Proposal to National Science Foundation CISE Directorate Division of Experimental and Integrative Activities Information Technology Workforce (ITW) FY 2001 Competition Stretching Horizons: Upward Bound Programs in Stimulating Information Technology Education and Career Aspirations among Underrepresented Minorities Project Focus This project is an attempt to understand why minority group members and women are severely underrepresented in information technology careers. The extent of minority and female underrepresentation is apparent in Department of Commerce figures. In 1996 women made up 30 percent of the IT work force, black people 5 percent, and Hispanic people less than 5 percent (NSF, 2000a, 2000b). In the U.S. population, women are 51 percent, African Americans 12 percent and Latinos more than 12 percent. Women and minorities are underrepresented in college IT majors, but that discrepancy begins as early as high school (Benning, 1998). Anita Borg, founder of the Institute for Women and Technology, has asserted that if women had been going into IT at the same rate as men since 1984, no shortage of IT workers would currently exist (Olsen, 2000). The focus of our research proposal is the information technology higher education pipeline, and why so few male and female African American and Latino/a students are studying computer science at the college level. We specifically will research the experiences, opportunities, attitudes and aspirations with respect to mathematics, science, computer science and information technology of underrepresented students in the federally funded Upward Bound and Math/Science Upward Bound programs. Besides comparing the experiences afforded students by the two versions of Upward Bound, we will compare these experiences and opportunities with those of students in non-Upward Bound high schools. Our threepart research model is synthesized from the pipeline concept, social cognition theory, and the status attainment model. First, the pipeline concept focuses on the interplay between high school educational environments(the availability of course offerings, prerequisites, resources, teacher training, and mentoring), and the psychological and cultural factors that influence male and female college-bound underrepresented minority students’ interests, motivations, and experiences regarding computer science, access and home computing environments, cultural assumptions of who will and will not succeed in computer science, race and gender stereotypes, peer dynamics). While we will consider the educational environments, cultural assumptions of who will and 1
will not succeed in computer science, race and gender stereotypes, peer dynamics). While we will consider the educational environments in high schools attended by our students, our emphasis will be on the additional experiences made available through Upward Bound and the educational and vocational aspirations of students. The status attainment model will guide the selection of variables and analytic approaches used to study the educational and career aspirations of our target populations and how they are influenced by the Upward Bound experiences. Through a variety of qualitative (focus group and individual interviews) and quantitative (sample survey) methodologies, we will study the educational and vocational aspirations of high school students as they relate to computer science and information technology college majors and, ultimately, to postgraduate careers. In addition to the focus on high school experiences and attitudes about computer science and information technology, we also will study graduates of Upward Bound in college and former Upward Bound students who have traveled through the pipeline to graduate from college. These students and former students will offer perspectives on decisions to take courses in computer science and information technology. The sites of our study are schools in the three-county Southern California region IX of Upward Bound, Los Angeles, Ventura and Orange Counties. Why Upward Bound? Given the critical underrepresentation of African Americans and Latino/as in the information technology workforce and in the college classrooms educating that workforce, we thought that an appropriate frame would be a national program designed to stimulate the skills and motivation in first-generation or low-income high school students for success in postsecondary education. Since Upward Bound students are from the lowest socioeconomic quintile and/or have parents who have not attended or graduated from college, they are the epitome of "underrepresented students." While Upward Bound was not expressly set up to target underrepresentation in the information technology workforce, its emphasis on assisting nontraditional high school students to attend college could indirectly produce more college computer science students from this segment. The Upward Bound Program is one of several federally funded programs known as TRIO programs. In order to support a commitment to providing educational opportunities for Americans across race, ethnicity, and economic background, the United States Congress established a series of programs to help lowincome Americans enter college and graduate. As indicated previously, the Upward Bound Program assists high school students from low-income and/or first-generation
2
college families (where neither parent graduated from college) prepare for higher education. Participants take courses in literature, foreign languages, composition, mathematics, computers, and science after school, on Saturdays, and in the summer in six-week residential programs at cooperating colleges and universities. Nationally, about 50,000 high school students are currently served by 579 projects (Federal TRIO Programs, Upward Bound, 2000). Programs are housed at colleges and universities where students take courses. Projects are required to keep records and maintain contact with Upward Bound students and graduates during high school, college attendance, and after graduation from college. Consequently, it is possible to use existing Upward Bound data and to collect additional data to investigate variables that may influence decisions to pursue high school prerequisites for college admission and IT careers following college. Upward Bound targets schools that have high populations of educationally at risk students, who come from families without college going parents and/or whose families are low-income. For example, the six target Upward Bound high schools in Ventura County have high dropout rates. Upward Bound is designed to -increase high school graduation rates -increase competency in