Running head: PEER-LED SUPPLEMENTAL SESSIONS 1

The Effects of Peer-Led Supplemental Sessions on Students in an Undergraduate

Calculus Course

Keith Williams
EDFS 209: Professor Sean Hurley
University of Vermont
December 3, 2015
 Peer-Led Supplemental Sessions 2

Abstract

Research and practice support the idea that undergraduate, peer-driven resources, such as

teaching assistants and tutors, are viable methods to improve academic performance in higher

education settings. Further, research demonstrates the connection between these programs

and higher grades, superior general competencies, and higher rates of graduation. Often, such

as the widely used Supplemental Instruction, programs are voluntary and allow for self-

selection. The proposed research addresses this gap, and attempts to address two primary

related questions. First, what is the effect of a required peer-led recitation compared to voluntary

sessions and or no support at all beyond the lecture? Also, what kinds of students, and in what

circumstances, self-select for optional resources? Through a pre- and post-assessment with the

Learning and Study Strategies Inventory (LASSI), students across all sections of a Calculus I

course will be identified and tracked across the course of a semester to determine the levels of

success based on varying levels of academic support outside of the lecture.

 Peer-Led Supplemental Sessions 3

Introduction and Literature Review

Supplemental Instruction is a formalized, peer-led, group review system used at

institutions of higher education all around the world. Founded at the University of Missouri at

Kansas City, it has proven successful in supporting higher grades and superior graduation rates

at schools across the US, Canada, the UK, South Africa, Australia, Hong Kong, and several

other countries (Bowles, McCoy, & Bates, 2008). The model has been applied to courses at the

introductory, intermediate, and advanced level courses, across the disciplines, and it has a

proven and consistent track record.

The SI program incorporates an SI coordinator along with instructors and one SI leader

per participating section. The SI leader, a student him or herself, having completed the particular

course prior to participating, attends the lecture for a second time. Outside of class time, the SI

leader hosts two office hours and two review sessions per week. Students are informed that

these resources are available through a number of methods, but there is no tangible reward and

their attendance is not required in any way (Fayowski & MacMillan, 2008).

Bowles, McCoy, and Bates surveyed a number of first-year courses at Utah State

University during the 2001-2002 academic year (2008). They compared participating students

against the students in the same courses who did not use the program. The researchers used a

single equation regression model and a treatment effects model, incorporating high school GPA

and ACT scores. Their results indicate that attendance of SI sessions during first-year courses,

increases the probability of graduation within approximately four years by 0.1075 or 10.75%

(Bowles, McCoy, & Bates, 2008, p. 856). Through their statistical analysis, the self-selection

bias is addressed. It is acknowledged that additional explanatory variables may support the

study, but that any attempt would miss unobserved, and likely unmeasurable, characteristics

that affect SI attendance and graduation rates (Bowles, McCoy, & Bates, 2008).
 Peer-Led Supplemental Sessions 4

Fayowski and MacMillan explored SI in a first-year calculus course, with a focus on

performance when course grades are adjusted statistically to account for both gender and

motivation (2008). Their study considers 869 students of one particular cohort, enrolled in nine

different sections of a particular first-year math course, designed for non-math majors. The

students are all members of a single institution in the Pacific Northwest of North America. They

also consider data from 390 students who completed the course in a previous semester, prior to

the SI implementation. The design itself is quasi-experimental, as the students could not be

randomly assigned to SI or non-SI groups. A covariate consisting of students prior GPAs is

used, to statistically adjust for under- or over-performing students gravitating to the self-selected

nature of SI sessions. In the end, SI participants earned higher grades than both non-

participants and pre-treatment students. The Pearson Chi-square test confirmed that a

statistically significant difference existed (Fayowski & MacMillan, 2008). These significant

differences in performance remained even after correcting for prior GPA and gender. While

meaningful, the results are still unable to address the precise element, or elements, of the SI

program that contributes to improved academic performance in students.

Ning and Downing address the potential impact of SI on not only academic performance,

but on general learning competencies (2010). To do so, they incorporate the Learning and Study

Strategies Inventory (LASSI) as a pre- and post-treatment assessment. They look at 430

undergraduate business students at a university in Hong Kong. They use structural equation

modeling procedures to consider the LASSI scores in relation to academic grades and whether

or not students attended the voluntary SI sessions. Their findings support the idea that SI not

only supported higher grades, but more growth on the factors in the LASSI. Despite the

researchs obvious limitations, such as its small sample size from within one, highly focused

program, the results, corroborate the view that supplemental instruction is a feasible tool for

enhancing students learning competence and academic performance (Ning & Downing, 2010,

p. 935).
 Peer-Led Supplemental Sessions 5

While there is a good deal of literature available for the efficacy of SI, much of is based

on an international audience. The phenomenon is significantly less studied in the US, especially

when compared to peer institutions in Canada and the UK, among other countries. The inability

to generalize findings across international, as well as cultural, boundaries proves an obstacle in

arguing for the overall effectiveness of the model.

Further, qualitative research offers support that peer education models offer benefits to

both learners and educators, though it often fails to take on a large enough groups, or a

sufficiently representative sample, to provide data for other regions, communities, or institutions.

