Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2023 Jun 26;100(7):2495–2503. doi: 10.1021/acs.jchemed.2c00635

Perceived Barriers to Equitable Participation in the Learning Assistant Program

Ronia N Kattoum †,*, Ibraheem Abbood †, Khristina E Huff †, Mark T Baillie †
PMCID: PMC10340102  PMID: 37455797

Abstract

graphic file with name ed2c00635_0003.jpg

To build a more diverse STEM workforce, institutions seek to increase the representation of diverse groups in faculty and mentoring positions. The Learning Assistant (LA) near-peer student support program has the potential to bring diverse students into highly visible and impactful mentoring roles early in their college careers, benefiting both LAs and students in LA-supported courses. However, the demographic characteristics of potential students interested in the LA Program and the subsequent barriers to entry have yet to be investigated. This short-term longitudinal case study revealed that even though students from historically underserved groups (HUGs) started the semester equally as likely as non-HUGs to see themselves as future chemistry LAs, an inequity developed later in the semester. Similar trends were not detected based on students’ gender or age group (traditional/nontraditional). Qualitative data indicated that regardless of demographic group, the most prominent barriers to students seeing themselves as future LAs include a lack of time and self-efficacy in chemistry content knowledge/social skills. The trends observed at this diverse metropolitan research institution indicate that further research is needed to find and lower barriers for students to have the opportunity to become LAs, especially those from HUGs.

Keywords: learning assistant, first-year undergraduate/general chemistry, chemical education research, collaborative/cooperative learning, minorities in chemistry, women in chemistry

In the past decade, there has been an increase in the number and shares of STEM (Science, Technology, Engineering, Mathematics) degrees awarded to students from Historically Underserved Groups (HUGs: Black, Hispanic/Latina/o/e, and Multiple/Other Race)1 and women.2 However, disparities still exist, and many organizations are calling to make deliberate changes to increase the diversity of the STEM workforce.3,4 A recent US report focused on Science and Engineering (S & E) found that the percentage of S & E bachelor’s degrees awarded to Black students remained at a stagnant 8% from 2008 to 2018, with the disparity even more significant in the earth and physical sciences.2 In 2018, only 5.43% of earth and physical science bachelor’s degrees were awarded to Black students.2 Women earned half of bachelor’s degrees in S & E fields and 40.6% of bachelor’s degrees in the physical and earth sciences, with chemistry leading the way in the shares of degrees awarded to women in 2018 (50.8% at the bachelor’s level, 45.4% at the master’s level, and 39.0% at the doctoral level).2 While this is promising for women, the trend does not seem to carry over in S & E employment, where men held more than twice as many S & E positions as women in 2018.2 This is especially concerning in academia, where full-time faculty representation in US higher education continues to be predominantly white and male in all fields, with an astonishing disparity in the physical sciences: 77.7% White, 14.5% Asian, and 7.7% HUG and 29.4% female.5

Since college is the training ground for future STEM professionals, the lack of faculty diversity has created a negative feedback loop in which women and students from HUGs do not see themselves reflected in academia, decreasing their sense of belonging and triggering stereotype threats that lead to underperformance.6−9 Current faculty members are those individuals who succeeded in the traditional lecture setting and may not fully grasp the struggles and strengths of their students, who bring a vast array of valuable life experiences to the classroom.10 Providing diverse near-peer mentors who reflect the student population may help narrow the gap in perceived diversity in leadership and encourage students to pursue STEM careers who otherwise may not have seen themselves reflected in STEM fields.11 The Learning Assistant (LA) model is one such example of a near-peer mentoring model that could help to increase the diversity of the teaching team.

The Learning Assistant (LA) Model

The Learning Assistant (LA) model originated at the University of Colorado Boulder in the Department of Astrophysics and Planetary Sciences in 2001 and has since spread to over 500 institutions across the U.S. and abroad.12 Different from the Supplemental Instruction model, where student mentors typically have very little training or mentorship,13 the LA model is built upon three main principles: (1) training with the supervising faculty member on content knowledge, (2) training on and reflection about teaching through pedagogical training, and (3) serving as a facilitator in the classroom.12 Each of these principles informs the other, creating a structure that results in (1) better classroom outcomes, especially for students from HUGs, (2) improvement in faculty teaching skills, and (3) better outcomes for the LAs.14

Physics education researchers have found that students’ DFW rates (percent of students with a grade of “D”, “F”, or withdrawal from the course) tend to be lower in courses that are supported by LAs.15,16 Although not heavily studied, some evidence suggests that LAs increase their content knowledge compared to non-LAs and have content mastery closer to that of Graduate Teaching Assistants (GTAs) than undergraduate students.17 Physics LAs have been shown to develop a strong physics identity, seeing themselves as “physics people”.18,19 Being an LA increases “professional identity” in which LAs shift toward a mastery perspective (focused on the process of learning) versus a performance perspective (focused on earning the highest grade).20

LAs can be a seminal part of classroom reform as they help faculty implement evidence-based teaching practices,14 including active learning, that has been shown to improve student outcomes and narrow equity gaps (the differences in achievement between students from HUGs and the majority group).21−25 The LA model can serve as a bridge between faculty and students to give faculty more insight into the student’s unique perspectives and varied experiences.14 It also serves as an opportunity to expose students to a diverse teaching team that is more reflective of the student body, which may funnel down into a more diverse STEM workforce.11

