A Q factor analysis approach to understanding female college students’ attitudes toward multiple STEM disciplines

Abstract

Research on gender disparities in STEM (Science, Technology, Engineering, and Mathematics) has paid little attention to the fact that not all STEM disciplines experience the same degree of gender imbalance. Previous research has primarily examined a single STEM discipline or combined STEM disciplines in their analyses. This study addressed some of the limitations of previous research using an innovative statistical approach, Q factor analysis (QFA). QFA is used to explore multifaceted human perceptions, behaviors, and experiences. It enables researchers to categorize people based on their pattern of responses and opinions on a certain topic, in contrast to the more commonly used R factor analysis that categorizes variables. QFA was applied to a sample of 98 female undergraduate students who were enrolled in introductory STEM courses. Participants competed a survey that assessed their attitudes, experiences and beliefs about math, science, and computers. Questions tapped into constructs typically used in social cognitive models of academic and career choices. Two typologies emerged from the analyses. The math-computer group had favorable attitudes and beliefs toward math and computers and less interest in science; whereas the science group had more favorable attitudes and beliefs towards science. Participants? major choice and self-reported academic support aligned with the two groups in ways that were consistent with the groups? interests. The study demonstrates the potential for QFA to be applied with various types of data on a wide range of topics and to address questions that are not easily answered using traditional statistical approaches.