Welcome to women, minorities, and technology!
Jacquelynne Eccles and Pamela Davis-Kean (PI's)
Oksana Malanchuk (Presenter)
Oksana Malanchuk (Presenter)
goals of wmt
Why are women and minorities underrepresented in the IT labor force? We will address this issue in terms of three sub-questions: (a) what are the psychological mediators of both gender and ethnic group differences in entry into and persistence in the IT labor force? (b) What are the family and school forces that underlie the gender and ethnic group differences in these psychological mediators? (c) How do experiences in tertiary educational settings and in IT work settings influence gender and ethnic group differences in entry into and persistence in the IT labor force? We propose three data analytic projects designed to answer these questions. In 1983, Eccles et al. outlined a theoretical framework based on expectancies, values, and achievement behaviors to study questions such as these. This framework has two components: a psychological component and a socialization component. For 17 years, we have used this framework to study educational and occupational choices using longitudinal survey designs that include multiple indicators from children, adolescents, young adults, parents, teachers, and school records. We have 3 large longitudinal data sets that we will use to assess the utility of the Eccles et al. frameworks for understanding the under-representation of females and minorities in occupational fields related to informational technology. We will use these data sets to test the following specific sets of predictions:
- To test the utility of the psychological components of Eccles et al. theoretical framework for understanding the psychological mediators of gender and ethnic group differences in activity and task choices (such as computer use and enrollment in IT courses or programs) related to entry into information technology jobs.
- To test the utility of the socialization (experiential) components of the Eccles et al. framework for understanding the social origins gender and ethnic group differences in the psychological mediators identified under #1.
data sets
Children and Beyond (CAB): Data collection began in 1987 with three cohorts (kindergarteners, and 1st and 2nd graders) of white middle class children, their parents, and their teachers. We have gathered 9 waves of data on these cohorts (4 during the elementary and beginning middle school, 3 during high school and 1 two years post high school graduation). We are seeking funding from NICHD to continue to follow this sample of young people through college and both post high school and post college employment. We have surveys, teacher assessments, and school record data and other publicly available information on the schools and communities across most waves on approximately 850 children; we have surveys from about 65% of their parents. The project focuses on 3 basic issues: (1) the ontogeny of competence beliefs, task value, and task importance for math, language arts, physical science, and computers; (2) the relation of these self and task beliefs to adolescents' academic and non-academic activity choices, performance in these activities, and long term educational and occupational planning; and (3) the relation of both current and previous experiences in the home and school to the ontogeny of these self and task beliefs, activity preferences and competencies. We gathered extensive information about all of the experiences the children had related to computers both at home and in school at each wave. This information will allow us to model the role of experiences in the home and at school on the development of gender differences in children's and adolescents' interest in computers and information technology.
Michigan Study of Adult Life Transitions (MSALT): Data collection for this study began in 1983 with 1700 fifth and sixth graders (and their parents and teachers) recruited from 10 different school districts in Southeastern Michigan. This sample represents white working and middle class families. We have collected 8 waves of data from these adolescents (4 waves in grades 6 and 7, two waves in 10th and 12th grade, one wave at age 20 and one wave at age 24-26 when the sample size is around 1400). We are now collecting a 9th wave of data at age 27-29. The study was initially designed to evaluate the impact of the junior high school math class environments and family environment on early adolescents' achievement-related self-concepts, interests, and motivation for math, physical science and other school subjects. We measured ability self-concepts and anxieties for math and physical science as well as many characteristics of math classroom environments. Over time, the study expanded to include information on educational and occupational planning, decisions, and choices. Throughout, the project was guided by the Eccles' theoretical framework. Thus, we have measures on self perceptions of one's competencies, interests in, and enjoyment of, various school subjects and extracurricular activities (including those linked to information technology); personal and vocational goals and values; gender-role stereotypes; educational, occupational and family plans; sources of information considered in making these plans; perceptions of the available opportunities and options; perceptions of the usefulness and intrinsic value of various options and of the likelihood of obtaining each option; and perceptions of the links between various adult roles (e.g., family roles and work roles). We have gathered information on the adolescents' perceptions of occupational fields involving information technology (e.g., their confidence and interest in their ability to work in fields that use computers), as well as information about what they think has driven them away from fields related to math and physical science. In addition, we have comprehensive information on youth's experiences in school, collected from the youth, their teachers, their principals, and school public records. These measures include indicators of academic achievement and performance on standardized tests; involvement in a variety of positive activities including work and extracurricular teams and clubs; exposure to different types of peers; and the nature of their secondary schools. Finally, we have extensive information about their experiences in various jobs either related to or not related to IT. These data will allow us to model comprehensively the psychological mediators outlined in Figure 1 of gender differences in entry into and persistence in the IT labor force.
