Name: Mawar Uzlil Fatil jahra
Nim: E1D02310110
Class: 4E
CHAPTER 3
DESIGNING CLASSROOM LANGUAGE TESTS
A. FOUR ASSESSMENT SCENARIOS
This four scenarios are designed to real world assessment situations.
1. Scenario 1: Reading Quiz
The first context is an intermediate-level class for secondary school students learning English in
Brazil. The students have been assigned for homework a two-page short story to read on their own,
and the teacher has decided to begin class the next day with a brief “pop quiz”: 10 short-answer
written comprehension questions. The purpose is to give students sense of how they understand the
material, and the result of the quiz will not be noted in teacher’s record book.
2. Scenario 2: Grammar Unit Test
This test is at the end of a three week unit in a grammar focus course at a high beginning (Level 2)
class in an adult school in the United States. The grammar unit has covered verb tenses. The
curriculum specifies that the 45-minute test is to be divided into three sections: multiple-choice items,
fill-in-the-blank (cloze) items, and a grammar editing task (where students must detect errors in
several written paragraphs). The test will be handed in, graded by the teacher, and returned to
students a few days later.
3. Scenario 3: Midterm Essay
The midterm essay is an opportunity for students to demonstrate their ability to write a coherent essay
with relatively few grammatical and rhetorical errors. With the goal of each student to revise his or
her essay, followed by a student-teacher conference after a revision has been turned in.
4. Scenario 4: Listening/Speaking Final Exam
Children in the fifth grade of a private school in Japan have been taking a 15-week course in oral
communication skills (listening and speaking). Their fluency would be described as very low, and
their grammatical accuracy is perhaps passable with the minimal amount of language they can handle.
B. DETERMINING THE PURPOSE OF A TEST
As you gain experience, you can attempt bolder designs. In that spirit, next consider some practical steps
in constructing classroom tests.
1. Test Usefulness
The first and perhaps most important step in designing any sort of classroom assessment is to step
back and consider the overall purpose of the exercise that your students are about to perform. Based
on what Bachman and palmer says that the purpose of an assessment refer to as test usefulness, or the
extent to which a test accomplishes its intended criterion or objective.
2. Reading Quiz
The quiz is designed to be an instructional tool to guide classroom discussion during one classroom
period. Its significance is minor but not trivial when viewed against the backdrop of the whole course.
Because it is a surprise test and a tool for teaching and self-assessment, the results will justifi ably not
be recorded, and so one student’s performance compared with that of others is irrelevant.
C. DEFINING ABILITIES TO BE ASSESSED
Establishing appropriate objectives involves a number of issues, from relatively simple ones about forms
and functions covered in a course unit to much more complex ones about constructs to be represented on
the test. In other words, define or state the abilities you intend to assess. Remember that every curriculum
should have appropriately framed, assessable constructs stated in terms of performance by students. An
objective that states “Students will learn tag questions” or simply names the grammatical focus of “tag
questions” cannot be tested. Your first task in designing a test, then, is to determine the ability or abilities
(that is, in assessment terms, the construct) that you want students to demonstrate.
1. Grammar Unit Test
Elsewhere in the curriculum, “appropriate contexts” are described as a continuation of the material
introduced and practiced in the other two (listening/speaking and reading/wniting) classes, so you're
left with a sketchy but workable set of objectives on which to base your unit test. You certainly need
to flesh these out in more detail before you can be satisfied that you have clearly specified, assessable
constructs. Where do you begin? In this grammar course, students equally use all four skills as they
work with the grammar forms/structures. Therefore, to achieve content validity, the test should
require the students to perform all four skills and sample all four verb tenses.
D. DRAWING UP TEST SPECIFICATIONS
Test specifications, or “specs,” for classroom use can be an outline of your test what it will look like. To
design or evaluate a test, you must make sure that the test has a structure that logically follows from the
unit or lesson it tests. The class objectives should be present in the test through a variety of appropriate
task types and weights and a logical sequence. Think of your test specs as a blueprint of the test that
include: The skills/abilities assessed, a description of its content, item types (methods, such as multiple-
choice and cloze), tasks (e.g., written essay, reading a short passage), skills to be included, specific
procedures to be used to score the test an explanation of how test results will be reported to students.
It’s important to note here that test specifications are much more formal and detailed for larg-scale
standardized tests that are intended to be widely distributed and therefore are broadly generalized. Such
secrecy is not a part of classroom assessment; in fact, one facet of effectively preparing students for a test
is giving them a clear picture of the types of items and tasks they will encounter.