Broadening access to the LA role may help close the equity gap in undergraduate STEM disciplines (like chemistry) by increasing the representation of HUGs in STEM teaching teams while also increasing the capabilities, confidence, and likelihood of diverse students in the LA role to consider pursuing a STEM career path.11 Entry to the LA role is typically filtered by a recruiting process organized by an LA Program Director and/or the course instructor.26 Faculty recruit students from their classrooms to apply by targeting individual students or making announcements more broadly to the entire class.26 Current LAs often encourage students in their classes to apply for the role. LAs are generally paid a stipend near minimum wage salary or given course credit and must enroll in a pedagogy course during their first semester as an LA.12 However, the recruitment and selection processes may hold unintended barriers for diverse student entry that decrease the likelihood that these students will benefit from the LA role. Systemic inequities, including unconscious bias on the part of the recruiters and stereotype threat on the part of the students, may inadvertently perpetuate the equity gap in the recruitment process.7

Therefore, it becomes prudent to investigate if there is equity in students’ perceptions of themselves in this role and the subsequent barriers preventing them from taking advantage of the opportunity to apply to the program. According to the Theory of Planned Behavior (TPB), students’ intention to enroll in a program (or engage in any other behavior) is strongly influenced by their perception of the control they have over that behavior (self-efficacy, time, resources).27 That is, to benefit from the LA leadership experience, students must perceive they can fit that role and have full volition to engage in it before they intend to apply to the program. In some cases, limited perceived behavioral control overpowers students’ intent to engage in a behavior and is worthy of exploration.27 This project investigates inequities in which students see themselves as future chemistry LAs by demographic group and identifies barriers to becoming LAs that prevent students from accessing the benefits of the LA Program.

Purpose of This Study

LA programs are typically designed to increase success for all students in LA-supported courses that may serve as a gateway to the STEM industry and create a sense of belonging and community for the LAs.28 Although some basic recruiting strategies exist in the LA literature,26 instructors/program coordinators may inadvertently perpetuate the equity gap if they are unaware of which groups may experience inequities in perceiving this opportunity as attainable and valuable. This short-term longitudinal case study seeks to evaluate if demographic (race, gender, age group) inequities arise over a semester in students seeing themselves as future chemistry LAs at a moderately selective institution with a diverse student population. In addition, qualitative surveys were conducted to determine why some students do not self-identify as future LAs and are less likely to benefit from this support system. To date, no work has explored which students see themselves in this leadership role. Hence, this work aims to (1) identify if there are demographic inequities in students seeing themselves as future LAs and (2) determine barriers that prevent students from seeing themselves as LAs. The implication of this work is to help faculty and LA program coordinators develop approaches to minimize future obstacles and promote equity in recruiting practices as well as encourage researchers to examine equity of participation in LA Programs across the nation.

Furthermore, the experiences of students from HUGs at selective, Predominantly White Institutions may not reflect the experiences of those at institutions with a diverse student body. This data is collected from students at a metropolitan institution serving a diverse population: 61% nontraditional students (>22 years old), 43% HUGs, and 60% female.29 In addition, prior research on the LA Model has been derived mainly from physics courses within a small group of researchers.14 This work seeks to expand on previous research in the chemistry context and contribute to the broader picture of inclusion at urban institutions. These findings may help spur more research that will lead to the development of site-specific recruiting strategies tailored to the discipline of chemistry at more diverse institutions where there is a greater representation of women and students from HUGs.

This study will attempt to address the following research questions:

  • RQ 1: How do students from various backgrounds perceive the opportunity of becoming a chemistry LA?

  • RQ 2: What barriers do students identify when considering becoming future chemistry LAs?

Positionality

This study’s principal investigator was RNK; her advisor was MTB. RNK identifies as a cisgender first-generation woman from a historically underserved group (Middle Eastern). MTB identifies as a continuing generation cisgender White man.

Methodology

This exploratory mixed-methods study helps identify inequities in whether students see themselves as future chemistry LAs and attempts to find commonly perceived barriers to entry into the LA Program. Data were collected in the spring semester of 2021. Institutional Review Board (IRB) approval was obtained. Student participants were enrolled in first-year chemistry courses at a diverse metropolitan research university in the southern United States.

Courses Surveyed

Two courses (3 sections total) taught by the first author (RNK) were surveyed for this study in the spring semester of 2021 after obtaining IRB approval. Since RNK was also the lead researcher, the coauthors (IA and KEF) collected the data and provided the deidentified student data to the first author to avoid conflict of interest and biases in student grading.

The Fundamentals of General, Organic, and Biological Chemistry (GOB, 85 students) course is a core university course composed primarily of nursing students and nonchemistry majors who are not assumed to have prior chemistry knowledge. There is no required placement test or chemistry prerequisites. The General Chemistry 1 (GC1, 55 students) course is a university core course comprised primarily of chemistry and preprofessional majors. There is a required placement test or a minimum of a 24 on the ACT science reasoning test for entry to GC1; therefore, students are assumed to have some basic math and chemistry knowledge upon entry into the course. The population of these two courses differs, which is further discussed in the Results and Discussion section.

Both courses are four credit hours. All students must enroll in a 3-h weekly lab session led by a Graduate Teaching Assistant (GTA) or faculty instructor and a weekly 1-h workshop session facilitated by an LA. All the classes, laboratories, and workshops in this study were taught during the COVID-19 pandemic using the same course modality: synchronous virtual sessions via Zoom. In addition, the instructor employed active learning techniques in class, including iClickers (live student response system) and smaller breakout group sessions where students worked through challenging problems facilitated by an LA or GTA. Class participation was part of the course grade. The weekly workshop aims to provide students with additional opportunities to collaborate as they work through a Process Oriented Guided Inquiry Learning (POGIL)30 or similar activity that the instructor has selected to supplement student learning and address misconceptions.

