Expressions of students’ disagreement in life sciences team-based online discussions, Thesis of Evolutionary biology

Download Expressions of students’ disagreement in life sciences team-based online discussions and more Thesis Evolutionary biology in PDF only on Docsity! Expressions of students’ disagreement in life sciences team- based online discussions Honors Thesis Presented to the College of Arts and Sciences Cornell University in Partial Fulfillment of the Requirements for the Biological Sciences Honors Program by Priyanka Menon December 2022 Supervisor: Dr. Abby Drake 2 ABSTRACT Studying disagreement in STEM classrooms is one way to better understand students’ sense of belonging in a learning environment by linking the frequency of disagreement with how comfortable they are contributing their true opinions. I determined whether students’ frequency of expression of disagreement in an online STEM course is affected by group composition (gender, size, topic). Disagreement is measured in this study directly by quantifying how many students replied to prompted discussion questions with disagreement statements or when students ask and answer questions to clarify the disagreement. I predicted that students would express more disagreement when they were male, were in groups with less than four people, and discussed topics involving evolutionary processes and phylogeny based on conclusions from previous literature. The results showed that there was not a difference in disagreement rate across gender, there was more disagreement for "Population Genetics" and "Other Evolutionary Processes" compared to "Phylogenetics" and "Biodiversity," and there tended to be more disagreement for teams of five compared to smaller sized groups. These findings can be applied to future biology education research to improve the ways group activities are structured to be more inclusive of students of all backgrounds and to maximize true expression of opinions. 5 and express their true beliefs, including those of disagreement. Therefore, I propose that expression of disagreement is linked with a sense of belonging (having “comfortability” in one’s academic setting) and self-efficacy, or a student’s belief in his or her own ability to contribute a correct answer; when one is more comfortable in their group learning environment or is confident in their ability to answer a discussion question, they are more likely to express a dissenting opinion (Doménech-Betoret et al., 2017). Disagreement is measured in this study directly by seeing how many times students' responses to prompted online group discussion questions are categorized under Phases I- VI responses based on the Interaction Analysis Model (IAM) (Gunawardena et al., 1997). Phase II, disagreement, is recorded when there is dissonance or inconsistency among participants and when students ask and answer questions to clarify the disagreement (Gunawardena et al., 1997). In this study I measured gender, team size, and topic complexity factors because they are relevant to group participation (Takeda et al., 2013; Shaw, 2012). There exists a clear gender gap between male and female representation (choosing to participate and successfully completing a specific course or major) in large introductory STEM courses, with women more likely to feel lower confidence, imposter syndrome, and a tendency to associate themselves with stereotypes (Sullivan et. al, 2018). Therefore, I predict male students to express more disagreement in group interaction compared to female students. Group size has also been shown to have a clear impact as larger groups have been linked with lower participation; in fact, increasing group size can lower group performance for more complex problems involving low demonstrability (Amir et. al, 2018). In a study conducted by Amir et al., participants of groups in an online setting were assigned to four conditions with varying types of problem demonstrability and difficulty levels; they were presented with three possible solutions to the problem including one that was correct, one that was 6 wrong, and one that was their own solution. When presented with problems where it was difficult to recognize a correct solution to the problem (low demonstrability), groups with more students performed worse. The reason these larger groups may have performed poorly could be that not all members are able to share their thoughts, since a greater number of people can be intimidating. In fact, larger groups are more likely to confirm to the same ideas or thoughts, which comes in the way of dissenting viewpoints from being expressed; with larger majorities, an individual with a different opinion becomes more aware of being different, which causes a greater need to agree with a larger opinion (Jhangiani et al., 2022). Thus, I predict that students will express more disagreement when they are in smaller groups, namely those with less than four members, because they are more likely to express their true beliefs in a less intimidating setting to achieve a correct answer to a problem. A students' familiarity with a topic can influence how often they participate and express their true opinions. I predict that students may feel more comfortable discussing generally simpler, more familiar, introductory topics than those that are more complex including evolution and phylogenetics. In summary, this study seeks to analyze what factors concerning group composition and environment will maximize true expression of opinions to ensure all students have “comfortability” and are well-adjusted to their learning environment. MATERIALS AND METHODS Student Population Freshman through senior