Concerns regarding student retention in STEM (science, technology, education, and mathematics) education in the United States have sparked numerous initiatives designed to increase the number of college students entering and persisting in the STEM disciplines (President’s Council of Advisors on Science and Technology, 2012). However, increased student enrollment can present challenges in large, lecture-based introductory science courses, which have traditionally acted as barriers rather than gateways to students pursuing science majors (Tai et al., 2005). Increasing the ratio of students to instructor may prevent goals for course transformation toward more student-centered instruction (American Association for the Advancement of Science, 2009) from being realized. Without additional instructional support, these courses may dissuade students from pursuing science majors, as many cite uninspiring and non-inclusive classroom environments (President’s Council of Advisors on Science and Technology, 2012) and a lack of support in large courses (National Academy of Sciences, National Academy of Engineering, and Institute of Medicine, 2007) as the main reasons for leaving STEM majors. To address these concerns, instructors may consider reformed teaching approaches to transform their courses from passive traditional instruction to student-centered instruction (Froyd, 2008; Wood, 2009).

Peer learning assistant (PLA) programs offer one approach that many universities have implemented across science disciplines to aid in course transformations toward student-centered active learning. PLAs are undergraduate students who have successfully completed a course and subsequently return to provide additional instructional support in the classroom (Otero et al., 2006). PLAs complete a pedagogical training seminar in which they learn how to engage students in active learning by using a variety of evidence-based teaching strategies, such as guided questioning, probing for reasoning, and formative assessment (Blackwell et al., 2017; Philipp et al., 2016a). PLA programs can help foster collaborative, active-learning environments by redirecting the traditional linear transmission of information from instructor to student into a multidirectional flow of information among students, the PLAs, and the instructor. In these environments, students can meaningfully engage with course material and actively construct knowledge with peers (Knight & Wood, 2005; Otero et al., 2006).

At our institution, PLAs earn course credit for their first semester and are paid an hourly wage for subsequent semesters. PLAs only assist students during regularly scheduled class time and have no grading responsibilities. During their first semester as a PLA, they complete a pedagogical training seminar that meets once per week. In this course, PLAs learn about epistemology, student conceptions, questioning techniques, and equity. PLAs also receive content training through weekly meetings with their assigned instructor to review content, group activities, and student challenges that will be encountered in upcoming classes. PLAs then put into practice what they have learned from their pedagogy training and content review meetings in the classroom to guide their interactions with students.

The value of peer learning is best supported by social constructivist theories of learning (Driver et al., 1994; Vygotsky, 1978). Because peer learning occurs in the classroom, student learning cannot be separated from the social context, as the community plays a central role in the “meaning making” of information (Driver et al., 1994). PLAs serve as “more knowledgeable others” and use scaffolding to help students move through the “zone of proximal development” to a higher level of conceptual understanding (Vygotsky, 1978). Additionally, because PLAs are cognitively congruent (Lockspeiser et al., 2008) with students, they are able to explain course concepts in familiar terms and are aware of misconceptions and challenges associated with those concepts (Bulte et al., 2007). In their interactions with students, PLAs ask guiding questions, elicit student thinking, and probe for reasoning to help students identify and address what they do not know. PLA-student interactions are characterized by a cycle of questioning and explaining, as the PLA guides the students’ sense-making and knowledge building (Furtak & Ruiz-Primo, 2008).

There is evidence that students in PLA-supported science courses perform better on exams and earn higher overall grades than students in traditional science courses (Chapín et al., 2014; Chasteen et al., 2011; Philipp et al., 2016c; Pollock, 2005; White et al., 2016). Students in PLA-supported courses express more positive attitudes toward STEM (Talbot et al., 2015) and are more likely than those not supported by PLAs to enroll in subsequent science courses (Philipp et al., 2016b).

However, although numerous studies have shown that PLAs positively impact student performance, few have investigated how PLAs support student learning. Talbot, Hartley, & Liddick (2015) used social network analysis to show that PLAs were the central figures in the network and fostered interactions between students. Knight et al. (2015) found that different methods of questioning by PLAs influenced the quality and quantity of students’ discussions during clicker questions. These studies highlight that PLAs play an important role in the classroom, as their actions contribute to active learning and student engagement. Thompson et al. (2020) recently developed a tool that outlines PLAs’ actions while they are engaged with students, although it has not yet been applied to understanding the impact of those actions on student learning.

