on Syllogistic

Reasoning," Quarterly Journal of Experimental Psychology Human
Experimental Psychology, v36A, n3, p519 -30, Aug. 1984.

Lipman, Matthew; Sharp, Anne Margaret; and Oscavan, Frederick, Philosophy in

the Classroom, Universal Diversified Services, Inc., 1977.

1 (ed.) Growing Up with Philosophy, Temple University Press, 1978.

1 "CT: What Can It Be?" Educational Leadership, v461 p38, Sept. 1988.

Lohman, David F., "Predicting Mathemathanic Effects in the Teaching of Higher-

Order Thinking Skills," Educational Psychologist, v211 n3, p191 -208,
1986.

Mandic, Peter, and Stojakovic, Peter, "Contemporary Educational Psychology in

Yugoslavia: A Review of Research", Contemporary Educational Psychology,
v9, n3, p207-13, July 1984.

Martin, David S., "Restructuring Teacher Education Programs for Higher-Order

Thinking Skills," Journal of Teacher Education, v40, p2, May-June 1989.

Marzano, Robert J, and Jesse, D., "A Study of General Cognitive Operation on
Two Achievement Batteries and their Relationship to Item Difficulty,"
Unpublished paper, Mid-Continent Regional Educational Laboratory, Aurora,
Colorado, 1987.

Marzano, Robert J, et. al, Dimensions of Thinking: A Framework for Curriculum

and Instruction, Assoc. for Supervision and Curriculum Development,
Alexandria, VA., 1988

Marzano, Robert Ji and Costa, Arthur L., "Question: Do Standardized Tests

Measure General Cognitive Skills? Answer: No," Educational Leadership,
v461 pp 66-71, May 1988.

McKee, Saundra J., "Impediments to Implementing CT," Social Education, v52,

p443, Oct. 1988.

McPeek, John, Critical Thinking and Education, St. Martin's Press, New York,
NY, 1981.

Mebane, John S., "Teaching Interpretive Skills Through Testing in Literature

Class," Exercise Exchange v32, nl, p7 -10, Fall 1986.

Modieski, Richard B., and Michael, William B., "An Evaluation by a Panel of
Psychologists of the Reliability and Validity of Two Tests of CT"
Educational and Psychological Measurement, v431 n4, p1187 -97, Winter
1983. (The tests reviewed were the "Watson-Glaser CT Appraisal" and the

4-
4rs
 "Cornell CT Test."]

Morante E. A., and Ulesky, A. "Assesmment of Reasoning Abilities," Educational

Leadership, v42, p71 -74, 1984.

Moss, Pamela A., and Petrosky, Anthony R., "A Proposal for Meastiring Critical
Thinking," A revision of paper presented at the Annual Meeting of the
American Educational Research Assoc., Montreal, Canada, April 1983.
Pittsburgh board of Public Education, PA, Report No: T-83-2. 23p, Sept.
1983.

Moore, Brooke, and Parker, Richard, "The Written Debate -- A Technique for
Honing CT Skills," HeNsletter on Teaching Philosophy, American
Philosophical Association, p8, Sumter 1987.

Morgenstern, C. F., and Renner, J. W., ''Measuring Thinking with Standardized

Science Tests," Journal of Research in Science Teaching, v21, p639 -648,
1984.

Meyer, Chet, Teaching Students to Think Critically, San Francisco, Jossey-

Bass, 1986.

Nickerson, R. S., "Kinds of Thinking Taught in Current Programs," Educational

Leadership, v42, p26-36, 1984.

Nickerson, R. S.; Perkins, D. N., and Smith E. E, The Teaching of Thinking,

Lawrence Erlbaum Associates, Hillsdale, NJ, 1985.

Norris, Stephen P. Test on Appraising Observations, Institute for Educational

Research and Development, Memorial University of Newfoundland, 1963.

"The Choice of Standard Conditions in Defining CT Competence,"

,

Educational Theory, v35, p97 -107, 1985.

"Evaluating CT Ability," History and Social Science Teacher, v21, n3,

1

p135-146. Spr. 1986.

, "Controlling for Background Beliefs When Developing Multiple-Choice

CT Tests," Educational Heasurepent v7, n3, p5 -11, Fall 1988.
,

1 "Research Needed on CT," Canadian Journal of Education,, v13, p125-

37, 1988.

1 "Verbal Reports of Thinking and Multiple-Choice CT Test Design,"

Technical Report Ho. 447, Champaign, IL: Center for the Study of Reading,
University of Illinois. (ERIC Doc. No: ED302826.)

1 "Effect of Eliciting Verbal Reports of Thinking on CT Test

Performance," Journal of Educational Heasureeent, v27, nl, 1990.

1 "Informal Reasoning Assessment: Using Verbal Reports of Thinking to

Improve Multiple-Choice Test Validity," in Inforaal Reasoning and
Education, D. N. Perkins, J. Segal, and J. F. Voss (Eds.), Lawrence
Erlbaum Publishing, Hillsdale, NJ, in press.

46
4
 "Can We Test Validity for CT?" Educational Researcher, in press.

Norris, Stephen P., and Ennis, Robert H. Evaluating CT, Midwest Publications,
Pacific Grove, CA, 1989.

Norris, Stephen P., and King, Ruth, "Observation Ability: Determining and
Extending its Presence," Infornal Logic, v6, n3, p3 -9, 1984.

Norris, Stephen P., Phillipe, L. M., "Explanation of Reading-Comprehension:

Schema Theory and CT Theory, Teaching College Reading, v89, n2, p281-306,
1987.

Norris, Stephen P., and Ryan, James, "Designing a Test of Inductive

Reasoning," in Proceedings of the International Conference on
Argulentation 1986, Van Eemeren, Frans H., and Grootendorst, Rob (Eds.),
Faris Publications, Dordrecht, Holland, 1988 (?).

Passmore, J., "On Teaching to be Critical," The Concept of Education,

Routledge and Kegan Paul, London, p192-:11, 1967.

Patrick, John J., "Core Content, CT, and Civil Values: Issues on Education in
the Social Studies," in Trends and issues in Education, 1986. Flaxman,
Erwin, (Ed.), Council of ERIC Directors, ERIC Clearinghouse on Urban
Education, New York, Jan. 1987.

Paul, Richard W., "Teaching CT in the Strong Sense: A Focus on Self-

Deception, World Views, and a Dialectical Mode of Analysis," informal
Logic, May, 1982.

"The CT Movement: A Historical Perspective," National Fortin, Winter

1985.

1 "CT Research: A Response to Stephen Norris," Educational Leadership,

v42, p46, 1985.

1 CT Handbook Rohnert Park, CA, Sonoma State University Center for CT

and Moral Critique, 1986.

1 "Dialogical Thinking: Critical Thought Essential to the Acquisition

of Rational Knowledge and Passions," Teaching Thinking Skills, Perkens,
et. al. (eds.), Lawrence Erlbaum Associates Inc., 1987.

1 "CT in North America: A New Theory of Knowledge, Learning, and

Literacy," ms. Center for CT and Moral Critique, Sonoma State U, Rohnert
Park, CA., 1989.

, "35 Strategies for Infusing CT Principles and Applications," ms.

Center for CT and Moral Critique, Sonuma State U., Rohnert Park, CA,
Jan. 1989.

1 "CT: What, Why, and How," ms. Center for CT and Moral Critique,
Sonoma State U, Rohnert Park, CA, April 1989. (Draft for CT: Educational
Isperative New Directions for Cosaunity Colleges, Jossey-Bass.)

Pecorino, Philip, "Cl Bibliography," Newsletter on Teaching Philosophy

47 dr-0
 American Philosophical Association, p18, Summer 1987.

Perkins, D. N.,"Post-Primary. Education Has Little Impact on Informal

Reasoning," Journal of Educational Psychology, v771 p562 -571, 1985.

"Reasoning As It Is and As It Could Be: An Empirical Perspective,"

,

Presented to the American Educational Research Assoc., San Francisco, CA,

1986.

Knowledge as Design: Critical and Creative Thinking for Teachers and

1

Learners, Hillsdale, NJ, Erlbaum Assoc., 1986.

Duellmalz, Edys S., "Designing an Integrated Instructional and Assessment

System for CT SKills," Presented to the American Educational Research
Assoc., Montreal, Canada, 1983.

"Needed: Better Methods for Testing Higher-Order Thinking Skills,"

Educational Leadership, v43, p29-35, 1985.

Ramirez, Paul M., "Valett Inventory of CT Abilities," The Reading Teacher,

v41, p348, Dec. 1987.

Raths, Louis Edward, et. al., Teaching for Thinking Theory and Application,
Columbus, Ohio, C. E. Merrill Books, 1967, (2nd ed.) 1986,

Resnick, L. W, Education and Learning to Think, National Academy Press, 1987.

(ed.) Cognition and Instruction: Issues and Agendas, Lawrance Erlbaum

1

Associates Inc., (in press).

Ross, Elinor P., "Teaching--Not Just Testing--The Main Idea," Reading florid,
v24, nl, p84-88, Oct. 1984.

Ross, John A., and Maynes, Florence J., Teaching Problem-Solving The Ontario
Institute for Studies in Education, Ontario, Canada, 208p, 1982.

"Development of a Test of Experimental Problem-Solving Skills,"

1

Journal of Research in Science Teaching v20, p63 -75, 1983.

Rud, A. Jr., "CT and Pre-College Philosophy," Presented to the American

G.
Educational Research Assoc., San Francisco, CA, 1986.

Ruggiero, V. R., Teaching Thinking Across the Curriculum, Harper and Row, New
York, 1988.

Sabini, John, and Silver, Maury, "CT and Obedience to Authority," National
Forum, p13-171 Winter 1985,

Scherer, Donald, and Facione, Peter A., "A Pre/Post Test for Introductory
Logic Courses," Netaphilosophy, v81 n4, p342 -347, Oct. 1977.

Schievella, P. S., Critical Analysis: Language and Its Functions, Humanities

Press, New York, NY, 1968.

, Hey! Is That You, God? Sebastian Pub. Co, Port Jefferson Station NY.

41.
Striven, Michael, "Critical for Survival," National Forum, p9 -12, Winter 1985.

"Multiple-Rating Items," draft ms., U of Western Australia, Nedlands,

1

Western Australia, Oct. 1988.

Sheffler, Israel, Reason and Teaching, Bobbs-Merrill, 1973.

Shemesh, Michel, and Lazarowitz, Reuven, "The Interactional Effects of

Students' Cognitive Levels and Test Characteristics on the Performance of
Formal Reasoning Tasks," Research in Science and Technological Education,
v6, nl, p79-891 1988.

Shipman, Virginia, New Jersey Test of Reasoning Skills, Institute for the
Advancement of Philosophy fur Children, Montclair State College, NJ,
1983.

Siegel, Harvey, "CT as an Educational Ideal," The Educational Forum, p7-23,

Nov. 1980.

1 Educating Reason: Rationality, CT, and Education, Routledge, 1988.

Smith, E. L., "Effects of Training and Use of Specially Designed Curriculum

Materials on Conceptual Change Teaching and Learning," paper to National
Assoc. for Research in Science Teaching, Washington DC, 1987.

Smith, Charles W., "Verbal Behavior and Classroom Practice," 36p, Presented at
the International Conference on Thinking, Cambridge, MA, Aug. 1984.

Sternberg, Robert J., Handbook of Huean Intelligence, Cambridge University

Press, 1982.

1 "Teaching CT Part I: Are We Making Critical Mistakes?" Phi Delta

Kappan, v67, p194 -198, 1985.

"Teaching CT Part II: Possible Solutions," Phi Delta Kappan, v67,

1

p277-280, 1985.

Beyond IG: A Triarchic Theory of Human Intelligence, Cambridge

1

University Press, New York, 1985.

"In Defense of CT Programs," Education NeeA Commentary, 19, Oct.

1986.

1 "Teaching CT: Eight Easy Ways to Fail Before You Begin," Phi Delta
Kappan, v68, p456 -459, 1986.

1 (Ed.) Teaching Thinking Skills: Theory and Practice 1987.

1 Its Nature, Measurement, and Improvement," National Institute of

"CT:
Education, Washington, DC, 37p, 1986.

Sternberg, Robert J., and Baron, Joan B., "A Statewide Approach to Measuring
CT Skills," Educational leadership, v43, n2, p40-43. Oct. 1985.

4::.
Cr, e")
11.1
Sternberg, Robert J., and Bhana1 K., "Synthesis of Research on the
Effectiveness of Intellectual Skills Programs: Snake-Oil nemedies or
Miracle Cures," Educational Leadership, v44, p60-67, 1986.

Sternberg, Robert J., and Wagner, Richard K., Understanding Intelligence:

Mhat's in it for Educators, National Commission on Excellence in
Education, Washington, DC, 82p, July 1982.

Stewart, B. L., "Testing for CT: A Review of the Resources," Rational Thinking
Reports Caber 2, University of Illinois, Champaign-Urbana, IL, 1987.

Stovall John E., "Document-Based Questions and Testing for CT," Social Science
Record, v24, nl, p11-12, Spr. 1987.

Swartz, Robert, "What is CT and Haw Can We Teach It?, ms. U. Mass. Boston,
June, 1989.

Talbot, Jan, "The Assessment of CT in History/Social Science Through Writing,"

Social Studies Review, v24, n2, p33 -41, Winter 1986.

Thompson, B., and Melancon, J. G., "Validity of a Validity Measure of CT

Skills," Psychological Reports, v60, p1223-1230. 1987.

Tucker, Carol B., "Examples of Question Types Used to Test Reasoning,"

Educational Testing Service, Princeton NJ, Apr. 1986.

1 "Some Multiple-Choice Formats Used to Test Reasoning," Educational

Testing Service, Princeton NJ, Apr. 1986.

"Working Paper on Defining CT for Purposes of Testing," Educational

Testing Service, Princeton, NJ, Summer 1988.

Van Der Bogert, M. Rebecca, "The Evolution of the Concept of CT: A Literature
Review," Ms. Harvard University, 1988.

Watson, Goodwin, and Glaser, Edward M. Matson-Glaser CT Appraisal, The

Psychological Corporation, San Antonio, TX, 1980.

Weddle, Perry, "CT in California: The Department of Education Testing Program

in Social Studies," History and Social Science Teacher, v21, n3, p147-
149, 151-154, 156-157, Spr. 1986.

Whimbey A., and Lochhead, J., "You Don't Need a Special 'Reasoning Test' to
Implement and Evaluate Reasoning Training," Educational Leadership, v43,
p37-391 1985.

50
 MEELaILS
Ibelleaphis,zghRes
a Lette_gg

Over course of this research sixteen "Delphi Letters" were

sent to the experts participating in the APA Committee on Pre-

College Philosophy Delphi Research Project on CT assessment.

Eight letters constituted the specific interactive Delphi

research rounds. The eight others were for purposes of planning,

clarifying procedures, providing information, and sustaining

involvement.

Feb. 11, 1988 ROUND 1

Mar. 1, 1988 Information and Plans

Mar. 14, 1988 ROUND 2
Apr. 14, 1988 Plans and Procedures
May 4, 1988 ROUND 3
May 18, 1988 Procedures and Involvement
June 28, 1988 Information and Involvement
Sept. 1, 1988 Procedures and Information
Sept. 23, 1988 ROUND 4

Nov. 22, 1988 Plans and Involvement

Feb. 7, 1989 Information and Involvement
Feb. 28, 1989 ROUND 5A
Mar. 6, 1989 ROUND 5B
Mar. 10, 1989 ROUND 5C
May 9, 1989 Information and Involvement
Sept. 25, 1989 ROUND 6

51
 California State University, Fullerton
Fullerton, California 92634

Department of Philosophy Feb. 11, 1988

(714) 773-3611
ROUND

Dear Colleague,

Gary Matthews, Chair of the APA Committee on Pre-College

Philosophy, asked me to head up an ad hoc committee on testing
critical thinking. To get the project rolling he suggested severe
names of people interested in the question of 1%Pqr to validly and
reliably test critical thinking skills. I adoea.a few more. Here are
some of the particulars.
1) By using a modified Delphi approach, I think the necessity for
actual meetings can be largely, if not entirely eliminated. Committee
members' contributions will involve sending their reasoned and timely
responses to questions, given their particular background and
expertise.

2) There are, no doubt, a great number of other people who are

interested in the question of testing critical thinking and who have
valuable expertise which would help us in in dealing with this
question. They should also be invited to participate.
3) Since our charge is rather vague, I propose that among the
first things we should do is agree on priorities. To do this, using
the Delphi process, let me lead off with some assumptions and
questions:

First assumption: Most of the members of our group will come at

the issue of testing critical thinking with the orientation of
philosophers or logicians who teach at the post-secondary level.
rather than as K-12 educators, psychologists, or personnel
directors (all of whom also have legitimate theoretical and
practical interests in assessing critical thinking). Given the
interests of the Amer4can Philosophical Association, thisi.s
acceptable.

