Transitioning to a Flipped Model in Introductory Statistics: The Role of Proctored In-Class Quizzes

Transitioning to a Flipped Model in Introductory Statistics: The Role of Proctored In-Class Quizzes Dr. Zaher Kmail1, Dr. Vanessa Peters Hinton2 1University of Washington Tacoma, 1900 Commerce Street, Tacoma, WA 98402 2Digital Promise, 702 Marshall Street, Suite 340, Redwood City, CA 94063 Abstract This study examined whether adding brief proctored in-class quizzes to a flipped introductory statistics course using digital courseware improves performance and engagement. Across three academic quarters with the same instructor and curriculum, we varied accountability: baseline with in-class quizzes, first term courseware with online-only quizzes, and courseware plus weekly proctored in-class quizzes. Exam means and attendance were compared using ANOVA, and student questionnaires and interviews described student experiences. Restoring in-class quizzes raised average midterm and final exam scores from 77.29% and 79.91% to 84.68% and 85.92%, respectively, and attendance returned to its previous baseline level. Students reported that courseware worked best for reinforcement after instructor explanation. Results align with evidence on formative practice and active learning and with design research on aligning tool usability with learning tasks. Key Words: flipped classrooms, digital courseware, gateway courses, high impact practices, student engagement, formative practice, feedback 1. Introduction Gateway courses are central to degree completion because they are required for progression and often taken early in a student’s program, and performance in these courses predicts retention and persistence (Bloemer et al., 2017). Gateway courses like introductory statistics frequently fulfill requirements across majors that include nursing, business, and communications. In TMATH 110 at the University of Washington Tacoma, introductory statistics satisfies a general education mathematics requirement and serves a broad mix of majors, which makes design decisions about preparation, accountability, and in-class support consequential for student progress. Given these stakes, instructors in gateway courses have increasingly turned to flipped instruction as a way to improve preparation and engagement. Flipped instruction shifts initial content exposure to the out-of-class space and repurposes class meetings for guided practice and feedback. In higher education, the model has gained traction as a means to support active learning and to expand opportunities for problem solving and explanation during class. Studies in STEM courses indicate that low-stakes checks aligned with in-class activities can strengthen preparation, while formats that rely only on unproctored online tasks often produce uneven engagement and mixed achievement results (Bossaer et al., 2016; Eichler & Peeples, 2016; Wozny et al., 2018). The effects vary by discipline and implementation choices. These choices include how pre-class preparation is structured, how accountable students are for that preparation, and how in-class time is used (Strelan et al., 2020; Zhang et al., 2024). Accountability is a central design challenge in flipped courses. When preparatory work is encouraged but not required, many students delay or skip it, which undermines the value of in-class activity. Digital courseware can organize readings, videos, and practice. However, structure alone rarely ensures engagement; students respond most strongly when the system includes clear expectations and consequences for preparation. The presence and format of assessments that check for preparation is often decisive. Prior studies show that lowstakes checks aligned with class activities increase preparation and improve outcomes (Eichler & Peeples, 2016; Wozny et al., 2018). Conversely, settings that rely only on unproctored online tasks can produce uneven engagement and mixed achievement results (Bossaer et al., 2016). Still, even well-designed accountability structures can produce temporary declines in performance when new routines or technologies are introduced. Change theory suggests that adopting new tools or workflows often leads to short-term performance declines before improvement occurs (Fullan, 2007). This “implementation dip” reflects the learning curve faced by both instructors and students. Implementation science research shows that initial disruption, unclear workflows, and unfamiliarity with new technology can temporarily lower performance and participation, even when long-term outcomes are positive (Durlak & DuPre, 2008). Related research on recorded lectures shows that providing captured content can reduce attendance and, in some cases, lower achievement compared to in-person instruction (Artz et al., 2022). Randomized