Table 3: P-values for 2016 Data Table 1: Mean Percentages Enrollment and Retention of Women in Undergraduate Engineering Programs Olive Tuyishimire, Kelsey Shelton   Abilene Christian University Department of Engineering and Physics 1201 E Ambler Ave. Abilene, TX 79601 kxs15c@acu.edu Abstract In this research we seek to understand what, if any, meaningful differences exist in the enrollment and degrees awarded to women at Christian universities compared to other (not specifically Christian) universities, as well as in small and large programs. We analyze enrollment and degrees awarded data from ASEE using a t-test for a difference in means. Our overall goal is to better understand program characteristics that lead to higher enrollment and retention of women. Introduction As female engineering students, we have a great interest in the enrollment and retention of women in engineering programs across the nation. As a starting point, we chose to research differences in enrollment and degrees awarded over a number of years. So far, we have analyzed data from the American Society for Engineering Education (ASEE) for two years: 2010 and 2016. For our analysis of Christian vs. other schools, we looked at the mission statement of each school in the ASEE data to determine which group was appropriate for each school. Our decision on the size of the school was based on the assumption that a program of 200 students would have class sizes of approximately 50 students, and a program of 400 students would have class sizes of approximately 100 students. Once we determined how we would group our data, we began analysis of each group. 3. Methodology Our next step in analyzing this data was to perform t-tests on each group, to determine if any of the differences in mean percentages observed in Table 1 are statistically significant. There are two kinds of t-tests for a difference in means: 1) two-sample assuming equal variances and 2) two-sample assuming unequal variances. Another test, called the F-test, is first used to determine which t-test to apply. Once this was determined, we ran the appropriate t-test. The results of the t-tests tell us whether or not the difference between the means of two groups is a reflection of a meaningful difference in the population from which each group was sampled. A reported p-value of less than 0.05 means there is less than a 5% chance that an observed difference in means is due to random variability in the samples rather than a meaningful difference in the populations. Group Enrollment 2010 Degrees Awarded 2010 Christian vs Other 0.981 0.946 Large vs Small (200) 0.0545 0.792 Large vs Small (400) 0.00677 0.342 Table 2: P-values for 2010 Data Group Enrollment 2016 Degrees Awarded 2016 Christian vs Other 0.946 0.528 Large vs Small (200) 0.285 0.670 Large vs Small (400) 0.031 0.538 Group N (Sample Size) Enrollment 2010 Degrees Awarded 2010 Enrollment 2016 Degrees Awarded 2016 Christian 41 17.9 17.35 31 21.7 21.1 Other 303 17.3 266 21.6 20.05 Small (<200) 43 15.7 17.0 10 17.4 22.4 Large (>=200) 301 18.2 288 20.0 Small (<400) 91 15.95 16.6 32 17.7 18.7 Large (>=400) 253 18.6 17.5 22.05 20.3 Table 3: P-values for 2016 Data 3. Interpretation and Conclusion In Table 1, for small schools the mean percentage of degrees awarded is greater than the mean percentage of enrollment, and the opposite is true for large schools. This indicates that small schools for the years 2010 and 2016 did a much better job of retaining and graduating women than large schools. In Tables 2 and 3, the highlighted boxes are of the p-values that are equal to or less than 0.05, the commonly used cut off used in statistical analysis, indicating that there is a significant difference in the data. According to these results, there is no significant difference in the proportion of women that are enrolled at or have received degrees at Christian institutions compared to other institutions. However, these results suggest that there may be a meaningful difference in the enrollment and retention of women in small vs. large universities. This is an interesting finding, and it calls for further investigation of the differences between small and large universities. Table 1: Mean Percentages 2. Initial Observations The table above includes the average percentages of women in each of the indicated groups. The most interesting values are found in the enrollment and degrees awarded for small and large schools in 2016. For the small vs. large enrollment with small defined as less than 200 students, we see that the mean percentage of women enrolled at small schools is quite a bit smaller than the mean percentage of women enrolled at large schools. Yet, small schools outperform large schools in the percentage of degrees awarded to women in that same year. As we continued our research, we used these observations to guide us to where we would continue our investigation. Proceedings of the 2018 ASEE Gulf-Southwest Section Annual Conference The University of Texas at Austin April 4-6, 2018