Second assumption: Critical thinking can be defined operationally

to the extent that it can become a dependent variable in a valid
and reliable assessment tool.

QUESTIONS:
1. Do you agree with the two assumptions? How would you
amend/clarify them? Why?

2. To which educational level (from Kindergarten through post-

Baccalaureate) should the committee give priority? Why so?
3. After looking the attached preliminary list, whom else
would you recommend be added to our committee?
the Cal!fornia State Unp.ersity

Appendix C: Delphi Research Letters, PAGE 52

 For the Delphi process, which is very interactive, to function
optimally, reasonably quick turn around is needed. Let's target a
Feb. 29.postmark. Is that possible for you?

Please send responses, suggestions, comments, etc. to:

Ad Hoc APA Committee on Critical Thinking Testing

c/o Dr. Peter A. Facione
Professor of Philosophy and Education
Dept. of Philosophy
California State University, Fullerton
Fullerton, CA 92634

Like math and composition, 'at many colleges throughout the

country CT is being built into the curriculum. For example, the
California State University system, which enrolls hundreds of
thousands of undergraduates, has implemented a system-wide critical
thinking requirement as part of its general education package. If
those of us who teach critical thinking were able to agree on a way or .

ways it could be tested, what a positive contribution that could be to

the quality of that curriculum.
I sincerely hope you will agree to become an active participant
in what promises to be a most interesting and important effort.
I've included some reference material from The Ninth Mental
Measurements Yearbook regarding published instruments which purport to
measure critical thinking and/or related cognitive skills.
Yollpensincerely,
.0/
Pete Facione

cc. Gary Matthews, University of Massachusetts, Amherst

Steve Tigner, University of Toledo, Ohio

ATTCHMENTS: Preliminary List of Committee Members

Quick Survey of Published Instruments

TO BE DEVELOPED: Bibliography on Testing Critical Thinking

Appendix C: Delphi Research

Letters, PAGE 53
 California State University, Fullerton
Fullerton, California 92634

Cepa Iment of Philosophy

(714) 773-3611

March 1, 1988

Dear Colleague,

Early :responses to the first Delphi round are very encouraging!

Some of you know what I'm talking about, others, new to the effort,
need to be brought up to speed. Let me back track a bit and explain.

I am writing to you because you were nominated by one or more of

your colleagues as a person interested in the question of testing
critical thinking. You are being invited to participate in the work of
an ad hoc American Philosophical Association subcommittee concerned with
the problem of testing critical thinking. In its boldest form, our aim
is to find ways to validly and reliably test critical thinking, or find
solid reasons why such a goal is not achievable. Using the Delphi
process, I will serve as coordinator of the effort. At the moment we
are very near the beginning of Phase 1.

Please review the preliminary plan outlined below and decide to

become actively involved in what promises to be an intriguing effort to
shed light on an important pedagogical and profession concern.

ad hoc APA Sub-Committee on Testing Critical Thinking

Draft Preliminary Plan

Phase 1: Start the Inquiry

The five objectives of this phase are:

(a) Initiating the Delphi process. This is a method of achieving

reasoned consensus among a group of experts with regards to a given
problem or issue. The core of the strategy is to make inquiries,
uther each expert's responses and their reasons, then summarize
and share those with the group. After "hearing" what other experts
think, people hale the opportunity to refine their responses or
defend those responses. The interactive rounds continue until
reasoned consensus is achieved (or communicationsbreak down).

(b) Developing the "List of Experts" who will take part in this
inquiry. Many of you were nominated by those who replied to my first
letter, (2-11-88). That letter was sent to an original group of
about twenty-five experts and interested persons suggested by
the APA committee that conceived of this project and asked me to
coordinate it. Az any time if someone is no longer interested in
continuing, just drop me a note. I plan to send out updated
rosters of participants periodically. We are now up to fifty.
5 I"
Appendix C: Delphi Research Letters, PAGE 54
Th r. Pt, 1.1.1, et.. t
 (c) Developing a bibliography on testing critical thinking. Work
on this is progressing well. Many of you have sent me items to
include, and I appreciate that. I will send you a copy of the
bibliography later this semester.

(d) Planning the subseqent phases in our process of responding to

the general question of whether or not critical thinking (whatever
CT is) can be validly and reliably tested at some educational
level or levels. If the consensus is "Yes, at level X," then we
will focus on the question: How? If the consensus is "No, at leant
not at level X," then we will focus on, "Why not?"

(e) Agreeing on basic assumptions.

In the 2-11-88 letter two assumptions were put to the group:

Assumption 41: "Most of the members of our group [of experts

participating in the Delphi process] will come at the issue of
testing critical thinking with the orientation of philosophers
or logicians who teach at the post-secondary level, rather
than as K-12 educators, psychologists, or personnel directors
(all of whom also have legitimate theoretical and practical
interests in assessing critical thinking). Given the
interests of the American Philosophical Association, this [is
an] acceptable [orientation] ."

Assumption #2: "Critical thinking can be defined

operationally to the extent that it can become a dependent
variable in a valid and reliable assessment tool."

People were asked (1) if they agreed with the two assumptions
as stated, or whether they would reject them or rephrase them
somehow. Naturally, people were invited to explain why. They
were also asked (2) to identify the educational level (K-post
baccalaureate) to which our committee should give priority,
and why they would recommend that level.

Both questions have genereted controversy, as you will see in my

next letter. If you haven't had the opportunity to respond to these
questions, you will be invited to respond when the first round of the
Delphi is reported back to you.. Delphi is not about vote-counting, it
aims at reaching agreement on the basis of reasons and common
assumptions. In the Delphi method people are supposed to share their
premises, not just their conclusions.

As conceived at the moment, our work can be divided into 4 phases.

These are not in stone! I welcome your suggestions, amendments,
alternatives, etc. WE WILL USE THE DELPHI PROCESS TO AGREE ON OUR PLAN
OF INQUIRY. Because the plan should be. amended as a result of your input,
only goals, not detailed objectives, have been developed so far for
the next three proposed phases.

Appendix C: Delphi Research Letters, PAGE 66

 Phase 2: Define "Critical Thinking"
The goal of this phase is to define "critical thinking" with
sufficient clarity and precision to ask and answer the
question of whether or not CT can be tested. Is CT
fundamentally a set of skills, concepts, procedures,
attributes, behaviors, outcomes, dispositions, aptitudes, or
what? Even if we cannot reduce CT to an equivalent
operational definition, how might we express what CT is with
sufficient operational precisions to permit us to justifiably
infer things about the relative CT abilities of students?

Phase 3: Recommendations
The goal of this phase is to communicate our findings about
what CT is and whether there is an adequate way of
characterizing CT operationally so as to permit its being
tested as some educational level. Depending on our results in
Phase 2, we will recommend either that programs aimed at
testing CT be abandoned, or:that they be focused in certain
ways. If this is the direction Phase takes, then we will
also try to come to consensus on recommendations regarding the
relative importance of different kinds of CT sub-skills and
possible strategies for accessing and measuring those sub-
skills.

Phase 4: Design and Validation of Model Testing Strategies

Contingent on the results of earlier phases, the goal, if it
were considered achievable in principle, would be to construct
and evaluate different approaches to testing CT at some
appropriate educational level or levels. We might find
ourselves breaking into sub-committees to achieve this goal,
although all work will have to be guided by the agreements
reached in earlier phases and as well as by the special
expertise of those who understand the intricacies of
designing, piloting, forming and validating educational tests
at specific educational levels.

As I mentioned, you have been nominated as person who might be

interested in this project and could make a strong contribution to the
work of this ad hoc sub-committee. I hope you will agree to participate
actively, because, as you must know, the quality and utility of our
effort is directly related to the involvement of concerned persons like
yourself.

Sincerely,

Pete Facione

5D
Appendix C: Delphi Research Letters, PAGE 56
 California State University, Fullerton
.
Fullerton. California 92634

Department of Philosophy
(714) 773-3611
March 14, 1988
RO vat D
Dear Testing CT Colleagues,

Let's give Phase I, Round 2 a shot!

Thanks for your responses to the first round of questions. Nineteen of the
twenty-six or so who received the original 2-11-88 letter for Round 1 were able
to respond. Round 2 invites everyone (which now includes just over 50 people)
to review the results of Round 1 and coament on the agreements and controversies
that are emerging.

Round 1 focused on three issues: (1) The composition of our ad hoc

committee in view of the interests of the berican Philosophical Association,
(2) the assumption that critical thinking can be operationally defined, and (3)
the educational level to which we should give priority.

In regard to the composition of our committee, we are in decent shape,

particularly since our group has been greatly expanded as per your
recommendations. In regard to an operational definition of CT, we generally
agree on the possibility but many would add various caveats. In regard to the
educational level to which we should give priority, we have disagreements.

The following pages cover each of the three questions in turn. You'll find
restatements of the original questions and several representative quotes and
summaries of your comments. AFTER EACH SUMMARY, A SECOND ROUND QUESTION WILL BE
PUT TO YOU. The new questions take the form of stating a position and asking
your opinion, now that you have had an opportunity to consider what our
colleagues have to say. In all, there are three new questions. IF YOU COULD
GET YOUR RESPONSES TO ME WITHIN TEN DAYS OF RECEIVING THIS LETTER THAT WOULD BE
GREAT! (I wish we all had electric mail, or unlimited phone budgets, but...)

Several people noted that our task was huge, yet were willing to give it a
try. In contrast, one person wrote a major critique of the entire enterprise.
This person argued that trying to test CT was a serious mistake. So that his
opinion is not lost in the shuffle, at the end of this package I have provided
extensive quotations from his letter. If you find yourself in agreement with
his views, then let me know and we will take up any "prior questions" we must.
If you don't agree, then we will press on along the path we are charting for
ourselves.

Think you in advance for your partiAioation.

6()
s ..s!nC-yuly
Appendix C: Delphi Research Letters, PAGE 57
 a.

PHASE I, ROUND 2, FIRST ISSUE

CURRENT STATUS: CLOSE TO CONSENSUS

COMPOSITION OF THE AD HOC COMMITTEE: We were asked if we agreed with the

assumption that, although most of the membe.'s of our ad hoc committee would come
at the issue of testing with the orientation of philosophers or logicians who
teach at the college level, this orientation still would be acceptable in terms
of the interests of the American Philosophical Association.

Almost everyone agreed, however some qualified their responses in terms of

our collective professional interests and abilities, or in terms of educational
level to which we should give priority. Here are representative responses:

"I agree." "I have no problems with this assumption."

"I don't see why this is a problem. First we're concerned about students
acquiring the thinking skills required for college work,... Second, we're
concerned that they learn the standards of good reasoning; I do not believe we
need to know a lot about psychology to:achieve this purpose."

"I agree, with reservations. We need to avoid tunnel vision. It is

acceptable that most members be philosophers, but there should be a generous
sprinkling of 'outsiders' for the insights they will bring and to give our
findings greater credibility outside the APA."

"We are what we are! This is an appropriate place to begin. We are

starting from what we know best and with what we can deal with most easily.
This is not to suggest that we shall forever ignore other orientations, or that
we really know that we can define all aspects of CT operationally."

"I agree, this is acceptable; but it is unnecessarily narrow. Since so

many of the tests are created by cognitive and educational psychologists, I
think some of them should be included..."

"It should not be too quickly assumed that those who teach at the post-
secondary level are therefore knowledgeable and competent with regard to testing
at the elementary school level."

"I agree, but we should make a serious effort to inform ourselves of

approaches to CT in pre-college and non-academic settings... Assuming our
primary focus is everyday reasoning skills, we should not allow college CT
instruction to be fundamentally different from pre-college CT instruction nor to
become idiosyncratically colored by our own traditions.

One person disagreed but did not give a reason. And one urged "Pete, get a
proof-reader!"

ROUND 2 ONE: In view of the above comments, and in view of the

additional names added as the result of your recommendations, can we agree that
th ad hoc committee, as listed on the attachment, is sufficiently well-
constituted for us to move on with our main task? As you can see, it still has
its original orientation toward philosophers teaching at the college level, but
it also includes several people from other relevant disciplines backgrounds,
including psychology and education.
6i

Appendix C: Delphi Research Letters, PAGE 58

 PHASE I, ROUND 2, ISSUE 2
CURRENT STATUS: AGREEMENT SEEMS ACHIEVABLE

OPERATIONALIZING CRITICAL THINKING: We were asked if we agreed with the

assumption that CT can be defined operationally to the extent that it can become
a dependent variable in a valid and reliable assessment tool.

Here, too, most people agreed and were ready to get on with the work. Yet,
some crucial ambiguities, concerns and caveats emerged. Here are some
responses.

"I expect we will argue about the details of any definition proposed, but I
do not object to the assumption that we shall need some such instrument if we
are to get any comparisons of interest."

"This is a tautology because of the to the extent phrase. Perhaps this is

how we should leave it..."

"Sounds ok. ...I'm not a statistician, so I'm not quite sure what
dependent variable means -- but if you'are asking whether CT can be tested,
then, yes, I agree!"

"I don't understand what this assumption is supposed to mean:"

"I would agree only if we amended it to say at least some components of CT

can be defined operationally to the extent that they can become dependent
variables in a valid and reliable assessment tool... I do not accept as analytic
the proposition that CT can be defined operationally... I think some important
aspects of CT, such as making judgment calls and weighting nuances may resist
operational definition."

"As the term operational definition is generally used by philosophers and

education researchers, I do not think CT can be operationally defined... but I
do think that part of the operational spirit can be employed in formulating
reduction sentences (that do not reduce!)."

"I agree, but there will probably have to be a variety of sub-definitions

because CT is not one thing, but many. It somewhat resembles IQ in that." '

"There are several definitions of CT floating around... Some lend

themselves more to operational definition than others.... If we are to get
anywhere, we will have to become clear in our own minds as to how CT is to be
distinguished from other kinds of thinking..."

"Is this a normative, definitional, conceptual, or planning assumption?"

ROUND 2 QUESTION TWO: Without hanging ourselves up on the word "operations,"

can we agree that: (1) Even if CT cannot be reduced entirely to an equivalent
set of operations [or performances, behaviors, processes, outcomes, or skills,]
(2) it is possible to conceptually analyze CT so as to describe a set of
relevant and important CT operations, such that (3) -using these descriptions,
[competent] investigators could, on a consistent basis, gather sufficient
evidence to draw conclusions, with high degrees of confidence, regarding the
relative CT abilities of a group of people, (everything else being equal, of
course) .

A.0

Appendix C: Delphi Research Letters, PAGE 59

 PHASE I, ROUND 2, QUESTION 3
STATUS: CONTROVERSY

PRIORITIES: We were asked which educational level (kindergarten through

post-baccalaureate) should be the priority for our committee. And why so?

Responses were split. Here are some representative examples:

A person with considerable experience in the area of Cl testing wrote: "I

think we should concentrate on high school at first. Since this is a
subcommittee of the pre-college committee, all [levels] above that [are] ruled
out. Furthermore, the younger the population, the more difficult the problems.
Let's start with the easiest ones first -- and they are very difficult."

By contrast, it was argued, "Priority should be given to the post-secondary

level. One should examine end-products first, and then work backwards if
needed. Find out first if the car doesn't run before attempting to determine
where the problem is. If a good test of CT revealed no CT deficiencies on the
part of graduating seniors -- (no doubt a counter-factual assumption) -- then I
would think the APA might not wish to pursue the issue down in K-12."

Noting that we are a sub-committee of the APA pre-college committee one

person argued: "We must give priority to K-12; that is our mission."

However, the person who will assume the chair of the APA pre-college
committee for the next three years wrote: "It makes sense to start by playing
from the APA membership's greatest experience and strength, which is surely
college freshman level logic."

Taking note of the interests of the APA, one person argued, "Since our ad
hoc committee is convened within the structure of the APA, our focus should
perhaps be primarily on the improvement of post-secondary education..."
However, this person also suggested, "... that our assessment tool should be
usable in secondary schools as well as at the post-secondary level..."

Some people did not offer an opinion, but did note important distinctions.
For example: "There are really two areas. One is the whole K-12 integration of
thinking skills into the curriculum. The other is the single CT course,
typically the approach followed in post-secondary education. The single college
CT course offers exceptional opportunities for measuring gains in thinking
skills, while the effort to incorporate thinking skills into the [K-12]
curriculum may offer much greater potential for actually improving student
skills."

Another person, experienced in the pre-college arena, wrote: "It may be

necessary to think of four tests, one for grades K-3 (one should not have high
expectations for reliability at this level); one for grades 4-8 (the level at
which testing might have the maximum impact, even though the maximum impact for
the teaching of CT might be at K-3); one for grades 9-12, and one for 13-16."

Some were tentative: "Perhaps we should give priority to CT at the college

level, at least to start with, since the large majority of APA members teach at
the college level. Later we might wish to broaden our focus."