and quasiexperimental studies find that live instruction often leads to better outcomes than recorded delivery, particularly for lower-achieving students, underscoring the importance of maintaining in-person accountability as new systems are adopted (Edwards & Clinton, 2019; Setren et al., 2020). This study examines a transition in TMATH 110, an introductory statistics course at the University of Washington Tacoma, across three quarters. The winter 2024 quarter used Pearson MyLab Statistics and included in‑person, proctored quizzes. In winter 2025 the course adopted Lumen One courseware and moved quizzes online and were unproctored. Spring 2025 continued with Lumen One and added short, proctored in‑class quizzes at the start of the week to strengthen accountability. The same instructor, curriculum, and assessments allowed for quarter‑to‑quarter comparisons. The purpose of this paper is to describe and analyze how integrated accountability, specifically the addition of brief proctored in‑class quizzes alongside digital courseware, relates to exam performance, attendance, and students’ learning processes in a flipped statistics course. Guided by the literature and the local context, our analysis focuses on three research questions: 1. How does the addition of proctored in-class quizzes (vs. online-only quizzes) impact student performance on summative assessments? 2. Does introducing proctored in-class quizzes change student attendance and engagement behaviors relative to online-only quizzes? 3. How do students describe their learning processes across digital and in-person modalities in a flipped statistics course? 2. Literature Review and Background 2.1 Flipped Learning in Undergraduate Statistics Flipped instruction reallocates content delivery to pre-class learning so that class time can be used for application and feedback. Meta-analyses across disciplines show small to moderate gains in achievement outcomes, with variation linked to design and context (Strelan et al., 2020). A recent bibliometric review highlights the field’s rapid expansion and confirms that implementation choices remain the strongest predictors of effect size (Zhang et al., 2024). Experimental findings are more mixed. A randomized trial at West Point found short-term gains in mathematics but no overall effect in economics, with larger achievement gaps in the flipped condition, and underscores the need for explicit supports for preparation and in-class participation (Setren et al., 2020). These findings align with learning theories that emphasize cognitive load, motivation, and instructional alignment as conditions under which flipped formats improve learning (Abeysekera & Dawson, 2015). 2.2 Accountability and Grading Structure in Formative Practice In flipped courses, accountability measures often determine whether students engage meaningfully with formative practice, the preparatory work intended to support in-class problem solving. The structure of these systems is a consistent predictor of course effectiveness because it determines how expectations are communicated, how feedback is delivered, and how students perceive the value of preparatory work (Chen, 2023). When students expect timely assessment of preparatory work, they engage more reliably and arrive better prepared for in-class activities. Evidence from chemistry shows that low-barrier flipped modules paired with readiness checks and aligned activities improve course grades relative to traditional formats (Eichler & Peeples, 2016). Similarly, when accountability structures are clear and reinforced, students demonstrate more consistent preparation. In contrast, when instructors rely only on unproctored online tasks, students sometimes minimize effort or focus on meeting minimum completion requirements, resulting in limited or inconsistent performance gains. In pharmacy education, a cohort comparison found that a flipped pharmacotherapy module without enforced pre-class accountability produced lower exam performance than an interactive lecture baseline (Bossaer et al., 2016). A randomized trial in econometrics further showed that flipping results in modest or null average gains without careful integration of assessment and in-class practice (Wozny et al., 2018). Taken together, these studies indicate that the format of formative practice and its associated grading structure are critical to sustaining engagement. Short, proctored in-class quizzes can strengthen preparation by raising the salience of retrieval and aligning assessment with upcoming in-class activities. 