This study seeks to contribute to the research documenting the ways in which PLAs affect student learning in undergraduate science courses, particularly those that incorporate student-centered, active learning. To determine if PLAs affect student learning, we compared student performance in an undergraduate introductory biology course across two sections when PLAs were present versus absent. Additionally, we surveyed students to better understand how they perceive the role of PLAs and how they believe PLAs support student learning in that course. We hypothesized that students would perform better when PLAs are present and would perceive PLAs as valuable for their learning.

Methods

Study context

This study was conducted across two sections of an introductory biology course (Principles of Biology I) at a large, public university in the southeastern United States during spring 2019. This course is required for science majors and the first in a two-course sequence. Major topics include biological molecules, energy processes, information flow inside cells, and trait inheritance. Both sections of this course (A and B) were taught by the same instructor and met twice each week for 75 minutes. Each section started with an enrollment of 150 students, and the overall DFW rate (i.e., the course percentage of students who receive a D or F or withdraw from the course) was 10%. One semester of introductory chemistry was a co-requisite for enrollment. Demographically, students who participated in the study were 55% White, 73% female, 65% first-year students, and 28% second-year students. To compare student performance when PLAs were present versus absent, a block design was implemented such that a group of five PLAs (three seniors and two juniors, four females and one male) alternated attending these two sections for each of the four units taught in this course (Table 1).

Opportunities for active learning in groups were incorporated throughout this class in the form of clicker questions and open-response group activities. Clicker questions were multiple-choice, higher-order questions that required application of knowledge, and students were encouraged to discuss their ideas with their peers before responding. Multiple clicker questions were posed throughout class, and students were given approximately 5 minutes to respond. Open-response group activities presented a novel scenario with open-response questions that often included multiple steps. These activities were incorporated on average once per week, and students were given approximately 20 to 25 minutes to engage with them. When present, PLAs would circulate among student groups to provide clarification, probe for reasoning, and ask guiding questions on clicker questions and group activities to assist student learning. Across the two sections, 86% of students consented to participate in the study (Section A = 139 students; Section B = 121 students; total n = 260). There was no difference in average attendance between sections (p = 0.637) or average final grade score (p = 0.688).

Student performance

For each of the four units, unit exams included five questions that aligned with clicker questions from that unit and five questions that aligned with open-response group activity questions from the same unit. Aligned questions had similar deep structures with varied surface details. Examples of each type of question are presented in Table 2. To assess the impact on student performance when PLAs were present versus absent, we compared student responses to the unit exam questions that aligned with clicker questions and group activity questions using either independent t-tests or Fisher’s exact test depending on how exam questions were scored. Scores were then compared between the two sections for each unit to determine if the effect of PLA presence on student performance was statistically significant. Exam questions were scored blindly by the instructor and two student assistants who were employed as graders. Assistants focused on questions that were straightforward to score, and the instructor graded open-response questions.

Student perceptions

An online survey was used to collect data on students’ perceptions of PLA impact on their learning. The survey was administered during class and was voluntary. Nineteen of the questions had Likert-scale items (7 = strongly agree to 1 = strongly disagree) to investigate how students perceived their interactions with PLAs in the course (Table 3). One question asked students to rank the resources in the order they find most useful for learning course content (1 = most useful to 6 = least useful). One multiple-selection question asked students to identify the strategies that PLAs used to assist their learning (Figure 1).

Results and discussion

Student performance

There was no significant difference (p > 0.05) in student performance on unit exam questions that aligned with clicker questions across sections when PLAs were present versus absent. However, students performed significantly better on nine out of the 20 unit exam questions that aligned with open-response group activities when PLAs were present across sections. Figure 2 shows the level of significance for each exam question. In the figure, the first bar always represents Section A and the gray bar indicates PLA presence, so PLAs were present in Section A during Units 1 and 3 and in Section B during Units 2 and 4. The unit exam questions that showed significantly higher performance when PLAs were present were evenly spread between the two sections, with Section A scoring significantly higher on four questions across two units and Section B scoring significantly higher on five questions across two units when PLAs were present. Therefore, the observed difference in performance is not likely a result of exam question difficulty or student demographics. The comparisons also indicated that the presence of PLAs never lowered exam scores.