Sample Demographics

Student self-identified demographics, including race, gender, and age, were collected after obtaining consent. Race was then condensed into two categories, students from (1) HUGs (Black, Hispanic/Latina/o, and Multiple/Other Race) or (2) non-HUGs (White and Asian).31 Because the Asian experience may not parallel the White experience, descriptive statistics were conducted to justify the categorization used in previous literature. Students who self-identified as Asian had a higher median course grade (N = 8, median = 86.7, IQR = 5.19) than their peers who self-identified as White (N = 56, median = 79.8, IQR = 11.3) or from HUGs (N = 29, median = 75.7, IQR = 21.0). The descriptive comparison also showed Asian students were more likely to see themselves as LAs at the end of the semester (identified as LA Score that is described in the Methodology section below; N = 8, median = 4.5, IQR = 1.8) than their peers who self-identified as White (N = 39, median = 3, IQR = 2) and HUGs (N = 18, median = 2, IQR = 2). While the complex dimensions of human identity cannot be oversimplified, based on historical context, previous literature,25,31 and some descriptive trends within the study sample, it was determined to categorize Asian students as non-HUGs for this study. Lastly, if students identified as White/Asian, they were classified as non-HUGs. But if they identified as multiracial with one or more identities falling into the HUG category, they were categorized as HUGs.

While the survey had a nonbinary set of gender options (Male, Female, Nonbinary/third gender), the number of nonbinary student participants was small enough to limit reporting on this sample for privacy concerns. The authors recognize that sex refers to the individual’s genetic makeup, and gender refers to how that individual identifies based on social constructs. When administering the survey, the researchers intended to have students self-identify their gender. They understood that female/woman and male/man might be used interchangeably if the researchers explicitly asked students for gender, not sex. One may identify as male/masculine/man, although he/they may have the genetic makeup (i.e. sex) of a female. Although it would have been more consistent with associating man/woman with gender (and will be done for future studies), the dynamic and changing landscape of the topic may not have imposed a detrimental limitation to the interpretation of results in this study.32 Lastly, age was condensed into two categories (1) traditional age (18–22) and (2) nontraditional age (>22).31

LA Score Data Collection and Method of Analysis

By week three, students would have learned about what the LA does and settled into a routine. At this time (referred to as “Early-Semester” or ES), students were asked a Likert-type question that is referred to as their “LA Score. " It read: “In the future, I could see myself working closely with a professor or graduate student as a Learning Assistant,” which was adapted from previous work on belonging and involvement.33 Students were asked to rate their LA Score on a scale from 1–6 (1, strongly disagree; 2, disagree; 3, somewhat disagree; 4, somewhat agree; 5, agree; 6, strongly agree). The LA Score question was then followed by an open-ended prompt to explain their choice regarding whether they see themselves as future LAs. Ninety-nine students consented to participate and responded to the ES survey and the open-ended response (GOB = 58, GC1 = 41). For qualitative analysis of the ES open-ended responses, all participants were deidentified by replacing the student ID numbers with a number from a random number generator. The datasets were then randomly assigned to four subsets that were analyzed sequentially by the research team. The research team used the first data subset of the ES data to develop emerging themes while documenting the rationale in separate spreadsheets.36 The research team met several times to discuss the emerging themes and refine them through deliberation. Once the themes from the first data set were finalized, the research team used the newly established themes to code the second data subset independently. Because it was possible to have more than one theme per student, some students had multiple observations.

To assess inter-rater reliability for independent coding, Fliess’s Kappa was calculated using RStudio.37 Fliess’s Kappa value for interobserver reliability for m number of raters is more psychometrically sound than percent agreement because it accounts for random theme selection.38 A kappa value of 0.61–0.80 means substantial agreement, and 0.81–1.00 is near-perfect agreement.39 Initial analysis of the second ES data set yielded fair agreement for three raters and 38 observations (kappa = 0.292, z = 8.89, p < 0.0001). The research team met again to discuss the data and re-evaluate until substantial agreement was reached (kappa = 0.95, z = 7.99, p < 0.0001, N = 38). The process was repeated for the remaining two ES data sets with two raters. Again, discussions were held until a high level of agreement was reached (third data set: kappa = 0.93, z = 9.02, p < 0.0001, N = 43; fourth data set: kappa = 0.98, z = 10.22, p < 0.0001, N = 44).

The same Liker-type LA Score question was asked during week 12 of 14, referred to as “Late Semester” (LS). Seventy-five students responded to the questionnaire: (GOB = 44, GC1 = 31), with 46 students citing they do not see themselves as future LAs (GOB = 29, GC1 = 17) by answering 1–3 on the Likert-type question. These students were then asked to select why they did not see themselves as future LAs with a list of options to choose from (multi-select) based on the themes developed from the ES responses.

Sixty-nine students responded to both ES and LS surveys that was used as the final sample for the quantitative longitudinal analysis. This included thirty-nine 39 GOB (45.88% response rate) and thirty 30 GC1 students (54.55% response rate). The final sample (N = 69) demographics were 29.85% HUGs, 64.18% female, and 34.78% nontraditional students. Since neither normality nor variance assumptions can be made about the distribution of the discrete variable, nonparametric tests were conducted for all quantitative analyses (one- and two-sample Wilcoxon test and Wilcoxon effect size).34–39

Results and Discussion

To address the first research question of whether there was equity of perceived opportunity in being part of the LA Program over the course of the semester, only students that completed both the ES and LS LA Score measures were compared, resulting in a smaller but matched data set (N = 69). This included 39 GOB students (45.88% response rate) and 30 GC1 students (54.55% response rate). Because the course can be a confounding variable that may determine whether students see themselves as future LAs, the LA Score was compared between both courses at each time point. There was no significant difference in LA Score either early (W = 481, p = 0.202) or late (W = 456, p = 0.113) in the semester based on the course, Table 1. Hence, it was justified to aggregate both courses for further analysis to determine how the LA score changed over the semester and how that differed based on demographic factors.