Appendix C: Delphi Research Letters, PAGE 60

 Some were direct, "College and university level."

Others were focused, buL concerned not to overlook anything important,

"College level -- but someone ought to look at the high school level."

One person declared for the college freshman level and ergued against going
any higher saying, "There are few if any thinking skills possessed by people
beyond this level not also possessed by well-prepared college freshman.
Graduate school and professional life chiefly consists in the ability to
persistently apply these skills in more and more recondite subject matters.

Some narrowed the range, but still left us with a choice: "I would say
grades 9-12 and freshman/sophomore level in college."

Another argued we should give priority to the introductory baccalaureate

level saying, "First, it is the area where most philosophy departments have
numerous classes actually being taught. Second, it is taught at a level which
will have the most connections in other areas and at other levels."

ROUND 2, QUESTION THREE: To get started let's give priority to the college
freshman/sophomore level. Do you agree? If not, is your disagreement based on
pedagogical and theoretical concerns or on concerns relating to our charge as a
subcommittee of the APA committee on pre-college philosophy?

Please try to get your responses to me within ten days. Thanks.

Appendix C: Delphi Research Letters, PAGE 61

 PHASE I, A CHALLENGE RAISED AGAINST THE ENTERPRISE

One person wrote: "I would like to comment that teaching CT skills... is a
far more significant matter than testing for them. Since CT is not an inherited
trait, as is intelligence, the quality and extent of the CT is dependent on and
proportional to the degree to which children have learned to, or have been
taught to, think critically about their experiences and knowledge.

"Our educational system ... is an abysmal failure. Most students are

unable to recognize assumptions, not alone question or examine them. ... With
rare exceptions, they show an almost total absence of recognition of even the
simplest of logical/mathematical/linguistic/philosophical/scientific facts and
concepts needed to be able to think critically. They have been so nurtured in a
world of superficial "education" ... that to think of testing them on the basis
of that "education" is an exercise in futility or at best an attempt to
determine how inadequately they think critically as opposed to how much -- which
if taken literally amounts to the same thing.

"I am willing to contribute my expertise to the teaching of CT. I have

been doing so for over thirt, years. But until I see considerably more evidence
of students being able to think critically without such teaching, I see little
point in testing them for the insignificant amount of critical acuity they may
have acquired haphazardly.

... Teaching CT must precede and supersede testing for CT. Testing for CT
cannot be considered to be an enterprise separate from teaching it. Testing for
such skills and concepts presumes prior teaching of them. Psychologists and K-
12 teachers as well as other educators show interest in CT. The problem
remains, however, that most of them have only superficial, naive, and
conflicting concepts, of what critical thinking entails. Even we philosophers
can't agree on what it is...

"From my understanding of the term, testing for CT means testing based on

what I teach CT is. ... Any tests that I would, and have designed, are
predicated upon the version of CT I have taught.

"My comments will undoubtedly reflect those which you will receive from my
colleagues equally concerned with the problems facing our educational system,
particularly as they relate to the teaching of CT skills and concepts."

REMINDER: If you believe there are issues (suggested by the above or otherwise)
which our ad hoc sub-committee must address before we can move ahead, please let
me know. On the other hand, the above challenge may represent a view which is
not widely shared, or may raise questions which, in your view, do not fall
within the scope of our work or do not warrant our attention at this time. If
that is the case, then, for the present, no response on this item is necessary.

Thanks again for your participation.

Appendix C: Delphi Research Letters, PAGE 62

uJ
 California State University, Fullerton
Fullerton, California 92634

Department of Philosophy
(714) 773-3611

April 14, 1988

Dear Colleagues,

Thanks for your help with the addresses. I also appreciate the
notes, suggestions and other helpful comments many of you are
sending.

So far 17 responses to R.Jund Two have been received. It would

help if we had more, particularly since it there may still he a
split over Question Number 3, the one:about which level to pursue
first. Drop me a note, with your reasons and opinions. Thanks.

One of the responses to Round Two proposed an alternative to

actually trying to come up with our own CT assessment instrument
a goal some of us hope to achieve, but others of us are
extremely skeptical about. This person suggested that we
articulate the best list of CT skills we can, then let people go
their own way with regard to building testing instruments. I'll
expand on that idea when I summarize Round Two responses. I

mention it here because perhaps we all should be thinking ahead

and trying to chart the most reasonable path for ourselves.

Some of you have electronic mail. I don't, at least not yet.

So, I'll be using snail mail and telephone to try to reach you
for particular questions or clarifications. If you want to phone
in your views on the questions in Round Two you can reach me at
714-773-3742 (office) 08:30-10:00 MV or 09:30-11: '0 TTh (PDT).
If those times are inconvenient, call the department secretary at
714-773-3611 and leave your phone number. I'll get back to you.

I will be attending the First National Conference on Assessing

Thinking in Baltimore on May 6 and 7. This conference is
sponsored by the Maryland State Department of Education and the
Association for Supervision and Curriculum Development. Although
plans are to discuss all educational levels, the participating
organizations, (over 35 professional associations, centers and
government agencies) are concerned primarily with K-12 education
in some way or another. I'll report on what promises to be a
most interesting gathering.

within the week you should receive two its-.s interest. One is a
partial bibliography CT with emphasis on testing CT, the other is
an updated listing of some of the existing tests whch purport to
:ensure CT or closely related reasoning s:;111s.

Appendix C: Deipni Research Letters, PAGE 63

The California Slate University
 AID

California State University, Fullerton

Fullerton, California 92634

Department of Philosophy
(714) 773-3611

RovAID
May 4, 1988
Dear Colleagues,

I hope this letter finds you happy and well.

Round 3 of our Delphi -- the round where we finally It's time for
get to the
heart of the matter: What is CT? Also, for those who are new
to our effort, this letter includes a brie!: overview
of who we
are and what we are about. The last three pages summarize
Round 2 results. our
Please send Round .3 responses by June 25.
As a result of nominations in Round 1 and Round 2, sixty
persons, including some of the most eminent names in the field,
are now invited to participate in this effort.

By way of background, in January the American

Philosophical
Association Committee on Pre-College Philosophy asked me to chair
an ad hoc sub-committee on testing critical thinking.
with an initial group of APA nominees and asking Beginning
them for
additional recommendations, the "sub-committee" has grown to
include people from a variety of academic disciplines and
professional affiliations. Our unifying concern is in testing
CT. However, we do not necessarily share the same
conceptualization of what CT is nor do we necessarily agree on
how it might best be tested.

It is to resolve precisely these two things that we have

undertaken the Delphi process. In Round 1 (Feb. 11, 1988) and
Round 2 (Mar. 14, 1988) the focus was on establishing group
membership and agreeing on preliminary working assumptions --
such as the assumption that CT could be operationalized to the
extent that valid and reliable assessments of importallt. and
relevant CT skills could be made. Starting with Round 3 we will
focus on what those CT skills are and eventually we will decide
on recommendations regarding testing, based on any Delphi
consensus we achieve. To assist with the conceptual work that
must be done, I developed and circulated two items, a list of
existing CT tests and a CT-Testing Bibliography (Apr. 19, 1988).

If you want copies of any of materials mentioned or if you

wish to have a copy of the mailing/membership list, just drop
a line. me

Last week I presented a workshop on testing CT at Sacramento

State. At that time Perry Weddle agreed to publish the CT-
testing bibliography and the list of CT tests in a fall issue of
CT News. So, please get any corrections, additions, or deletions
to me as soon as you can. MID NOW ON TO ROUND THREE!
Appendix C: Delphi Research Letters,
PAGE 64
1i Ca!cir StVI,
67
 **** ROUND THREE ****
QUESTION: WHAT OPERATIONS ARE CENTRAL TO CT?
Response requested by June 25.

The sole task for Round 3 is for each of us to come up with

that list of operations (or performances, behaviors, processes,
outcomes or skills] which we understand to be at the core of the
concept of Critical Thinking.
Although many of us have published on this question, since
there are potentially sixty in our group and since I have no
assistants, it will be most helpful to me if you would take the
time to distill your views and send a list of what you in
the central CT operations to be, fell free to indicate which are
the more general and which are the sub-operations. Naturally you
are welcome to include justifications for the items on your list.

In thinking about this, please keep in mind that in the two

preliminary Delphi rounds we have narrowed our focus for now to
CT :operations" understood as performances, behaviors, processes,
outcomes or skills which could be tested validly and reliably at
the college freshman/sophomore level. But keep in mind that we
very likely will extend the question downward to K-12 later.

In Round 4, which I will initiate in Sept., you will be

given combined lists and invited comment on the wisdom of
excluding, retaining, or amending the descriptions of specific
items. If the results of Round 3 are clean enough, Round 4 will
also invite you to begin rating items in terms of how more or
less important, crucial, central, integral etc. they are to the
concept of CT.

If you do not intend to respond to Round 3, for whatever

reasons, please drop me a line so that I can keep track of
participation levels.

Matt Lipman suggested that we might have-an excellent chance

of working with the APA and the Assn. for Informal Logic and CT
to secure the use of the Wingspread Conference Center. Please
let me know if you think it would be productive to get together
in that setting. How might a conference be organized to most
effectively use our valuable time? What kinds of issues,
problems, tasks might we address? What kinds of solutions or
desirable results might we achieve by meeting which couldn't be
achieved (at all or as well) using the Delphi?

Since Round 3 asks the "big" question, please take the time
to respond.

Sincerely,

Appendix C: Delphi Research Letters, PAGE 65

68
 SUMMARY OF RESPONSES TO ROUND TWO
Overall response rate: 23 of a possible 51.

Question 1 of Round 2 asked if cur sub-committee was sufficiently well

constituted to move on with our task. As you can infer, the consensus answer
was "Yes." A small number of additional names from psychology and education
were recommended and strong cases for adding ther were made. That is how we
moved up to sixty members -- presuming the new nominees agree to join in. I'll
send you an updated list soon.

Question 2 of Round 2 asked if we agreed with this claim:

"(1) Even if CT cannot be reduced entirely to an equivalent set of

operations [or performances, behaviors, processes, outcomes, or
skills,] (2) it is possible to conceptually analyze CT so as to
describe a set of relevant and important CT operations, such that (3)
using these descriptions, [competent] investigators could, on a
consistent basis, gather sufficient evidence to draw conclusions,
with high degrees of confidence, regarding the relative CT abilities
of a group of people, [everything else being equal, of course]."

There was sufficient consensus on this to move ahead. The majority of

responses (18 were strongly to moderately positive, 2 were negative and 3 did
not respond to this question.)

However, to avoid misunderstanding, let us keep L.. mind that the above
statement should not be interpreted to imply that construct validity can be
determined strictly in an a priori manner (by simply coming up with our list).
Nor should the words "relevant and important" be interpreted to imply that we
can come up with an exhaustive list. Nor should the purpose in (3) be
interpreted to mean that we have set our sights on actually writing a CT test,
or, for that matter come to any agreement about what recommendations regarding
testing we are likely to make.

Most of the positive responses (13 of 18) were vury short, "Yes," or
"Agree," or "Yes, this is an acceptable working hypothesis." Here are two of
the longer positive responses I found interesting:

"I agree with the statement offered. I am compelled to note till' the
tasks described in the statement are going to be quite difficult to complete.
There will not be universal acceptance. There will be criticisms (legitimate
and illegitimate) of the results for a long time to come and many will .an a
course similar to criticisms of attempts to define and measure intelligence."

"(11 YftS, Pven thmtigh pve T 1"/4:7C :c c..,

broadly
satisfying reduction... (2) agreed here, (3) agreed here. Therefore, yes. And
rather than get hung up on this question, I'd prefer to jump right in and see
whether we can do it. If we can, terrific! If we can't, well, then the
doubting T's will have a field day; but I'm prepared to take that risk..."

One of the negative responses was expressed this way: "I'm.sorry, but I
can't help but get hung up on 'operations' -- the term so psychologizes and
jargonizes... the question. Why not go for 'principles'?..."

The strongest negative wa$ registered by a person who said, "I fear this

Appendix C: Delphi Research Letters, PAGE 66

fi
 proposal to define CT operationally may have the effect of ruling out, by
definition, one major position on CT, a position with which many in the CT
field may be in sympathy, at least to some degree." This person was concerned
that defining CT as a list of skills would focused on weak-sense-CT and missed
strong-sense-CT which relates to a person's character -- "being self-critical,
seeking to overcome blind spots, biases, prejudices..., (being] critics of
one's society,... (seeking] what is of value in auother's position..." The
person asked, "Can these character traits be defined operationally?"

In contrast, note this response: "Yes, [I agree with the statement]

particularly if we confine ourselves to abilities and ignore dispositions..."

Even though agreeing that "A subset of the processes that constitute CT
can be assessed using the multiple-choice format that I assume is being
sought," one person expressed serious concern saying, "I fear the creation of
an instrument promising more than it can deliver -- an instrument touted not
for what it is, an assessment device measuring certain important, but
rudimentary, CT activities, but rather as a valid and reliable assessment tool
for CT." He maintained that "As is so commonly true when discussing
assessment, the instrument and its characteristics would then dominate the
social construction of what is being assessed, in this case CT. The initiation
of CT activities, generation of appropriate CT strategies, and defense of a
tentative reasoned judgment are.not susceptible to the type of assessment
legislators are willing to finance or faculty are willing to undertake. The
extensive writing or oral argument required to demonstrate CT, as I understand
it, are not practical inclusions in an assessment instrument."

Four who agreed with the approach mentioned important factors which relate
to construct validity. One mentioned the role of background knowledge in CT,
another the role of divergent assumptions, another the relationship of CT
skills to reading skills, and a fourth spoke to the need to validate any list
of CT operations we might agree. There is much in the research literature
about these problems. Steve Norris, in particular, has been working on
strategies to respond to precisely these kinds of problems.

Although I mentioned all the negatives, the positive responses were far
more numerous than the negatives. A consensus to move ahead exists. But we
must not forget the warnings and concerns of our colleagues. A great deal
depends on what we come up with when we actually sit down to answer the
question for Round 1.1. since both the positives and the negatives were based on
our ideas about what CT is.

Rou. 2 question 3 asked if people would be willing to agree my proposal that

to get started by giving priority to the college freshman/sophomore level.

The responses ran: 14-yes, 3-no, 2-both, 3-abstain, and one that I could
not figure out. Since the question was about priority and was not intended to
exclude working at the K-12 level, which is, after all, what the APA Pre-
college Committee is charged with doing, I believe we have sufficient consensus
to focus initially on the lower division post-secondary level.

Here are some "Yes, give priority to the frosh/soph level" comments:
"Most if not all of the CT we teach is directed at this level"
"I doubt philosophers should take the lead...when it comes to K-12."
"I'm still unpersuaded to reverse my forpgrlv expressed views --

Appendix C: Delphi Research Letters, PAGE 67

70
 intro. undergraduate level first, until we get square there,
where we live."
"I feel rather strongly that we should begin on the post-secondary
level. I agree with those who say that that is where the
strength of our membership lies and that that is where the vast
majority of our members teach."
"Yes. That's where most of the pedagogical action is; that's where
the students we're interested in testing and have relatively
easy access to are to be found. Later, if we succeed at all
here, we can extrapolate to other levels."
"I agree in the light of what was said in Round 1."
"Why not? It's what we know best, and we can always move on to other
levels later."

0n the negative side:

"[Your] recommendation seems arbitrary and not consistent with the
fact that this is a pre-college committee."
"No, based on the name of the committee."
"I would prefer to begin at the K-3 level. My position and
opposition is based upon pedagogical and theoretical concerns
which I assign a higher priority to than to political concerns
related to the officers and membership of the APA.... I am
willing to accept that the sub-committee begin with the college
level but the project will have to be extended downward then..."

Two people expressed the concern that this was a difficult question to
answer until one knew the purposes for testing. [A point well taken.]

Another suggested that our goal should be to make "contributions to the

criteria for a college test...(but that] individuals should be encouraged to
make up their own [assessment instruments] and try them out, obtain
correlations with other tests and with outcomes, and then subsequently compare
notes with one another as to what worked and what didn't."

Thanks to all who responded to Round 2. Your

letters were most interesting and thoughtful.
Don't forget -- Round 3 by June 25!

Appendix C: Delphi Research Letters, PAGE 68

 California State University, Fullerton
Fullerton, California 92634

Department of Philosophy
(714) 773-3611

May 18, 1988

Dear Colleagues,

I hope this letter finds you well, happy and looking forward
to a restful as well productive summer. In my last letter I
mentioned that I would be attending the "First National

Conference on Assessing Thinking" in Baltimore and that I would

be sending along a. report. The conference was very rewarding and
the promised report is attached.

