Teaching assistants (TAs) are playing a more important role in undergraduate education, especially in science, technology, engineering, and mathematics (STEM) disciplines. Specifically, large introductory STEM classes tend to utilize TAs to instruct laboratory and discussion sections. Although the exact number is not known, 97% of higher education institutions report using TAs to instruct undergraduate students, and 88% of biology TAs were assigned to teach introductory biology courses (Reeves et al., 2016). Additionally, for many STEM courses, retention drops during the first series of core discipline classes and can be a predictive source of students who will continue in that major throughout their college career (Kendall, Niemiller, Dittrich-Reed, & Schussler, 2014). It is during these STEM introductory courses that the importance of developing an effective TA training model becomes apparent.

In many higher education institutions, course-based research experiences (CBREs) are being introduced. CBREs encompass both undergraduate (CUREs) and nonundergraduate course-based undergraduate experiences. The course described includes both undergraduate and high school students participating in a CBRE. Auchincloss et al. (2014) distinguished CBREs from other undergraduate laboratory experiences including traditional, inquiry-based, and internship experiences. CBREs afford five key dimensions of laboratory education to a similar magnitude as internships, yet do so in a far more cost-effective manner given courses have the capacity to impact many more students than internships per instructor. These dimensions include (a) the use of scientific practices, (b) discovery, (c) broader relevance or importance, (d) collaboration, and (e) iteration (Auchincloss et al., 2014).

CBREs are designed as part of formal coursework to involve students in authentic research experiences (Weaver, Russell, & Wink, 2008). Laboratory courses facilitate students’ opportunities to engage in investigation and inquiry, but they can also encourage metacognition through collaboration and reflection (Ryker & McConnell, 2014). Rodenbusch, Hernandez, Simmons, and Dolan (2016) showed a significant impact of CBRE participation early in undergraduate education on graduation rates and majoring in STEM fields, though there was no effect on grade point average (GPA). CBREs are inherently student-centered courses and have been shown to increase student achievement as well as deeper understanding of the nature of science (Ryker & McConnell, 2014). Similarly, studies of undergraduate research experiences have been shown to increase the development, recruitment, and retention of undergraduate students particularly among minority students (Weaver et al., 2008). CBREs serve as unbiased points of entry into research opportunities as they provide access for all students, not just those who are invited into active research laboratories as undergraduates (Bangera & Brownell, 2014). An investigation into the longer term impacts of three semesters of CUREs on undergraduates revealed that they were more likely to graduate with a STEM degree compared with peers who had not engaged in CUREs for three terms, and they were more likely to graduate within 6 years, although there was no difference in GPAs (Rodenbusch et al., 2016).

Despite the vast literature singing the praises of CBREs, there are some drawbacks that can limit extensive implementation. First, many studies of CUREs in the literature are potentially impacted by selection bias as students usually chose to enroll in these courses (Auchincloss et al., 2014; Brownell, Kloser, Fukami, & Shavelson, 2013). Constraints to widespread implementation include potential prohibitory costs, variation in receptiveness in different student populations within and among diverse universities, and variation in effectiveness among instructors (Brownell et al., 2015).

Bangera and Brownell (2014) called for universities to mandate CUREs as introductory courses, yet the ability to scale these courses can be limited by the variation of instructor effectiveness (Brownell et al., 2015). As a result, this movement toward including more and more STEM classes to implement CBREs, especially at the introductory level, puts an even greater burden on TAs and requires TAs to become facilitators of the research process.

The aim of this study was to develop a model for teaching CBRE instructors, primarily TAs, how to guide students through an authentic research experience. This requires the TAs to operate their class as a primary investigator would operate their research lab; designing and investigating open-ended scientific questions. The researchers were offered the opportunity, through their institution, to design a community-based course that served both an undergraduate population and a local high school population with a biology experience. The community-based course utilized undergraduates as TAs (UTAs) and research mentors, and high school students as the undergraduate mentees. The researchers were the instructors of this course and acted as mentors to both the undergraduate and high school populations.

