Supporting data for "Getting Students through the Gates: How Self-efficacy and Interest Trajectories Drive Learning Outcomes in Large Mathematics Courses"

Gateway courses are foundational prerequisite courses that undergraduate students must complete prior to enrolling in major courses (e.g., first-year mathematics, chemistry, psychology, statistics). Gateway courses often have high enrolment, and provide less support, structure, and feedback compared to previous experiences (e.g., secondary school). Declines in students' motivation and performance are common. This PhD project investigated two sources of engagement and motivation: self-efficacy and interest across two mathematics gateway courses. In particular, factors related to how self-efficacy and interest changed during the courses were examined across the studies.

Four studies were conducted across five offerings of these two courses from 2020-2022. Participants were students enrolled in these courses. Study 1 (n=175; Sept-Dec 2020; Course 1) was conducted in an online (pandemic) setting. The interplay between students' (amounts of) self-efficacy, interest, and performances (i.e., quizzes) across the course was investigated. Study 2 (n=349; Sept-Dec 2021; Course 1) was conducted the next year, and examined how overall self-efficacy changes, and how those changes were associated with performances across a course, and interest at the end of the course. Study 3 (n=313; Sept-Dec 2021; Course 2) investigated short-term changes in interest, and how they were related to performance, and self-efficacy. Lastly, Study 4 contained two studies (n=299; n=407; Studies 4a, 4b; Courses 1 & 2) that investigated the interplay between perceived difficulty on performance tasks (i.e., quizzes), short-term changes in self-efficacy, performances, and interest (in the second study).

The data files are the datasets used to conduct the analyses across the four studies. These included students' responses on formative quizzes, and self-reported data on self-efficacy, interest, perceived difficulty, and gender. These data were used for quantitative analysis using MPlus and other software. Each folder contains the relevant files each study (presented in the respective chapter of the thesis).

1) Chapter 3 - Study 1 contains the dataset used for the first study. This study is already published.
2) Chapter 4 - Study 2 contains the datasets used for the second study, including for the full model, invariance and reliability testing, and dataset for IRT.
3) Chapter 5 - Study 3 contains the datasets used for the third study, including for the full model and dataset for IRT.
4) Chapter 6 - Study 4 (Studies 4a and 4b) contains the datasets used for the last study, including those used for the full model, dataset for IRT, and perceived difficulty.