2.3 Engagement, Attendance, and the Role of In‑person Structure Attendance and engagement are sensitive to how students perceive the value of showing up prepared for class. Studies of lecture capture reveal a similar pattern. In a matched-cohort study, the introduction of recorded lectures was associated with lower attainment, mediated by reduced attendance. Viewing recorded lectures did not offset the loss of in-person participation once attendance was controlled (Edwards & Clinton, 2019). A classroom randomized trial in economics found that students taught via live lectures outperformed peers assigned to recorded lectures, with particularly negative effects for lower‑achieving students (Artz et al., 2022). These findings support the idea that in‑person accountability, including brief proctored checks, can help stabilize participation and readiness in courses that routinely rely on self‑regulated pre‑class work. 2.4 Implementation Dip as a Lens on Course Redesign Implementation research predicts a worse-before-better pattern when new learning tools or class routines are introduced. The implementation dip (Fullan, 2007) is a temporary decline in performance and confidence while instructors and students learn new workflows. Broad evidence shows that levels of implementation and the availability of supports are strongly associated with outcomes during adoption phases (Durlak & DuPre, 2008). In the context of courseware adoption and flipping, early terms may feature unclear expectations, misalignment between platform pacing and class pacing, and uncertainty about how online work translates to graded outcomes. These factors can diminish engagement and exam performance. As classroom routines normalize and accountability structures are fine-tuned, outcomes often recover. 2.5 The Current Study TMATH 110 at the University of Washington Tacoma provides a natural setting to examine accountability design in a flipped introductory statistics course. Across three consecutive quarters, the instructor implemented different courseware and quiz structures: Winter 2024 used Pearson MyLab Statistics with inclass proctored quizzes, Winter 2025 adopted Lumen One with online unproctored quizzes, and Spring 2025 retained Lumen One with brief proctored quizzes at the start of the week. The instructor, curriculum, and assessments were consistent across quarters, allowing for comparison of outcomes and experiences across these conditions. The study integrates quantitative indicators of performance and attendance with student questionnaires and interviews that explored how digital and in-person modalities supported learning. 3. Methods 3.1 Study Participants A total of 88 undergraduate students participated in the study across the three quarters. No additional eligibility criteria were applied, and all students enrolled in the course during these terms were included in the analysis. None of the students had previously attempted the course. 3.2 Study Context TMATH 110 fulfills a general education mathematics requirement and enrolls students from a range of majors, most in their first or second year. The class met twice weekly for two hours and offered support through the Teaching and Learning Center and weekly instructor office hours. 3.3 Instructional Conditions 3.3.1 Winter 2024: Baseline with Pearson MyLab Statistics and in-class quizzes Winter 2024 served as the baseline condition. Students used Pearson MyLab Statistics, which included access to StatCrunch and multiple-attempt online homework. The class followed a traditional structure: students attended lectures first, then completed assigned homework outside of class. Quizzes were administered in person and proctored during class sessions. 3.3.2 Winter 2025: First implementation of Lumen One with online quizzes In Winter 2025, the course shifted to Lumen One, marking its first use in the class. Students completed an online study plan before weekly class sessions, typically on Sundays. Rather than grades for correctness, students received a completion score that was applied to their overall grade. Study plan progress was monitored throughout the quarter, especially in the first several weeks to help students become adjusted to using Lumen One. Class meetings focused on reviewing and reinforcing concepts. Quizzes were moved online through Lumen One, were unproctored, and usually due at the end of the week. 3.3.3 Spring 2025: Second implementation of Lumen One with online and in-class quizzes In spring 2025, the course used Lumen One again with the same study plan format before class. In an effort to help students organize the large volume of class handouts, a course packet was created for student purchase at a minimal cost at the beginning of the quarter. The packet included a mix of activities provided by Lumen One as well as formulas and activities that were curated by the instructor for previous iterations of TMATH 110. Students continued completing online quizzes through Lumen One on their own time. In addition, short in-class quizzes were introduced at