Maryland Adolescent Development In Context Study (MADICS): This sample of approximately 1000 is comprised of African-American (61%) and European-American (35%) adolescents and their families of comparable SES distributions (making it possible to look independently at the impact of family income and ethnic group membership on occupational choice). We have 5 waves of data (beginning in 1991 when the students were in the 7th grade and again after grade 7, after grade 8, during grade 11, and one year post high school) gathered from the adolescents, their caregivers (parents, guardians, or others), and their school records. We have just completed gathering a sixth wave of data (at three years post high school) specifically focused on college experiences with a major focus on the students' perceptions of factors likely to either enhance or undermine their interest in information technology, as well as other college majors and occupations. Information was collected from the adolescents, their parents, and their schools on: (1) psychological adjustment and mental health; (2) social, musical, athletic, and academic (math and other subjects) competencies and performance; (3) self perceptions, sense of efficacy, and identity formation, (including both personal and social identities linked to: ethnicity, gender and religion); (4) educational and occupational goals and expectations; (5) expectations regarding discrimination and other barriers to achieving one's goals, as well as beliefs regarding one's ability to overcome these barriers; (6) involvement in leisure activities related to computer use, as well as other skill domains like athletics and instrumental music; and (7) complete educational [including courses taken and college major] and employment histories. We have added a series of items directly linked to information technology to wave 5 and our current wave. These indicators include stereotypes about the types of individuals who go into information technology, stereotypes about jobs in information technology, confidence and interest in various types of information technology jobs, experiences in computer science, information technology, engineering and math courses at college, and exposure to either encouragement or discouragement about majoring in an information technology subject or aspiring to a job in information technology. This data set will allow us to begin to test the applicability of the Eccles et al. frameworks for understanding ethnic group differences in entry into and persistence in the IT workforce.
Michigan Study of Adult Life Transitions (MSALT): Data collection for this study began in 1983 with 1700 fifth and sixth graders (and their parents and teachers) recruited from 10 different school districts in Southeastern Michigan. This sample represents white working and middle class families. We have collected 8 waves of data from these adolescents (4 waves in grades 6 and 7, two waves in 10th and 12th grade, one wave at age 20 and one wave at age 24-26 when the sample size is around 1400). We are now collecting a 9th wave of data at age 27-29. The study was initially designed to evaluate the impact of the junior high school math class environments and family environment on early adolescents' achievement-related self-concepts, interests, and motivation for math, physical science and other school subjects. We measured ability self-concepts and anxieties for math and physical science as well as many characteristics of math classroom environments. Over time, the study expanded to include information on educational and occupational planning, decisions, and choices. Throughout, the project was guided by the Eccles' theoretical framework. Thus, we have measures on self perceptions of one's competencies, interests in, and enjoyment of, various school subjects and extracurricular activities (including those linked to information technology); personal and vocational goals and values; gender-role stereotypes; educational, occupational and family plans; sources of information considered in making these plans; perceptions of the available opportunities and options; perceptions of the usefulness and intrinsic value of various options and of the likelihood of obtaining each option; and perceptions of the links between various adult roles (e.g., family roles and work roles). We have gathered information on the adolescents' perceptions of occupational fields involving information technology (e.g., their confidence and interest in their ability to work in fields that use computers), as well as information about what they think has driven them away from fields related to math and physical science. In addition, we have comprehensive information on youth's experiences in school, collected from the youth, their teachers, their principals, and school public records. These measures include indicators of academic achievement and performance on standardized tests; involvement in a variety of positive activities including work and extracurricular teams and clubs; exposure to different types of peers; and the nature of their secondary schools. Finally, we have extensive information about their experiences in various jobs either related to or not related to IT. These data will allow us to model comprehensively the psychological mediators outlined in Figure 1 of gender differences in entry into and persistence in the IT labor force.