I've also attached the current list of persons invited to

respond to Round 3 in our Delphi process. We will have to freeze
the list at this point. According to what I've read about the
Delphi method, once the central debate is joined, it can be

disruptive to try to add people who have not the benefit of

earlier rounds in the dialogue. With the circulation of the
Round 3 question we have reached that point.

Many thanks to those who have already sent there responses

to Round 3. Don't panic if you haven't yet, though. The target
date is June 25.

Have a good summer. I look forward to hearing from you.

Sincerely,

/' .?

Fete Facione

Appendix C: Delphi Research Letters, PAGE 69

The California State University

 11111111MINOINIMMINV
41111111MENOMMINEMIMMINEMONNI
California State University, Fullerton
Fullerton, California 92634

Department of Philosophy
(714) 773-3611

June 28, 1988

Dear Colleagues,

Thanks for the large number of interesting

responses to the
Round 3 of our critical thinking Delphi
process. Judging from
the length and sophistication of what
many of you sent, it should
take me a fair, amount of time to
organize and synthesize the
material and then to frame fruitful questions for
our next round.
I expect to be working on this for several weeks,
so if for
some reason you haven't had,the opportunity
to respond to Round
3, please know that your ideas are more than welcome.

Have a good summer and thanks again for

so much high quality
participation it's very encouraging.

Sincerely,

Pete Facione

Appendix C: Delphi Research Letters, PAGE 70

The California State Unwerliry

 mumalimmimmennumpimw
California State University, Fullerton
Fullerton, California 92634
111 11.01.
Department of Philosophy
(714) 773-3611

Sept. 1, 1988

Dear Delphi Colleagues,

I'm working through the 25 responses to Round 3, which asked for your

list of sore CT skills. The scope, quality and care evident in your responses

is most impressive. Right now I'm analyzing, synthesizing, and organising the

ideas so that I can play them Lack for your reaction and rectification.

Expect the Round 4 letter in about two weeks.

A Delphi operational rule I've adopted is to drop persons off the Delphi

mailing list if they have made no contact whatsoever after receiving

invitations to respond to at least three Delphi rounds. It's fair to say a

person's sustained silence signals he or she wishes not to be included.

Included here is something Phil Pecorino shared. it lists (a) Critical

Reasoning and Informal Logic Texts, (b) Related Texts, (c) Logic Texts, (d)

Newsletters and Journals related to CT, and (t) CT Centers and Organizations.

Also, you might also want to write to James Bell, Howar:1 Community College,

Columbia, Maryland, 21044 and ask for a copy of his 132 page Guide to CT for

Maryland Social Scientists. It includes a wealth of material on how CT is

seen from different disciplinary backgrounds and it lists a number of ideas

and resources.

Thanks again for so much high quality participation in Round 3 -- it's

very encouraging. You'll be hearing from me soon.

Sincer

Pete Facione
Appendix C: Delphi Research Letters, PAGE 71
The Caiilornia State University

74
 California State University, Fullerton
Fullerton, California 92634
.1111.11M.

Department of Philosophy Sept. 23, 1988

ROUND 4
Dear Delphi Colleagues,

Round 4 seeks verification of a list of CT skills and sub-skills.

Please accept, reject, amend, and comment on the group's responses to
round 3. Remember, the goal in this phase of our project is to arrive
at an accord regarding the skills we understand to be central to CT.

Your responses to round 3 yielded 200+ pages. Some sent previous

publications, some sent lists and commentaries, some wrote new pieces
of clearly publishable quality. You tended to approach the question
of identifying core CT skills and sub-skills four ways: (a) by appeal
to your own experience and understanding, (b) by citation and comment
on what other CT authorities (including others in our Delphi list)
have written or said, (c) by describing the key characteristics of
persons who have internalize CT, and (d) by consideration of what
should reasonably be taught or included in a CT curriculum. In
addition to differences of opinion, there were variations in
disciplinary orientation, vocabulary, and emphasis. There were also
differences in the specificity, depth, and scope of responses. Some
were extremely general, others very specific.

Distilling your opinions, positions, views, ideas, lists,

descriptions, explanations, examples, counter-examples, caveats, .

credos, and course outlines was one of the most intellectually

interesting and stimulating experiences I've ever had the pleasure of
attempting. Although I've been - aching and researching CT for two
decades, I noticed that my own views on the range and character of CT
expanded greatly as a result what you contributed in round 3.

In naming and describing CT skills for Round 4, I intend to rely

on standard English usage and to avoid technical or discipline-
specific vocabulary. Your responses emphasized generic skills.
Distilling your responses, I name and describe six generic CT skills
and give two or three sub-skills under each. Clearly additions or
deletions might be needed. To avoid prejudicing your responses at
this crucial time, I do not indicate the numbers of persons who may
have agreed on any given point. Areas of agreement and controversy
will come out in round 4. In this round you are invited to make a
number of kinds of responses to a variety of questions. However,
because organizing the material and framing the'issues was such a
delicate and complex task, I urge you to read the whole package and
get an overview of the terrain prior to starting to reply.

Thanks again for the high quality responses to Round 3. To

insure we are talking about the same things when we refer to CT
skills, we'll need maximal participation in round 4. If you could
consider this material and reply in 15-20 working days that would be
wonderful. If you need more time, or want to discuss any aspect of
this project, call me at CSUF (714) 773-3742, [office] or 372-3611
[dept.], or 993-1356 [home].
5
Appendix C: Delphi Research Letters, PAGE 72
APA Delphi Project * RESEARCH IN PROGRESS -- NOT Ell PUBLICATION * PAGE 2

There are five parts to Round 4:

(1) The listing of CT skills and sub-skills.
(2) A model diagramming the flow of CT skill input and output.
(3) A list of things some of you said CT is not,
(4) A description of what is meant by "skill".
(5) A list of caveats and comments you shared.

Round 4: Part 14 CT skills

Comment: We've all seen examples, like the duck-rabbitl'of how comprehension
combines imposing an order on reality as well as discovering an order inherent
in reality. Applying that lesson to the problem at hand, there are many ways
CT skills and sub-skills could be organized. Even among those of you who
essentially agree, the variations in your responses to round 3 illustrate
this. After considering a large number of ways of organizing things, I've
decided on the configuration you will find below. True, the skills and sub -
skills could have been named using other labels or grouped in other ways. To
this add that some sub-skills may come into play in more than onegeneral
area, some can operate on the products of others, some presuppose others, some
are almost always employed while others may not come into play except under
special circumstances, and you have an even more conceptually complex
situation. Now add that some of us might.exclude one of the more generic
groupings, to say nothing of how we might want to add, suhtract, amend or
rearrange the sub-skills, and the complexity of our task takes on greater
magnitude. There were other ways to "see" the Round 3 data. So, beside
asking yourself if each skill and sub-skill belongs in the list, and if
anything central is missing, another question to ask is, like the duck-rabbit,
can you see CT this way?

Instructions: BeloN you will find a.list naning and describing six core CT
skill: and sub-skills. Read and consider the entire list. Then,
(1) Considering. each skill and mark each "Yes" if you verify
it to be a core CT skill or sub-skill and "No" if you mould argue to exclude
it. State your, reason for exclaim any_ marked "No".
(2) Make needed substantive amendments to the descriptions of. La sub-
:kill, and explain why. the change is needed. Add au missing,sub-sAill, name
and describe it, locate it within one of the sig CT skills. Considering the
sub-skills within each of the sig CT skills, indicate which sub-skills, if
any" should moved to some other skill.
(3) Amend the list of sig CT :kills, If you delete a skill, indicate
what to, do with its sub-skills. If you add a minim CT skill, name and
describe it name and describe its sub-skills, explain how it is central to
q.t. how it differs from any_ of the six skills on the current list,, and whv it
merits being listed at the level of a skill rather than a sub-skill.
(4) Consider the. NOTE regarding the role and composition of a human's
knowled_IL ha e, verif2L amend comment on any_ aspect of this.
(5) Make ap r needed editorial changes._
(6) Comment, if y.E1 wish on the entire list of skills, its organization,
utility, what have you.

I urge ,.ou to read the entire list of

skills and sub-skills before beginning to
respond to any specific item. Thanks!

Appendix C: Delphi Research Letters, PAGE 73

APA Delphi Project * RESEARCH IN PROGRESS -- ER PUBLICATION * PAGE 3

PROPOSED: Core Critical Thinking Skills

TAXONOMY
1. Interpreting -- Observing, Decoding, Clarifying.
2. Inferring CLu Conjecturing, Drawing Conclusions.
Z. Analyzing -- Locating A.-guments, Parsing Arguments.
4. Evaluating -- Verifying Claims, Assessing Logical Strength.
b. Expressing -- Stating Results, Describing Procedures.
U. Monitoring -- Regulating, Reviewing, Correcting.

DESCRIPTIONS OF SKILLS AND SUBSKILLS

YES/NO

1. INTERPRETING: To comprehend the significance of a wide variety of

experiences, situations, judgments, beliefs, rules, procedures and
criteria.
1.1 OBSERVING: To detect, attend and correctly perceive
experiential input with particular focus on input that conveys or
is intended to convey data, information, or inferential
relationships.
1.2 DECODING: To detect, attend to and correctly perceive
the informational content, rules, procedures, criteria, and
inferential relationships expressed in various convention-based
communication systems, such as language, social behaviors,
drawings, numbers, signs and symbols.
1.3 CLARIFYING: To make explicit, through stipulation or
description, the contextual, conventional and/or intended
meanings of words, ideas, concepts, statements, behaviors,
drawings, numbers, signs or symbols; to remove confusing.
vagueness and ambiguity; to facilitate communication.

2: INFERRING: Ti' secure eleaents needed to sake inferences and to

deteraine the inferential relationship: between or flcuiing frog
statements, descriptions or representations.
2.1 QUERYING: At any .point in the CT process, to recognize
the need for evidence or information of some kind, and to
formulate and execute a strategy for seeking and gathering that
evidence or information.
2.2 CONJECTURING: To formulate alternatives, to develop
hypotheses, to postulate suppositions.
2.3 DRAWING CONCLUSIONS: Given a set of statements,
descriptions or representations, to educe their inferential
relationships and to educe the consequences which they
support, warrant, imply or entail.
,7
Appendix C: Delphi Research Letters, PAGE 74
APA Delphi Project * RESEARCH IN PROGRESS -- EEL pt PUBLICATION, * PAGE 4

*.
NOTE: As many of you argued, all CT skills, but particularly sub-
skills like 2.1, 2.2 and 2.3 presume a knowledge-base. A human's
knowledge-base i9 composed of at least these things:
(a) a world view which includes one's understandings of
* what is real to metaphysics),
* how knowledge is gained and refined (an epistemology),
* what is important or valuable to value theory);
(b).a data base. including one's opinions, beliefs,
experiences, etc. as filtered through the world view;
(c) an inference engine which includes
* general rules for drawing logical inferences to logic),
* sets of procedures and criteria appropriate for making
reasonable judgments within specific areas of human
thought and inquiry (discipline-specific rules.)

Executing sub-skills 2.1, 2.2, and 2.3 as well as 4.1 and

4.2more effectively can be achieved by learning how to think
logically, by expanding one's repertoire of sets of procedures and
criteria used in different areas of human thought and inquiry,
and increasing one's base of relevant data.
An implication of this analysis of CT skills and sub-skills
is that they transcend specific disciplines, but executing them
demands background knowledge, some of which is specific to how one
goes about making reasonable judgments in different realms.
Becoming adept at CT .involves learning CT skills and
learning to use those CT skills more effectively in different
contents -- hence the importance of a liberal education to go
along with one's CT ability.
* * *

3. ANALYZING: To identify the inferentlal relationshIps betNeen

statements, descriptions or representations which express experiences,
situations, judgments, beliefs, or opinions.
3.1 LOCATING ARGUMENTS: Given a set of statements,
descriptions or representations, to determine whether it does
express or was intended to express a reason or reasons in support of
some claim, opinion or point of view.
3.2 PARSING ARGUMENTS: Given a the egpression of i reason
or reasons in support of some claim, opinion or point of view, to
identify: (a) the intended conclusion, (b) the premises and
reasons advanced in support of that conclusion, (c) additional
unexpressed elements of that reasoning, such as intermediary
conclusions, unstated assumptions, and (d) for exclusion, any
items contained in body of expressions being parsed which are not
intend to be taken as crucial to the reasoning being expressed.

4. EVALUATING: Ti assess the credibility of statements, descriptions or

representations; and to asses the strength of the expressed inferential

Appendix C: Delphi Resenrch Letters, PAGE 75

APA Delphi Project * RESEARCH IN PROGRESS -- EL FM. PUBLICATION * PAGE 5

relationships between such statements, descriptions or representations.

4.1 VERIFYING CLAIMS: To assess the degree of confidence
to place in a given statement, description or representation.
4.2 ASSESSING LOGICAL STRENGTH: To determine the nature
and quality of expressed inferential relationships; to judge
whether the assumed truth of the premises of a given argument
justify one's accepting as true, or very probably true, the
expressed conclusion of that argument.

5: EXPRESSING: To state, describe or represent to one's self or others

the results of one's CT activities and the way one went about producing
those results.
5.1 STATING RESULTS: To produce accurate statements,
descriptions or representations of the results of one's CT
activities so as to analyze, evaluate, infer from, monitor or
remember those results, or so as to communicate them effectively to
others.
5.2 DESCRIBING PROCEDURES: To produce accurate statements,
descriptions or representations of how one applied and executed
any CT skill or sub-skill so as to evaluate or monitor one's
proficiency, or so as to communicate to others about how one went
perfcrming a given CT skill or sub-skill.

6: MONITORING; To regulate all aspects of one's own CT activities, the

eleaents used in those activities, and the results produced by those
activities, particularly by applying the skills of analyzing, and
evaluating to one's own inferring with a view toward confirming,
validating and/or correcting the results.
6.1 REGULATING: To sequence one's execution of CT skills
and sub-skills.
6.2 REVIEWING: To examine one's own CT activities and verify
both the results produced and the correct application and execution
of each CT sk'll and sub-skill involved.
6.3 CORRECTING: Where errors are found in one's own CT
activities, to ccwrect those errors and remedy their causes.

Appendix C: Delphi Research Letters, PAGE 76

APA Delphi Project * RESEARCH IN PROGRESS -- NOT Ea PUBLICATION * PAGE 6

Round Part :71.. An Input/Output Model of CT Skills

Comment: There's no denying that the model I've come up is heavily

influenced, for good or ill, by my research in computer science and
artificial intelligence. The arrows indicate the direction of the
flow of input and output as it circulates between and is operated upon
by the six CT skills, here conceived of as functions.
In this model c;Elru.sin.g, receives output in the form of-
,
.
inf ormation.from the knowledge base, the results of other CT
functions, and lists of the steps performed during other CT functions.
Cxpreinq then formats that: output for use by the next CT function,
for storage in the knowledge-base, or for transmission out of the CV
cycle.

Within the CT cycle, monl±orknq receives material from expressing

and, as the regulating and self-correcting function, determines vlhere
it should go next. .It can route things into or out of any of the five
other CT skills or the knowledge-base. For example, monitoring can
loop material back through any skill. Thus, it can send the results
of one's own inferr.ing, for review by routing them to eya/qAt.thg before
allowing yxpresing. to transmit them to others. Ur, it can route
information from the knowledge base to 2y411y.Atincl to help it verify a
claim, or to irligrcipq to help it draw a conclusion using criteria
specific to a given discipline.

The four LT skills many of us spend so much time helping students

bocomo proficient aLl namely koivraetIngL Ang1,YZI.O.P3,
0±PrrjPg. receive material routed to them by the monitoring +unction.
They operate on this material. And then they send the results to the
expressing function to be formatted for delivery elsewhere. "Mese
skills also output a record o+ the steps they performed in coming to
those results. This record is crucial i+ the monitoring function is
to work correctly, since it must: check not only what was achieved but
how it was achieved.

The kno.l_pdge t?...se. is a storehouse from which interpreting,

analyzing, evaluating and inferring draw resources. it stores the
output 0+ any CT function, when directed to do so by the monitoring
+unction. The kns:Ilegge bs!e also +liters raw external input and,
thus, influences the interueting, of that we are observing, decoding
or clarifying.

Instructions: Consi_der the model pn the next LIAog%

PPeS it Ta.e. 1:0 vPu "?

J1.4 ate?,cur
u.Seft,A.Y7.
TOSLY P.LLPOP find Pictori_?1 (2)001s tIPARtY1 in
understanding comalgx call yqp qlw
fl!kub.t tp!.4. kf YPA EftPemgle(10 gar). YCN. PrPRPEP

Appendix C: Delphi Research Letters, PAGE 77

 =
AN INPUT/OUTPUT FLOW MODEL OF
=
CRITICAL THINKING SKILLS
1
O
O.

ri

EXPRESSING
K
U)
N
0

L
...... .
z
E
D

E
i
INTERPRETING
\
ANALYZING EVALUATING Ga

fi=

A
\-
S
E
#.