challenging subject matters including English, mathematics, science, literature, and foreign language -encourage more students to pursue programs that lead to careers in math and science -gain parental support in the social, emotional, and academic growth of their children. Given the critical role of math and science as prerequisites for computer science, the emphasis by Upward Bound on increasing competency in such challenging subject matter may prove beneficial in understanding how to widen the narrow pathway to information technology careers. Theoretical Models in the Study The Pipeline Model The starting point for investigating the intricate dynamic of factors that influence students' decision is the pipeline model developed by Oakes (1989) for examining reasons for the low numbers of women and non-Asian minorities in math and science education. The math/science pipeline may best be understood as a phenomenon where, beginning in middle/junior high school, the number of female students, students with lower socioeconomic status, and students of color in proportion to White males in advanced math and science progressively declines during high school
3
(Oakes & The Rand Corporation, 1990, p. 154). There is evidence from research that the pipeline continues to function in college (Holland & Eisenhart, 1990). This model takes into account the structure of schools, including tracking, as well as the relevant psychological and sociological influences, such as confidence and racial-identity issues by examining a) opportunities to learn, b) achievement and c) decisions. Following Margolis and Oakes (2000) we believe that students' engagement and disengagement with computer science involves a myriad of cultural, social and educational issues unique to the field of computer science and different from those in the math and science research. While there is likely a complex interaction between students' self-perceptions and the context of the social, cultural, and structural constraints of computer science education, the strengthening of the sense of self in Upward Bound programs should be a key factor. The role of pre-college outreach programs, including Upward Bound, has been the subject of sustained research interest (Blake, 1998; Kaplan, 1999; McLure & Child, 1998; Thomas, Farrow & Martinez, 1998). Research done on these precollege tutorial and counseling programs suggests that the resilience and self-esteem fostered by such programs is a key benefit (Kaplan, 1999). We propose to apply these research approaches to understand how to widen the pipeline to more underrepresented students in computer science and information technology. Following, Moses et al. (1999), we will focus on course-taking patterns, student attitudes toward pipeline courses, student aspirations, and the context of choice as these are influenced by participation in Upward Bound. Upward Bound and Math/Science Upward Bound provide counseling to students who normally might fail to choose math and science courses because they are either effectively gender segregated or unlikely to have much representation by lowSES students or students. Further, girls and minorities are likely to stereotype mathematics and science as a White male domain. In this context, Oakes observes that “considerable work suggests that girls and minorities show a greater interest in people, whereas white boys show more interest in things” (Oakes & The RAND Corporation, 1990, p. 171). Student aspirations are a key to persisting in courses that open the pipeline to college, particularly in mathematics and science. Lack of female role models and the belief that careers in scientific field make it difficult or impossible to raise a family may suppress female aspirations to mathematics, scientific, or computer science careers (Eccles, 1987; Michigan Department of Education, 1990; Gaskell, 1984; Oakes & The RAND Corporation, 1990; Holland & Eisenhart, 1990). Finally, students of lower socioeconomic status (SES) often operate within social contexts that are less likely to lead them into the pipeline. Apparently, collegeeducated parents expect their children to go to college. Therefore, they follow
4
through to ensure that their children continue in the pipeline, while parents with less formal education might not, due to a variety of different factors (Oakes & The RAND Corporation, 1990). The context of choice is characterized by two kinds of factors that serve to filter girls and lower-SES and students of color out of the pipeline: (a) various sorting mechanisms – counseling, testing, and tracking, for instance – and (b) students’ own choices. Following Moses et al. (1999), we are strongly intrigued by the latter. As sorting is increasingly rejected, the fact that students freely decide to exclude themselves from the pipeline is used to justify unequal patterns. This justification is one that Moses et al. (1999) encountered in their work in public schools and which is found in the “current backlash against the goal of equity” p. 578 (Sommers, 1996). Through their survey, Moses and her colleagues sought to confront and refute it. In that survey, they found that girls and students of lower SES exhibited more negative attitudes about math, science, and computer courses. Social Cognition Theory The pipeline phenomenon is the first contribution of three theoretical formulations to the framework we propose in our study. The second model is the social cognitive theory of career and academic interest, choice, and performance (Lent, Brown, & Hackett, 1994). Mau, & Bikos, 2000 used this model in a longitudinal study of race and sex differences as predictors of aspirations. This model is concerned with the effects of self-efficacy, beliefs, expected outcome and goal mechanisms, and the interplay of these variables with gender, contextual, experiential and learning factors. In our research, we will link the social cognition theory to the aspirations and choice components of the pipeline model. The Status Attainment Model In addition to social cognition theory, we will combine the status attainment model (Blau & Duncan, 1967; Sewell, Haller, & Portes, 1969), along with the pipeline model to fashion a theoretical framework to study why racial and ethnic minorities decide to opt in and out of the math/science pipeline. The status attainment model proposes that family and cognitive variables influence social psychological processes, which in turn affect educational and occupational attainment. Mau & Bikos (2000) used both of these models to choose four clusters of variables for their study: psychological variables, family variables, schools variables, sex, and race.