the beginning of the first class each week. These in-class quizzes were proctored, paper-based, and students were allowed to use a single reference card for formulas. An overview of the flipped model used in spring 2025 is shown in Figure 1. Figure 1: Flipped classroom model used in spring 2025. In addition to using Lumen One’s online quizzes, a paper-based graded quiz was added to the first 10 minutes of the first class of the week to simulate test conditions and increase student accountability. While cohorts differed slightly between quarters, the instructor, course design, curriculum materials, and assessments remained consistent across all three quarters. This consistency supports quarter-to-quarter comparisons by minimizing instructional variance as a confounding factor. 3.3 Data Sources Quantitative data were drawn from course-level measures of student performance, including average grades on midterm exams, final exams, homework, and quizzes, as well as average attendance rates. These metrics provided a basis for comparing student outcomes across the different instructional conditions used across quarters. Qualitative data were collected from a questionnaire given to students in winter 2025 and spring 2025 (27 students and 30 students, respectively). In the questionnaire, students were asked to explain what they most liked and disliked about using Lumen One courseware. They were also invited to share any other thoughts they had about their experiences using Lumen One. In addition, semi-structured interviews were conducted with five students enrolled in the course in spring 2025. The interviews asked students about their experiences with the Lumen One courseware, their perceptions of online versus in-class activities, and the ways these modalities (online vs. in-class) complemented or conflicted with one another. Together, these data sources supported a mixed-methods analysis of learning outcomes and student engagement. 3.4 Data Analysis 3.4.1 Quantitative analysis Course-level data were analyzed descriptively to compare performance and attendance across the three instructional conditions. Metrics included average scores on midterm and final exams, homework, and quizzes, as well as average attendance rates. To assess whether differences across terms were statistically meaningful, PROC GLM procedure in SAS was used to perform analyses of variance (ANOVA) for each outcome variable, followed by post-hoc pairwise comparisons where appropriate. 3.4.2 Qualitative analysis Both the questionnaire data and interview transcripts were analyzed using thematic analysis (Braun & Clarke, 2012). Coding was conducted in cycles to identify themes related to course accountability, engagement, and perceptions of preparedness. To ensure trustworthiness, themes were collaboratively reviewed by the researchers and triangulated with quantitative results to provide deeper contextual understanding and enrich the interpretation of findings. 4. Results Study results are organized by the three research questions. Section 4.1 presents quarter-over-quarter outcomes on summative assessments (RQ1). Section 4.2 examines patterns of student engagement and attendance (RQ2). Section 4.3 describes students’ learning experiences (RQ3), with subsections 4.3.1–4.3.3 highlighting their students’ study approaches, views of instructional materials, and perceptions of how accountability structures and grading shaped their engagement with the courseware. 4.1 Impact of In-Class Quizzes on Summative Assessments (RQ1) Using the winter 2024 quarter as a baseline for assessment outcomes, results show that the addition of proctored, in-class quizzes during the spring 2025 quarter was associated with improved performance on summative assessments (see Table 1). The in-class quiz format appeared to help prepare students for both the midterm and final exams, resulting in comparable performance across the two assessments. This stands in contrast to the winter 2025 quarter, which saw a decline in exam performance when only online quizzes were used. Table 1: Mean Student Performance Across Quarters Measure (%) Winter 2024 (n = 31) (Pearson MyLab Statistics with in-class quizzes) Winter 2025 (n = 27) (Lumen One with online quizzes only) Spring 2025 (n = 30) (Lumen One with online and in-class quizzes) Midterm Exam 86.86 77.29 84.68 Final Exam 81.76 79.91 85.92 Homework 94.02 89.43 92.69 Quiz Average 85.22 85.60 89.15 Attendance 88.53 85.07 88.83 Note. Significant differences were observed between winter 2025 and spring 2025 midterm scores (p = .044) and between winter 2024 and winter 2025 midterm scores (p = .010). During winter 2024, students using Pearson MyLab Statistics achieved the highest mean midterm score across quarters (86.86%) and a final exam average of 81.76%. In winter 2025, when Lumen One was implemented with online quizzes only, mean midterm scores dropped significantly to 77.29%, a difference of 9.57% (p = .010), and final exam performance declined to 79.91%. In spring 2025, after the introduction of in-class quizzes, midterm scores increased to 84.68% (p = .044), and final exam averages improved to 85.92%. Although the difference was not statistically significant, it exceeded the winter 2024 quarter. 