The course designed for this study reflects the methodology used to train laboratory TAs at the researchers’ institution. Typically, laboratory TAs meet once a week with a laboratory coordinator to run through the lab for the following week. This course was designed to test the researchers’ proposition that TA motivation for student-centered and self-regulated learning can be facilitated through an effective TA training model that fosters TAs’ identities as research mentors. To that end, this study describes the researchers’ attempt to effectively design a mentor model that supports CBREs.

Because of the aim of the course design, relevant literature includes scholarship examining CBREs, including CURE literature, research mentorship, and scaffolding.

Teaching material to another person intrinsically motivates the teacher and leads to greater conceptual learning and engagement in the material (Philipp, Tretter, & Rich, 2016). In Philipp et al.’s (2016) study of STEM UTAs, intervention showed that an undergraduate teaching assistantship program was correlated with increased self-confidence, communication skills, deeper content knowledge, and better-defined career goals (2016). Further, Feldon et al. (2011) described significant increases in overall capabilities to perform research by early-career STEM graduate students who teach compared with those who only do research. Specifically, they showed improvement in their ability to write testable hypotheses and to design experiments to test those hypotheses. TAs teaching CBREs would be mirroring the same procedure their principal investigators perform with mentoring their undergraduate and graduate students. From this view, TAs are guides facilitating the research process through iterative feedback rather than more traditional styles of laboratories and lectures. TAs embodying research mentors in CBREs should value their teaching experience as preparation for their future careers as STEM faculty (Feldon et al., 2011).

As Vygotsky (1978) described, scaffolding is the process of providing and adjusting support for learners with regard to their zone of proximal development. Scaffolding therefore becomes necessary for allowing students to achieve higher level functioning by aiding students when they are struggling and removing those aids as they become more competent and reach mastery (McCaslin, 2009). It also is known to increase students’ expectations of their own success and encourages persistence in learning and problem solving (Perry, Hutchinson, & Thauberger, 2008).

The UTAs are paired to encourage confidence and persistence during their teaching. For the first 4 weeks, the course design scaffolds the UTAs’ experience by allowing the instructors to model the lesson for the UTAs before the UTAs implement the lesson themselves. For example, during the course, the instructors perform the lesson for the UTAs on Monday, and on Friday the UTAs perform the lesson with the high school students. In this way, the UTAs learned by experiencing the lessons as students; they could ask questions and troubleshoot possible complications for their own instruction. Additionally, each week the UTAs were encouraged to discuss any issues of student engagement or pedagogy that arose during their instructions during the previous session. The instructors facilitated an open discussion so that both the researchers and the other UTAs could provide feedback. This weekly feedback session and the course’s weekly written reflections provide UTAs time to reflect on their theory and practice as recommended by Jacobs (2001).

Teaching orientations can be predictive of teaching practices and are thought to be the conceptual maps for instructional decision making (Gilmore, Maher, Feldon, & Timmerman, 2014). In Gilmore et al.’s (2014) study of STEM GTAs, they quantified teaching orientation into a longitudinal score that was compared with training, mentorship, prior teaching experience, and prior research experience; among these variables only mentorship showed statistically significant positive correlation with change in teaching orientation. The instructors began the course by asking the UTAs to reflect on their past teaching experiences as well as their experiences as a student. The instructors emphasized the importance of mentoring students as producers of knowledge rather than consumers of information. UTAs were instructed to ask questions rather than just giving students the answers and to seek out information with students rather than just saying, “I don’t know.” TAs were encouraged to adapt a student-centered orientation to their own personalities and teaching styles.

