Reforming early undergraduate instruction influences long-term retention and attrition in STEM majors

Background/Question/Methods

Despite national calls for increasing literacy and education in the areas of Science, Technology, Engineering, and Mathematics (STEM) in order to meet contemporary societal challenges, changes in science education to respond to this call have been slow.  Numerous studies cite improved learning gains for students in the context of reformed courses, yet there remains a paucity of data about longer term impacts, including whether reformed instruction influences students’ decisions to persist in earning degrees in traditional STEM disciplines.

We reformed a large-enrollment, introductory biology course (Bio1) for life science majors using Backward Design principles, existing theory about how people learn, and national and local goals for student learning.  Between Spring 2008 and Fall 2011, 9 instructors adopted the reformed version of the course (n=2240 students) whereas, 6 instructors continued with traditional versions of the course (n=2938 students).  We compared the declared majors of students during their enrollment in the introductory course to their declared major at the conclusion of the Fall 2011 semester. Students that were STEM majors who were still STEM majors at the end of Fall 2011 were classified as “retained”.  Students that had no declared major or were declared as non-STEM majors and were now STEM majors at the end of Fall 2011 were classified as gained.  Those students who switched from STEM majors to non-STEM majors were classified as lost. Non-STEM majors who stayed in non-STEM majors in Fall 2011 were excluded from the study. 

Results/Conclusions

Preliminary results indicate that students from the reformed course switch into STEM majors and stay STEM majors more often than those who took traditional versions of the course (X²=10.531, df=2, p-value=0.005167).  This trend may be driven by lower performing students (lower 1/3 of entering GPAs) from reformed sections who are retained and gained in STEM majors at a higher rate compared to traditional sections (X²=6.506, df=2, p-value=0.03866).  Although women do not differ statistically between treatments, men from the reformed sections are less likely to leave STEM majors compared to men  from the traditional sections (X²=6.975, df=2, p-value=0.03058).  We are continuing to explore the role other explanatory demographic and/or academic variables in order to more fully explain patterns of attrition and retention in this case.