There is a shortage of STEM (science/technology/engineering/mathematics)-trained individuals in the United States, and entire groups of individuals (e.g., females) are underrepresented in the STEM fields. The majority of research focuses on why females do not pursue STEM as careers, with researchers agreeing that a key component is self-efficacy, which is one’s belief in one’s capability of achieving a goal. Moreover, research demonstrates that self-efficacy or self-confidence in one’s abilities leads to an increased chance of pursuing a career in STEM. For this reason interventions have been developed to increase self-efficacy for STEM careers, and often these interventions are targeted towards females. However, there is disagreement about the effect of self-efficacy once one begins pursuing the goal of obtaining a career in STEM. Social cognitive theory proponents argue that self-efficacy will promote effort; whereas, control theory proponents argue that self-efficacy might undermine effort. STEM majors provide a unique population in which to study the relationships among the variables in a real world setting. In particular, this study investigates the relationship among gender, self-efficacy, effort, and performance in STEM-focused college courses.
To develop a set of hypotheses, the literatures on self-efficacy, which is a type of expectancy, were reviewed with a focus on the differences between social cognitive theory and control theory. Due to the theoretical and empirical evidence, control theory was used to generate the hypotheses. In particular, the expected comparatively lower self-efficacy of females versus males for STEM-related classroom goals was hypothesized to encourage effort and enhance performance towards classroom goals such as exam grade goals. Additionally, alternative explanations and important contextual variables in regards to the hypotheses were considered due to the field study methodology in the current study.
Data were collected using a sample of engineering, mathematics, and physics majors in mathematics courses required for their major. Self-efficacy, effort, and performance were measured multiple times over the length of one semester. Performance measures were collected from instructors, and ability indices (i.e., standardized test scores) were collected from the university, both based on permission from participants.
As hypothesized, gender was found to relate to self-efficacy and effort such that females had lower self-efficacy for mathematics exams than males and also studied more hours for the exams (e.g., M = 15.22, SD = 10.72) than males studied (e.g., M = 10.42, SD = 10.15). Further, self-efficacy was found to fully mediate the relationship between gender and effort (95% CI: 0.081 to 6.270). Females also reported a lower goal on the exam than males. There was no difference between the genders on exam performance, possibly due to stereotype threat concerns undoing the advantages of extra studying. Indeed, there was no difference between males and females on ability as assessed via standardized test scores. Other possible reasons for the lack of a performance effect were explored in the discussion. For instance, differences in classroom networking or study habits between males and females (e.g., access to old tests), as well as issues of diminishing returns for additional studying were discussed. Also, the broader impacts of the results on future STEM interventions that include self-efficacy or confidence-building components were discussed. Overall, this study demonstrated that self-confidence (i.e., self-efficacy) does not always have a positive effect on motivation. Furthermore, females’ lower self-confidence for science, technology, engineering, and math-related goals appeared to motivate studying behavior in mathematics courses.