Date of Award

Fall 2020

Project Type


Program or Major

Biological Sciences

Degree Name

Master of Science

First Advisor

Melissa L Aikens

Second Advisor

Carrie L Hall

Third Advisor

Daniel R Howard


The field of biology is becoming increasingly reliant on quantitative tools, methods, and techniques, driving a need for incoming biologists to have robust quantitative skills. However, efforts to incorporate more quantitative skills at the undergraduate level are hampered by low student engagement with math in biology. Students’ motivation towards quantitative biology can provide insight into how best to increase their engagement and thus performance with these topics. This thesis examines students’ motivation towards math in biology through two key constructs: 1) students’ self-efficacy, through the theoretical lens of Social Cognitive Theory; and 2) students’ task-values, through the theoretical lens of Expectancy-Value Theory.

In Chapter 1, I explore how students’ self-efficacy towards quantitative biology problems is impacted by their experiences when working together in small groups to tackle mathematical problems in a biological context. In two sections of an introductory biology class, I surveyed students about their self-efficacy before and after completing two separate group work assignments about evaluating Hardy-Weinberg Equilibrium and modeling population growth, as well as asked them to report through short responses their experiences during those assignments which increased or decreased their confidence towards these kinds of problems. I qualitatively coded students’ short responses and found that students draw from a breadth of experiences to evaluate their self-efficacy. In particular, students reported many mastery experiences which increased their self-efficacy, through opportunities to practice solving these problems, confirming their success with them, or even being able to teach and guide their peers through the problems. Students also valued how group work fostered an availability of help and support from their peers which built their self-efficacy, through discussion, collaboration, and being able to simultaneously receive and seek help from their peers. I performed logistic regression to find that students’ self-efficacy level before entering each group work assignment predicted their likelihood of reporting mastery experiences or help availability from peers as the source of their increased self-efficacy, with higher self-efficacy students more likely to report mastery experiences and lower self-efficacy students more likely to report the availability of help from their peers.

Meanwhile, I found that while most students did not report any experiences which decreased their self-efficacy, those who did described a wide range of specific experiences. Most commonly, a lack of mastery decreased self-efficacy, ranging from simply not understanding the problem or making mistakes on the problem, to being unable to complete the assignments due to a lack of time or their group rushing ahead of them, to groups not even checking their answers or progress. Some students also described a lack of availability of help from their peers or instructors, with some groups failing to communicate openly or fully collaborate to group members simply being unable to help them with no one else around for support. Students also described a handful of experiences where they compared themselves unfavorably to their peers, feeling like they were falling behind or otherwise lacking in skill, as well as a general sense of anxiety from working in groups. I performed a logistic regression to find that students’ self-efficacy level before entering each group work assignment also predicted their likelihood of reporting a lack of mastery which decreased their self-efficacy, with lower self-efficacy peers more likely to describe a lack of mastery than their higher self-efficacy peers.

In Chapter 2, I explore how an alternative, multidimensional model of task-values compares to a more traditional model of students’ task-values towards statistics, and how these task-values relate to their statistical understanding. I surveyed life-sciences students at two institutions about their task-values towards statistics and measured their performance on an assessment of their understanding of biological variation in an experimental design context. I performed confirmatory factor analyses to find that students’ task-values towards statistics are better represented using a multi-dimensional model which differentiates the four canonical task-values—intrinsic value, attainment value, utility value, and cost—into multiple task-value ‘facets’, each capturing a specific aspect of each task-value, such as ‘utility for school’ or ‘emotional cost’. After excluding attainment value due to its poor fit, my model of task-value facets includes: 1) intrinsic value, with no facets; utility value with five facets (‘utility for school’, ‘utility for daily life’, ‘social utility’, ‘utility for career/job’, ‘utility for future life’); cost with three facets (‘effort required’, ‘emotional cost’, ‘opportunity cost’). Using multiple linear regression, I found that students’ utility value for statistics for school and emotional cost of statistics predicted their performance on the statistical assessment; students with higher utility value for statistics for school performed better than their peers with lower utility value for statistics for school, and students with lower emotional cost of statistics performed better than their peers with higher emotional cost of statistics.

My findings show how exploring students’ motivation towards quantitative biology can be a helpful lens for better understanding how students engage with math in biology. I reveal a mechanism by which in-class experiences can impact students’ confidence, highlighting a need for more focused work into how these specific experiences arise and how they relate to and interact with each other to shape students’ self-efficacy beliefs. Understanding this mechanism may reveal more effective and positive ways to increase students’ engagement with quantitative biology and reinforce their quantitative skills. Furthermore, I show how a more focused model or characterization of students’ task-values can predict their performance, providing a useful tool for educators and instructors to develop lessons or interventions to bolster their students’ values to increase their performance. Future work into students’ values about statistics should center around exploring this multi-dimensional model of task-values in a variety of circumstances with students of different backgrounds and experiences to broaden our understanding of how these values relate to their performance and understanding.