Searching for Explanations for the Gender Gap in STEM: Methodological Approaches from across K-16 Presentations Linda J. Sax, UCLA Beth Casey, Boston College Karen Kim, CSU Fullerton Breakout groups 1. National datasets 2. Intervention studies 3. Qualitative research
Using the Freshman Survey to Study the Gender Gap in STEM ( or, Like Being a Kid in a Candy Shop) Linda J. Sax, UCLA
Women Overrepresented Across All Fields, but Underrepresented in STEM Proportions of Bachelor’s Degree Recipients, by Gender 100% 90% 80% 43 70% 65 60% 50% 40% 30% 57 20% 35 10% 0% All Bachelor's Degree Recipients STEM Bachelor's Degree Recipients Women Men Source: National Center for Education Statistics, Digest of Education Statistics, 2011
Women’s Relative Representation in STEM Varies by Field Proportions of Bachelor’s Degree Recipients, by Gender 100% 80% 42 57 59 60% 82 83 40% 58 43 20% 41 18 17 0% Engineering Computer Science Physical Sciences Mathematics/Statistics Biological Sciences Women Men Source: National Center for Education Statistics, Digest of Education Statistics, 2011
NSF Research Part I: Disaggregating the Gender Gap in STEM (NSF HRD #1135727) How has the gender gap in aspirations to major in different STEM majors changed over time? Freshman Survey • Nation’s largest and longest - running study of entering college students, housed at UCLA’s HERI • Between 1971 and 2011, responses from over 8 million students entering over 1,000 baccalaureate institutions • Study focuses on 5 STEM fields
Engineering Stable(ish) Gender Gap 25% 20% Men 15% 10% Women 5% 0% 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Source: Cooperative Institutional Research Program Freshman Survey, Higher Education Research Institute, UCLA
Computer Science Fluctuating Gender Gap 15% 14% 13% 12% 11% 10% 9% 8% 7% Men 6% 5% 4% 3% 2% Women 1% 0% 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Source: Cooperative Institutional Research Program Freshman Survey, Higher Education Research Institute, UCLA
Physical Sciences Diminished Gender Gap 15% 14% 13% 12% 11% 10% 9% 8% 7% 6% 5% Men 4% 3% 2% Women 1% 0% 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Source: Cooperative Institutional Research Program Freshman Survey, Higher Education Research Institute, UCLA
Math/Statistics No Gender Gap 15% 14% 13% 12% 11% 10% 9% 8% 7% 6% 5% 4% Men 3% 2% 1% Women 0% 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Source: Cooperative Institutional Research Program Freshman Survey, Higher Education Research Institute, UCLA
Biological Sciences Gender Gap Reversal 15% 14% 13% 12% 11% Women 10% 9% 8% 7% Men 6% 5% 4% 3% 2% 1% 0% 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Source: Cooperative Institutional Research Program Freshman Survey, Higher Education Research Institute, UCLA
NSF Research Part II: Evolving Predictors of STEM Major Selection How have the determinants of men’s and women’s STEM major selection changed over time? Example from computer science… Data from 1976, 1986, 1996, 2006, and 2011 18,830 first- year “intended” computer science majors 904,307 first-year students from all other majors Logistic regressions run separately by gender Dependent variable: Intent to major in CS Independent variables blocked according to Social Cognitive Career Theory Changing salience examined through “Variable x Time” interactions
Positive Men Women Father’s Career in STEM + + Asian/Asian-American + + African American + + Status Striving Orientation + + Negative Men Women Social Activist Orientation - -
BOTH GENDERS Positive predictors Math Self-Ratings Weaker positive Scholarly Orientation Stronger positive Negative predictors Family Income Weaker negative Leadership Orientation Stronger negative Goal: Raising a family Stronger negative WOMEN ONLY Scientific Orientation Stronger positive Artistic Orientation Weaker negative
NSF Research Part III: Decomposition Analysis What are the roots of the gender gap over time? Decomposition Analysis Used logistic regression and mean replacement to identify the extent to which gender differences in CS major choice can be attributed to: differences in average characteristics between men and women differences in the salience of those characteristics Conducted for 1976, 1986, 1996, 2006, 2011
What Matters Most in Explaining the CS Gender Gap? Proportions of the Computer Science Gender Gap Explained by Characteristics versus Coefficients 1976 1986 1996 2006 2011 Source of Gender Gap (%) (%) (%) (%) (%) Gender Differences in Characteristics 78.5 44.1 39.1 35.5 34.9 Gender Differences in Coefficients 21.5 55.9 60.9 64.5 65.1 Bottom line : The gender gap is increasingly explained by gender differences in the salience of variables, rather than the mean values of the variables.
Methodological Considerations Sample Upside : With 8 million students, possibilities for disaggregation are endless (unique STEM disciplines, institutional type, student demographics, etc.) Downside : Ditto! Variables Upside : Large number of variables with some consistency over 40 years Downside : Not 100% consistency over time (limitations in available variables); available variables not a perfect fit for theoretical framework Other Major aspiration ≠ degree earned
Recommend
More recommend