1. Grammar Unit Test
The test specifications design might comprise the following four sequential steps:
A broad outline of how the test will be organized (already specified in the curriculum; see
above)
Which of the eight subskills you will test (if not all)
What the various tasks and item types will be
How results will be scored, reported to students, and used in future classes (washback)
These decisions are not easy ones to make. Even though Steps 1 and 2 could be fairly easy, 3 and 4
present genuine challenges.
2. Midterm Essay
The specifications for this assessment might look like this in Midterm Essay Specifications:
Provide clear directions.
Write a prompt for either a narrative or a description essay.
The prompt must be on a familiar topic that students will, with a reasonable level of
confidence, be able to write about coherently.
Assign an expectation of a full page, handwritten, and no more than
two full pages.
Students will be given a 90-minute time limit.
In the prompt, include evaluation criteria: content, organization,
rhetorical discourse, and grammar/mechanics.
The final grade is to include four subscores: content, organization,
rhetorical discourse, and grammar/mechanics.
The more meticulously you specify details of an assessment procedure, the more likely the
assessment will provide students an opportunity to perform well. Other test specs may be more
complex. Suppose students will perform two or more skills, as in the listening/speaking final exam
for Japanese fifth graders (Scenario 4).
E. DEVISING TEST ITEMS
Let’s look at the midterm essay scenario again this time more specifically in terms of item design:
1. Midtern Essay
This could be one of the easiest kinds of test tasks to create, because only one “item” is involved,
student responses are open ended, and evaluation criteria have already been covered well in previous
instruction.
2. Listening/Speaking Final Exam
A limited number of modes are applicable to elicit responses (that is, to prompt) and to respond on
tests of all types and purposes. Consider the options: the test prompt can be oral (student listens) or
written (student reads), and the student can respond orally or in writing. As you design your final
exam, it’s important to consider the age of the students. Fifth graders are approximately 10 years old,
and at this age explicit form focus is appropriately not a part of the curriculum. Listening
Comprehension Section Because of the constraints of your curriculum, the listening part of the final
exam must take no more than 20 minutes, as already noted. Oral Production Section Your curriculum
allows you to design your own oral interview protocol, so you draft questions to conform to the
accepted pattern of oral interviews
G. DESIGNING MULTIPLE-CHOICE ITEMS
Two principles stand out in support of multiple-choice formats: practicality and reliability (of course).
With their predetermined correct responses and time-saving scoring procedures, multiple-choice items
offer overworked teachers the tempting possibility of an easy and consistent process for scoring and
grading.
1. Design Each Item to Measure a Single Objective
2. State Both Stem and Options as Simply and Directly as Possible
3. Use Item Indices to Accept, Discard, or Revise Items (Optional)
4. Listening/Speaking Final Exam
H. ADMINISTERING THE TEST
Once the test has been created and is ready to administer, students need to feel well prepared for their
performance. An otherwise effective, valid test might fail to reach its goal if the conditions for test taking
are inadequately established. Here’s a list of pointers:
1. Pre-test considerations (the day before the in-class essay)
2. Test administration details
I. SCORING, GRADING, AND GIVING FEEDBACK
1. Scoring
scoring plan reflects the relative weight place on each section and on the items in each section. In
the four scenarios we have been discussing in this chapter, scoring (in mathematical terms) is a
factor in only two of them, the grammar unit test and the listening/speaking final exam for
Japanese fifth graders.
2. Grading
Grading is such a thorny issue that all of Chapter 12 is devoted to the topic. How assign letter
grades to this test is a product of the: country, culture, and context of the English classroom,
institutional expectations (most of them unwritten), explicit and implicit definitions of grades you
have set forth, relationship you have established with this class, student expectations that have
been engendered by previous tests and quizzes in the class
3. Giving Feedback
In the four example scenarios we have been referring to in this chapter, a multitude of options exist
for giving feedback: Reading Quiz. The primary if not exclusive purpose of the reading quiz was
to prompt self-assessment and class discussion. Grammar Unit Test. The most salient form of
feedback is the total score and subscores, but perhaps the most useful feedback could come in the
form of diagnostic scores, a checklist of areas needing work, and class discussion of the test
results. Midterm Essay. All the types of feedback listed are feasible and potentially useful, but
perhaps the kind of feedback that would contribute most to beneficial washback would be the
subsequent peer conferences and individual conferences between student and teacher.
Listening/Speaking Final Exam. The children in this class will eventually receive a letter grade for
the course, which may include scores and subscores of the final examination, with little else
possible within the system.