MONITOR I NG

00

f rl

82
Si
''.APA Delphi Project * RESEARCH IN PROGRESS -- NOT EOR PUBLICATION * PAGE

Round 4: Fart What CT is Not

Comment: Many of you distinguished CT skills from other closc:ly

related things. Below is a list of what various persons said was
Dot_ Cl'.Just because something is on this list does not mean a
person does not use Cr in doing the thing nor that a person might.
not do the thing before, after or during CT.

Instructions: ElgAge onnsider this Usti verify fad

coll2mnt Pn

CRITICAL THINKING IS NOT

YES/NO
1. Sensing, (Seeing, Touching, Hearing, etc.)
2. Reading. Listening,

3. Speaking or writing,

4. Motivating, persuading, selling,

5. Interrogating, cross-examining, petitioning,

o. Physically investigating the world around,

7. Trouble-shouting, problem-solving, puzzle solving,

0. Decision-making, selecting, choosing, egercisintj one's will,

9. Planning, defining goals and objectives,

10. Finding or ascribing a meaning to art, life, .or events,

11. Defending an opinion or belief, arguing a case,

12. Managing, administrating, cr governing persons or things,

13. Philosophizing,

14. Conducting research within any particular- discipline,

15. Experiencing, feQling, emoting, or empathizing,

16. Communicating using lanyuaqe.

Appendix C: Delphi Research Letters, PAGE 79

83
 'APA Delphi Project * RESEARCH IN PROGRESS -- En FOR PUBLICATION * PAGE 9

Round Part Concept of a Skill

Comment: It became clear from your responses that it would be useful to have
an understanding about what a skill is. Although there may be very little
disagreement about this, some of you mentioned subtleties others may or may
not accept. Based on your contributions I've written a little narrative.

Instructions: Levi se edit and comp] ete the f ol 1 oiling_ narrati ve:

A skill the ability to do something well. Having a skill includes knowing

what to do, when to do it, and how to do it. That is, being skilled at

something involves knowing a set of procedure:, judving when to apply those

procedures, and being proficient or adept at executing those procedures.

Skills can i aught in a variety of ways inch ding, but not limited to,

coaching, demon:;trating, and training. Part of the teaching involves making

the procedures explicit and showing when and how they are employed. Skills

can be learned through a combination of observation, guided practice, drill,

and self-correctitn. Persons can be judged as being more or less proficient

in a given skill. The first way of assessing is to observe the skill as it is

being performed. A second nay is to compare the outcomes (if any) that result

frog executing a given skirl against so:e set of criteria. A third way is to

query persons and receive their descriptions of the procedures and judgments

they are using as they perform that skill, would use if they were to perform

that skill, or did use when they performed that skill. . n ce we cannot

directly observe the performance of mental skills the way we can physical.

skills, only the second and third ways are available to those who would assess

CT. The second way is its not) superior to the third because

provided tha'

841

Appendix C: Delphi Research Lettere, PAGE 80

.!
APA Delphi Project * RESEARCH IN PROGRESS -- GT ER PUBLICATION * PAGE 10

Round 4: Part fi_s_ Caveats and Cautions

Comment A number of you sent comments. Some were intended as caveats or

cautions, other as encouragement. I appreciate them all. thought some
I

might be good to share; In many cases to save space have paraphrased.

I You
might agree or disagree with what your colleagues have said. Or, reading this
'list might prompt you to pass along a contrary view. Let's find out.

Instructions Consider these comments. Make any remarks you wish, whether in
agreement or disagreement. Use these remarks, assoring boards for your own
reflections.

AGREE/DISAGREE

1. Keep in mind that our goal in defining CT is to do

some assessment. But you don't just start testing people.
Assessment needs focus and purpose.
2. I agree with working at the college level only because
I want to have an idea about what we should be doing with kids
in K-12. I think we can use what we say about CT for college
frosh/soph to guide curriculum development in K-1.. CT. Of
course, we will have to adjust reading levels, background
knowledge expectations, and lots of other things..
3. Skills are not the same as operations. Resist
behaviorism! Behaviors give evidence that a person has a
skill or ability, but a skill is not a set of behaviors.
4. Don't trap yourself into using the jargon or
vocabulary of any one discipline (especially philosophy) when
you describe CT.
5. Although no word will escape criticism, don't use
"deduction" or "induction." Avoid semantic spats.
6. If CT is a set of attitudes as well as a set of skills
that poses no problem for assessment because we can just
develop ways to assess the CT attitudes, too.
7. CT is thinking skills. Saying CT is a set of
attitudes may be a way of describing what people who are good
at CT are like, but it is not a way of describing critical
tt. king itself.
B. Even if we agree on what CT is, we still have to face
the problem that any student might get the right answer on a
CT test but for the wrong reason, or might get a problem wrong
but have done a good Job of CT,
9. When assessing CT we should not duplicate efforts with
areas already well tested by existing instruments, such as
covered by reading or intelligence tests.
10. You don't have to test every ingle CT sub-skill to
decide that P,person is good at CT.
11. I looked at that list of expe,,.5 and you have all the
big names I can think of, but you can't possibly expect those
people to agree. If they did agree, even on what CT is, that
would really be something. Good luck!

FEEL FREE TO DISCUSS THIS MATERIAL WITH COLLEAGUES, IF YOU WISH.

PLEASE DCN'T REPRESENT ANY OF THIS LETTER AS THE OPINION OF OUR DELPHI GROUP.
Appendix C: Delphi Research Letters, PAGE 81
the Enc ** Thanks!
85
 California Stat.: University, Fullerton
Fullerton. California 92634
wissown

Department of Philosophy
(714) 773-3611
Nov. 22, 1988
Dear Delphi Colleagues,

During this Thanksgiving season and I want to express my gratitude to

you for your generous participation in this research during 1988. The many
responses to the very long and dthicult ROUND 4 have been most gratifying.
There won't be any more "ugly-long" rounds like that, I promise.

Let me pass along this quick review of what we accomplished in 1988.

First, working under the auspices of the Ameri"an Philosophical Association
Committee on Pre-College Instruction in Philosophy, we built the Delphi
list of experts. By your recommendations, during Rounds 1 and 2 (Feb. 11 &
Mar. 14) we expanded the original APA list of about ten names to a working
group which numbers around 45 active participants and which, I am proud to
say, includes many of the most important people in CT research today.

During Round 1 and 2 we agreed that whatever CT is, we would be

possible for us to make the concept operational to the extent that
important parts of CT could be assessed validly and reliably. We also
agreed to begin by identifying the core elements of CT expected at the
Frosh/Soph. general education college level. We agreed to use this college
level theoretical-construct of CT to guide what is said about CT the K-12
levels. Round 4 (Sept. 23) sought to verify the concept of CT which
emerged from Round 3 (May 4). A quick look at the results of Round 4 is
most encouraging!

Along the way we shared journal articles, lists of existing CT tests,

CT bibliographies and other items of mutual interest. The Delphi, however
is not a substitute for the fine work being done by journals, newsletters
and the many centers for CT that have emerged in recently.

While I work on analyzing the results of Round 4, I invite you to

consider where do we go from here. Last spring I outlined a four phase
Delphi project. When we achieve consensus on the core list of CT skills
expected at the lower division college level, we will have completed the
two of the four phases. Originally phases 3 and 4 were described this way:

Phase 3: Recommendations
The goal of this phase is to communicate our findings about what
CT is and whether there is an adequate way of characterizing CT
operationally so as to permit its being tested at some
educational level. Depending on our results in Phase 2, we will
recommend either that programs aimed at testing CT be abandoned,
or that they be focused in certain ways. If this is the
direction Phase 3 takes, then we will also try to come to
consensus on recommendations regarding the relative importance of
different kinds of CT sub-skills and possible strategies for
accessing and measuring those sub-skills.

Phase 4: Design and Validation of Model Testing Strategies

Contingent on the results of earlier phases, the goal, if it were
considered achievable in principle, would be to construct and
evaluate different approaches to testing CT at some appropriate
educational leVel or levels. We might find ourselves breaking
in-o.sub-committees to achieve this goal, although all tmork will
Appendix C: Delphi Research Letters, PALS 82
 have to be guided by the agreements reached in earlier phases and
as well as by the special expertise of those who understand the
intricacies of designing, piloting, norming and validating
educational tests at'specific educational levels.

To date I've done little regarding preparing to communicate our

findings. Two CT newsletters interested in publishing something regarding
our results have contacted me. Also, the Pre-College Committee scheduled a
session at the March 1989 APA meetings at UC Berkeley. At that session
I'll be outlining our Delphi process and what we have agreed on by tnat
time. More suggestions are most welcome.

For many reasons I am extremely skeptical about actually developing a

good CT test using a Delphi process. Once we declare consensus regarding
the theoretical construct of CT for the general education (lower division)
college level our choices include at least these three, and maybe more.

(1) We could move on to consider questions like these: Given what we

understand CT to be at the college level, what does CT mean a` different
grade levels in K12? What is the relative importance of the skills or
sub-skills in our college-level CT construct in terms of testing, say
junior high school students? How might one write a question which assesses
a given sub-skill in, say 5th graders?

(2) Having declared consensus on a conceptualization of CT for use at

the college level, we might recommend that test makers at all educational
levels be guided by our conceptualization. But we, ourselves, might decide
to leave the matter of writing specific tests for specific age groups to
others, better qualified than ourselves for developing and validating such
instruments. Those of us interested in specific grade levels could be put
in contact with one another.

(3) When we reach consensus on the CT concept as it applies at the

college level, we might recommend examples of how questions framed
to address these skills and sub-skills in college students or K-12
students. We could share these example questions and evaluate them. Those
which we think a priori might be good to assess certain skills or sub-
skills, could be included in recommendations we make regarding CT
assessment. These questions would not be a CT assessment tool. At best we
might think of them as models of how to conceive of questions that might be
included in a CT assessment.

Note: Even with questions which a priori seem to address the proper
concept of CT and avoid other difficulties (like relying on special
background knowledge or esoteric vocabulary), there is still the problem of
a posteriori verification. Steve Norris has done important work on how to
overcome the "construct-validation" problem, namely determining, for any
given test item, if students get right answers because of good CT skills
and wrong answers because of inferior CT skills.

Give the issues of what do to next some thought. I would welcome

hearing from you on this. Have a joyous holiday season and thanks again
for contributing so generously during phases 1 and 2 of this project.

Sincerely,

*".41-
Appendix C: Delphi Research Letters, PAGE. 83
87
 California State University, Fullerton
Fullerton, California 92634

Department of Philosophy
(714) 773-3611

Feb. 7, 1989
Dear Heroic Delphi Colleagues,

34 people responded to Round 4. A great return, our biggest so far --

both in numbers of persons and in numbers of pages! will tally, analyze,
I

distill and share what people said on each chur% of Round 4. That way you
will have the benefit of an overview of what others in the Delphi are
thinking. But, intending to spare all of us any more horrendously
budeilsome rounds, when it comes to formulating the next set of questions,
I'll keep them as short and focused as possible.

particularly want to thank the many people who sent me detailed,

I

thoughtful, (and even footnoted) responses. Several ran near ten pages
single spaced. Not intending to diminish the value of the brief "YesiNo"
responses as we approach consensus on crucial points, I must acknowledge
and convey my appreciation for the many extra hours of work several of you
are putting in. Also, I've learned a great deal from your sensitive and
sensible comments.

Yes, we are approaching consensus. My first two readings of the input

on Round 4 is that we have a great deal of accord on the list of CT sk.11s,
although there will be scores of adjustments and amendments to be made.
Also, your comments on what a skill is and how a cognitive skill can be
assessed suggest we are close there too. But I'll be summarizing all that
and more for you very soon.

In the next few weeks I'll be sending you a few quick short rounds.
My plan is to focus each brief letter on one discrete aspect of our work.
We should try to complete several mini-rounds this spring. That way 1'11
be able to rough out a first draft of our report to the APA Pre-College
Philosophy Committee during the summer. Be thinking about recommendations.

I've encloseU a list of the people who are participating in our Delphi
research project. The "RI" "R2" "R3" and "k4" symbols indicate the rounds
to which the person has contributed. The "l" symbol means the person
communicated an interest to be involved, but has not responded yet to any
rounds. Currently there are 56 names on the list of people being invited
to respond. A few have never responded in any way. So, when it comes to
makiny our final report, expect there to be around 50 of us in the group.
I

be in touch with a summary of Round 4 and some mini-rounds 5, 6

'.'11
etc. very soon. Thanks again for all your work.

Sincerely

Peter A. Fatione
Appendix C: Delphi Research Letters, PAGE 84

The California State University

%.0
;.,;''
 41111=11.1111111=1.1.111111111111111111M111111011110

California State University, Fullerton

Fullerton, California 92634

Department of Philosophy DELPHI ROUND 5-A

(714) 773-3611 Feb. 28, 1989

Dear Delphi Colleagues,

One of you wrote, "I'm beginning to think that to have done all that Ewe
have done] and not have tried to define CT may turn out to be a mistake."

Also in response to ROUND 4 three or four of you commented like this: "I
have no major quarrel with any parts of your organization, Cbut3 your
emphasis leaves out a major component of CT -- the dispositional
component and the set of values inherent in being a critical thinker.... I
think it is a good working model of CT skills, but an incomplete picture
of being a critical thinker."

In view of the many positive responses to ROUND 4, yet sensitive to the

concerns raised by comments like the above, ROUND 5 begins by building on
our success in articulating a decent first draft list of CT skills. In
this letter we start right in on the question: For purposes of ner ral
education assessment at the college lower division level, what do
experts recommend be included as a core critical thinking skill?
In addition to asking your endorsement of a revised draft of the skills
dimension of CT, this letter also shares some key ROUND 4 results and
some of the many useful comments you sent.

The next letter, ROUND 5-B, works on the two other aspects of CT you
commented on in your ROUND 4 responses -- namely CT's dispositional and
the normative dimensions. Some of this also finds its way into the
revised skills statement -- see 6.1! ROL. J 5-C picks up the remaining
pieces of ROUND 4 and asks your approval of an outline of our report to
the APA Committee on Pre-College Philosophy.

The table at the end of this letter shows that over 85% of us, (23 of
26), could be described as fundamentally in accord with our first listing
of CT skills. The second draft you are now being asked to consider
endorsing was prepared in view of the many helpful comments and
suggestions you sent in. I am very optimistic about the revised
statement of CT skills, first because ROUND 4's draft was approved by
such a solid plurality, and second because your suggestions helped me
substantially strengthen and enrich that statement.

Since we are very close on so many things, your approval or disapproval

of the expressions of our views presented in ROUND 5 should clarify
things enough for me to start putting together our report to the APA
Committee on Pre-College Philosophy. Where we have consensus our report
will say so. Where we diverge, it will say that as well.

I truly appreciate all that you have already contributed, and I realize
you you are all very busy folks. Vet I beg your ,:ontinued indulgence.
Please respond to the three round 5 letters with all reasonable dispatch.
All responses are welcome, no matter how brief or selective.

With sincerest gratitude,

The Cali lo:nia Stela university Appendix C: Delphi Research Letters, PACE 85
0.
89
 Half of this letter shares comments regarding defining and testing CT
made in response to ROUND 4. Before digging into our revised statement
of CT skills, you might jump to page 6 and look through the comments or
examine the tabular results of our earlier work on page 10. learned,
I

for example, that not listed in our original statement was a skill the
majority feels is part of CT -- arguing. And, given what this Delphi
project is all about, how on earth could I have omitted the CT skill
analyzing an idea from the first draft?

ROUND 5-A, PART 1 -- CT SKILLS

INSTRUCTIONS: Consider the following amended description of CT skills along

with the accompanying statements. Starting with the title and preamble, make
any needed changes, deletions, or additions (editorial or substantive). After
working through the descriptions and statements you will be asked specific
questions regarding endorsement. Please respond to those question as well.

Skills Dimensions of Critical Thinking

For purposes of general education assessment at the college, lower

division level, we understand CT to include the cognitive skills of
interpreting, inferring, analyzing, evaluating, expressing and monitoring.
Because of our collective conviction regarding their centrality to CT, we urge
those persons interested.in assessing the skills dimensions of CT focus on
these six abilities. However, since CT can be subclassified in a number of
legitimate ways, our subclassification should not be interpreted as an
educational taxonomy nor as implying or presupposing any psychological,
logical or epistemological order or sequence of skills. While including those
skills we take to be central to CT, we do not claim' that our list is
exhaustive in either breath or detail.

Critical thinking involves actively interpreting one's experiences and

self-consciously making and expressing one's analytical, evaluative and
inferential judgments regarding what to believe or do. As such, critical
thinking is a pervasive and multi-dimensional human phenomenon involving both
dispositions and skills. Without diminishing the vital importance of
cultivating CT dispositions throughout the K-12 and post-secondary educational
process, we have here chosen to focus our attention on listing and describing
CT abilities. As a goal statement of .what a generally educated college level
critical thinker should be able to do, we hope our consensus description of CT
skills will assist in CT assessment and CT curriculum development both at the
college and the K-12 levels.

