Abstract
This work-in-progress research study examines the response patterns of first-generation college students (FGCS) to the engineering identity measures compared to non-first- generation college students (non-FGCS). This work answers the following research question, “Do FGC and non-FGC engineering students interpret the engineering identity measurement items in a conceptually different manner?” We explore if FGCS respond to engineering identity items similarly to non-FGCS and the fairness of using these instruments for FGCS to make claims about this group. The data for this work are from a survey instrument completed by 2,916 first-year engineering college students from four U.S. institutions. We hypothesize that quantitative measures constructed for the general engineering student population (non- FGCS) may not function the same for a FGCS subpopulation in engineering. Using extensions to the confirmatory factor analysis, we tested for measurement invariance of engineering identity constructs between FGCS and non-FGCS. Our comparative analysis of FGCS and non-FGCS found weak measurement invariance within the engineering identity constructs (i.e. interest, recognition, and performance/competence) indicating a similar factor structure and factor loadings, but different uses of the identity item scale. This research raises questions on the use and fairness of normative measures in engineering education for populations that fall outside the majority engineering student population.
Keywords
first-generation college students, engineering identity, confirmatory factor analysis
Date of this Version
2017
Comments
Final paper published in the 2017 FIE Conference proceedings.
(c) 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.
The version of record is available at http://doi.ieeecomputersociety.org/10.1109/FIE.2017.8190616.