For the majority of TAs, instructing undergraduate courses are their first teaching experiences (Starr & DiMartini, 2015). In addition, TAs may be influenced by their personal experiences with teaching practices of their instructors. Each of these factors may play a significant role in TAs who are assigned to CBREs in which previous teaching experience may help or hinder how they engage with students. In addition, prior research experience may also provide a better foundation for TAs to scaffold students’ learning in CBREs. The course instructors surveyed the UTAs to assess their previous teaching experiences and motivations. UTAs were paired on the basis of the instructors’ understanding of their familiarity with facilitating learning and science content strength. Each teaching pair had at least one UTA who had experience teaching and one UTA who had taken science laboratory courses.

Research has shown that positive experiences with instructors may lead to increased retention, better student attitudes, and greater learning success (Kendall et al., 2014). Kendall et al.’s (2014) study of biology GTAs showed that student perceptions of TA teaching effectiveness were predicted by teaching techniques (being calm, being interested, keeping students’ attention, using good examples, providing relevant materials, interacting, and presenting information well) and interpersonal rapport (being compassionate and empathetic, being flexible and laid-back, being approachable and relatable). According to research performed by Hartnett, Romcke, and Yap (2003), students’ performance is positively associated with the instructor’s approachability. Research supports that TA behavior and approachability can positively affect student motivation to learn and lead to improved student performance (Hartnett et al., 2003). UTAs were encouraged to discuss topics beyond the scope of the course with their mentees, specifically about college life, majors, and professional aspirations.

There have been repeated requests for better TA trainings, and some institutions have implemented TA training courses (Zehnder, 2016). However, there is a wide range of formats and degree of effectiveness that vary by institution and departments (Gilmore et al., 2014). Despite the variety, TA trainings have been shown to increase teaching confidence and reduce anxiety (Zehnder, 2016). In addition, a TA training that supports social constructivist ideas, such as (a) TAs as learning guides, (b) TAs as aids to scaffold students’ learning, (c) TAs that support collaborative work, and (d) TAs that advocate open-ended learning, would best serve to support the utilization of CBREs in undergraduate education. Henry and Bruland (2010) argued that training TAs in positional reflexivity can enhance learner-centered teaching approaches. To that end, the researchers designed the course to implement their understanding of best practices for TA training by providing weekly modeling of the lessons for the high school students as well as weekly written reflections and in-class discussions.

The course described in this study is the first round of implementation of this CBRE. The CBRE focused on the scientific method and aimed to have students develop and design their own experiments. The course ran for an entire 10-week quarter during which there were two 2-hour meetings weekly, a Monday and a Friday class. The UTAs met twice a week (Mondays and Fridays)—once with only the researchers who were the instructors for the course (Mondays) and once with the high school students and instructors (Fridays). The UTAs met 20 times, and the high schoolers met 9 times.

The first 4 weeks of the 10-week course included the UTAs leading the high school mentees through a set of designed research experiments, which allowed both cohorts to become familiar with the model organism, Wisconsin Fast Plants, and the conceptual understanding and essential vocabulary to design and perform carefully controlled experiments. The second half of the class, Weeks 5–10, required the groups to design and test their own research project. Each group had unique scientific questions and refined their projects as they tested and collected data. The groups were encouraged to collectively ask a relevant scientific question that could be addressed within the remainder of the course. Each group had to define their scientific question and scope of the proposed experiments to the instructors. Groups then designed, executed, and interpreted well-controlled experiments.

During the first 4 weeks, on Mondays, when the UTAs would meet with the instructors, the instructors would model the upcoming lesson for the UTAs. The UTAs were then expected to lead the exact lesson with the high schoolers that following Friday. The instructors would actively aid the UTAs in their Friday lessons. During the last half of the course, Weeks 5–10, the instructors acted as guides and gave control to the UTAs, who would plan what their groups were doing each week. The course culminated with each group presenting their findings to the rest of the class, their parents, and administrators from both the undergraduate and high school institutions.

