Math-Stat Modeling in Non-STEM Classes: Project outcome report

The broader goal of TLU?s M²AC project is to ensure that all undergraduates gain the quantitative literacy skills they need to advance in their related fields and to live informed lives. We designed and tested a new collaborative model for computer-based mathematical modeling projects that are: (a) developed within an interdisciplinary context by teams of faculty from math and statistics in support of faculty from art, applied sciences, social sciences, and humanities, (b) offer a range of mathematical and statistical principles and applications, and (c) are embedded within existing courses of non-math majors. The modules are archived at www.tlumathcsis.org. Participating faculty administered pre- and post-surveys, as well as a short three-question narrative survey, to gauge attitude change of students.

Across AY 20-21 and AY 21-22, 400 students participated in classes that incorporated mathematical or statistical modules. Disciplines included Biology (N=47), Economics (N=25), Kinesiology (N=84), Philosophy (N=31), Political Science (N=46), Psychology (N=81), Theology (N=73), and Visual Arts (N=13). Course grades for the 400 participants were analyzed using ANOVA with discipline, and gender as the independent variables. These results were mixed; however, it should be remembered that the primary focus of the study was on changing attitudes toward mathematics and statistics. The survey which was created in AY 20-21 included four factors : 1) Mathematical/Statistical Disposition (9-items), 2) Comfortableness in Using Math/Statistics (9-items); 3) Applications of Math/Statistics (8-items), and 4) Value of Quantified Data (7-items). Reliability of these factors were validated using Cronbach?s alpha index. Analyses were conducted where we had both a pre- and post- survey for each student. This resulted in 234 cases being analyzed. Repeated Measures 2X2 ANOVAS with gender as an independent variable, and with the four factors (time 1 and time 2) as the repeated dependent variables, revealed that ?Mathematical/Statistical Disposition? was higher for males than females at time 1, and that males had statistically significant improvements over time, whereas females did not. The same pattern was observed for? Comfortableness using Math/Statistics.? There were no statistically significant effects for ?Applications of Math/Statistics,? although there was a trend for males to improve in this area as well ( p-value =.092.)  For ?Value of Quantified Data,? males showed improvements over time, and females showed decrements. Collectively, these analyses suggest that the modules were impactful for improving attitudes of males but were less likely to do so for females. However, analyses that examined females who scored higher in ?Mathematical/Statistical Disposition? at time 1, demonstrated that these females revealed statistically significant improvements with ?Comfortableness Using Math/Statistics? over time. Therefore, it may be that some pre-existing openness to math and/or statistics of females made further enhancements in attitudes more likely to occur.

Mixed results are to be expected. Fear of quantitative reasoning will not disappear after a few hours in a quantitative module. It occurred to us that surveying attitudes immediately after the module would be more apt to measure a small change in attitude. To this end, starting  in AY 21-22, a survey was administered to 226 students immediately after completing the module in their classes. Students were asked to ?tell us something that influenced your attitudes toward Math/Statistics in a positive way? and ?tell us something that you found challenging.?  Students spontaneously noted appreciation of: the modules for addressing real life issues (24.37%); use of graphs and visualizations (17.65%); and for analysis and interpretation (14.29%). An example of a comment addressing real life issues was ?I have learned how important math/statistics really is in the real world and how math can be applied to almost anything.?  Challenges were noted regarding technology (29.94%), e.g., ?I?ve never used Python or coded before so getting through all the errors was challenging? and analysis and interpretation (29.34%), e.g., ?I found some of the solutions challenging to understand even when the math was correct.?  Some students recognized the intent of the project even though it was never explicitly stated, e.g., ?I find it really inspiring that the aim of this exercise is to integrate the use of math and stats in other classes more readily. I find that often students leave behind these subjects after the required courses but fail to realize how applied math (is) in every field. Showing these applications may be more work but necessary and helpful if not done to an excess.? The attitudinal shifts and spontaneous observations provided by students collectively reveal that the modules appear to be having their intended effects, although additional work to further attitudinal shifts of males and females who start with lower mathematical and statistical dispositions is still needed.

Last Modified: 10/25/2023
Modified by: Reza O Abbasian