Table 1. Comparison of Students’ LA Scores Early (ES) and Late (LS) in the Semester Based on Course and Demographic Factors (Race, Gender, Age Group).

Time Point Course/Demographic N Median IQR 2-Sample Wilcoxon Statistic p-value Wilcoxon Effect Size
ES GOBa 39 3.00 2.00 481 0.202 0.154 (small)
GC1a 30 4.00 2.00
LS GOBa 39 3.00 2.00 456 0.113 0.192 (small)
GC1a 30 3.00 1.75
ES HUGs 20 3.00 2.00 368 0.158 0.173 (small)
Non-HUGs 47 3.00 3.00
LS HUGs 20 2.00 2.00 236 0.00114 0.398 (moderate)
Non-HUGs 47 4.00 3.00
ES Femaleb 43 3.00 2.00 457 0.436 0.0960 (small)
Maleb 24 3.00 2.25
LS Femaleb 43 3.00 2.00 406 0.142 0.180 (small)
Maleb 24 3.50 1.00
ES Nontraditional 24 3.00 3.00 592 0.509 0.0804 (small)
Traditional 45 3.00 2.00
LS Nontraditional 24 3.00 3.25 502 0.635 0.0318 (small)
Traditional 45 3.00 2.00
Open in a new tab

An inequity in LA Score between HUGs developed late in the semester (LS), although it did not exist earlier(ES). Equity in LA Score early and late in the semester was also examined across gender and age, but no significant differences were detected.

b

To remain consistent with the questionnaire that asked students to identify their gender as male/female/nonbinary, the authors use these terms in this paper but recognize that gender/sex are different constructs.

a

Because the LA Score did not differ between the two courses at either time point, both courses were aggregated for subsequent analysis.

Overall, there was no statistically significant difference between the ES LA Score (median = 3, IQR = 3) and LS LA Score (median = 3, IQR = 2, W = 576, p = 0.205, N = 69, Figure 1A). In aggregate, students did not experience a significant change in their perception of fitting the LA role at the end of the semester compared to the beginning (Figure 1A). However, further inspection revealed a more nuanced understanding of the variation of that trend based on demographic factors over the semester (Figure 1B and 1C).

Figure 1.

Figure 1

Open in a new tab

Longitudinal analysis of LA score distributions over the semester and delineated by race. Only matched ES (Early semester) and LS (Late-semester) LA Scores were used for the longitudinal analysis (N = 69, GOB = 39, GC1 = 30). (A) In aggregate, students’ LA Score did not show a change across the semester. (B) When separated by race, a disparity develops at the end of the semester between HUGs (N = 20) and non-HUGs (N = 47) that was not detected at the beginning of the semester (Table 1). (C) The change in LA Score distributions for non-HUGs significantly differed from HUGs over the semester (W = 320, p = 0.0334, N = 69, Wilcoxon’s effect size = 0.261, small). While LA Scores for students from non-HUGs remained unchanged, students from HUGs experienced a significant decline in LA Score over the course of the semester (Table 1).

Early in the semester, students from HUGs (N = 20) and non-HUGs (N = 47) started as being equally likely to see themselves as LAs. However, toward the end of the semester, an inequity in LA Score among students from HUGs and non-HUGs developed (Figure 1B, Table 1). Students from non-HUGs were significantly more likely to see themselves as LAs at the end of the semester (median = 4, IQR = 3, N = 47) than their HUG counterparts (median = 2, IQR = 2, N = 20, W = 236, p = 0.0011) with a medium effect size (Wilcoxon effect size = 0.398). No such disparity developed based on gender or age group (Table 1).

To assess if there was an equitable change in LA Score over the semester, the raw difference between students’ ES/LS LA Score was calculated and compared across races (Figure 1C). The change in LA Score distributions for non-HUGs was significantly different from the change in LA Score distribution for HUGs (W = 320, p = 0.0334, N = 69, Wilcoxon’s effect size = 0.261, small). In other words, regardless of their starting point, students from non-HUGs did not experience a significant change in seeing themselves as LAs over the semester (a median change = 0, IQR = 2, Figure 1C). However, students from HUGs experienced a significant decline in seeing themselves as LAs (a median change = −0.5, IQR = 1.25, Figure 1C). Hence, an inequity in how students perceived the LA opportunity developed for students from HUGs after one semester of chemistry, suggesting that they are less likely to have the intention of applying to the LA program and reap the associated benefits. There was no significant difference in the change of LA Score based on gender (W = 192, p = 0.71) or age group (W = 224, p = 0.225).

Perceived Barriers to Participation in the Learning Assistant Program

To answer the second research question on the perceived barriers to entry into the LA program that may influence students’ intent to enroll, researchers used the early semester open-ended responses to code the reasons students gave for their LA Score (Table 2). During the thematic analysis of student responses, the research team found seven overall themes: Time, Self-Efficacy: Knowledge, Self-Efficacy: Social Skills, Utility Value: Benefit, Interest, and Fit. The codes were then condensed into positive themes (i.e., strength in content knowledge) and negative themes (i.e., lack of content knowledge). “Time” was the only response not coded as positive or negative (Table 2). Since this paper intended to find barriers to entering the LA program, the focus was on the negative responses (reasons students did not see themselves as future LAs) to help identify areas to be remedied or addressed in the recruiting/program development process.