Among the many ways one might improve one's CT are by reflecting on one's
reasoning processes and learning how to think more analytically, objectively
and logically, by expanding one's repertoire of those more specialized
procedures and criteria used in different areas of human thought and inquiry,
and by increasing one's base of information and experience. An implication of
our analysis of CT skills, however, is that CT skills per se transcend
specific disciplines, yet executing them successfully in certain contexts
demands background Knowledge, some of which may be specific to how one makes
reasonable judgments in that context. Since becoming adept at CT involves
learning to use CT skills effectively in many different contexts we cannot
overemphasize the value of a solid liberal education to supplement the honing
of one's CT skills and the cultivating of one CT dispositions.

Appendix C: Delphi Reeearoh Le6kAns, PAGE 86 90

 1.

1..'4.01)1

Li

Names of :"ore CT Skills ar,d

1. Interpreting - Categorizing, Investigating, Decoding, Clarifying.

2. Analyzing - Analyzing Ideas, Identifying Arguments, Analyzing Arguments.
3. Evaluating - Assessing Claim., Assessing Arguments.
4. Inferring - Querying, Conjecturing, Concluding, Developing Reasons.
5. Expressing - Stating Results, Describing Procedures, Stating Arguments.
6. Monitoring - Self-examination, Self-correction.

Descriptions of Core CT Skills and Sul-skills

1. INTERPRETING: To comprehend the meaning or explain the significance of a

wide variety of experiences, situations, data, events, judgments, conventions,
beliefs, rules, procedures and criteria.
1.1 CATEGORIZING: To formulate categories, distinctions, frameworks or
questions, and to describe experiences, situations, beliefs, events, etc., so
that they take on comprehensible significance or meaning, as for example to
recognize a problem and define its cnaracter without prejudice to inquiry.
1.2 INVESTIGATING: To actively seek, attend to, discriminate and describe
experiential input relevant to a given situation, problem or concern; to
gather input that conveys or is intended to convey data, information, or
inferential relationships, as for example to gather evidence relevant to
solving a problem in the light of how that problem is defined.
1.3 DECODING: To actively detect, attend to and correctly understand,
the informational content, effective purport, directive functions, intentions,
purposes, symbolic significance, values, views, rules, procedures, criteria,
or inferential relationships expressed by others in conventionbased
communication systems, such as in language, social behaviors, :16..ings,
numbers, signs and symbols.
1.4 CLARIFYING: To explain, paraphrase or make explicit, through
stipulation, description, analogy or figurative expression, the contextual,
conventional or intended meanings of words, ideas, concepts, statements,
behaviors, drawings, numbers, signs, symbols, rules, events or ceremonies; to
an extent proportionate with the purposes at hand, to use stipulation,
description, analogy or figurative expression to remove confusing, unintended
vagueness and ambiguity, or to design a reasonable procedure for so doing.

2. ANALYZING: To identify the intended inferential relationships among

statements, questions, concepts, descriptions or other forms of representation
intended to express beliefs, judgments, experiences, information, opinions.
2.1 ANALYZING IDEAS: to identify expressions used in communication and
determine the role they are intended to play in arguing or persuasion, as for
example to identify a phrase intended to trigger a sympathetic emotional
response and induce an audience to agree with an opinion; to identify related
judgments, views, or concepts and to determine the conceptual similarities and
differences between them; to identify issues or problems, determine their
component parts, and identify the conceptual relationships of those parts to
each other and to the whole.
2.2 IDENTIFYING ARGUMENTS: Given a set of statements, descriptions,
questions or representations, to determine whether it does express or was
intended to express a reason or reasons in support of or contesting some
claim, opinion or point of view.
2.3 ANALYZING ARGUMENTS: Given the expression of a reason or reasons
intended to support or contest some claim, opinion or point of view, to

Appendix C: Delphi Research Letters, PA3E 87

21
 identify: (a) the intended main conclusion, (b) the premises and reasons
advanced in support of the main conclusion, (c) further premises and reasons
advanced as backup or support for those premises and reasons intended as
supporting the main conclusion, (d) additional unexpressed elements of that
reasoning, such as intcrmcdizry conclusions, unstatcd assumptions or
presuppositions, (e) the overall structure of the argument or intended chain
of reasoning, and (f) any items contained in body of expressions being
examined which are not intend tc be taken as part of the reasoning being
expressed or its intended background.

3. EVALUATING: To assess the credibility of statements, descriptions,

questions or other representations expressing experiences, situations,
beliefs, judgments, or opinions; and to assess the strength of the expressed
inferential relationships among such statements, descriptions, questions or
other forms of representation.
3.1 ASSESSING CLAIMS: To assess the degree of credibility to ascribe to
a source of information or opinion; to a:.) sr the relevance of questions,
information, principles, rules or procedcr.:, ;,*ections to a given issue or
concern; to assess the truth or the level J idence to place in any given
representation of an experience, situation, judgment, belief or opinion.
3.2 ASSESSING ARGUMENTS: To determine the nature and quality of
expressed inferential relationships; to judge whether the assumed truth of the
premises of a given argument justify one's accepting as true, or very probably
true, the expressed conclusion of that argument; to anticipate and raise
questions and objections, and then to assess whether these point to
significant weakness in the argument being evaluated; to determine whether an
argument relies on false or doubtful assumptions or presuppositions to judge
how crucially these affect its strength; to judge between reasonable and
fallacious inferences; to judge the probative strength of an argument's
premises and assumptions with a view toward determining the acceptability of
the argument; to determine and judge the probative strength an argument's
intended and unintended implications with a view toward judging the
acceptability of the argument; to judge the extent to which additional
information would strengthen or weaken an argument.

4: INFERRING: To identify and secure elements needed to make inferences and to

determine the inferential relationships between or flowing from statements,
descriptions, questions, or other forms of representation on the basis of
which inferences can be drawn.
4.1 QUERYING: to recognize the need for evidence or information of some
kind, in particular to recognize which statements, including those offered as
premises, need justification, and to formulate and execute a reasonable
strategy for seeking and gathering that evidence or information.
4.2 CONJECTURING: Given a problem, question or point of view on an
issue, to formulate multiple alternatives, develop hypotheses, or postulate
suppositions, and to design reasonable strategies for determining their
plausibility, viability or relative merit; to objectively draw out the
presuppositions and the consequences of decisions, positions, beliefs or views
with which one might agree or disagree.
4.3 CONCLUDING: Given a set of statements, descriptions, questions or
other forms of representation, to educe with the proper level of logical
strength, their inferential relationships, both deductive and inductive, to
educe the consequences or the presuppositions which they support, warrant,
imply or entail; to successfully employ iar'ous sub-species of inductive or

Appendix C: Delphi Research Letters, PAGE 88

c:
92
deductive reasoning, as for example to reason analogically, arithmetically,
dialectically, scientifically, etc.
4.4 DEVELOPING REASONS: Given a question to be answered or a position on
an issue, use appropriate inductive or deductive modes of inference to
articuleze reasons for answering the question one way as opposed to another,
or for supporting or for opposing the position.

5: EXPRESSING: To state, describe or represent to one':, self or to others the

results of one's reasoning and the way one went about producing those results.
5.1 STATING RESULTS: To produce accurate statements, descriptions or
representations of the results of one's reasoning activities so as to analyze,
evaluate, infer from, or monitor those results, or so as to accurately and
effectively recall or represent those results to one's self or to others.
5.2 DESCRIBING PROCEDURES: To represent as clearly as possible how one
came to one's interpretations, analyses, evaluatiWor inferences, so that one
might accurately record, evaluate, describe or justify those processes to
one's self or to others, or so as to remedy perceived deficiencies in the
general way one executes those processes.
5.3 STATING ARGUMENTS: To present arguments which communicate one's
grounds for accepting some claim, their logical force in supporting that
claim, and, as necessary, meeting objections to the premises one relied on or
the reasoning one employed.

6: MONITORING: To self-consciously regulate one's cognitive activities, the

elements used in those activities, and the results produced by those
activities, particularly by applying analyzing and evaluating to one's(own
inferring with a view toward confirming, validating, correcting or questioning
either one's reasoning or one's results.
6.1 SELF-EXAMINATION : To reflect carefully on one's own reasoning and
verify both the results produced and the correct application and execution of
the cognitive skills involved; to make a thoughtful meta-cognitive self-
assessment; to reflect on the extent to which one's thinking is influenced by
deficiencies in one's knowledge, or by stereotypes, prejudices, emotions or
any other factors which constrain one's objectivity or rationality; to reflect
on one's motivations, values, attitudes and interests with a view toward
determining that one has endeavored to be unbiased, fair-minded, thorough,
objective, respectful of the truth, reasonable, and rational in -owing to
one's interpretations, analyses, evaluations, inferences, or expressions.
6.2 SELF-CORRECTION: Where self-examination reveals errors or
deficiencies, to design reasonable procedures to remedy or correct, if
possible, those mistakes and their causes.

*******,
ROUND 5 -A, PART 2 -- FNDORSEMENTS,

1) Do you endorse the above statement as useful for purposes of

asst sing tht; skills dimension of CT at the lower division college
level?

2) Would you be willing to lave your name listed in association

with the above description of CT skills as a contributing member
of the Delphi research project which generated it?
*****

Appendix C: Delphi Research Letters, PACE 89

 ROUND 5-A CART 3 -- Reanns.
The responses to ROUND 4 included some telling observations, thoughtful
objections and well focused criticisms. Sharing these with others in
the Delphi group is essential to the Delphi process. I've included as
many as is feasible and as can be understood outside the context of
whatever ROUND 4 item may have prompted them. Rather than use of your
time reading positive comments, of which there were many, I've stuck
chiefly to the critical ones.

INSTRUCTIONS: Read and consider what our colleagues are telling us. In
addition to the adjustments already incorporated into the above draft
statements regarding CT and CT assessment, what other responses and
improvements should we make? In the light of these comments and other
concerns that come to mind, what specific recommendations should we
include in our report to the APA Committee on Pre-College Philosophy?

SELECTED RESPOPSES OR CT TESTS AND TESTING CT

* "A test cannot be considered in the abstract, without working

out its intended use and intended users, the specific population to be
tested, and the discriminations the test would be required to make...
The domain the test covers is governed by this context, and concepts
that constitute sub: :lassifications of the domain are arranged in
different ways from the ways in which they might be ac ranged for a test
with the same name but a different purpose... CT can be subclassified
in a number of legitimate ways, with any of the same elements
recurring in different places in different classification schemes..."

* "There's hardly anything we now need more than improved CT

tests. I've used ... in pre-post testing for several years. It's the
best I know of. (But...)"

* "...to makesense of CT we must sake sense of the critical

thinker. The cognitive must be discussed in relation to the affective,
and both must be discussed in relation to their roles in the real
world. Curriculum and assessment must be put into some broad context.
There is no one right definition of CT, and in testing we should
never confuse testing for micro-skills with testing for CT itself.
Most CT tests are micro-skill tests only. They are valuable, but only
in a qualified way. The Delphi project seems well on its way toward
confusing the part with the whole."

* "The 4 list of skills) seems fine to me.

(Round 'there are
definite limits to armchair analysis, and until soPeone actually starts
trying to easure these things, it is difficult to knoN just hou to
revise the list of CT skills and sub-skills."

* "Please interpret my responses rautiously. I have not spent

much time trying to define CT. My business is trying to tdo)
assessment once you guys arrive at clear and useful definitions!"

e "We oust focus on the purpose and target audience of an; given

Appendix C: Delphi Reeearch Lettere, PAGE 90

94'
CT test. Only that way can we fine tune our definition of CT, or
should I say our "theoretical construct" of CT."

* than limit our conception of CT to achieve a certain

"Rather
kind of test, why not simply make more modest claims about the test?
Why not say that you are testing aspects of CT? This swtil, more
justified. Otherwise the test becomes the tail that wags the dog... Be
honest. about what We are and are not testing by multiple choice tests.
Let's not reduce rich and complex realities like CT to that which can
be directly tested in the multiple-choice format."

* [Some of your descriptions] are difficult to test without open-

ended items... (For example] it's difficult to test formulation of
strategies, MC exams focus more on recognition of best 'trategy.

* categorization
"The system...gives little guidance on what
precisely to teach and test. For instance, under "evaluating" is "to
assess the credibility of statements." But what should be taught when
doing this, and what should be tested when trying to find out what
skills students have? If teachers are supposed to act based on what we
produce, I believe we need to include criteria for assessing
credibility. Again I refer you to..."

* "CT is deeply connected in those individuals who are successful

critical thinkers to a pervasive self consciousness about one's own
thinking and reasoning processes. Such self consciousness should be
deliberately cultivated in our students and should therefore be
included somewhere in the overall description -- it is an intrinsic
part of the (CT] process. It monit' s the selection, application and
interlinkage of the various relevant processes."

SELECTED RESPONSES ON CT AHO DEFINING CT

it "Untilwe have a theory of reasoning (a combined normative

theory of informal logic along with a descriptive theory of cognitive
processes) we will not be able to spell out CT skills non-arbitrary
with completeness and precision.... We are producing...a framework,
(that is] a list of concepts used for understanding a domain. If that
is so, it will play hell out of our attempts to assess CT skills. I

doubt that we will be able to manage construct validity for any

conventionAl MC test with our (list of CT skills). Yet, what we are
coming up with is extremely valuable if we focus on performance
assessment. Our framework gives us a passably good set of criteria by
which one would judge good performance on CT tasks. The criteria are
developed by experts -- us -- and we are currently judging whether we
accept them or reject them, another step in the process of developing a
gaol performance assessment. The next step would be to distribute
typical essays, (good, bad and ugly ones) and aim at some consensus in
telling the good ones from the bad ones."

"We should resist the assertion that CT is domain dependent."

* beginning to think that to have done all that (we have

"I'm
done] and not have tried to define CT may turn out to be a mistake...
What makes (the list of] important, indeed basic, intellectual skills

Appendix C: Delphi Research Letters, PAGE 91

95
(in Round 4 3 a list 0 critical thinking skills? It could just as
well be a list of rational thinking skills, or logical thinking skills
or higher order cognitive skills. fail to see how this list
captures the force of the word "critical". (Etymologically) the
meaning of "critical" is judging, evaluating, estimating the worth of
something... A critical thinker is someone whu renders an opinion on an
intellectual product.. by assessing the strengths and weaknesses of
that product... Doing se requires the capacity to elicit and apply
standards, principles and criteria. None of the (Round 4) list is
really this skill. If we asked for a list of problem solving skiPs,
would we get the same list (as in Round 43? If then either there
is no conceptual difference between the two, which I think is wrorg, or
else the list fails to capture what is distinctive about CT skills. CT
also connotes "crucial". (Here) the (Round 4) list eres better,
because...these six skills are crucial essential for
intellectual survival. The problem is that the list is so bread and
wide-ranging that it is not clear what intellectual skills have been
excluded."

* "If one identifies CT as that which makes a critical person to

be what he is, then (your) narrow concept of CT is inadequate. CT
is... an answer to the general problem of conformity, prejudice,
narrow-mindedness, and irrationality in the world. CT is what one does
to achieve autonomy and independence of thought, to lesson one's
prejudices, to broaden one's perspective, and to become more rational."

* "The overall worry is that the categories are such too

main
broad. When we get down to testing it will not be for something like
"assessing logical strength" or "clarifying" but very specific skills
such as "recognizing whether something is a necessary or sufficient
condition; recognizing the difference between if p, then q's and if q
the p's etc. The lines of demarcation are very unclear e.g. between
psychological and logical (epistemological) criteria conerning say
observation."

"I have no major quarrel with any parts of

* your organization,
(but) your emphasis leaves out a major component of CT -- the
dispositional component and the Iet of values inherent in being a
critical thinker... I think it is a good working model of CT skills,
but an incomplete picture of being a critical thinker."

it"There is no attention... to the dispositions... characteristic

of CT. These... are as essential to CT as are the knowledge and
information used in the processes."

* dispositions are at
"The least as irportant as the skills. I

suggest their addition."

"CT works by recognizing and criticizing sources of information,

*

by drawing implications from given materials, identifying assumptions,

noticing relationships of consistency, inconsistency, implications and
contradiction, inferring interesting consequences, recognizing,
analyzing and evaluating arguments and constructing them as well. Of
course, there's a lot more to it.'

Appendix C: Delphi Research Letters, PAGE 92

 * "Arguing is not listed as a separate skill!"

"No mention is made of understanding another's

* purpose... this
would fall under interpretiny, I suppose."