The UTAs were paired based on their teaching and laboratory experiences within the first week of the course. They were grouped with five high school students whom they were responsible for instructing and guiding through a research project. At the beginning of the course the instructors informed the undergraduate students that they would be acting as mentors and TAs. Figure 1 depicts the course’s tiered mentor model; the instructors served as the foundation for research mentoring, primarily providing support to the undergraduate TAs, but also mentoring the high school students, most significantly during the first 4 weeks of the course. The high school students primarily received research mentoring from the UTAs who helped design and test their research questions. Additionally, the instructors served as teaching mentors to the UTAs, and the UTAs provided the majority of the teaching to the high school students, with the instructors relinquishing their teaching role to the UTAs during the progression of the course.

The conceptual framework and course design led the researchers to ask the following research questions toward designing TA trainings.

1. Do TAs’ perception of their teaching skills change after participating in this tiered mentor model?
2. Do TAs report mentoring skills gains after participating in this tiered mentor model gain skills for mentoring students in a CBRE?
3. Do TAs perceive gaining a better understanding of scientific content and process after participating in this CBRE?

Each week, both sets of students, the UTAs and the high schoolers, were asked to reflect on specific aspects of that week’s lesson. Additionally, each week the instructors led a discussion with the UTAs about group improvements and challenges. The UTAs were encouraged to confer about instructional and content-specific issues. The instructors’ goal with this discussion was to allow for peer feedback and to offer guidance, as necessary. The discussions consisted of three framing questions: (a) what worked; (b) what didn’t work; and (c) what, if any, changes did the UTAs want to make in the upcoming lesson for the mentees. IRB approval was received through the researchers’ institution.

Five out of the six undergraduate students enrolled in the course participated in this study. UTA demographics are displayed in Figure 2. There were two females and three males. Fifteen high school students were selected by their principal to be in the course. Due to the research questions for this study, only the UTA reflections were analyzed.

Data analysis consisted of deductive coding based on the described conceptual framework (Creswell & Creswell, 2015). Data was organized, read and coded, and reread and recoded for a priori themes (Creswell & Creswell, 2015). Data sources included the UTAs’ final reflection assignment. Table 1 displays the four question prompts from the final reflection assignment. In vivo hand coding was used by one of the researchers.

Three main themes were analyzed from the qualitative analysis of the UTAs’ reflections: (a) teaching and instructional gains, (b) science content and process gains, and (c) mentoring gains. Table 2 displays the codebook created from the qualitative analysis.

All of the UTAs reported feeling much more confident teaching and learning how to effectively teach. Additionally, many of UTAs continued to explore teaching and mentoring after the course. One of the UTAs reflected: “I do feel more confident about my abilities to teach and work with students on science, which is something I would love to do as a career.”

One UTA reflected on scaffolding their mentee’s learning, writing:

I really learned how to efficiently guide them to a specific answer…instead of just telling them the answer…I would have them repeat it to me in their own words so they could better understand it. I found this method of teaching to be extremely beneficial in not only the students learning but also for myself in making sure I am explaining concepts in a clear, concise way.

The UTAs’ reflections suggest that they gained teaching and instructional skills. They report being able to explain concepts and check in with their mentees’ understanding. One UTA wrote that “the ability to be able to work with multiple groups with different backgrounds effectively is a hard but incredibly important skill to have.” Their reflections suggest that they were able to adapt their teaching to better suit the needs of their mentees and provide student-centered instruction.

All of the UTAs reported that they felt like they were mentors to the high school students. Reflecting on the course design, one UTA wrote, “I also think it was great that each group was able to explore a different concept and be in charge of their own learning,” and another wrote, “This course has given the [UTAs] a tremendous amount of leadership experience; having been mentors to the high school students and have them look up to the mentors as a role model was a phenomenal idea.”

Some of the UTAs focused on the trust they were able to build through their relationship with their mentees:

By being able to have both the teacher–student relationship on top of the student–teacher relationship, I saw how great the flow of work turns out and the level of trust that is between all parties.

Not only did I want to seem as their teaching assistant but I also wanted to be their mentor/friend. I felt that having that kind of environment made it more easy for students to not feel judged when asking questions but also not being afraid to speak up when they don’t understand something.