Table 2. Emergent Themes Describe Why Students Do Not See Themselves as Future Chemistry LAs.

Theme Descriptors Sample Quotes
Time Time constraints, heavy course load, work, family, etc. “I feel as though I barely have time as it is for work and school. So I don’t think I could realistically find the chance to work as a LA.”
Self-Efficacy: Knowledge Lack of chemistry knowledge “I feel as though I will not know enough to help others.”
Self-Efficacy: Social Skills Lack of communication skills, social skills, and/or confidence “I am not good at teaching things, even things I feel like I know well.”
“I’m not the most outgoing person. I’m not a complete introvert either, so it’s not out of the realm of possibility.”
Utility Value: Benefit Lack of benefit to the career path and/or reward “The reason for my answer is because my career path is nursing and I just do not plan to work alongside a professor.”
Fit Lack of fit “I do not have the confidence to work as a Learning Assistant. It’s not my thing.” (Also coded under “Lack of Confidence”)
Interest Lack of commitment and/or interest “I’m not really interested, but it also depends on the subject.”
“I don’t want to put in the time.”
Open in a new tab

“Time” was used to describe (1) lack of availability (heavy course load) and (2) life circumstances (family life or work outside of school), indicating that overall time constraints were perceived as a barrier to joining the LA program. “Self-Efficacy: Knowledge”, or lack thereof, was another theme that emerged to describe the feeling that students “will not know enough to help others”. “Self-Efficacy: Social Skills” was used to code students who identified lacking communication skills or confidence in leading others or simply not enjoying working with others. A code emerged related to students’ fit as it applies to seeing themselves as a future LA. For example, “I do not have the confidence to work as a Learning Assistant. It’s not my thing” was coded under “Self-Efficacy: Social Skills” and “Fit” because it implies that the student does not see themselves fitting that role. However, this was not coded under “Knowledge” because it was not apparent from the statement that the lack of confidence resulted from a lack of chemistry knowledge. Some of the early coding categories, like lack of benefit (not beneficial to degree path) and lack of reward (not a rewarding experience), were binned together as one theme (“Utility Value: Benefit”) since the overarching principle is whether the experience is beneficial and valuable to the students, despite “reward” being more intrinsically driven than “benefit.” The last emerging theme was “Interest,” which described students who were not interested in the subject of chemistry but perhaps could be interested in being an LA for another subject: “I am not really interested, but it also depends on the subject.” Students that did not “want to put in the time” was interpreted as lack of interest in either being an LA or the subject of chemistry (as opposed to did not have the time) and, therefore, binned under “Interest.”

At the end of the semester, students were again asked to rate whether they see themselves as a future LA and select a/the reason(s) for their answer from a multiple-select question that included all the subthemes that the research team identified from coding at the beginning of the semester (Table 2). For example, if students responded with some level of disagreement (LA Score of 1–3), they were directed to select from the subthemes of (lack of) Time, Knowledge, Benefit, Social Skills, Fit, Interest, or other.

A frequency plot was generated to compare the barriers most selected by students to describe why they perceived the chemistry LA role as something they did not want to engage with. (Figure 2). Although there was a total of 46 out of 75 students (GOB = 29, GC1 = 17) who responded as not seeing themselves as future LAs, there were 150 total codes because many students selected more than one response. Therefore, percentages were calculated based on the total number of observations to assess overall trends. For example, 47 observations were cited by students in the time category out of 150 responses (31.33%). So, for all demographic groups, the lack of time was the most common theme (31.33%) that emerged in students not seeing themselves as LAs, followed by Self-Efficacy: Knowledge (16.67%), Self-Efficacy: Social Skills (16.00%), Utility Value: Benefit (12.67%) and Interest (12.67%), and Fit (8.67%).

Figure 2.

Figure 2

Open in a new tab

Frequency of perceived barriers to participating in the LA program. Summary of Late-Semester thematic responses, delineated by race, specifically students from HUG and non-HUG (N = 46, 150 observations). Percentages were calculated based on each group’s total number of observations [i.e. 20 out of 72 (27.78%) observations from students from HUGs cited time as a factor that prevented them from seeing themselves as future LAs].

Since an inequity in the change of students’ perceptions in seeing themselves as LAs was detected for students based on race (not gender or age group) in the longitudinal analyses, differences in students' reasons were also examined based on race. Percentages were calculated based on each group’s total number of observations. For example, 20 out of 72 (27.78%) observations from students from HUGs cited time as a factor that prevented them from seeing themselves as future LAs. There was a minimal difference in the LS reasons for students not seeing themselves as future chemistry LAs. However, students from non-HUGs tended to cite a lack of “Time” (34.62%) more frequently than students from HUGs (27.78%). In addition, students from HUGs tended to cite a lack of “Interest” more often (11.11%) compared to non-HUGs (6.41%). Otherwise, the remaining themes were mentioned equally for both groups. A chi-squared test showed that the reasons for students not seeing themselves as future chemistry LAs were not associated with race (X2 = 1.98, p = 0.9611, df = 7).

Conclusions

Summary of Findings

The Theory of Planned Behavior posits that students are more likely to engage in a behavior (such as joining the LA Program) if they have perceived behavioral control (time, opportunity, resources, etc.) and intent in engaging in that behavior. This short-term longitudinal case study explores students’ perceived behavioral control by focusing on whether students see themselves in this role and the perceived barriers that prevent them from taking advantage of the benefits of this opportunity.