students at Cornell University participated in an online, asynchronous evolutionary biology course. This study collected data from students who participated in the Summer 2019, Fall 2019, and Spring 2020 semesters, which had 28, 38, and 24 7 students, respectively. In total, data was ultimately collected for 90 students across the three terms, with 65 females and 25 males participating. The topics covered in this evolutionary biology course fell into five main categories: Adaptation & Speciation, Phylogenetics & Macroevolution, Biodiversity, Population Genetics, and Other Evolutionary Processes (Table 1). Table 1. Each subtopic under the overarching category, including Adaptation & Speciation, Phylogenetics & Macroevolution, Biodiversity, Population Genetics, and Other Evolutionary Processes. Category Topic Adaptation & Speciation Adaptation & Speciation Genetics of Adaptation Phylogenetics & Macroevolution Adaptation & Speciation & Macroevolution Macroevolution Macroevolution & Phylogenetics Phylogenetics Biodiversity Biodiversity Biodiversity & Evolutionary Development Biodiversity: Prokaryotes and Protista Population Genetics Conservation Genetics of the Prairie Chicken Population Genetics Population Genetics & Adaptation & Speciation 10 Phase IV Testing and modification: testing the proposed new knowledge against existing cognitive schema, personal experience, or other sources. Phase V Phrasing of agreement and applications of newly constructed meaning: summarizing agreement and metacognitive statements that show new knowledge construction. Phase VI Methodological clarification: questions and discussions about class assignments or technical issues. Statistical Analyses After the coding process, the interactions were analyzed based on the different IAM phases. R-Studio (4.2.0) was used to collect summary statistics and observe the initial data. Microsoft Excel 16.54 was used to analyze the data by factors of interest (gender, topic, and team size) and to generate relevant figures using the pivot table function. The gender and topic variables were compared against number of disagreement and no disagreement interactions. For team sizes, the number of students in each group was compared against the number of students who expressed disagreement in that group; this was to see if a certain team size was linked with a student ever expressing disagreement. Statistical analyses were performed through a chi-square test (chisq-test function in R) which performs a test for equality of proportions with continuity correction (B=10,000) using the libraries psych (Revelle, 2022), lsr (Navarro, 2015), and car (Fox & Weisberg, 2019). 11 RESULTS 1. Gender Analyses There were 65 female students and 25 male students who participated in the study. A single interaction for these students was whenever they had any comment or response on VoiceThread or Slack. There were a total of 2315 student interactions for females, with only 52 of those having expressions of disagreement (coded to Phase II of the IAM scale). There was a total of 962 student interactions for males, with only 22 of those being expressions of disagreement (Fig. 1). On average, each female student expressed disagreement within interactions 2.44 +1.8 (SD) times, with a median of 2 times. For male students', on average each student expressed disagreement within interactions 1.93 +1.24, with a median of 2 times. While there was a much greater number of females interactions compared to males’ interactions (given that there were more females than males), the proportion of disagreement expressed across both groups was approximately the same (2-sample test for equality of proportions with continuity correction p-value > 0.05), with females disagreeing about 2.24% of the time and males disagreeing about 2.28% of the time (Fig. 2). 12 Figure 1. Number of interactions for female and male students showing the number of disagreement (orange) and not disagreement (blue). Female students had a greater number of interactions including those of disagreement compared to male students. Figure 2. Frequency of interactions for female and male students showing the proportion of disagreement (orange) and not disagreement (blue). Female and male students had approximately the same proportion of disagreement (2.24% and 2.28%, respectively). 2. Topic Analysis When comparing the frequency of interaction for expressions of disagreement by class topic (Table 1), there appeared to be the most interactions for Population Genetics and Biodiversity. In fact, Population Genetics had the greatest expressions of disagreement (n=35). The topic ‘Other Evolutionary Processes’ had a greater proportion of disagreement interactions out of the total number of interactions for that topic. However, the proportions of disagreement across all topics did not show a significant difference (5-sample test for equality of proportions with continuity correction p-value > 0.05). 