* disagree with the

"I list of CT skills as described.
Interpreting is obviously a cognitive operation. But it is not a CT
operation. Expressing is essentially low level communication, not
generative in the sense of CT. Monitoring is meta-cognitive. The
problem is that this description is so broad "critical" thinking gets
lost in all the other kinds of thinking. This blurs the nature of CT
beyond recognition... CT is "judging the worth, accuracy or
significance of something."

* "There are several items I mess overall. They may be subsumed in

some of the processes you have listed, but very few people will be
conscious of them unless they are brought out explicitly. One is the
capacity for arithmetical reasoning with ratio and di.ision -- it
begins with word problems in 5th and 6th grade arithmetic and carries
up to exactly similar reasoning with concepts such as density,
composition, contraction, rates of change, in more sophisticated
settings. It includes the ration reasoning that goes with scaling
areas, volumes rates, etc. This capacity is profoundly important in
any CT that involves numerical information (whether it be scientific,
economic, sociological, psychological... I'm talking about arithmetic
and not sathesatics at the level of calculus or even algebra. EA
second capacity to include is] "correlational reasoning." Finally I

miss explicit inclusion of the process of translating symbols (e.g.

graphs, numerical data, histograms) into words or words into
corresponding symbols. Such translation is essential to much CT."

* "[The should
concept also
of CTinclude] discriminating
explicitly between the factual or experiential input and the inferences
drawn, identifying gaps in available information and identifying
irrelevant or superfluous information, and it should include the
capacity to consider some situation in the abstract and, by applying
relevant governing principles or constraints, arrive at reasonable and
plausible conclusions about the outcomes that would result from the
imposition of some change -- hypothetico-deductive reasoning."

* something about oppositional reasoning -- taking the

"Include
part of whom one disagrees.
one with Also include hypothetical
reasoning -- reasoning from suppositions and hypotheses.

"Taxoilomy" is not a good word far what we've got; it's more
* like
a list. kiLving a taxonomy in biology and education is to have a
hierarchical set of categories such that each subsequent step in the
hierarchy subsumes the steps below it. We don't have that here.

a "[Conjecturing] is creative thinking not CT.

"[To have included seeking and gathering evidence] is ambiguous.

*

We are not scientists.

* "[Regarding expressing,] I don't see any strong rationale for

Appendix C: Delphi Research Letters, PAGE 93

 extending the focus to communicating the results of CT.

* "None of [your sil:J listed items are skills -- let alone sub-
skills. They are general categories into which many (at least dozens
of) distinguishable skills may be lumped. This important.

TABLE OF RESPONSE ON THE ROUND 4 CT SKILLS LIST

Agree Disagree Unsure
I. INTERPRETING 24 n
4.

Observing 20 3 3
Decoding 23
,.. 3
Clarifying 26 -
2. INFERRING 23 3
Querying 24 1 1

Conjecturing 23
4.1 2 n
1

Drawing Conclusions 24 - n
4
3. Analyzing 23 - 3
Locating Arguments 23 1 2
Parsing Arguments 23 - 3
4. Evaluating 24 - n
,
Verifying Claims 25 - 1

Assessing Logical Strength 26 GIP OD

5. Expressing 21 3 3
Stating Results 21 3 3
Describing Procedures 71 3 3
6. Monitoring - n
?I ,
Regulating 23 - 3
Reviewing 23 - 3
n
Correcting 24 - 4

RESPONSES TO 'MNAT CT IS HOT"

To help delineate CT through comparisons and contrasts, I offered a list of
activities that bore some family resemblances to CT. Each of them depends on
CT. But whether any of was CT per se was .the issue. Here's what you said:

Isn't Is Partly.
CT CT CT
1. Sensing, (Seeing, Touching, Hearing, etc.) 16 2 5
2. Reading, Listening, 13 4 6
3. Speaking or writing, 14 3 6
4. Motivating, persuading, selling, 14 4 5
5. Interrogating, cross-examining, petitioning, 12 6 6
6. Physically investigating the world around, 13 4 6
7. Trouble-shooting, problem-solving, puzzle solving, 13 4 6
8. Decision-making, selecting, choosing, exercising will, 14 3 6
9. Planning, defining goals and objectives, 10 7 6
10. Finding or ascribing a Peaning to art, life, or events, 13 5 6
11. Defending an opinion or belief, arguing a case, 7 7 8
12. Managing, administrating, governing persons or things, 16 2 5
13. Philosophizing, 12 3 8
14. Conducting research within any particular discipline, 12 3 8
15. Experiencing, feeling, emoting, or empathizing, 15 3 5
16. Communicating using language. 15 2 6

Appendix C: Delphi Research Letters, PAGE 94

9.8
 California State University, Fullerton
Fullerton, California 92634

Department of Philosophy
(7i4) 773-3611

DELPHI ROUND 5B
Mar. 6, 1989
Dear Delphi Colleagues,
This letter follows the outline in the Round 5-A letter --
questions first, background information second. The focus here is on
fleshing out our .conceptualization of CT beyond the revised list of CT
abilities presented in Round 5-A. Specifically this letter responds to
your comments regarding those dispositional and normative dimensions
many include when describing CT. The implications for K-12 and college
level assessment, curriculum development and pedagogy of including
either of these dimensions are crucial for our effort. You're asked for
your ideas about this, too.
To get things started, on the next page you'll find a draft
statement regarding the dispositional dimension of CT. Following that
is a draft sta;ement regarding the normative dimension. These two draft
statements are based on your comments regarding needing a fuller
conceptualization of CT and an analysis of commonly referenced concepts
of CT which appear in the literature. Have at those draft statements.
Amend, edit, comment, accept, reject -- whatever you think will help us
be able to present an intellectually credible and educationally useful
conceptualization of CT.
Coming soon: Round 5-C focuses on our report to the APA Pre-College
Philosophy Committee. It will include a proposed outline of that
report, showing how the various pieces of the assessment puzzle we've
worked on for over a year now will be incorporated. It also picks up
the two pieces from Round 4 which haven't been addressed yet, namely the
input-output model Cwhich we rather roundly rejected], and the statement
of what a cognitive skill is and how cognitive skills might be assesaed.
I'll be speaking on March 24 at 1:00 p.m. at the Pacific Division
meetings of the APA in Oakland CA, sharing a little of what we've been
doing and the direction things seem to be taking. If you happen to be
in the neighborhood, please stop by so we can visit.
Knowing you are very busy, I beg mercy and ask you to reply as soon
as is reasonably possible. A quickie note with general comments is
helpful, if you won't have time to go through things in careful detail.
Thanks for your continuing support and involvement.
Yours sincerely,
irk
Appendix C: Delphi Research Letters, PAGE 95

..Tha ralitriprrtis Rims Ilnivomilv Pete Facione 95

 '

Part 1: The Dispositional Dimension of CT

INSTRUCTIONS Reflect on the following statement. Edit, amend, or
revise as you see Lit. Some background information is presented in the
second half of this letter. Please respond to the following questions:
1) Should our final report include a statement on the dispositional
dimension of CT? Why, why not?
2) What implications or recommendations for K-12 and college lower
division level assessment, curr.culum development and pedagogy follow
from including a dispositional element in our conceptualization of CT?
CT -- The Dispositional Dimension
In addition to it cognitive skills diaension, CT also involves certain
cognitive dispositions, personal traits or intellectual virtues which are
crucial to its broad and successful use. Just as with the cognitive
abilities dimension of CT, when conceiving of CT assessment or an

instructional program in CT, it is important to consider ways of developing

materials, teaching strategies and assessment tools which focus on CT
dispositions. The cultivation of CT cognitive dispositions is particularly
important in insuring the use of CT abilities outside the narrow CT
instructional setting. Persons who have developed the intellectual virtues
and personal traits listed below are such sore likely to apply their CT
skills than are those who know the skills but lack the intellectual
discipline to use thee.
The critical thinker is one who possesses and cultivates the cognitive
dispositions, personal traits or intellectual virtues of
* seeking a clear statement of issues, questions or problems,
* curiosity in exploring issues and seeking information,
* eagerness in seeking and in employing CT abilities,
* openness to seek and to consider divergent views or alternatives,
* trust in the processes of reasoned inquiry,
* perseverance, diligence and discipline in keeping well - informed,
* honesty and humility in appraising one's own reasoning and views,
* willingness to seek, use and aention credible sources,
* prudence in suspending or making judgments and in taking or changing
one's position,
* wisdom and persistence in the use of appropriate, defensible and
relevant criteria,
* ee ort to address issues reasonably and to remain relevant to the
basic concern or problem.
* precision, to the extent the subject permits,
* orderliness in the treatment of complex issues and processes.

Appendix C: Delphi Research Letters, PAGE 96

109
.
 Part 2: The Normative Dimension of CT
INSTRUCTIONS Reflect on the following statement. Edit, amend, or
revise as you see fit. Some background information is presented in the
second halt of this letter. Please respond to the following questions:
1) Should our final report include a statement on the normative
dimension of CT? Why, why not?
2) What implications or recommendations for K-12 and college lower
division level assessment, curriculum development and pedagogy follow
from including a normative element in our conceptualization of CT?

CT -- The Normative Dimension

In addition to its cognitive skills and dispositional dimensions, :T
also involves certain normative features which govern its proper use.
Understanding that in making this statement we are going beyond a purely
descriptive analysis, we judge it vital to include this normative component
in our conceptualization of CT. In a free and rational society education
lust be more than skills training and more than the inculcation of a set of
personal dispositions. Education must include the nurturing of those civic
and personal values which insure that the heritage of intellectual fair-
mindedness and political freedom will be passed to future, genera ions.
Properly used, CT contributes to the fair-minded analysis anc resolution
of issues, rather than the blind, sop.istic, or irrational defense of views
known to be intellectually defective or biased.
Properly used, CT unites those who would reason together objectively in
a reflective and intellectually sound process, even if their final judgments
and analyses should disagree.
Properly used, CT promotes rational autonomy, intellectual freedom, and
the objective investigation of any issue or concern whatsoever.
Properly used, CT treats all affected persons with sensitivity and with
equal respect and dignity -- it is never exclusively self-interested,
abusive, coercive, or without integrity and good faith.
Thus, in a free, open and rational society, when conceiving of CT
assessment or an instructional program in CT, one ought to consider ways of
developing materials, teaching strategies and assessment tools which insure
the normative dimension of CT will be developed in students, in addition to
CT skills and dispositions.

Appendix C: Delphi Research Letters, PAGE 97

 Part 3 --BACKGROUND

For your consideration, here are the views of six widely cited persons
regarding what CT involves.

1) CY is 040
. "Active persistent and careful consideration of a belief or
supposed form of knowledge in light of the grounds that support it and the
further conclusions to which it tends" John Dewey. 1909

2) CT "is reasonable reflective thinking that is focused on deciding what to

believe or do." In addition to 12 CT abilities, CT also includes 14
dispositions. Namely; "to seek a clear statement of the thesis or question,
to seek reasons, to try to be well informed, to use credible sources and
mention them, to take into account the total situation, try to remain relevant
to the main point, to keep in mind the original or basic concern, to look for
alternatives, to be open-minded, to take a position when the evidence and
reasons are sufficient to do so, to seek as much precision as the subject
permits, to deal in an orderly manner with the parts of a complex whole, to
use one's CT abilities, to be sensitive to feelings, level of knowledge, and
degree of sophistication of others" Robert Ennis, 1987.

3) "CT is an attitude of being disposed to consider in a thoughtful way the

problems and subjects that come within the range of one's experience;
knowledge of the methods of logical inquiry and reasoning; and some skill in
applying those methods..." E.G. "recognize problems, find ways to meet them,
marshal pertinent data, recognize assumptions use language well appraise
evidence evaluate statements, see logical relationships, draw warranted
inferences, test one's own conclusion, reconstruct one's beliefs based on
experience, render accurate judgments." Edward Glaser, 1941.

4) "CT can be characterized as skeptical, radical, disillusioned, problem-

seeking, holistic, judgment-oriented, non-algorithmic, constructive,
comprehensive, empathic, meta-cognitive, higher-order, reflective reason-
governed, logical, self-correcting, context-sensitive, criterion-referenced,
crisis-oriented, normative, inferential, pragmatic, purposive, relational, and
open- minded." Matthew Lipman, 1988.

5) ."A given student, S, is a critical thinker in some area requiring mental

effort, X, if S has the disposition and skill to solve problems in X using
some subset of the available evidence E pertinent to activities in area X.
"CT" has an identifiable meaning but the criteria for its correct applica,tion
vary from field to field." John McPeck, 1981.

6) CT is disciplined, self-directed thinking which exemplifies the perfections

of thinking appropriate to a particular mode or domain of thinking. It comes
in two forms. If thinking is disciplined to serve the interests of a
particular individual or group, to the exclusion of other relevant persons and
groups,... it is sophistic or weak sense critical thinking. If the thinking
is disciplined to take into account the interests of diverse persons or
groups, it is fair-minded or strong sense critical thinking." Richard Paul,
1988.

Appendix C: Delphi Reas4roh Letters, PAGE 98

.1.02
 Here are your comments regarding statrment 06 in the caveats and cautions
section of Round 4, "If CT is a set of attitudes as well as a set of
skills, that poses no problem for assessient because Ne can just develop
Nays to assess the CT attitudes too."

-- "Difficult to do!"
-- "[This] seems too facile. To be sure, rhetoricians and social
scientists study attitudes and so have developed ways to ascertain or measure
certain attitudes. But does this mean that all attitudes can be so tested? How
can we talk about whether certain attitudes can be tested until we have
enumerated those attitudes and so know specifically what we are trying to test?
Also, can't some aspects of proficiency in CT, or beinn a critical thinker,
involve both skill and attitudes combined? If CT involved recognizing one's own
prejudices and stereotypes, one must have an attitude of willingness to confront
one's stereotypes and the skill to delineate and recognize them. Attitude and
skill would go hand in hand. We don't just have a set of skills and as set of
attitudes. This is too atomistic a description."
- - "Agree" (with (6) as stated, no comment.] 6 people.
- - "Agree, he should teach skills and dispositions, the attitudes are
minor."
-- "Agree, but attitudes are harder to test."
-- "Agree. CT is more a set of attitudes than a set of skills... It is
easier to change attitudes in a single course than to improve skills
substantially. Changing attitudes and fostering self-monittiring should lead to
lifelong improvement in a student's CT performance. So it's actually very
important to say simething about assessing CT attitudes..."
-- "Disagree. The attitude/skill distinction isn't that sharp."
"How?"

-- "Good luck!"
-- "Unlikely."
-- "There is a very great difference between skills and attitudes (although
there are elisions at the boundaries of both concepts). Richard Paul has taught
us most effectively (that] there is, or should be, a causal relation between
skill and disposition such that the disposition might/should elicit, motivate,
and work as side-constraints for the skills."

Appendix C: Delphi Research Letters, PAGE 99

1O3
 -tr

Here.are your comments regarding statement $7 in the caveats and cautions

section of Round 4, "CT is thinking skills. Saying CT is a set of
attitudes pay be a Nay of describing that people rho are good at CT are
like, but it is not a Nay of describing CT itself.

- - "Good point!"

- - "(This] simply expresses a position dogmatically, and in the light of

discussions by Richard Paul, Harvey Siegel, and others this is just not
acceptable. To dismiss thes'e positions without argument is just not
philosophical!... [This] simply rules out a major position on what CT is.
- - "Agree" (with (7) as stated -- no comment]. 3 people.
- - "The attributes required for CT car, be ins:ifled given enough time and
the right emphasis of subject matter in CT, but not all CT subject matter
instills proper attitudes.
- - "Disagree" [with (7) as stated -- no comment]. 3 people.
- - "Disagree, (we] need to include some of the attributes."
- - "Good move. Go for the (CT] itself!"
-- "Agree, but I often find that performance does not change until I get a
change in attitude. That the attitude has changed is best seen in the improved
performance... The most important attitude is "I really want to find out what
you believe, and what your reasons for believing it are. Then when I am sure
that you agree that I have understood you, I will think about and state where I

differ from you."

- - "Agree, but our ultimate concern is to encourage people to think
critically when it is appropriate to do so. And that requires a critical spirit
(critical attitudes) as well as thinking skills and knowledge."
- - "Disagree. The attitude/skill distinction isn't that sharp."
-- "Of course, some descriptions of dispositions are ways of describing
what people who are good at CT are like, but I fall to see how it follows from
this that they are not also ways of describing r7 itself."