Methods

The proposed project will be quasi-experimental in nature. It will explore the

impact of required and structured academic review, outside of the conventional lecture.

The research will utilize an instrument to assess students broader study strategies, as

well as more conventional academic measures. The group of students enrolled in a

particular course will be tested using this instrument at the beginning and end of the

semester. Throughout the semester, the entirety of the group will have optional review

sessions available to them, while half will be required to attend one session every week.

Overall attendance will be tracked. Results over time - final grades, first to second year

retention, and four year graduation rates, in addition to the standardized assessment -

will be analyzed in light of the requirement, and the ultimate frequency in which a

student uses the optional review sessions.

 Peer-Led Supplemental Sessions 6

Participants

Broadly, all participants will be students enrolled at the University of Vermont, a

medium-sized, public research institution in the northeastern United States. Beyond

this, there will be substantial diversity in prior preparation and content knowledge,

reason for taking the course, and individual biographies and demographics.

The participants for the study will focus on all students enrolled in a particular

course at the University of Vermont, MATH 019. MATH 019 is the first semester of a

two-semester Calculus I sequence. It includes twenty different sections in any given fall

semester, typically ranging from approximately 30 to 60 students per section. Students

are predominantly first-years, but the enrollment is not restricted and certainly includes

some more advanced undergraduate students.

MATH 019 is a quantitative course, based on developing proficiency in various

procedural problem-solving tasks. It is a standard three credit course, with students

spending two and a half hours in lecture, distributed as either two, seventy-five minute

sessions or as three, fifty minute sessions. It stands out for its high rates of DFW

(grades of D+, D, D-, a failure, or a withdrawal from the course). It is a popular

prerequisite, as students are required to complete it in virtually any natural science

discipline, any business concentration, or any of various social sciences, overall

including, but not limited to, Biology, Economics, and Accounting.

Research Design and Implementation

 Peer-Led Supplemental Sessions 7

The Learning and Study Strategies Inventory (LASSI) contains 90 items, all

statements with Likert Scale responses; the statements range across ten categories:

Anxiety, Attitude, Concentration, Information processing, Motivation, Scheduling,

Selecting main ideas, Self-testing, Study aids, and Test strategies (Weinstein,

Zimmerman, & Palmer, 1988). The LASSI has been compared to other tests or

measures, focusing on similar elements. It was designed as a result of the perceived

need to more rigorously assess students entering American colleges and universities by

their academic skills (Weinstein, Zimmerman, & Palmer, 1988).

The LASSI has been studied in a number of contexts. Olaussen and Braten have

confirmed its reliability and suggested generalizability (1998), as they compared their

results with Norwegian students to those of other researchers working with American

students. Further, Obiekwes work reports alpha coefficients that reasonably matched

those in the LASSI manual, with some subscales having Cronbach alpha that are

slightly better than reported (2000).

All participants will complete the LASSI at the beginning of the semester and

again at its conclusion. The pre-test will be implemented on the first day of classes,

either August 29 or 30, depending on the section. The post-test will be run on the last

day of classes, either December 8 or 9. For the actual assignment to the required

session, each class section will be coded based on its schedule; classes starting from

8:00 am to 12:00 pm will be coded together, along with those between 12:00 pm and
 Peer-Led Supplemental Sessions 8

4:00pm, and those from 4:00 pm and later. Assignment per section will consider, as

much as the discrete and individual sections allow, a balance in major courses of study,

class status at the institution, identified gender, and section schedule disparity (i.e.

classes running on either Monday, Wednesday, and Friday or Tuesday and Thursday).

This required recitation session will be a once weekly, fifty minute session led by

a peer educator. That position will be filled by a junior or senior, who will have taken the

same course at UVM, received at least a B+ or better, acquired a faculty

recommendation from the Mathematics Department, and participated in a twenty-hour

training series during the prior semester, including theory related to learning, teaching,

developmental psychology, and motivation, as well as practical case work on how to

clearly present information, improve motivation in pupils, develop academic skills, and

facilitate group work.There will be approximately 25 total peer educators, and

assignments will be based on their availability.

The recitation sessions themselves would offer approximately twenty-five

minutes on re-lecture, which includes presenting information from lecture in a different

manner or perspective, guided practice through concepts and or problem-solving, or

question and answer sessions. The remaining half of the session would be taken up by

structured group problem-solving. Leaders would facilitate dividing groups, assigning

problems, and promoting successful study habits.

 Peer-Led Supplemental Sessions 9

Each leader would be observed twice throughout the semester. In part, these

would be opportunities for the leaders, students themselves, to engage in a professional

development activity of receiving and acting on feedback. More relevant to the research,

however, will be ensuring the session leaders are meeting minimum expectations.

Students will also provide feedback at the end of the semester, reflecting on both the

recitation program generally, as well as the leader specifically. Observation data and

feedback can assist in analysis, as the results will most certainly vary from leader to

leader.

