2025 Collaborative Network for Engineering & Computing Diversity (CoNECD)

In this research paper, we describe our initial development of a holistic measure of intersectional socioeconomic inequality that accounts for both the inequalities that lead to and stem from differences in income, impacting (non)engineering students in and out of the classroom.

Low-income students’, henceforth referred to as socioeconomically disadvantaged students (SDS), challenges traversing engineering extend beyond their access to income specifically. Research suggests that SDS, especially those from multi-marginalized groups, are impacted by various overlapping forms of (non)socioeconomic inequality. This overlap leads to unequal access to resources and experience that, otherwise, supports STEM success. However, traditional socioeconomic measures do not reflect realities of inequality, instead measures focus on income or capital which are bare of these relationships. Yet, how socioeconomic disadvantage is defined alters how SDS are impacted by policy (e.g., financial aid), including how researchers and practitioners create a more equitable engineering education (Atwood et al., 2020). Thus, a measure that accounts for the complex processes that lead to resource deprivation is needed.

In this work, we consider inequalities that impact students and members of their communities amongst their home, neighborhood, and school environments, hypothesizing that these inequalities come together to create an “ecosystem of disadvantage” that adversely impacts SDS. Our process and final measure reflect this existence. To develop the measure, we obtained restricted-access to the US National Center for Education Statistics’ Education Longitudinal Study of 2002 (ELS:2002) which tracked the trajectories of 16,200 (non)engineering students from their 10th grade year in high school (2002) to their 8th year out of high school (2012). The restricted-access set contains finite geographic, demographic, and occupational data which we used to ascertain a more nuanced valuation of students’ various socioeconomics. Such nuance was determined by pairing the data with publicly available statistics from the US Bureau of Labor Statistics, Census Bureau, and Departments of Agriculture and Education, and statistics from the American Association of University Women. Exploratory factor analysis (EFA) was used to identify potential underlying factors amongst the socioeconomic data. Thereafter, structural equation modeling (SEM) was used to confirm the structure of the latent factors, and to model latent variables and relationships to other socioeconomic variables in tandem.

Using EFA, we identified that three latent factors, Parent Educational Involvement, Household Educational Resources, and School Hindrances existed that could be used to better describe students’ experiences in STEM. Using SEM, we not only confirmed the structure of these factors, but were also able to model them with several other traditional and non-traditional socioeconomic factors. The final model suggests that differences in Parent Educational Involvement, Household Educational Resources, and School Hindrances across the sample are an outcome of broader socioeconomic inequality driven by racism, sexism, and classism. This paper discusses the development of this model, our findings, and potential implications for future research, including our larger project which uses the measure to predict engineering student application to, and enrollment and persistence in, engineering.