***

104
Appendix C: Delphi Research Lettere, PAGE 100
 ,mmilimmmilmummimammlimmi
California State University, Fullerton
Fullerton, California 92634

Department of Philosophy
(714) 773-3611

DELPHI ROUND 5-C

Mar. 10, 1989

Dear Delphi Colleagues,

This letter asks you to consider three things::

(1) An outline of our report to the APA, (page 2),
(2) A comment on skills and skill assessment, (page 3),
(3) Specific recommendations we should make, (page 4)

Your ROUND 4 comments

on what a skill is and how a skill can be
taught and assessed are very revealing. Take a look, for example, at
the differences of opinion over the relative superiority of two of the
strategies of assessment. The research Steve Norris is doing, whether
or not idiot savants can be said to be skilled, and the basic conflict
between practical efficiency and accuracy in assessment, all found there
way into your responses. (See pages 5 and 6.) This is another example
of when an apparently off the wall question stimulated some useful
ideas.
Early responses to ROUND 5-A are coming in already. That's great.
Please respond to the three round five letters as soon as is
reasonably pissible for you. I plan to start putting our report
together this semester. Naturally you'll have a shot at it before
final revisions. If all goes well we'll be ready to submit our report
to the Pre-College Committee in the fall.
Depending on the quality of the recommendations that come, we
could be very close to wrapping things up. Thanks for your continuing
support and active involvement.
Yours sincerely,

Pete Facione

Appendix C: Delphi Research Letters, PAGE 101

'rho California State Unworsity

105
 Part 1: Proposed Outline of Our Report
INSTRUCTIONS: Consider the following outline of our Delphi
research report to the American Philosophical Association
Committee on Pre-College Philosophy. Comment, make any and all
additions, deletions, amendments, changes you think reasonable.
CT -- A Theoretical Construct for Purposes of Assessment
Submitted to the Pre-College Philosophy Committee
American Philosophical Association
EXECUTIVE SUMMARY
A. Delphi Conceptualization of CT
B. List of Delphi Recommendations
C. Endorsement of Members of Delphi Research Panel
REPORT
I. Introduction
A. The Concern for CT Assessment
1. National interest and large scale assessment
2. CT assessment in the inciividual classroom
2. APA mempers concern, college and K-12 'level
B. Formation the Delphi Research Project
1. Charge to the project director
2. Description of Delphi Research Methodology
3. Building the Delphi Panel of Experts.
II. Delphi Findings, Points of Agreement and Disagreement
A. Preliminary assumptions, [Rounds 1 and 2]
B. Conceptualization of CT, [Rounds 3, 4, and 5]
1. CT -- the skills dimension
2. CT -- the dispositional dimension
3. CT -- the normative dimension
C. General Comment on Assessing a Skill. [Rounds 4 and 5 -C]
III. Recommendations
A. General Considerations Regarding Educational Assessment
1. Validity, Reliability
2. Difficulty and Discriminability
3. Purposes of a CT assessment
4. Characteristics of persons being assessed.
B. Strategies for Classroom CT assessment
1. Some questions to ask one's self
2. Assessment, pedagogy and curriculum development.
3. Suggestions on putting together a classroom CT test.
C. Comments on Large Scale CT assessment
IV. Appendices
A. A Quick List of Purported CT Assessment Tools
B. A CT Bibliography with emphasis on Assessment
C. Delphi letters from each Round
D. Response rates to each Delphi and other tabular data
E. List of Delphi Panel of Experts

Appendix C: Delphi Research Letters, PAGE 102

 z.

Part 2: General Comment on Assessing A Skill

INSTRUCTIONS: This revision is based on you Round 4 comments. Please

edit. You migh' first read your colleagues' comments, on pp. 5 and 6.

A skill is the ability to perform certain processes or procedures more or

less Nell. Having a skill includes being able to do the right thing at the
right tine. So, being skilled at something involves knowing, perhaps
implicitly or without the ability to articulate this knowledge, a set of
procedures and when to apply those procedures. It also involves having some
degree of proficiency in executing those procedure:. Reflecting on and
improving one's own skills involves judging when one is or is not performing
Nell, and considering Nays of improving one's performance.

It is generally thought that skills, particularly cognitive skills, can

be taught in a variety of Nays including caking the procedures involved
explicit, describing how they are to be applied and executed, explaining and
modeling their correct use, and justifying their application. Teaching
cognitive skills also involves exposing learners to situations where there are
good reasons to exercise the desired procedures, judging their performance and
providing the learners with constructive feedback regarding their proficiency
and Nays of improving it. Instruction might start with situations that are
artificially simple, but it should culminth in situations that are realistic
and complex. Teaching is not everything.' The learners must contribute a
measure of effort, attention/ practice) desire, and self-monitoring. Teaching
skills involves motivating learners to achieve higher levels of proficiency
and, particularly in the case of CT, independence. It also may involve
coaching learners on how they can achieve those goals.

Persons can be judged as being more or less proficient in a given skill.

The first Nay of assessing is to observe the person performing the skill and
make a judgment regarding the degree to which the person possesses the general
skill in question. A second Nay is to compare the outcomes (if any) that
result from executing a given skill against some set of criteria. A third way
is to query persons and receive their descriptions of the procedures and
judgments they are using as they perform that skill, would use if they were to
perform that skill, or did use when they performed that skill. A fourth Nay
is to compare the outcomes (if any) that result from performing another task
against some set of criteria, where the performance of that other task has
been shown to correlate strongly with performance of the skill of interest.
No matter which Nay is used, it is important to ensure that the test
conditions foster an attitude in which the test-takers are disposed to use
their skills as well as they can, and are not constrained or inhibited from
doing so. It is highly advantageous to cross check the results of any one way
of assessment against the results of other mays.

Appendix C: Delphi Research Letters, PAGE 103

ot7
 Part 3: Recommendations

INSTRUCTIONS: Lonsidering what we know and don't know about CT

assessment today and considering the direction our Delphi effort has
taken us, what advise, suggestions, recommendations should we make
regarding CT assessment in the classroom or larger scale enterprises,
(e.g. naticnal, state, district wide, college wide CT assessment)?
Mention your intended audience, and be az precise about the
recommendations as possible. [e.g. To ETS and ACT we should say, ...;
to teachers of primary grades we should say,...; tc.; professors' who teach
CT at the college level we should say,...; to the researrlh community in
cognitive psychology we should say,...].

To help you think through this most important matter of making

recommendationi, I've drafted some questions. You need not respond to
them. They're only to stimulate thinking. Rather, formulate specific
recommendations. Oh! A plea for mercy: I have no staff, so please don't
ask that I search for things you might have once said about this
someplace else.)

1. What questions have you found it useful to ask yourself regarding your
own assessing of CT in teaching or profeSsional work?
2. How might a college age person who possessed the CT abilities we have
described be able to show that she had those abilities?
3. What about showing that she has the designated CT dispositions or
normative attitudes, (should we decide to include those in our final report)?
4. How might a person in elementary school, junior high or high school
show that she has those CT abilities, dispositions or normative attitudes?
5. What questions or tasks might you ask a group of persons to undertake
if you wanted good evidence regarding which persons were better at CT than
which others?
6. If you were doing a workshop for teachers at some grade level, what
would you tell then about CT assessment?
7. If you were serving as a consultant to some organization which wanted
to initiate large scale CT assessment, like a school district, university, or
state department of education, what would your recommendations be?
8. If a colleague and friend asked what you really thought about how CT

can best be taught, learned and measurpd, what would you tell your friend to
do or avoid doing?
9. What makes you optimistic about CT assessment?
10. What do you most fear about CT assessment?

Appendix C: Delphi Research Letters, PAGE 104

108
 Here are some of your comments on the idea of a skill and how
skills can be taught and assessed.
-
"... Your ways of teaching skills are conspicuous in their lack of
explicit emphasis on educational values including the use of morally justified
procedures in teaching. Similarly your ways of learning lack such
educationally important activities as explaining, understanding and
justifying."

* "Mental skills are no less directly observable than physical skills.

Skills are abilities, and abilities can't be observed. Thus, observing the
skill as it is being performed is impossible for any skill."

* "I have my doubts about ... the implicit mind/body dualism.

Performances can be measured directly (e.g. by observing a surgeon's technique
and results) and indirectly (e.g. by a paper and prncil test of knowledge of
anatomy). Both constitute a sample from which inferences are made to the more
general skill, and this is all we need to concern ourselves with..."

* "Your description exclude any effort, visualization, mental rehearsal,

or using of any inner resources a person brings to skill learning and skill
performance -- concentration, automaticity, goal directedness etc."

* "(The narrative] presupposes that if one has a skill, one consciously

knows the intricacies and interworking of that skill. Often this is simply
not the case. ... Indeed idiot savants may be said to possess skills with
little or no understanding of how they perform skilled activities. ... (Those
who] understand Chow they arrive at the correct result] are better critical
thinkers (than those arrive at the result] with no real understanding of the
rationale.

* "...Don't we observe performances which we interpret as being skillful

or being evidence of skill at a certain degree or level."

* "It seems odd to talk about performing a skill, given the initial
equating of a skill with the ability to do something well."

*** (The second way is superior to the third...]

"because the third way can be employed only after it has been
settled that the person has the skill, and that can be settled only by
comparing outcomes against criteria, i.e. by employing the second way. In
short the third way is parasitic on the second. Beside, a skilled person may
not even be able to describe the procedures and judgments used when executing
the skill."
"for the assessment of average people, because (1) CT skills are
generally employed unconsciously so the third way just won't work for average
people, and (2) it's fast, cheap and yields unambiguous (if not wholly
. trustworthy) answers... But the third way is superior when dealing with
experts, whom we can expect to be conscious of their procedures. If we can
get en account of their actual thought processes, then we can assess those
processes directly rather than indirectly."
"if instead of outcome we also include how the answer is arrived at;
thus an idiot savant can produce answers to complex equations but not by any
process we understand (or she does); if we limit ourselves to outcomes in

Appendix C: Delphi Research Letters, PAGE 105

1.09
judging skill, we'll miss the distinction between the highly skilled
arithmetic operator an the idiot savant. Since I would want to claim that the
idiot savant is not a case of skill at all, this presents some difficulties."
"because performande is what we want, provided that outcomes include
the steps taken. That is, if a student makes a good judgment on someone's
reasoning, I would want to know how he/she got to that judgment."
"because it is much easier to carry out in practice, provided that
there are enough items to compensate for accidental correct and incorrect
outcomes..."
"because it may be less time-consuming, and because it would be
easier to design instruments for..."
"because it does not depend completely on the testimony of the
individual about what occurred. That's a strength because the literature on
self-reports suggests enormous difficulties with the agent's own accounts of
their activities. If we have the product, and the account of the process
which yielded it, were in better position to judge the degree of skill."

*** [The second way is not superior to the third.]

"because the second is a prerequisite for carrying out the third;
that is, to do a credible job of querying a person must fist have completed a
strong comparison and examination ni the argument"
"because, as far as CT is concerned, [the second] does not reveal
the understandings and reasoning behind the answers given, but just assesses
overt answers against some criterion. Clearly one could come to "wrong"
answers depending on his understandings of the situatioo, and yet this
reasoning not be a violation of, or deficiency in, CT. Suppose someone asks
how many piles of beans we may divide 27 beans into, if we put 5 beans in each
pile. Suppcse someone says 6, because he allows that we can have small piles,
the remainder can constitute a pile in its own right. 'Given his
understanding, he has given the "right" answer. Does having his understanding
mean he is not thinking critically? ...Obviously, the third way has its
drawbackscsince it is far more consuming than checking outcomes against a set
of criteria. A truly superior method of testing CT would incorporate both."
"because there are often other possible explanations of the
outcomes."
"because the outcome could have been accidental or a consequence of
good luck, or bad luck, or in the case of testing, copied."
"because the third method might reveal that a person has good CT
skills where the second method suggested poor skills."
"[if one cannot specify] the criteria against which to judge the
process-outcome."

' "I do not regard the second and third ways [of assessing] as
constituting a hierarchy with one way superior to the other. I use both ways
in assessing students and check the two against each other, whenever I can.

* "Part of the problem with analyzing CT into skills is that a good deal
of the knowledge that makes up CT (quite apart from the so-called knowledge
which it presupposes) propositional knowledge and not a skill. One cannot
spot an argument or assess it validly without being able to define what an
argument is and what valid is. Forcing CT into skills is basically confused."
***

ti()
Appendix C: Delphi Research Letters, PAGE 106
 California State University, Fullerton
Fullerton, California 92634

Department of Philosophy
May 9, 1989
Dear Delphi Colleagues,
Dozens of responses to rounds 5-A, 5-B, and 5-C have come in. Again
I find myself admiring your perceptive analyses and sound suggestions.
And I am not just saying that to make you fell good. Your responses
range from the outstanding to the excellent. Often I find myself writing
"good point" and "remember" in the margins of your letters.
Some of you have not responded yet the ROUND FIVE letters. But the
notes or phone calls from those too busy to jump in to round five were
certainly appreciated. If you still have round five on you desk and are
wondering whether or not to respond now that the semester is coming to an
end the answer is please, by all means. We need your input not just to
keep up the quantity of participants but also for your ideas and
insights. There is still time to reply before I start putting together
the draft of our final report, a project I will not undertake until
August or September. For the summer I've taken on a huge teaching load
to help three of my children and my wife continue their undergraduate and
graduate educations.
I hope to make two presentations at the Sonoma State conference in
Auyust. One is a workshop on CT assessment, the other is tentatively
called "The Marlboro Man and Broad-Shouldered CT" The paper, "Assessing
CT and Building Consensus on CT," delivered at the Pacific Division
meetings of the APA went well. I'm doing a similar presentation for the
Education Colloquium of UC Davis next week and again in Baltimore in
Sept. A couple of junior colleges and a couple of state universities in
California have asked me to do staff development presentations on CT
assessment. A favorite is "Thirty Ways to Mess Up a CT Test," another
is "What is an Elephant, or Defining CT for Practical Purposes."
The above is a not too subtle way of saying I'm available if you are
planning something on. CT.
CS'J Fullerton has given me some time next academic year to pilot
test a CT assessment instrument. The curriculum committee at our
university defines CT based on CSU executive order 338. For those of you
outside CA, that executive order gives a general specification of CT for
purposes of requiring a "CT course" be included in the general education
program of all twenty of state universities. That definition is not
inherently inconsistent with our Delphi findings, but it is far. less
sophisticated. If my research project goes along as planned, then by
this time next year I should know if the CT assessment tool we will have
put together is capable of detecting improvements in students' CT which
result from their taking a 3 unit required lower division CT course from
the Philosophy Dept. This is a dangerous question! Ironically, a key
assumption in our experimental design is that the Philosophy faculty
teaching CT are doing an effective job. The experiment is to find out if
the assessment tool is sensitive enough to detect the difference we
assume our CT instruction is making.
Have a good summer. ill.Sincierelyr7
The California State UnimArcilv
Appendix C: Delphi Research Letters, PAGE 107 Ae' D7-4.4
 h 21122=1111:27XLC..."" X45. G1

California State University, Fullerton

Fullerton, California 92634-4080

Department of Philosophy
*** FINAL ROUND ***
(714) 773-3611
DELPHI ROUND 6
Sept. 25, 1989
Dear Delphi Colleagues,

Please review_the enclosed final draft. Your

your comments, substantive or editorial, are
vital. Be sure to check the consensus
statements in the Tables and the final
recommendations. TO BE HELPFUL YOUR COMMENTS
AND SUGGESTIONS MUST REACH ME BY OCT. 30.

The final report to the APA Committee on Pre-College

Philosophy is due in November. Its intended audience is
educators, at arty level, interested in CT. To clarify, as
principle investigator my role is to express the Delphi findings
objectively, whether they be points of consensus, majority
positions, or minority opinions. I am not one of the 46 expert
participants. I hope you find the draft a clear and fair-minded
expression of the delicate confluence of your expert, opinions.

ROUNDS 5A, B and 5C were extremely ustful. Of the 46

experts, 26. 23 and 22 responded in these rounds. ROUND 5A
confirmed the strong consensus regarding the cognitive skill
dimensien of CT. ROUND 5B showed that 61% of experts hold
that CT includes affective dispositions, but 30% maintain CT does
not. However, over 80% would be willing to use the same list of
affective dispositions in describing the paradigm critical
thinker. Although everyone recognised the personal and civic
value of CT, only 17% argued that "CT" has a normative meaning.
ROUND 5C produced several thoughtful recommendations relating to
CT teaching and assessment. Be sure to consider these carefully
and_expand on them if_you think it would be of general use to N-12
Qx.colltgp educators.

In Delphi research once an expert expresses an opinion, even

a dissenting one, it becomes a factor in the mix and flow of all
subsequent argument and thought. Because of this, and because in
Delphi research it is reasonable to assume that silence from busy
experts is a sign of general accord with the direction of inquiry,
I will continue to operate on the principle that unless you
object., you find things generally acceptable.

My respect for your collective and individual wisdom and my

gratitude for your participation are so deep that I cannot
possibly express either as fully as I feel them. Thank you so
very much for being part of this two year adventure. I hope the
final report does justice to what you thinli.

I am most appreciatively yonrA.

Appendix C: Delphi Research Letters, PAGE 108
The California State University
keter A. Facion,::
112