5
Previous researchers (Farrell & Pollard, 1987; Mau, 1995; Shepard, 1992; Wilson & Wilson, 1992) have used procedures that incorporate a variety of variables or clusters of variables in their statistical analysis. Psychological Variables Briefly, Mau & Bikos (2000) reported results of research on the relationship between psychological variables and high school students’ educational and vocational aspirations. Researchers found that, compared with students with lower aspirations, students with higher aspirations were more likely to have higher self-esteem (Lay & Wakstein, 1985; Mau, 1995), a higher self-concept (Carpenter & Western, 1982), and an internal locus of control (Mau, 1995). Further, these students were more likely to assign more importance on having a high status job and spent more time thinking about their post-high school plans (Shepard, 1992). Family Variables The literature on high school students’ educational and occupational aspirations indicates researchers have devoted much attention to parents’ expectations. While there is a consensus that perceived parental expectations are positively correlated with the students’ aspirations (Conklin & Dailey, 1981; Farrell & Pollard, 1987; Mau, 1995; Shepard, 1992; Smith, 1991; Wilson & Wilson, 1992), researchers have interpreted the results differently. For example, Wilson and Wilson reported that maternal aspirations for the high school student may have the greatest effect on the student’s aspirations. Other interpretations focused on student’s agreement with parental expectations, positive change in parental expectation from the 9th to 11th grades, and the positive influence by significant people other than parents. Other significant family variables in the literature are the educational and financial status of the parents. Findings have established a positive association between high school students’ aspirations and the family’s SES (Farrell & Pollard, 1987; Mau, 1995), the father’s occupational status (Smith, 1991), and the educational level of the parents (Wilson & Wilson, 1992). Finally, research suggests that students in families with fewer siblings (Mau, 1995) and students whose parents place importance on having a day job (Shepard, 1992) were more likely to have higher expectations.
School Variables
6
Variables linked to school performance (Farrell & Pollard, 1987; Harris, 1970) appear to substantially affect high school students’ educational and vocational aspirations. Measures of these variables have ranged from grade point average (Mau, 1995), change in GPA over 2 years (Shepard, 1992); reading, science, and mathematics proficiency (Mau, 1995); to eighth-grade mathematics ability (Braddock & Dawkins, 1993). Other research has indicated that components of the school environment (Marjoribanks, 1985; Wilson & Wilson, 1992) and the perceived teachers’ aspirations for adolescents (Wilson & Wilson, 1992) have also influenced students’ aspirations. Sex and Race Variables According to Mau & Bikos (2000), research findings on the effects of sex and race on high school students’ educational and vocational aspirations have been mixed (p. 187). In a 10-year study, Hauser and Anderson (1991) found no significant differences in the aggregate trends in the aspirations of Black and White seniors. On the other hand, Wilson and Wilson (1992) concluded that Black adolescents had higher aspirations than their White counterparts did and that male adolescents had significantly higher aspirations than female adolescents did. Mau (1995) reported significant main effects and interaction effects for sex and race. Specifically, he pointed out that Asian American students had significantly higher educational aspirations than any other race-based grouping of students and that Native American students had significantly lower aspirations. Black and White male students had significantly higher aspirations than did Hispanic and Native American male students. In both racial-ethnic groups, female students exhibited significantly higher educational aspirations than did their male counterparts. Mau & Bikos (2000) suggested that perhaps the “most compelling evidence” about the effect of sex or race on the educational and career aspirations of high school students has been its tendency to interact with other variables p. 187). With respect to psychological variables, Lay & Wakstein (1985) reported that self-esteem affected the aspirations of Black students more than actual achievement. Carpenter and Western (1982) indicated that for male high school students, self-concept was one of the better predictors of college aspirations. Other interactions with race or sex have been reported with family variables, and with year in school. With respect to the latter, Black students started in early adolescence to become less inclined to adopt parental educational goals (Smith, 1991). Research Plan
7