15 Figure 5. Percentage of students for each category of team size that expressed disagreement (orange) or did not express disagreement (blue). Teams with five students had the greatest proportion of students expressing disagreement compared to the other group sizes. After performing a chi-square test for equality of proportions, this difference was not found to be statistically significant (p-value > 0.05) for teams of five versus teams of three and four, but it was statistically significant for teams of five versus teams of two (p-value < 0.05) DISCUSSION In analyzing student interactions in an asynchronous introductory biology course, I found that factors such as gender, topic, and team size can have different levels of influence in how often students express disagreement. The results of this study did not follow all the predicted trends, as they showed that there existed no significant difference in the number of expressions of disagreement by gender and that teams with more students (five members) had more students that expressed disagreement compared to teams with smaller sizes. While the differences in the number of disagreement interactions between topic studied were small, the greatest proportion of disagreement occurred for "Other Evolutionary Processes" (as predicted) but had the greatest number of interactions of disagreement for "Population Genetics," unlike for phylogenetic topics, 16 which was predicted. Ultimately, disagreement varied across topics with complex ones like Population Genetics having an increase in discussion among team members, and teams that had more students (specifically 5 members) being more likely to disagree. Initially, gender analysis revealed that female students were more likely to interact, and express disagreement compared to male students. However, when controlling for the much greater number of female students participating in the study (almost 3 times greater than male students), the proportion of disagreement expressed for the two sexes was almost identical and not statistically significant. The lack of difference in expressions of disagreement between males and females suggests that online, team-based learning environments may be effective in reducing the gender gap in STEM classrooms and can be correlated with improving female participation. Previous studies on team-based-learning have shown similar results, as one in an engineering environment promoted more gender inclusive teamwork (Erans & Beneroso, 2022). Yet another study which included 88 students in a microbiology course in India that was divided into a traditional, large lecture group and a team-based learning group, found that female students performed better in team-based lectures but the performance for males and females in a traditional lecture style was comparable (Harakuni et al., 2015). Thus, to effectively conclude that team-based learning can help improve female students' expressions of disagreement, there needs to be a direct comparison against the expressions of disagreement in traditional, lecture-style learning environments, as well as, determining if the students are not disagreeing because they do not feel comfortable or because they understand the topic. Analyzing the expressions of disagreement by topic revealed that there was no significant difference across topics studied (based on chi-square test for equality of proportions). In general, there was not much disagreement across the term (the proportions of disagreement for each topic 17 ranged from 1.14% to 3.03%). A few possible reasons are that students were not very comfortable expressing disagreement, or that students generally agreed on their answers to their group activities. "Population Genetics," which had the highest number of interactions including those expressing disagreement, has been found to be a complex topic that students have had difficulty in (Agorram et al., 2016). Population genetics is a rapidly developing field, but many students have had trouble understanding its "...abstract and difficult knowledge" (Agorram et al., 2016) that is more theoretical including topics like Genetic Drift and Allele Frequencies. These issues that are associated with understanding basic concepts in "Population Genetics," may have caused students to disagree with each other more on their answers and reasoning, as they did not have a solid grip of the course material. Furthermore, "Other Evolutionary Processes" had the highest proportion of disagreement, which have also been found to be complex topics for students to understand (Agorram et al., 2016). Evolutionary development is a newer approach in the field of biology, and its understanding has been influenced by religious or cultural objections (e.g., Creation story in biblical book of Genesis, Scopes trial that involved law prohibiting evolution instruction in schools, the concept of “Intelligent Design”) (Bertka et al., 2019). In fact, students have been shown to have misconceptions on evolutionary topics like speciation and accepted the theory of evolution for other species but not human evolution, as non-authoritative sources (television or parents) played a role in them learning inaccurate evolutionary mechanisms (Agorram et al., 2016). Therefore, students may be more likely to express disagreement over a topic that conflicts with their inherent beliefs or one that is more theoretical and newly developing, as the knowledge foundation is still being established. While disagreement interactions on these complex topics follow a known trend 20 2014), the Harvard-Panorama Student Perception Survey scale on Sense of Belonging (Gehlbach, 2015), and Yorke’s (2016) survey of student belongingness, engagement, and self-confidence. This study offers new insights into expressions of student disagreement in team-based- learning STEM environments and how it varies with team characteristics, which there is not much comprehensive literature for. In fact, when doing a search for disagreement in team-based-learning for undergraduates on Google Scholar, the results are only for team-based learning in an undergraduate setting; there is no information on how often students disagreed and the potential link to a sense of belonging in a classroom setting. These findings reveal the importance of studying disagreement in STEM learning environments because dissent is essential to a productive