The findings from this analysis demonstrate that PLAs had an effect on student performance and learning during certain types of classroom activities—namely, open-response group activities. There are several possible explanations for the difference observed in student performance on exam questions aligned with clicker and open-response activity questions when PLAs are present. It is possible that the additional instructional support provided by PLAs serves as a classroom management tool, such that with more instructors in the room, students are encouraged to spend more time focusing on the activity. However, this does not explain the difference in learning gains between the activity types (clicker questions versus open-response qustions), particularly because both require similar levels of higher-order cognitive processing (Bloom, 1956).

An alternative hypothesis considers the nature of the feedback provided by the PLAs to explain this observed difference. During clicker questions, students have limited time to interact with PLAs and ultimately receive the same mass feedback from the instructor after a few minutes of discussing with peers. Conversely, while working on group activities, students ask PLAs a variety of questions that lead to unique discussions in which PLAs elicit student thinking and provide guidance specific to each student based on the nature of their interaction. In this way, the feedback that students receive from PLAs during group activities is much more personalized and responsive to their own understanding. This hypothesis would explain why PLAs did not affect performance on exam questions aligned with clicker questions but did have an effect on some exam questions aligned with group activities.

PLAs’ ability to give individualized feedback to students is likely related to their pedagogical training and accessibility in the classroom. A key theme emphasized in PLA pedagogy training is the recognition and prioritization of students’ thinking and prior knowledge (Gray, 2013). PLAs learn that students enter the classroom with varying levels and types of prior knowledge, and they must be able to adapt their instruction to be responsive to the needs of each student. PLAs, like the instructor, recognize the value of student ideas, such that they serve as a starting point for building new knowledge. Formative assessment is often presented as the bridge PLAs use to identify students’ misconceptions and inform their approach to guide students’ conceptions into a complete and accurate framework for understanding (Top et al., 2018). By engaging in the formative assessment process with students, PLAs elicit student thinking and gather evidence about their understanding. They then use this evidence to provide necessary feedback to students. In many cases, this process can go through iterative cycles as the PLA helps the student move into a higher level of understanding (Furtak & Ruiz-Primo, 2008; Gray, 2013). This cycle may be particularly apparent when students are completing multistep, higher-order thinking activities that require frequent feedback from an instructor and subsequent discussion with peers.

Student perceptions of PLAs

This study also used a survey to explore students’ opinions regarding their interactions with PLAs and the role of PLAs in student learning (response rate: 90%). Survey questions centered around four major themes: students’ perceptions of PLA practices, students’ trust in PLAs, students’ value of PLAs, and study design. Results from the Likert-scale questions can be found in Table 3. Overall, more than 80% of the students agreed that PLAs used a variety of practices during their interactions that were helpful for learning. Students also reported more than 84% overall agreement with statements regarding their trust in PLAs, specifically that they were comfortable asking PLAs questions about course content and felt confident answering clicker questions after interacting with PLAs. Survey responses also indicated that students valued having PLAs in class, with only 7% of the students indicating that they would rather attend a class without PLAs. It is encouraging that students acknowledge that their interactions with PLAs are generally beneficial and an important aspect of their learning.

Responses to a multiple-selection survey question indicated that students observed that PLAs used a variety of strategies to support their learning. Total response count shows that students perceived that PLAs most frequently helped clarify confusing points and explain concepts in understandable terms, and PLAs gave students the right answer the least often (Figure 1). Almost all responses included multiple selections by students, suggesting that PLAs use a range of strategies during their interactions with students rather than relying on one strategy. Overall, the strategies that students identified are consistent with the pedagogical training PLAs receive regarding inquiry-based teaching practices that promote active learning. PLAs are trained to help guide students’ understanding rather than provide them with direct answers (Blackwell et al., 2017). However, the strategy most utilized by PLAs was offering explanations rather than questioning or prompting explanations from students, the latter of which is more aligned with inquiry-based practices. These findings suggest that PLAs are able to leverage their pedagogical training and content knowledge to engage with students and gain their trust, thus enabling them to participate in constructive discourse that students perceive as helpful for learning.

