A Social Cognitive Construct Validation: Determining Women's and Men's Success in Engineering Programs

Traditionally, women have comprised the majority of students in the humanities and arts. Recently, their numbers have grown in the life sciences and social sciences, until today, when their numbers are on par with or exceed those of men. However, the physical, computing, and engineering sciences provide a very different picture. Annually, women have earned approximately 25% of the degrees in computer science and less than 20% of the degrees awarded in physics (National Center for Educational Statistics [NCES], 2000). Moreover, in engineering, the percentage of women graduating from either undergraduate or master's programs is approximately 19% annually (National Science Foundation [NSF], 1998). After several decades trying to increase women's presence in doctoral engineering programs, women only receive slightly less than 20% of the engineering doctorate degrees (NSF, 1998; Gibbons, 2004). In 2000, the NCES reported an unprecedented turn of events. The document stated, "female students did not fall behind in the S & E pipeline; they actually did better than male students in degree completion and program switch. Women who do enter are likely to do well" (p. ix). This report creates a departure from past research where women had reported higher attrition rates and slightly lower GPAs. To understand the recent success of women in engineering, it might be enlightening to profile the women currently entering these programs. Specifically, we identify four noticeable changes in these young women as compared to those in the past: 1. They are at the top of the mathematics test score range (Brainard & Carlin, 1998; Nauta, Epperson, & Kahn, 1998; Turner & Bowen, 1999). 2. They are as likely as males to have taken the appropriate prerequisite mathematics, science, and physics courses in high school (NCES, 2000; National Institute for Science Education [NISE], 1998). 3. They are unambiguous about their academic and career choices (NISE, 1998). 4. They are confident in their academic abilities (Brainard & Carlin, 1998; Nauta, Epperson & Kahn, 1998; NCES, 2000). Besides solid academic preparation, healthy self-confidence, and lack of ambiguity about their choice of major, other explanations cited for this reversal in previous trends are strong family support and females' high expectations for success (NCES, 2000; NISE 1998). Despite the good news, the literature still strongly asserts that women experience more interpersonal difficulties in their science and engineering courses (NISE, 1998; NSF 1998; Seymour, 1995; Seymour & Hewitt, 1997). The lower number of female engineering majors--compared to women majoring in other sciences--indicates that the engineering pipeline may be especially intimidating for women (NCES, 2000; NISE 1998; Seymour, 1995). According to Zeldin and Pajares (2000), social cues can subtly dissuade women from pursuing studies in male domains such as engineering and related subjects. In past studies, women have criticized educators for being rigid, closed, and condescending (Seymour & Hewitt, 1997). These characteristics have been theorized to have lowered women's engineering self-confidence and less self-efficacy. Herein lies the discrepancy between the literature and the recent report from NCES (2000). It is this inconsistency that provides the basis for our research. We queried if women are still experiencing greater levels of discrimination and discontent in their engineering classes because previous studies have reported that unfriendly environments have decreased females' academic confidence, self-efficacy, and motivation to pursue engineering majors. In turn, this may have influenced their attrition into other majors. In order to measure the effect of environment on student behavior, we used Bandura's (1986) social cognitive model as the basis of this study. He classified three self-referent constructs of environment, self, and behaviors as self-reinforcing, symbiotic, and dynamically changing. …