4.2 Impact of In-Class Quizzes on Student Attendance and Engagement (RQ2) Patterns of attendance and engagement across the three quarters highlight the role of accountability structures in supporting student preparation within a flipped learning environment. During winter 2024, when students used Pearson MyLab Statistics, attendance averaged 88.53%, and students demonstrated strong engagement, frequently bringing questions from homework to class and participating in in-class discussions. In winter 2025, when Lumen One was implemented with online-only quizzes, mean attendance declined to 85.07%, accompanied by lower observed engagement. The instructor noted that students were less prepared for inclass activities and that online work did not consistently translate to active participation in class. In spring 2025, after the introduction of proctored in-class quizzes alongside the Lumen One online quizzes, accountability and engagement were restored. Attendance improved to 88.83%, and students were more consistently prepared for class sessions. The increased engagement reflected a sense of responsibility for learning and a stronger connection between pre-class preparation and in-class participation. Qualitative data from winter 2025 questionnaire responses provide additional insight. Although thirteen of the twenty-seven students (48%) appreciated the opportunity for ungraded practice problems in Lumen One, a little over a quarter (6/27%) raised concerns about the platform’s completion-based grading system. “The study plan is based on completion and not based on whether the student knows or got the answer correct,” explained a student. Another student noted, “I didn’t like how we did not get ‘homework’ or in other words, we didn’t actually get graded exactly on how we answered the question.” A third student described the grading system as easily bypassed, saying, “You don’t actually have to do the study guides. You can go straight to the self-check, put in random numbers, and get full points.” These perspectives are in contrast to those shared in the spring 2025 questionnaire, where no students raised concerns about the completion-based grading or getting points for incorrect answers. 4.3 Student Experiences of Digital and In-Class Learning in a Flipped Course 4.3.1 Misalignment between Lumen One content and in-class instruction In their questionnaire and interview responses, over half of students (59%) described challenges stemming from inconsistencies in the content between the Lumen One courseware and in-person instruction, particularly when content was introduced out of sequence or explained differently across the two settings. Several noted that Lumen One often presented topics before they appeared in class or used unfamiliar terminology, which led to confusion and reliance on their instructor for clarification. As one student explained, “Lumen will do things a little bit differently than we do it in class… And honestly, Lumen is more confusing. But again, maybe that’s just me learning better in person.” Another commented that “Sometimes it [Lumen One] just gets too ahead of itself... or it tells me, ‘Oh, in order to be ready, you have to finish the thing that you’re going to learn in this topic already, but we haven’t really learned it yet from the topic.’” Students also described variation in the pacing and content focus between Lumen One and their instructor’s approach, which reduced the perceived value of the courseware and required them to use class time to make sense of the material. One student described the issue as follows: Our class is at a different pace than Lumen and it makes things really confusing sometimes. It’s not super far off, but we’ll be a lesson ahead in Lumen than we are in class, and it just gets really confusing... And then we do problems differently in class than we do on Lumen, so we do different things. 