Maryland Adolescent Development In Context Study (MADICS): This sample of approximately 1000 is comprised of African-American (61%) and European-American (35%) adolescents and their families of comparable SES distributions (making it possible to look independently at the impact of family income and ethnic group membership on occupational choice). We have 5 waves of data (beginning in 1991 when the students were in the 7th grade and again after grade 7, after grade 8, during grade 11, and one year post high school) gathered from the adolescents, their caregivers (parents, guardians, or others), and their school records. We have just completed gathering a sixth wave of data (at three years post high school) specifically focused on college experiences with a major focus on the students' perceptions of factors likely to either enhance or undermine their interest in information technology, as well as other college majors and occupations. Information was collected from the adolescents, their parents, and their schools on: (1) psychological adjustment and mental health; (2) social, musical, athletic, and academic (math and other subjects) competencies and performance; (3) self perceptions, sense of efficacy, and identity formation, (including both personal and social identities linked to: ethnicity, gender and religion); (4) educational and occupational goals and expectations; (5) expectations regarding discrimination and other barriers to achieving one's goals, as well as beliefs regarding one's ability to overcome these barriers; (6) involvement in leisure activities related to computer use, as well as other skill domains like athletics and instrumental music; and (7) complete educational [including courses taken and college major] and employment histories. We have added a series of items directly linked to information technology to wave 5 and our current wave. These indicators include stereotypes about the types of individuals who go into information technology, stereotypes about jobs in information technology, confidence and interest in various types of information technology jobs, experiences in computer science, information technology, engineering and math courses at college, and exposure to either encouragement or discouragement about majoring in an information technology subject or aspiring to a job in information technology. This data set will allow us to begin to test the applicability of the Eccles et al. frameworks for understanding ethnic group differences in entry into and persistence in the IT workforce.
main activities undertaken by wmt
- Data collection of open-ended IT questions in the MADICS study has been concluded.
- Initial coding of open-ended IT items in MADICS grant has been completed.
- MSALT data collection of IT questions to be put on the web prepared and awaiting sample readiness.
- All data sets: occupation codes for IT-related occupations are under review and being defined for future analyses.
- Initial descriptive analysis for all datasets have begun.
- Initial papers on achievement outcomes and parental socialization are underway (see below).
- Papers have been submitted to 2 major conference regarding IT grant.
main results to date
Results from CAB (submitted for presentation to the International Society for the Study of Behavioral Development)
Parental Socialization of Children's Information Technology Activities - Sandra Simpkins and Todd Bartko
Gender differences in children's engagement in math, science, and computer activities are present before Kindergarten. Because these differences emerge before the school years, children must form these early differences in information technology activities, at least, in part from their parents and the home environment. The goal of this study was to examine the associations between different methods of parental socialization of activities and children's engagement in early information-technology activities. This study included children from second, third, and fifth grade who participated in CAB. Mothers and fathers separately reported their encouragement of children's activities, parent-child engagement in activities (henceforth labeled coactivity), parents' engagement in activities (i.e., modeling), and children's engagement in math, science, and computer activities. Children also reported the amount of time they engaged in those activities.
Results indicated that parents' encouragement, coactivity, and modeling of math, science, and computer activities were positively correlated with children's activity engagement. Parental modeling typically evidenced the weakest associations with children's activities. Gender differences were examined through disparities in the means and differences in the associations between parental socialization and children's activities. The means of parental socialization and the associations between children's activities and parental socialization were similar for both boys and girls. In conclusion, both mothers' and fathers' socialization of children's activities was positively associated with children's engagement in those activities during the elementary school period. Types of socialization in which parents directly interact with their children, such as encouragement and coactivity, were strongly related to children's activities. The results concerning gender differences suggest that boys and girls receive similar amounts of socialization from parents and that parental socialization has similar links with their activity engagement. Gender differences in children's activities may result from different aspects of parental socialization, such as provision of toys and materials related to information-technology activities. In addition, gender differences may emerge from the differential influence of parental socialization on children's confidence and self-esteem in these areas.
Results from MSALT (Submited for presentation at Society for Research on Adolescence)
Influences of Gender on Academic Achievement - Miriam Linver and Pam Davis-Kean
Growth curve models for adolescents' school math grades were estimated to address the following questions: (a) What do the average math grade trajectories look like, from 6th to 12th grade, by gender and by school track? (b) What impact does interest in math have over and above the effect of mothers' education and gender, by school track? Overall, there is a declining trajectory, but young women have higher grades than young men (within each tracking group). For all four groups, both the intercept and slope of the trajectory are significantly different from zero, so we are justified in estimating predictors of both parameters in our next analyses. The honors/college track starts with higher math grades, but these grades fall more dramatically than the regular/basic group. Gender, mothers' education, and interest in math are related to school grade trajectories. For the honors/college tracking group, mothers' education is related to both intercept and slope. Mothers with higher education levels have adolescents with a higher school grades intercept, but a slope that declines more quickly. For the regular/basic group, mothers' education level is unrelated to trajectories. For both tracking groups, young women have significantly higher intercepts; gender is unrelated to slope. Finally, higher math interest is also positively related to a higher intercept, but unrelated to slope for the honors/college group. For the regular/basic group, math interest is related to both intercept and slope: those with a higher math interest have a higher intercept, and their slope declines more slowly than those with a lower math interest. These results suggest that for both boys and girls, math grades fall over the course of junior high and high school. Math interest explains some of this decline, over and above students' gender and mothers' level of education.