Quantitative approaches such as the studies reviewed previously have much merit in clarifying the relationship between variables, but the qualitative approaches of educational anthropology and sociology have much to offer as well, particularly with respect to the role of an outreach program such as Upward Bound. For example, Kaplan (1999) studied adolescents’ perceptions of certain aspects of their experiences by using in-depth interviews to determine their ideas, desires, and beliefs about their educational experiences (p. 181). She discussed the students’ appraisals of a tutorial program in which they participated, the Youth Educational Program. She identifies a number of variables that “demonstrate why we should listen to inner-city Black and Latino children if we want to understand how their academic and personal objectives can change with collaborative efforts, for instance with programs such as YEP” (p. 182). The proposed research will be based on surveys, interviews, and focus groups with college-bound African American and Latino/a students at 10 to 12 public high schools in Upward Bound Region IX Southern California (Los Angeles, Orange, and Ventura Counties) about their plans to major in computer science in college. Additionally, these interrogations and methods will be used with college students who are Upward Bound Program graduates. Finally, interviews will be conducted with a small number of college graduates who also went through Upward Bound. The research is cross-institutional and is based on interdisciplinary perspectives of the computer science pipeline. We will be working with cooperating Upward Bound programs, the high schools and colleges and universities with which Upward Bound is affiliated. The traditional Upward Bound participants and graduates will be compared to both non-Upward Bound students and to Upward Bound Math Science students. The Math Science Upward Bound is a much newer program with fewer graduates, but investigation of its participants and graduates may yield useful data given, that the program is designed to focus on the math and science courses that serve as gatekeepers to computer science majors in college. Initial focus groups will be conducted with five groups of 10 to 12 current Upward Bound students during their summer residence programs. Kaplan and (Lingua & Roosa, 1992; Morgan & Spanish, 1984), have observed that focus groups are “good settings in which to interview teenagers because they allow participants to interact with others in an informal setting and to use their own language and their understanding of the world to set the tone of the interview and discussion.” (1997, p. 185). According to Mates and Allison (1992), the focus group is useful for gathering data and as a tool to generate new ideas (Lingua & Roosa, 1992). Such new ideas will
8
be used to understand how students are experiencing the math/science pipeline and how they perceive themselves in the context of high school and college computer courses and the information technology workforce. The focus groups can be used appropriately with the 60-student California Lutheran University Upward Bound and 50-student Math/Science Upward Bound groups spend 5 to 6 weeks on the CLU campus taking classes, getting cultural enrichment, and visiting other colleges. There will be an opportunity to focus group interview students from at least one outside Upward Bound group visiting the CLU campus in the summer of 2001. We recognize that the funding cycle may not be completed by that time, but we are willing to invest our resources in taking the first step in this project. These interviews will be used to formulate survey questionnaires that will be administered to 500 current Upward Bound students in 10 high schools in Region IX starting in the spring semester of 2002. Similarly, questionnaires will be administered to 500 students in 10 high schools without Upward Bound. Finally, 60-120 Upward Bound graduates, who are currently in college as well as former Upward Bound students who have complete their degrees will receive interviews to assess their experiences with math, science, and computer courses in college. Upward Bound students will also be interviewed once a year as part of their counseling sessions to get a sense of their experiences with the math/science/computer pipeline, their coursetaking, their educational and vocational aspirations, and their sense of resilience, self-efficacy, and self-esteem. The counselors will be trained to administer the instruments. Some of the students’ high school teachers and counselors will be interviewed. Finally, some of the students’ parents will be interviewed as well to learn what their expectations are for their children. Student Survey Purpose and Structure We have chosen a survey to gain a broad perspective on student perceptions and attitudes. If our inquiries suggest it, separate English and Spanish language versions of a survey instrument will be developed. Adapted from the work of Moses et. al. (1999), the instrument will have approximately 90 fixed-response items divided into sections on demographic information; attitudes toward math, science, computers, and vocational education; and perceptions of fairness of the treatment of various groups. Each questionnaire will also have about five open-ended items. These items will ask students to name the job they would like to have, the job they think they probably will have, the groups that were most picked on in their school, the groups of students with they felt most comfortable, and the groups of students with which they felt lest comfortable. The survey will be structured to determine the predictive power of four clusters of variables (e.g., personal/psychological characteristics, family
9