4.3.2 Preference for classroom learning and instructor-created resources Fifty-one percent of students expressed a preference for in-person instruction, particularly when it came to understanding mathematical and statistical content. Many felt that Lumen One’s digital interface lacked the clarity and support they needed to grasp complex ideas. As one student explained, “I kind of just waited until we learned it [a new concept] in class because he [the professor] explained it better. And then I would go back and do the Lumen stuff once I actually understood it.” This feeling was echoed by others who described online content as too abstract or difficult to follow without real-time explanation. For these students, professor-led instruction offered the structure and immediacy that digital tools could not replicate. In the words of another student: I just personally am not very good at learning things online, especially math things. For math, I have to be in person and seeing the other person do the math, seeing the teacher do the math, for me to learn it very well. Alongside a preference for face-to-face instruction, students also valued materials developed directly by their instructor. Rather than relying on Lumen One to learn statistics, students used a workbook designed by their professor specifically for their course. “Our professor had a workbook for us that we followed along with in class…you could buy it at the copy center or you could print it out yourself,” one student noted. “Another student explained, “I used it [workbook] mainly to study... I would use the examples we used in class to then practice using the Lumen tools.” A different student provided the following explanation about their preference for professor-led instruction and the workbook: He’s [the professor] really good at getting good detail, I guess, explaining the concept in a way that you can understand. And he didn’t like to move on from something until he felt like the whole class kind of collectively understood what was going on. And he gave a lot of different examples. And also, the follow-along workbook was really nice. He would have the workbook on a screen at the front of the class, and then he would highlight important things as we'd go along. And then he would have examples in the workbook that we would do together in class. 4.3.3 Lumen One as a space for knowledge reinforcement rather than pre-class learning Across the winter and spring 2025 quarters, 42% of students reported using Lumen One primarily as a tool for post-class practice rather than as a platform for pre-class learning. These students tended to delay engagement with online modules until after in-class instruction, relying on instructor explanations to clarify challenging concepts before returning to the platform. As one student explained, “I mostly did the self checks after class because by then I already knew what we were talking about in lecture, so it was just practice at that point.” Another student similarly noted, “I didn’t really use it [Lumen One] to learn the topic. I’d wait until we went over it in class and then go back into Lumen to finish the module or quiz.” Students also described using Lumen One to reinforce problem-solving processes demonstrated during lectures, aligning their practice with the pacing of classroom instruction. One student remarked, “After we did examples in class, I’d open Lumen later that day and try to redo similar problems there, it was good for seeing if I actually remembered how to do it.” For these students, Lumen One served primarily as a mechanism for reinforcement and consolidation of learned material rather than for initial concept acquisition. Consequently, some students engaged selectively with platform content, often disregarding items flagged as “needing review.” As one student explained, If I get an answer wrong [in Lumen], it just shows ‘needs review,’ which most of the time I don’t redo because I know we’ll go over it in class. Then I can go back afterwards and work on it in Lumen. 5. Discussion In this study, adding brief, proctored in‑class quizzes to a flipped introductory statistics course was associated with higher exam performance and a return to baseline attendance relative to an online‑only quiz format. In spring 2025, midterm and final means improved over winter 2025 and attendance increased, suggesting that credible weekly accountability made preparation more consistent and in‑class time more productive. Student accounts indicated that Lumen One was most useful for reinforcement after instructor explanation, which aligns with research on formative practice and active learning in undergraduate settings (Eichler & Peeples, 2016; Strelan et al., 2020; Wozny et al., 2018). The quarter‑to‑quarter pattern also reflects an implementation dip: outcomes decreased during initial adoption and then recovered as routines settled and proctored checks clarified expectations (Durlak & DuPre, 2008; Fullan, 2007). These findings are consistent with studies showing that flipped designs are most effective when out‑of‑class work is tied directly to what happens in class and when preparation is monitored with low‑stakes checks (Eichler & Peeples, 2016; Wozny et al., 2018). They also echo evidence that live, structured class time supports