MATH VALUES

View Original

Persistence of Black and Latina/o Students in STEM

By David Bressoud @dbressoud

The MAA’s calculus studies have highlighted problems with the retention of women in STEM disciplines, especially in the calculus sequence (see the October 2016 Launchings). Because the numbers are so small, we were not able to say anything about the persistence of Black and Latina/o students. But there are recently published and disturbing results on the persistence of these minority students (Riegle-Crumb et al, 2019), based on weighted data from the National Center of Education Statistics’ 2004/09 Beginning Postsecondary Students Longitudinal Study (BPS:04/09).

The authors approached the data with three research questions about students who either began in a 4-year undergraduate program or switched into a 4-year program from a 2-year college. I paraphrase their wording:

  1. Is there a difference between White students and Black or Latina/o students in the rate of persistence in a STEM major? Lack of persistence could occur either by switching to a major outside of STEM (labelled “switch”) or failing to earn an undergraduate degree within six years (labelled “leave”). STEM is defined as a biological science, computer science, engineering, a mathematical science, or a physical science.

  2. If there are differences, can they be explained by factors such as socio-economic status or high school preparation?

  3. How do the findings for students in STEM compare to those for students in business, social sciences, or humanities?

Figure 1 shows that there is very little difference in the percentage of students who choose a STEM major. In answer to the first question and partial answer to the third, Figure 2 shows that there are considerable differences in persistence rates among STEM majors, with less pronounced differences in Business and Social Sciences, and almost none in Humanities. Among STEM majors, Black students are far less likely to persist, with much higher percentages both for switching and leaving (p < .001).

Figure 1. Choice of college major by race/ethnicity.
Reprinted from Riegle-Crumb et al, 2019, p. 136. Source: BPS:04/04. N = 5,626.
* Indicates a significantly higher percentage of Black students chose to major in business, compared to both White and Latina/o students (p < .05).

Figure 2. Persistence patterns in chosen field by race/ethnicity.
Reprinted from Riegle-Crumb et al, 2019, p. 137. Source: BPS:04/04. N = 5.626.
*p < .05, **p < .01, ***p < .001.

 To answer question 2, the authors applied multivariate analysis, estimating logistic regression models predicting the likelihood of switching versus persisting and leaving versus persisting in each of the four categories of majors. Model 1 simply looked at race/ethnicity. Model 2 incorporated a wide variety of factors that included gender, socio-economic status, full or part-time employment, and characteristics of the post-secondary institution. Model 3 added four variables that measure high school preparation: SAT score, High School GPA, having taken Precalculus or Calculus in high school, and having taken four years of science in high school.

The results, which can be found on pages 139–140 of their article, are far too detailed to try to reproduce here. The main takeaway is the answer to question 2. Taking into account all of the additional factors, including high school preparation, the difference between White and Black students for switching remains high (p < .01). Between White and Latina/o students the difference essentially disappears once socio-economic factors are taken into account, even before factoring in high school preparation.  Differences from White students in terms of leaving remain high for both Black and Latina/o students (p < .01), even with all other variables factored in.

It is significant than in the three other categories of majors, differences between White and Black or Latina/o students in either switching or leaving essentially disappear once socio-economic factors have been included. Thus, there appears to be an additional mechanism at work in the STEM fields. The authors suggest that this might be explained by stereotype threat. In their words,

While such spaces [STEM degree programs] are challenging to navigate for most students, minority students experience these spaces while subjected to specific stereotypes about their presumed inferior cognitive and mathematical ability. Put briefly, in STEM contexts, the presence of stereotype threat is likely to be very high. (p. 142)

The authors reference two papers on stereotype threat (Beasley and Fischer, 2012; Woodcock et al, 2012). It occurs whenever a student is aware that a particular group to which she or he belongs is expected to perform at a lower level than the dominant group. This is well-documented to produce additional stress that lowers performance as well as strengthening any uncertainties about belonging in a particular program.

Faculty biases can be subtle, imperceptible to those of us who have them but easily reinforcing stereotype threats to our students. The more we and our students know about them, the better equipped we all are to recognize and deal with them.

References

Beasley, M.A. & Fischer, M.J. (2012). Why they leave: The impact of stereotype threat on the attrition of women and minorities from science, math and engineering majors. Social Psychology of Education, 15(4), 427–448. https://link.springer.com/article/10.1007/s11218-012-9185-3 

Bressoud, D. (2016). MAA Calculus Study: Women in STEM. Launchings October, 2016. http://launchings.blogspot.com/2016/10/maa-calculus-study-women-in-stem.html

Riegle-Crumb, C., King, B., & Irizarry, Y. (2019), Does STEM stand out? Examining racial/ethnic gaps in persistence across postsecondary fields. Educational Researcher. 48(3), 133–144. https://journals.sagepub.com/doi/abs/10.3102/0013189X19831006

Woodcock, A., Hernandez, P.R., Estrada, M., & Schultz, P.W. (2012). The consequences of chronic stereotype threat: Domain disidentification and abandonment. Journal of Personality and Social Psychology, 103(4), 635–646. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3779134/