variables, school variables, sex and race) on educational and vocational aspirations, particularly those having to do with computer science and information technology careers. The dependent variables will be college aspirations and occupational aspirations. Site Selection The research will take place in the three-county area of Los Angeles, Orange, and Ventura, which is Upward Bound Region IX. The home site of the researchers is California Lutheran University, which has hosted an Upward Bound program for 20 years and a Math/Science Upward Bound program for five years. Six public high schools in Ventura County are participating in the Upward Bound programs. Currently, 60 students are enrolled in the Upward Bound program and 50 students are enrolled in the Math/Science Upward Bound program. The students in the former program come from Ventura County and participate in after-school activities, tutoring, and counseling. The students in the latter program come from a larger region encompassing California and the Pacific Islands. Only those students from the local area participate in programs during the school year. They convene during the summer residential program. The probability sample for the survey will be drawn from the Upward Bound programs throughout Region IX. Non-Upward Bound students will be surveyed in non-participating high schools drawn in a similar fashion. Research Team The research will be conducted by a multi-disciplinary team consisting of faculty and administrators in computer science, education, communication, and economics. Relevant skills sets and experience include: inner city public school science teaching, comparative education, multicultural education, economic development, research methodology, and Upward Bound. First, the co-principal investigators: Dr. Ali Akbari is a professor of business with broad experience in research design and methodology. He did his dissertation in urban economics. Along with Dr. Jamshid Damooei, Dr. Akbari is co-founder and co-director of the Center for Economic Research at California Lutheran University. Dr. Myungsook Klassen is associate professor of Computer Science whose expertise. She has experience teaching computer science in the Far East and in the
10
United States. During her graduate studies, she taught limited English Korean immigrant students at a military community school in Ogden Kansas for one year as a volunteer. Dr. Russell Stockard is assistant professor of communication and business with an interest in new information and communication technology. He has taught high school dropouts in the inner city. He has directed market research in telecommunications and has specialized in qualitative research methods, particularly the design and moderation of focus group research. He is author of a handbook on African American Consumer Behavior.. The rest of the contributors to the project are as follows: Dr. Jamshid Damooei is a professor of economics with extensive experience in local and international economic development. He has broad experience in project management. He is experienced in working with international development agencies and rehabilitating war-torn regions. Dr. Hans Lingens is assistant professor of education is editor of a journal of comparative education, European Education and will be able to use that connection to help disseminate the results of the proposed research. He teaches a graduate course on cultural diversity and has directed a secondary school program on ethnic pluralism in the Los Angeles Unified School district.. He has taught science in inner-city high schools in Los Angeles. Ms. Diva Ward is co-director of the Upward Bound program at California Lutheran University. She also is a former computer science major and has worked in information systems. Data Dissemination The data will be disseminated through a number of channels. The conferences of the American Educational Research Association, the American Sociological Association, the American Psychological Association, the International Communication Association, and the National Communication Association are appropriate targets for conference papers. Educational Policy, Urban Education, and the Journal of Counseling and Development are some of the journals where we could submit papers from the research project. In addition, Dr. Lingens is editor of a journal of comparative education where we could submit papers.
11
Summary The proposed research stresses the role of underrepresented student decisions in entering and persisting in the math/science/computer pipeline. Upward Bound counseling strives to provide information that students may lack in their regular high school classes. Upward Bound may also support the raising of student educational and career aspirations through enhanced self-esteem. Given the climate undermining equal opportunity mechanisms the burden of choice is increasingly on high school students. They may not have all the information they need to properly decide without programs like Upward Bound and Math/Science Upward Bound. Furthermore, these programs are likely to provide the resilience to persist through the challenges of the math/science/computer pipeline. Upward Bound itself has been the subject of evaluations (Mathematica Policy Research, 1997) that have found that it strengthens student skills and persistence. Findings that the program can improve the aspirations and course-taking of underrepresented students in computer science could support it in future authorization. Findings from this research program could also help high school counselors, teachers, and parents in motivating students to consider more difficult courses of study. Computer courses and curricula could be redesigned to make them more inviting to women and students of color and of modest means at both the high school and college levels. Finally, the advantages of using Upward Bound to move more underrepresented students into the pipeline may also help in the development of alternative programs and collaborative college-high school efforts that extend the reach of Upward Bound.
12