Although students’ LA Score distribution did not change throughout the semester for the aggregated matched student population, further analysis revealed that the trend differed based on race. Students from HUGs started as equally likely to see themselves as future chemistry LAs at the beginning of their chemistry course. Yet, they were significantly less likely to see themselves as future LAs toward the end of the semester when compared to students from non-HUGs. This significant decline in the LA score distribution for students from HUGs over the semester was not observed for students from non-HUGs. Neither gender nor age group mirrored this trend in the study sample. Therefore, leaders/researchers in the Learning Assistant community may need to assess the equity of recruiting and participation in LA programs on a larger scale. If this trend is more prevalent at other institutions and disciplines, it may be worthwhile to hold sessions at the International Learning Assistants Conference (ILAC)12 to develop recruiting strategies that minimize systemic inequities in participation.

Qualitative data identified that lack of time was the most prevalent barrier for students not perceiving the LA role as an option. However, this study sample did not show variation in the reasoning behind students’ LA Scores based on race. A better understanding of the students’ lived experiences should be explored to further understand how to open the gateway for students that would benefit the most from these mentorship roles. It could be that these students are experiencing similar challenges that span across all races (socioeconomic status, childcare, work). Hence, similar interventions, such as financial compensation, could increase opportunities for all students. Providing competitive stipends will likely help students benefit from this opportunity, allowing those who typically cannot afford to volunteer their time to participate in this leadership opportunity in their educational career.

Lastly, qualitative findings (Table 2 and Figure 2) also suggest that future work needs to focus on assessing students’ Self-Efficacy, Utility-Value, Belonging, Interest, and performance in chemistry to provide more insight into the factors that influence the likelihood that students see themselves as future chemistry LAs. Perhaps the learning environment created by the instructor, both in faculty mindset and teaching practices, may result in different rates of decline in student LA Scores and should be studied further.40,41 Instructors and LA program coordinators may need to actively market the benefits of being an LA and make the path to becoming an LA transparent, especially for students from HUGs who may not see themselves fitting that role.

Implications

Higher education is currently undergoing a transformative phase by reflecting on the equity of existing systems and what parts of the system result in lower success for students from HUGs. LA programs are one potential tool that can be used to close the equity gap in chemistry and other STEM disciplines, providing support and mentorship not only to students in high attrition classes but also to those that have begun to move into higher-level courses (the LAs themselves). Examining how different groups of students see themselves engaging in these programs is becoming increasingly important. This involves learning what factors influence student decisions, especially at institutions with diverse student populations.

This pilot study indicates that students from HUGs may not be as likely to utilize this lever for success if they lack perceived behavioral control over their actions.27 That is, if they do not see themselves as fitting that role and lack the time or self-efficacy to participate in the LA Program, they are less like to make intentions to engage in that behavior.27 Thus, when institutions continue to operate under the assumption that all students are equally likely to engage in leadership experiences as long as they are provided to them, this may inadvertently propagate systemic inequities they are attempting to dismantle with such reform efforts. This study only begins to question the demographic characteristics of the students who may see themselves as future LAs and the barriers preventing them from applying to the program. Hence, future studies should examine if the demographic representation of the LA population is reflective of student populations at various institutions.

This study may increase the likelihood that faculty/directors may choose inclusive practices when recruiting students to participate in the leadership role of an LA. In addition, qualitative data suggests that time is a substantial factor for students that choose not to apply to be LAs, and administrators need to consider funding mechanisms to pay LAs a competitive rate so they may take on the role of a part-time job. This may allow students from diverse backgrounds to carve out a few hours a week from their other jobs that may be unrelated to their careers to develop their academic skills.

This work shows that there is much more to learn about the relationship between perceived opportunities or barriers to participation in the Learning Assistant program and whether that translates into actual inequities of representation.

Limitations

This study collected data from a single institution from a single instructor and aggregated student data from two courses (GOB and GC1) with different student populations. More data must be collected and compared with other instructors and courses across similar institutions to increase generalizability. Affective factors such as a sense of belonging, self-efficacy, utility value, interest, and course grade may help explain the factors that influence students seeing themselves as future LAs.

This data was collected during the global COVID-19 pandemic, in which classes were taught on Zoom, not in a historically traditional classroom. This may have impacted students seeing themselves as future LAs during a volatile time in which disruptions in life circumstances were prevalent.

Missing data presented a substantial challenge because the data were collected over a semester across multiple sections of students. The rate of missing data across sections varied from 14.4% to 38.7%, rendering a simple mean imputation for missing data inadequate. Multiple Imputations by Chained Equations (MICE) method was also considered to minimize bias in missing data. However, the primary output variable (LA Score) was an ordinal categorical variable (vs a continuous variable), the data were not normally distributed, and the method of analysis included repeated measures analysis. This limited the use of MICE imputation, especially for a small sample size. Future studies should focus on collecting data from students who withdraw from courses to avoid selection bias and use bigger sample sizes to increase statistical power and generalizability.

Acknowledgments

The authors thank the Promoting Active Learning & Mentoring (PALM) Network for funding undergraduate researchers for the project (RCN-UBE NSF Grant #1624200), the students who completed the surveys, Ruby Trotter, Dr. Michael Moore, Dr. Jack Barbera, Dr. Bud Talbot, and Zachary Stickley for providing their feedback on this project.

The authors declare no competing financial interest.