participation and achievement more reliably than recorded or unverified independent work, particularly for students who struggle (Artz et al., 2022; Edwards & Clinton, 2019; Setren et al., 2021). Together, the results highlight two design approaches for gateway statistics: tailor accountability to local context and align courseware pacing and terminology with classroom routines. Both actions ease the transition and strengthen the link between preparation and in‑class problem solving. This study is not without limitations. Comparisons are quasi‑experimental and limited to three cohorts at one institution with one instructor. Unobserved differences across cohorts cannot be ruled out, and outcome measures emphasize course‑embedded exams. Future work should analyze item‑level performance to distinguish procedural and conceptual learning and examine whether accountability structures have differential effects for student subgroups. 6. Conclusion In this partial transition to a flipped statistics course, adding short, proctored quizzes at the start of the week resulted in higher exam performance and a return to baseline attendance compared with an online‑only quiz format. Students reported that the courseware was most effective for practice after in-class explanations, and that the weekly in-class quiz increased the necessity of preparation. For instructors adopting digital courseware, two design principles follow. First, establish a simple and credible system for verifying preparation that aligns directly with how class time is used. Second, coordinate platform pacing and terminology with classroom materials so students experience a coherent workflow. These choices help mitigate the implementation dip that often accompanies new routines and support more consistent engagement and performance. References Abeysekera, L., & Dawson, P. (2015). Motivation and cognitive load in the flipped classroom: Definition, rationale and a call for research. Higher Education Research & Development, 34(1), 1–14. https://doi.org/10.1080/07294360.2014.934336 Artz, B., Johnson, M., Robson, D., & Siemers, S. (2022). Live or lecture capture: Evidence from a classroom random control trial. International Review of Economics Education, 40, 100240. https://doi.org/10.1016/j.iree.2022.100240 Bloemer, W., Day, S., & Swan, K. (2017). Gap analysis: An innovative look at gateway courses and student retention. Online Learning, 21(3), 5-14. https://doi.org/10.24059/olj.v21i3.1233 Bossaer, J. B., Panus, P., Stewart, D. W., Hagemeier, N. E., & George, J. (2016). Student performance in a pharmacotherapy oncology module before and after flipping the classroom. American Journal of Pharmaceutical Education, 80(2), 31. https://doi.org/10.5688/ajpe80231 Chen, T., Luo, H., Wang, P., Yin, X., & Yang, J. (2023). The role of pre-class and in-class behaviors in predicting learning performance and experience in flipped classrooms. Heliyon, 9(4), e15234. https://doi.org/10.1016/j.heliyon.2023.e15234 Durlak, J. A., & DuPre, E. P. (2008). Implementation matters: A review of research on the influence of implementation on program outcomes and the factors affecting implementation. American Journal of Community Psychology, 41(3–4), 327–350. https://doi.org/10.1007/s10464-008-9165-0 Edwards, M. R., & Clinton, M. E. (2019). A study exploring the impact of lecture capture availability and lecture capture usage on student attendance and attainment. Higher Education, 77(3), 403–421. https://doi.org/10.1007/s10734-018-0275-9 Eichler, J. F., & Peeples, J. (2016). Flipped classroom modules for large enrollment general chemistry courses: A low barrier approach to increase active learning and improve student grades. Chemistry Education Research and Practice, 17(1), 197–208. https://doi.org/10.1039/C5RP00159E Fullan, M. (2007). The new meaning of educational change (4th ed.). Teachers College Press. https://michaelfullan.ca/books/new-meaning-educational-change/ Setren, E., Greenberg, K., Moore, O., & Yankovich, M. (2021). Effects of flipped classroom instruction: Evidence from a randomized trial. Education Finance and Policy, 16(3), 363–387. https://doi.org/10.1162/edfp_a_00314 Strelan, P., Osborn, A., & Palmer, E. (2020). The flipped classroom: A metaanalysis of effects on student performance across disciplines and education levels. Educational Research Review, 30, 100314. https://doi.org/10.1016/j.edurev.2020.100314 Wozny, N., Balser, C., & Ives, D. (2018). Evaluating the flipped classroom: A randomized controlled trial. The Journal of Economic Education, 49(2), 115–129. https://doi.org/10.1080/00220485.2018.1438860 Zhang, F., Wang, H., Zhang, H., & Sun, Q. (2024). The landscape of flipped classroom research: A bibliometrics analysis. Frontiers in Education, 9, 1165547. https://doi.org/10.3389/feduc.2024.1165547