Many of the UTAs wrote about how they were able to establish trust in their groups and that their mentees were comfortable asking both science and general questions. They also reflected that they had control and freedom in their groups’ experiment and that this encouraged engagement from the mentees. Several UTAs also reflected on how they had established an environment for discussion and that they were learning alongside their high school mentees.

The UTAs also reported gaining experimental design, fundamental science, and experiment-specific content. One UTA wrote,

I was able to learn some new things about various sugars and the roles that they play in the life cycle of plants.…The most interesting thing that I learned from the results of our experiments was that there appeared to be a “sweet spot” for the concentration of sugar, much like how humans tightly regulate their blood sugar levels. Either too much or too little of a concentration of sugar appeared to be detrimental to plant growth. This was definitely a result that I was not expecting.

All the UTAs felt that they taught science content and skills efficiently to their mentees. They reported that their mentees were able to learn basic science, experimental design, and components of the scientific method. One UTA wrote,

Aside from getting to work in a lab, the students learned other useful skills such as notebook keeping, working in a team effectively, and learning how to present their findings to an audience. They learned how to calculate dilutions, which we all know is strangely difficult. But as much as I helped them expand their knowledge, they helped me expand mine.

Perhaps most important, the UTAs were able to share their passion for science. One UTA wrote,

I was pleased with how patient they were with the amount of quantitative data they had to take, and how well organized it was. They took excellent qualitative data (a weak point of mine), and kept good record of the dates.

Most of the UTAs reflected on the joy they felt seeing their mentees’ excitement for learning about science. A UTA noted: “We showed the students was that science can be really fun and exciting, but also that there’s a lot of parts that are very tedious, and detail oriented.” The UTAs all reported positive science content gains for themselves and the high school mentees.

These findings support the literature and the researchers’ proposition that TAs who consider themselves research mentors may lead to student-centered and self-regulated learning of their mentees. These findings reflect the self-reported gains of the TAs regarding their roles as teachers, scientists, and mentors. These factors are pivotal in addressing the challenges faced by CBRE implementation in higher education. Additionally, the researchers have described a novel tiered mentor model that employs a CBRE over a 10-week course. The researchers, who also served as instructors, seemed to successfully scaffold the UTAs’ orientation toward student-centered teaching. The UTAs, likewise, seemed to take on a mentor identity and self-regulate their own teaching strategies. The UTAs reflected on their own scaffolding techniques and their ability to relate to their mentees for better teaching opportunities. The UTAs also reflected on their ability to relay their love of science to their mentees. This is particularly important for developing TAs as research mentors in addition to teaching mentors. Although more research needs to be performed on this tiered mentor model, this study provides a foundation for developing an effective TA training model for TAs as research mentors specifically for CBREs.

This course was only 10 weeks long and consisted of six undergraduates and 15 high school students rather than the more common CBRE experience—GTAs instructing undergraduate students. Additionally, the UTAs were paired and the instructors were present throughout the course. Last, the UTAs were aware that their reflections were being used in this study, which may have altered their statements. The researchers support that the self-reported gains were reflected in the UTAs’ teaching practices.

This study was the first implementation of this tiered mentor model for a 10-week CBRE. The researchers in this study plan to expand this research to include qualitative data of experiences and perceptions from the UTAs, high school mentees, and instructors as well as quantitative data of reported science content gains. The goal of this study was to develop a successful model for training CBRE instructors, specifically TAs. For that goal the researchers plan to expand the tiered mentor model to include graduate students, who preliminarily serve as TAs. The researchers promote this model to be tested, refined, and refuted toward more CBREs.

Magdalene K. Moy (mkm99@drexel.edu) is a doctoral candidate in the School of Education, Penny L. Hammrich is an interim dean and professor in the School of Education, and Karen Kabnick is an associate teaching professor in the Biology Department, all at Drexel University in Philadelphia, Pennsylvania.