Parental Socialization of Children's Information Technology Activities - Sandra Simpkins and Todd Bartko
Gender differences in children's engagement in math, science, and computer activities are present before Kindergarten. Because these differences emerge before the school years, children must form these early differences in information technology activities, at least, in part from their parents and the home environment. The goal of this study was to examine the associations between different methods of parental socialization of activities and children's engagement in early information-technology activities. This study included children from second, third, and fifth grade who participated in CAB. Mothers and fathers separately reported their encouragement of children's activities, parent-child engagement in activities (henceforth labeled coactivity), parents' engagement in activities (i.e., modeling), and children's engagement in math, science, and computer activities. Children also reported the amount of time they engaged in those activities.
Results indicated that parents' encouragement, coactivity, and modeling of math, science, and computer activities were positively correlated with children's activity engagement. Parental modeling typically evidenced the weakest associations with children's activities. Gender differences were examined through disparities in the means and differences in the associations between parental socialization and children's activities. The means of parental socialization and the associations between children's activities and parental socialization were similar for both boys and girls. In conclusion, both mothers' and fathers' socialization of children's activities was positively associated with children's engagement in those activities during the elementary school period. Types of socialization in which parents directly interact with their children, such as encouragement and coactivity, were strongly related to children's activities. The results concerning gender differences suggest that boys and girls receive similar amounts of socialization from parents and that parental socialization has similar links with their activity engagement. Gender differences in children's activities may result from different aspects of parental socialization, such as provision of toys and materials related to information-technology activities. In addition, gender differences may emerge from the differential influence of parental socialization on children's confidence and self-esteem in these areas.
Results from MSALT (Submited for presentation at Society for Research on Adolescence)
Influences of Gender on Academic Achievement - Miriam Linver and Pam Davis-Kean
Growth curve models for adolescents' school math grades were estimated to address the following questions: (a) What do the average math grade trajectories look like, from 6th to 12th grade, by gender and by school track? (b) What impact does interest in math have over and above the effect of mothers' education and gender, by school track? Overall, there is a declining trajectory, but young women have higher grades than young men (within each tracking group). For all four groups, both the intercept and slope of the trajectory are significantly different from zero, so we are justified in estimating predictors of both parameters in our next analyses. The honors/college track starts with higher math grades, but these grades fall more dramatically than the regular/basic group. Gender, mothers' education, and interest in math are related to school grade trajectories. For the honors/college tracking group, mothers' education is related to both intercept and slope. Mothers with higher education levels have adolescents with a higher school grades intercept, but a slope that declines more quickly. For the regular/basic group, mothers' education level is unrelated to trajectories. For both tracking groups, young women have significantly higher intercepts; gender is unrelated to slope. Finally, higher math interest is also positively related to a higher intercept, but unrelated to slope for the honors/college group. For the regular/basic group, math interest is related to both intercept and slope: those with a higher math interest have a higher intercept, and their slope declines more slowly than those with a lower math interest. These results suggest that for both boys and girls, math grades fall over the course of junior high and high school. Math interest explains some of this decline, over and above students' gender and mothers' level of education.
next steps planned
- Finish MSALT data collection on the web about IT-related occupation choices.
- Design and complete 40 in-depth qualitative interviews with MSALT participants (male and female) who are currently in IT jobs (at age 30) or who earlier expressed interest in IT jobs but are not currently employed as such.
- Begin analyzing quantitative data to explore issues regarding African American's college majors and occupational aspirations.
- Begin major analysis across all three studies to examine predictors of IT participation in the workforce.
- Code and analyze MSALT qualitative data.
Key open research issues
- What is the impact of good math achievement on course selection in college and eventual occupational choice?
- Are those participants who are high in achievement the only ones who are going into IT jobs or are there other avenues?
- Do African Americans and women feel more discrimination in the IT workplace?
- What kinds of stereotypes exist involving IT jobs and are they more prevalent in certain genders or race?
- How does parental and peer encouragement and socialization impact on eventual IT career choices?
publications, presentations and reports
Coming soon.