References

  1. Green D. Historically Underserved Students: What We Know, What We Still Need to Know. New Dir. Community Coll. 2006, 2006 (135), 21–28. 10.1002/cc.244. [DOI] [Google Scholar]
  2. Women, Minorities, and Persons with Disabilities in Science and Engineering: 2021 | NSF - National Science Foundation. https://ncses.nsf.gov/pubs/nsf21321/report/about-this-report (accessed on November 12, 2021).
  3. Brewer C. A.; Smith D.. Vision and Change in Undergraduate Biology Education: A Call to Action; American Association for the Advancement of Science: Washington, DC, 2011; Vol. 81.
  4. National Academies of Sciences, Engineering, and Medicine. Call to Action for Science Education: Building Opportunity for the Future, 2021. https://www.nap.edu/catalog/26152/call-to-action-for-science-education-building-opportunity-for-the. [PubMed]
  5. Race and Ethnicity in Higher Education. https://www.equityinhighered.org/ (accessed on November 12, 2021).
  6. Master A.; Meltzoff A. N. Cultural Stereotypes and Sense of Belonging Contribute to Gender Gaps in STEM. Int. J. Gender Sci. Technol. 2020, 12 (1), 152–198. [Google Scholar]
  7. Cohen G. L.; Garcia J. I Am Us”: Negative Stereotypes as Collective Threats. J. Pers. Soc. Psychol. 2005, 89 (4), 566–582. 10.1037/0022-3514.89.4.566. [DOI] [PubMed] [Google Scholar]
  8. Good C.; Rattan A.; Dweck C. S. Why Do Women Opt out? Sense of Belonging and Women’s Representation in Mathematics. J. Pers. Soc. Psychol. 2012, 102 (4), 700–717. 10.1037/a0026659. [DOI] [PubMed] [Google Scholar]
  9. Rattan A.; Savani K.; Komarraju M.; Morrison M. M.; Boggs C.; Ambady N. Meta-Lay Theories of Scientific Potential Drive Underrepresented Students’ Sense of Belonging to Science, Technology, Engineering, and Mathematics (STEM). J. Pers. Soc. Psychol. 2018, 115 (1), 54–75. 10.1037/pspi0000130. [DOI] [PubMed] [Google Scholar]
  10. Grunspan D. Z.; Kline M. A.; Brownell S. E. The Lecture Machine: A Cultural Evolutionary Model of Pedagogy in Higher Education. Life Sci. Educ. 2018, 17 (3), es6 10.1187/cbe.17-12-0287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Kricorian K.; Seu M.; Lopez D.; Ureta E.; Equils O. Factors Influencing Participation of Underrepresented Students in STEM Fields: Matched Mentors and Mindsets. Int. J. STEM Educ. 2020, 7 (1), 16. 10.1186/s40594-020-00219-2. [DOI] [Google Scholar]
  12. Learning Assistant Alliance. https://learningassistantalliance.org/ (accessed on April 3, 2022).
  13. Dawson P.; van der Meer J.; Skalicky J.; Cowley K. On the Effectiveness of Supplemental Instruction: A Systematic Review of Supplemental Instruction and Peer-Assisted Study Sessions Literature Between 2001 and 2010. Rev. Educ. Res. 2014, 84 (4), 609–639. 10.3102/0034654314540007. [DOI] [Google Scholar]
  14. Barrasso A. P.; Spilios K. E. A Scoping Review of Literature Assessing the Impact of the Learning Assistant Model. Int. J. STEM Educ. 2021, 8 (1), 1–18. 10.1186/s40594-020-00267-8. [DOI] [Google Scholar]
  15. Van Dusen B.; Nissen J. Associations between Learning Assistants, Passing Introductory Physics, and Equity: A Quantitative Critical Race Theory Investigation. Phys. Rev. Phys. Educ. Res. 2020, 16 (1), 15. 10.1103/PhysRevPhysEducRes.16.010117. [DOI] [Google Scholar]
  16. Van Dusen B.; Nissen J. Equity in College Physics Student Learning: A Critical Quantitative Intersectionality Investigation. J. Res. Sci. Teach. 2020, 57 (1), 33–57. 10.1002/tea.21584. [DOI] [Google Scholar]
  17. Otero V.; Pollock S.; Finkelstein N. A Physics Department’s Role in Preparing Physics Teachers: The Colorado Learning Assistant Model. Am. J. Phys. 2010, 78 (11), 1218–1224. 10.1119/1.3471291. [DOI] [Google Scholar]
  18. Close E. W.; Close H. G.; Donnelly D.. Understanding the Learning Assistant Experience with Physics Identity; Philadelphia, PA, USA, 2013; pp 106–109. 10.1063/1.4789663. [DOI] [Google Scholar]
  19. Close E. W.; Conn J.; Close H. G. Becoming Physics People: Development of Integrated Physics Identity through the Learning Assistant Experience. Phys. Rev. Phys. Educ. Res. 2016, 12 (1), 010109. 10.1103/PhysRevPhysEducRes.12.010109. [DOI] [Google Scholar]
  20. Nadelson L. S.; Finnegan J. A Path Less Traveled: Fostering STEM Majors’ Professional Identity Development through Engagement as STEM Learning Assistants. J. Higher Educ. Theor. Pract. 2014, 14 (5), 29. [Google Scholar]
  21. Deslauriers L.; McCarty L. S.; Miller K.; Callaghan K.; Kestin G. Measuring Actual Learning versus Feeling of Learning in Response to Being Actively Engaged in the Classroom. Proc. Natl. Acad. Sci. U. S. A. 2019, 116 (39), 19251–19257. 10.1073/pnas.1821936116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Yoder J. D.; Hochevar C. M. Encouraging Active Learning Can Improve Students’ Performance on Examinations. Teach. Psychol. 2005, 32 (2), 91–95. 10.1207/s15328023top3202_2. [DOI] [Google Scholar]