Students also ranked course resources in the order in which they found them useful for learning. Ranking order showed that students perceive the instructor to be the most useful for learning, which was followed by course materials (e.g., PowerPoint slides). PLAs and fellow students were ranked third and fourth, respectively, but without much separation. The internet and textbook were ranked as least useful for learning, which was a surprising result considering students’ familiarity with technology. However, this may be due in part to the instructor’s limited use of the textbook for this course. It was expected that students would rank the instructor and course materials as the two most useful resources because they provide students with the content for the course. It also makes sense that students would rank PLAs and fellow students similarly, as both served as a feedback mechanism during class. Nonetheless, this was an exciting finding because it suggests that students recognize the benefit of collaborative learning and view it as a useful approach to support their learning.

Findings from the survey analysis are consistent with our predictions that students would perceive PLAs as valuable for their learning. This is supported by a concept called social and cognitive congruence, which suggests that students value learning from near-peers because they have similar and recent experiences with course content and understand the challenges of that course (Lockspeiser et al., 2008). These results are consistent with other studies that found that students in a PLA-supported course were more satisfied with the course (Talbot et al., 2015). Future research should focus on exploring PLAs’ perceptions of their role in student learning and evaluating their effectiveness in performing these roles.

Conclusions

Our results contribute to prior literature describing the benefits of PLAs in student-centered introductory college science classes. This study demonstrates that PLAs can have a considerable impact on student performance as they engage in active learning. However, the impact of PLAs may depend on the type of activity students are completing. Students performed significantly better on exam questions aligned with open-response activities than on those aligned with clicker questions. This observed difference could be due to the individualized feedback provided to students during these activities that is more responsive to students’ own conceptions. PLAs’ ability to provide such feedback may be informed by their pedagogical training. For this reason, student ideas, prior knowledge, and the role of formative assessment should be emphasized in PLA training. Furthermore, PLAs may be utilized during both types of in-class activities, but instructors should consider focusing more of their preparation efforts with PLAs on open-response group activities, as the support provided during these activities appears to be more beneficial than the support provided during clicker questions.

This study also shows that PLAs are capable of using a variety of strategies to support student learning. Specifically, PLAs employ several inquiry-based strategies such as asking guiding questions and eliciting student thinking. These strategies in particular may be useful for informing the personalized feedback they provide for students as they work on collaborative learning activities. Moreover, the strategies that PLAs use to support student learning are consistent with their pedagogical training, such that they mostly rely on approaches that prompt and guide student thinking rather than provide students with the correct answers. Students perceive these strategies to be useful for their learning and value the guidance provided by the PLAs. This is encouraging because it suggests that students recognize the importance of reflecting on their thinking as part of the learning process.

Results from this study are limited in that data were collected from one instructor during a single semester, and student performance was compared by the presence or absence of PLAs and the type of class activity. Future studies may explore why PLAs affect student performance for certain questions and not others (e.g., content area, cognitive level, type of PLA interaction), which would require additional data and a different experimental design. These future studies would further establish how PLAs can be used in ways that maximize the benefits for students.

Although this study was conducted in an introductory biology course, PLAs have been implemented in all STEM disciplines, so the implications of these findings may interest instructors engaged in college teaching across the disciplines. In particular, this research may help address concerns regarding challenges for student learning in traditional science courses as course transformations become essential for creating student-centered, active-learning environments. PLAs offer an efficient and successful avenue for supporting such course transformations by facilitating student learning through meaningful engagement with course content. Specifically, PLAs implement multiple strategies during their interactions with students to elicit student ideas and guide them toward a higher level of understanding. Students acknowledge the value of PLAs and the strategies they use to support learning in the classroom. This study also offers important considerations regarding PLA pedagogical training and the significance of individualized feedback. Our future contributions to this research will further investigate how PLAs can utilize inquiry-based strategies and feedback to support active learning and identify the activities in which PLAs have the strongest impact.

Acknowledgment

We thank Dr. Kristen Miller for her support and assistance with the planning, experimental design, and recruitment for this study.

Brittney A. Ferrari (baferrari13@uga.edu) is a doctoral candidate in the Department of Mathematics, Science, and Social Studies Education; Jonathan A. Dees is a lecturer in the Department of Plant Biology; Norris A. Armstrong is a professor in the Department of Genetics; and Julie M. Kittleson is an associate professor in the Department of Mathematics, Science, and Social Studies Education, all at the University of Georgia in Athens, Georgia.