2025 ASEE Annual Conference & Exposition

Advocacy and programming efforts for STEM representation often use data to illustrate progress - or the lack thereof - across gender and ethnicity. For example, at one large, public university in 2022, 46% of STEM undergraduates were women. This might lead an observer to conclude that great progress has taken place, but looking deeper into this data reveals that 28% of engineering students were women and 72% of Agriculture and Life Science students were women in 2022. The manner in which data is grouped may allow statistics to report desirable outcomes, but clearly, it has the potential to hide patterns we should further investigate, a phenomenon known as Simpson’s Paradox.

Looking at data within specific subpopulations can reveal trends and issues that help us identify where gaps still exist and where more effort is needed. Data, such as that presented in the National Center for Science and Engineering Statistics (NCSES) report, Diversity and STEM: Women, Minorities, and Persons with Disabilities [1] , are presented as a function of sex or of race/ethnicity, but not both. A closer look at such data shows interesting differences in the percentages of women in STEM fields among various ethnic groups.

Another area of interest in the data is the comparison of the number of STEM graduates to the number of persons employed in STEM, also divided by sex and race ethnicity. The commonly used sources for such data - the US Census and the National Science Board/ National Science Foundation report, Science and Engineering Indicators [2] - use specific definitions for STEM and non-STEM fields. Interestingly, these definitions do not always align with the definitions used to define and count STEM graduates from US institutions.

This paper seeks to compare trends in the data as reported and the trends observed when groups are divided into smaller, more specific subpopulations to uncover trends that may inform and motivate programming, recruitment, and student support organizations. We begin by discussing a case study of the College of Engineering at a public university, looking at program demographics as well as retention rates for each of the specified groups.

In addition, the paper examines differences in definitions of STEM and non-STEM fields that vary depending on data sources, and asks the question whether these hidden differences can also hide important trend information.

1 National Center for Science and Engineering Statistics (NCSES). 2023. Diversity and STEM: Women, Minorities, and Persons with Disabilities 2023. Special Report NSF 23-315. Alexandria, VA: National Science Foundation. Available at https://ncses.nsf.gov/wmpd.

2 National Science Board, National Science Foundation. 2019. Science and Engineering Indicators 2020: Science and Engineering Labor Force. Science and Engineering Indicators 2020. NSB-2019-8. Alexandria, VA. Available at https://ncses.nsf.gov/pubs/nsb20198/.