Background
Self-efficacy has been found to be an important factor in the success of women studying engineering (Blaisdell, 2000; Marra, Schuurman, Moore & Bogue, 2005). Self-efficacy is belief in ones capabilities to organize and execute the sources of action necessary to manage prospective situations (Bandura, 1986). Although efficaciousness applies to any situation, it is particularly important in choosing and executing constructive actions in situations that are perceived as negative or a barrier to success (e.g. lack of a meaningful role in a team project). Given that women are generally under-represented in engineering classrooms, a strong sense of efficacy can help them to persist in such situations.
This paper will report on a two year study longitudinal study of women engineering student self-efficacy with data collected from five institutions across the United States. We measured self-efficacy via the LAESE survey instrument (longitudinal assessment of engineering self-efficacy) (see aweonline.org). LAESE is a validated instrument developed via the NSF-funded Assessing Women in Engineering (AWE) project. LAESE provides results in six sub-scales. Our analysis examined the data for longitudinal differences in the subscales.
Methods
Subjects & Procedures Subjects were 165 undergraduate women studying engineering at Penn State University, Georgia Institute of Technology, University of Texas Austin, University of Arizona and the University of Louisville. There were 51 ethnic minorities in the sample. Early in the fall 2003 and 2004 terms, subjects were recruited via email, phone and other types of written communications. In all cases, subjects were women engineering students who had some affiliation with the WIE program at that institution .
Instrument
LAESE is tested and validated survey instrument designed to measure self-efficacy in engineering, feelings of inclusion and outcomes expectations (Marra et al., 2005; Marra, Moore, Schuurman, Bogue, 2004; aweonline.org). Results of our validity and reliability analyses showed that the survey measured several factors that are related to the concepts of self-efficacy, inclusion and outcomes expectations. This resulted in six subscales that are listed in Figure 1.
Subscales 1. Engineering self-efficacy I. 2. Engineering career expectations. 3. Engineering self-efficacy II. 4. Feeling of inclusion and having engineering role models 5. Confidence in being able to cope with difficulties. 6. Expectation that math is important for career and self worth. Figure 1. LAESE subscales.
Results
We examined the data to answer the following research questions: Are there longitudinal differences for the defined subscales for the overall data set? Are there longitudinal differences between institutions for the defined subscales? Are there longitudinal differences between students in different years for the defined subscales? Are there longitudinal differences between students of different ethnicities for the defined subscales?
For the overall data set, using paired t tests we found the following significant differences in the subscales: Longitudinal significant decreases 03 04 in feelings of inclusion (t = 2.37, p < .05). Longitudinal significant increases from 2003 to 2004 on second engineer efficacy, and coping self efficacy (t = -5.124, p < .01; t = - 3.874, p < .01).
When we disaggregated the data by ethnicity, school and year standing we found the following using a repeated measures ANOVA:
A main effect for inclusion with ethnicity (p = .034)
Post hoc analysis between different ethnicities shows no significant differences but African American means are lower than all other ethnic group means on inclusion.
When one performs a multivariate analysis (repeated measures) with a new ethnicity variable (African American or not) we find that African Americans do have significantly different inclusion means from all other ethnicity groups. In a further analysis we checked to see if this true for all non white groups and it was not so we can isolate this result to African Americans.
Math outcomes expectations Significant main effects and interactions for School year, Ethnicity, School (p < .01 for all three results). Post hoc analysis found that the difference in math for the different ethnic groups is between Asian students (higher) and all other races. Post hoc analysis on school year found that the difference is between 2nd year (higher means) and all other students.
The conference paper will provide further detail on the analysis that produced these results, discuss future data collection and analysis for this project and the implications of these results for WIE programs, and engineering education in general.
References:
Bandura, A. (1986). Social foundations of thought and action: a social cognitive theory. Englewood Cliffs, N.J. : Prentice-Hall.
Blaisdell, Stephanie (2000). Social cognitive theory predictors of entry into engineering majors for high school students. Arizona State University. Unpublished dissertation.
Marra, R.M., Schuurman, M., Moore, C., & Bogue, B. (2005). Women engineering students self-efficacy beliefs The longitudinal picture. Proceedings of the annual meeting of the American Society for Engineering Education Annual Conference a, 12 15 June, Portland, OR.
Marra, R.M., Moore, C., Schuurman, M., & Bogue, B. Assessing Women in Engineering (AWE): Assessment Women Engineering Students Self-Efficacy Beliefs. Proceedings of the Annual meeting of the Women Engineering Program Advocacy Network. June 6 9, Albuquerque, N.M. Available at http://www.x-cd.com/wepan04/prof14.html.