  23. Theobald E. J.; Hill M. J.; Tran E.; Agrawal S.; Arroyo E. N.; Behling S.; Chambwe N.; Cintrón D. L.; Cooper J. D.; Dunster G.; Grummer J. A.; Hennessey K.; Hsiao J.; Iranon N.; Jones L.; Jordt H.; Keller M.; Lacey M. E.; Littlefield C. E.; Lowe A.; Newman S.; Okolo V.; Olroyd S.; Peecook B. R.; Pickett S. B.; Slager D. L.; Caviedes-Solis I. W.; Stanchak K. E.; Sundaravardan V.; Valdebenito C.; Williams C. R.; Zinsli K.; Freeman S. Active Learning Narrows Achievement Gaps for Underrepresented Students in Undergraduate Science, Technology, Engineering, and Math. Proc. Natl. Acad. Sci. U. S. A. 2020, 117 (12), 6476–6483. 10.1073/pnas.1916903117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Freeman S.; Eddy S. L.; McDonough M.; Smith M. K.; Okoroafor N.; Jordt H.; Wenderoth M. P. Active Learning Increases Student Performance in Science, Engineering, and Mathematics. Proc. Natl. Acad. Sci. U. S. A. 2014, 111 (23), 8410–8415. 10.1073/pnas.1319030111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Bhattacharyya P.; Chan C. W. M. Can Undergraduate Research Participation Reduce the Equity Gap?. J. Scholarsh. Teach. Learn. 2021, 10.14434/josotl.v21i1.30462. [DOI] [Google Scholar]
  26. Schick C. P.Trying on Teaching: Transforming STEM Classrooms with a Learning Assistant Program. In ACS Symposium Series; Anna L. J., Higgins T. B., Palmer A., Owens K. S., Eds.; American Chemical Society: Washington, DC, 2018; Vol. 1280, pp 3–27. 10.1021/bk-2018-1280.ch001. [DOI] [Google Scholar]
  27. Ajzen I. The Theory of Planned Behavior. Organ. Behav. Hum. Decis. Process. 1991, 50 (2), 179–211. 10.1016/0749-5978(91)90020-T. [DOI] [Google Scholar]
  28. Goertzen R. M.; Brewe E.; Kramer L. H.; Wells L.; Jones D. Moving toward Change: Institutionalizing Reform through Implementation of the Learning Assistant Model and Open Source Tutorials. Phys. Rev. Spec. Top. - Phys. Educ. Res. 2011, 7 (2), 020105. 10.1103/PhysRevSTPER.7.020105. [DOI] [Google Scholar]
  29. Trojan Fast Facts - Institutional Research and Analytics - UA Little Rock. Institutional Research and Analytics. https://ualr.edu/institutionalresearch/trojan-fast-facts/ (accessed on March 21, 2023).
  30. Moog R. S.; Spencer J. N.. Process Oriented Guided Inquiry Learning (POGIL); ACS Symposium Series; American Chemical Society: Washington, DC, 2008; Vol. 994. 10.1021/bk-2008-0994. [DOI] [Google Scholar]
  31. Shortlidge E. E.; Rain-Griffith L.; Shelby C.; Shusterman G. P.; Barbera J. Despite Similar Perceptions and Attitudes, Postbaccalaureate Students Outperform in Introductory Biology and Chemistry Courses. Life Sci. Educ. 2019, 18 (1), ar3. 10.1187/cbe.17-12-0289. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Pryzgoda J.; Chrisler J. C. Definitions of Gender and Sex: The Subtleties of Meaning. Sex Roles 2000, 43, 553. 10.1023/A:1007123617636. [DOI] [Google Scholar]
  33. Knekta E.; Chatzikyriakidou K.; McCartney M. Evaluation of a Questionnaire Measuring University Students’ Sense of Belonging to and Involvement in a Biology Department. CBE Life Sci. Educ. 2020, 19, ar27. 10.1187/cbe.19-09-0166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Statistical Tests and Assumptions: The Best Reference. Datanovia. https://www.datanovia.com/en/courses/statistical-tests-and-assumptions/ (accessed on January 17, 2022).
  35. Burkholder E.; Walsh C.; Holmes N. G. Examination of Quantitative Methods for Analyzing Data from Concept Inventories. Phys. Rev. Phys. Educ. Res. 2020, 16 (1), 010141. 10.1103/PhysRevPhysEducRes.16.010141. [DOI] [Google Scholar]
  36. Saldaña J.An Introduction to Codes and Coding. In The Coding Manual for Qualitative Researchers, 3rd ed.; Sage: London, 2016; p 39. [Google Scholar]
  37. Fleiss’ Kappa in R: For Multiple Categorical Variables. Datanovia. https://www.datanovia.com/en/lessons/fleiss-kappa-in-r-for-multiple-categorical-variables/ (accessed on May 11, 2023).
  38. Watkins M. W. Interobserver Agreement in Behavioral Research: Importance and Calculation. J. Behav. Educ. 2001, 8. [Google Scholar]
  39. Stemler S. E. A Comparison of Consensus, Consistency, and Measurement Approaches to Estimating Interrater Reliability. Pract. Assess. Res. Eval. 2019, 10.7275/96JP-XZ07. [DOI] [Google Scholar]
  40. Canning E. A.; Muenks K.; Green D. J.; Murphy M. C. STEM Faculty Who Believe Ability Is Fixed Have Larger Racial Achievement Gaps and Inspire Less Student Motivation in Their Classes. Sci. Adv. 2019, 5 (2), eaau4734 10.1126/sciadv.aau4734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Turetsky K. M.; Sinclair S.; Starck J. G.; Shelton J. N. Beyond Students: How Teacher Psychology Shapes Educational Inequality. Trends Cogn. Sci. 2021, 25 (8), 697–709. 10.1016/j.tics.2021.04.006. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Chemical Education are provided here courtesy of American Chemical Society

RESOURCES