While ten of the sections would include this requirement, and would be

scheduled into their own sessions with other similar students, the remaining ten

sections would not have the requirement. However, open drop-in sessions would be

offered, which would be available for any MATH 019 student, whether from the assigned

group or the unassigned group. Those sessions would be led by the very same peer

educators as the required recitations. As much as possible, the structure would be

consistent across sessions, as long as attendance permits the structured group work

model. Attendance would be tracked for all sessions, whether of the required variety or

not.

This tracking will create four subgroups of study. First, there will be students who

are not required and choose not to attend the optional sessions. Second, there will be

students who are not required and choose to attend the optional sessions. Third, there
 Peer-Led Supplemental Sessions 10

will be students who are required and do not choose to attend any of the optional (and

for them, extra) sessions. Finally, there will be students who are required to attend and

also choose to attend the optional and extra sessions.

Analysis, Hypothesized Results, and Discussion

Analysis will consider student performance across several measures, compared

to the kinds and frequency of attendance. Those will include final grade for the course,

cumulative average for the semester, retention from one year to the next, and four-year

graduation rate, in addition to the pre- and post- assessment scores from the LASSI.

Research supports the idea that more time on task correlates with superior

performance in the academic world, especially when guided by a peer (Longfellow et al.

2008). It extends from that idea that students with greater frequency of attendance at

these review sessions, whether required or not, will perform better on average than

students with lesser frequency of attendance. Further, research indicates that structured

review will support increases in the performance on the LASSI (Ning & Downing, 2010).

With the different groups, we may be able to observe this disparity in magnitude of the

effect on performance associated with the total amount of structured review sessions.

Additionally, the LASSI includes components on Motivation, among other

qualities. Self-selected attendance could be compared, specifically, to the motivation

component in the LASSI instrument to determine the rate at which students attend

based on some prior state of personal motivation. Ultimately, what effects are there for
 Peer-Led Supplemental Sessions 11

attending these sessions, and how does the magnitude vary across frequency of

attendance? And also, how does motivation, among other prior skills assessed by the

LASSI, interact with the tendency for students to attend these sessions?

As individual students cannot be randomized either to receive additional

academic support or not, any results will be suggestive of correlations and not

necessarily demonstrative. Potentially significant results may support a varying

effectiveness related to the nature of academic support outside of the classroom. It is

expected that students who attend more sessions will do better. Further, it is expected

that the students who are required to attend the sessions will do better overall than the

sections with only the optional sessions, though confounding variables may diminish

this effect.

The hope is that the results will indicate whether required sessions offer a

meaningful positive impact on student success. This information could inform how

academic departments organize their teaching assistants, and how tutorial units offer

their tutors.
 Peer-Led Supplemental Sessions 12

References

Bowles, T. J., McCoy, A. C., & Bates, S. (2008). The effect of supplemental instruction
on timely graduation. College Student Journal, 42(30), 853-859.

Chapin, H. C., Wiggins, B. L., & Martin-Morris, L. E. (2014). Undergraduate science learners
show comparable outcomes taught by undergraduate or graduate teaching assistants.
Journal of College Science Teaching. 44(2), 90-99.

Fayowski, V. & MacMillan, P. D. (2008). An evaluation of the Supplemental Instruction

programme in a first year calculus course. International Journal of Mathematical
Education in Science and Technology. 37(9), 843-855.

Lockspeiser, T.M., P. OSullivan, A. Teherani, and J. Muller. (2008). Understanding the

experience of being taught by peers: The value of social and cognitive congruence.
Advances in Health Sciences Education: Theory and Practice. 13 (3), 36172.

Longfellow, E., May, S., Burke, L., & Marks-Maran, D. (2008). They had a way of helping that
actually helped: a case study of a peer-assisted learning scheme. Teaching in Higher
Education. 13(1), 93-105.

Leung, K. C. (2015). Preliminary empirical model of crucial determinants of best practice for
peer tutoring on academic achievement. Journal of Educational Psychology. 107 (2),
558-579.

Ning, H. K., & Downing, K. (2010). The impact of supplemental instruction on learning
competence and academic performance. Studies in Higher Education, 35(8), 921-939.

Obiekwe, J. C. (2000). The Latent Structures of the Learning and Study Strategies Inventory
(LASSI): A comparative Analysis. Presented at the Annual Conference of the American
Educational Research Association. April 24-28, 2000, New Orleans, LA.

Olauseen, B. S. & Braten, I. (1998). Identifying latent variables measured by the learning and
study strategies inventory (LASSI) in Norwegian college students. The Journal of
Experimental Education. 67(1), 82-96.

Topping, K. J. (1996). The Effectiveness of Peer Tutoring in Further and Higher Education: A
Typology and Review of the Literature. Higher Education. 32 (3), 321-345.
 Peer-Led Supplemental Sessions 13

Weinstein, C. E., Zimmermann, S. A., & Palmer, D. R. (1988). Assessing Learning Strategies:
The Design and Development of the LASSI. In C. E. Weinstein, E. T. Goetz, & P. A.
Alexander (Eds.), Learning and Study Strategies: Issues in Assessment, Instruction, and
Evaluation. Austin, TX: Academic Press, Inc.