discussion (Collins, 2018) and should be developed to ensure students can contribute their ideas effectively. This study shows what factors can foster expressions of disagreement and can serve useful to developing productive learning environments; this includes altering group composition to be more inclusive and gender-balanced, keeping groups smaller, and ensuring students are exposed to diverse backgrounds and varying levels of topic familiarity. ACKNOWLEDGEMENTS I thank Dr. Lina Arcila Hernandez for giving me the opportunity to conduct this research and for all the professional advice and support she has given me during my time at Cornell University, Dr. Abby Drake for agreeing to be my main faculty advisor for this project, and Dr. Kelly Zamudio for providing me the academic resources to do research for the last four years. Finally, I would like to thank my advisors at Cornell University’s Office of Undergraduate Biology, Dr. Laura Schoenle and Dr. Wojtek Pawlowski, for their guidance in the Biology Honors Program. 21 REFERENCES Agorram, B., Zaki, M., Selmaoui, S., Razouki, A., & Khzami, S.-E. (2016). Understanding of Population Genetics and Evolution Among University Students. Research Highlights in Education and Science 2016, 139–148. Aguillon, S. M., Siegmund, G.-F., Petipas, R. H., Drake, A. G., Cotner, S., & Ballen, C. J. (2020). Gender differences in student participation in an active-learning classroom. CBE— Life Sciences Education, 19(2). https://doi.org/10.1187/cbe.19-03-0048 Amir, O., Amir, D., Shahar, Y., Hart, Y., & Gal, K. (2018). The more the merrier? increasing group size may be detrimental to decision-making performance in nominal groups. PLOS ONE, 13(2). https://doi.org/10.1371/journal.pone.0192213 Ballen, C. J., Aguillon, S. M., Awwad, A., Bjune, A. E., Challou, D., Drake, A. G., Driessen, M., Ellozy, A., Ferry, V. E., Goldberg, E. E., Harcombe, W., Jensen, S., Jørgensen, C., Koth, Z., McGaugh, S., Mitry, C., Mosher, B., Mostafa, H., Petipas, R. H., … Cotner, S. (2019). Smaller classes promote equitable student participation in STEM. BioScience, 69(8), 669– 680. https://doi.org/10.1093/biosci/biz069 Bledsoe, T.S &, and Baskin J (2015). Strategies for Addressing Student Fear in the Classroom. Faculty Focus | Higher Ed Teaching & Learning, https://www.facultyfocus.com/articles/teaching-and-learning/strategies-for-addressing- student-fear-in-the-classroom/. 22 Bertka, C. M., Pobiner, B., Beardsley, P., & Watson, W. A. (2019). Acknowledging students’ concerns about evolution: A proactive teaching strategy. Evolution: Education and Outreach, 12(1). https://doi.org/10.1186/s12052-019-0095-0 Collins, B. R. (2018, November 28). Teaching students to disagree productively. Edutopia. Retrieved December 21, 2022, from https://www.edutopia.org/article/teaching-students- disagree-productively Fox, J., & Weisberg, S. (2019). An R companion to applied regression, Third Edition. Applied Regression 3E. Retrieved December 21, 2022, from https://socialsciences.mcmaster.ca/jfox/Books/Companion/ Erans, M., & Beneroso, D. (2021). Team-Based Learning: Promoting gender inclusive development of teamworking skills in engineering education. International Journal of Gender, Science & Technology, 13.3. Everywhere, P. (2020, June 22). The five simple methods that help teachers measure student engagement. Poll Everywhere Blog. Retrieved December 21, 2022, from https://blog.polleverywhere.com/how-to-effectively-measure-student-engagement/ Gehlbach, H. (2015). User Guide: Panorama Student Survey. Boston: Panorama Education. Retrieved from https://www.panoramaed.com/panorama-student-survey Gunawardena, C. N., Lowe, C. A., & Anderson, T. (1997). Analysis of a global online debate and the development of an interaction analysis model for examining social construction of 25 Shaw, R.-S. (2013). The relationships among group size, participation, and performance of programming language learning supported with online forums. Computers & Education, 62, 196–207. https://doi.org/10.1016/j.compedu.2012.11.001 Sullivan LL, Ballen CJ, Cotner S (2018) Small group gender ratios impact biology class performance and peer evaluations. PLOS ONE 13(4): e0195129. https://doi.org/10.1371/journal.pone.0195129 Takeda, S., & Homberg, F. (2013). The effects of gender on Group Work Process and achievement: An analysis through self- and peer-assessment. British Educational Research Journal, 40(2), 373–396. https://doi.org/10.1002/berj.3088 Yorke, M. (2016). The development and initial use of a survey of student 'belongingness,' engagement and self-confidence in UK higher education. Assessment & Evaluation in Higher Education, 41(1), 154-166. 26 SUPPLEMENTARY INFORMATION R SOFTWARE STATISTICAL ANALYSIS CODE: R SOFTWARE OUTPUT: 1. Gender 2-sample test for equality of proportions with continuity correction data: c(52, 22) out of c(2315, 963) X-squared = 4.7356e-30, df = 1, p-value = 1 alternative hypothesis: two.sided 95 percent confidence interval: [email protected] @.01120202 sample estimates: prop 1 prop 2 @.02246220 @.02284528 2-sample test for equality of proportions with continuity correction data: (490, 400) out of c(500, 500) X-squared = 80.909, df = 1, p-value = 1 alternative hypothesis: Less 95 percent confidence interval: -1.0000000 @.2131742 sample estimates: prop 1 prop 2 0.98 0.80 2. Topic 5-sample test for equality of proportions without continuity correction data: c(14, 7, 15, 35, 4) out of c(696, 497, 718, 1258, 132) X-squared = 3.8066, df = 4, p-value = 0.4328 alternative hypothesis: two.sided sample estimates: prop 1 prop 2 prop 3 prop 4 prop 5 @.02011494 0.01408451 0.02089136 @.02782194 0.03030303 3. Group Size 27