Education and Social Opportunity Major Grants


View by grant year

2008 | 2009 | 2010 | 2011 | 2012

2008 Grant Summaries

Paul Attewell

Other 75%: Policies & Practices Affecting Academic Progress and Opportunity Among Non-Traditional College Students
The Graduate Center, City University of New York

Attewell’s study addresses the crucial issue of how “non-traditional” students progress through college. Non-traditional students—i.e., students who are not full-time residential students financially dependent on their parents—account for 75% of entering two- and four-year college students in the U.S. While previous research on these students has examined the influence of individual risk factors on college retention and academic success, Attewell hypothesizes that these outcomes depend on the cumulative and interactive effects of individual attributes and institutional policies and practices.

There are two parts to this study. A first part uses data Attewell gathered through interviews with 100 non-traditional college students—half from a two-year community college, half from a four-year BA-granting institution—on a wide range of topics, including finances and financial aid, experiences in both paid work and in college, educational and career aspirations, family obligations, perceived challenges to completing their education, and perceived supports. Based on analysis of these data, he will produce student-centered portraits that attend to the multiple and interrelated student-level and institutional factors that shape outcomes. A second part of the study uses nationally representative data from the Beginning Post-Secondary Student Longitudinal Study (BPS) to test a series of hypotheses. The latest round of the BPS followed over 23,000 students attending 1,670 institutions from 2003/2004 through 2008. Analyzing these data will uncover the interrelationships among risks factors and non-traditional students’ educational outcomes and will offer a more nuanced understanding of the impact of institutional policies, rules, and regulations on their educational experiences. Overall, the results of this research can provide a clearer and more complete picture of the struggles nontraditional college students face and support reforms that counter the cumulative disadvantages these students often experience.

Heather J. Bachman

Identifying Child Care, Family, and Child Strengths that Promote Low-Income Children's School Readiness Skills
University of Pittsburgh

This study seeks a better understanding of the kinds of childcare low-income children receive prior to kindergarten and its impact on their development of academic and social skills. Bachman’s research in community-based childcare centers brings a rich variety of data and methods to bear on this under-examined issue, focusing on detailed aspects of teacher practice—skills teachers are targeting, amount of time devoted to instruction in different skills, instructional methods employed—and skill-specific outcomes of these practices.

Bachman and her team will study approximately 35 classrooms at 30 child care centers in the metropolitan Pittsburgh area, anticipating a total sample size of approximately 200 children and parents and 70 lead teachers over the course of two years. The study employs multi-method assessments of child, home, and childcare characteristics to relate early academic and social skills simultaneously to teacher practices and to child risks. Bachman and her team use a wide range of methodologies and instruments, including assessments of specific student skills and abilities; questionnaires for teachers, center directors, and parents on the students’ demographics, activities and behaviors; and extensive classroom observation with specific protocols for documenting behavior and activity, which builds on earlier work by Bachman and by other researchers.

Kathryn Borman

Dual Enrollment in Florida: Who Is Served and Who Benefits?
University of South Florida

Borman’s study focuses on high school students taking college courses for both high school and college credit, also known as “dual enrollment.” Although schools’ use of dual enrollment is widespread and growing, the small number of previous studies on the topic leaves many important questions unanswered. Borman aims to answer important questions about the causes and consequences of dual enrollment. With respect to causes, she asks, What kinds of schools are most likely to offer dual enrollment programs? What kinds of students typically take these classes? Do particular kinds of students tend to take different kinds of courses (for example, academic versus vocational)? Related to these questions are questions about the kinds of policies, programs, and practices used by high schools with high levels of participation and success for historically underrepresented students. To address the consequences of dual enrollment, Borman compares post-secondary experiences and achievements of dual enrollment students to other students with similar social and academic backgrounds, focusing in particular on historically underrepresented students, and to students taking other advanced coursework (e.g., Advanced Placement or International Baccalaureate).

To answer these questions, Borman uses data primarily from Florida, both statewide, student-level data—including social background information, course enrollments and grades (for both high school courses and postsecondary coursework), test scores, and postsecondary enrollments—and qualitative data gathered from case studies of six to nine high schools representing a range of outcomes for underrepresented students. These case studies are based on interviews with school staff and focus groups with dual enrollment and non–dual enrollment students and provide additional context for understanding patterns discovered through quantitative analyses and should help other schools improve their practices.

Parfait M. Eloundou-Enyegue
Peter Glick

Schooling and Labor Market Transitions in a Context of Fertility Decline: A Sub-Saharan Case Study
Cornell University

Eloundou-Enyegue and Glick’s study is a follow-up on a 1998 Spencer-funded study addressing determinants and trends in education in Cameroon, which received broad exposure in national and international forums, including the United Nations. The current study follows up ten years later with households from the earlier study, allowing a longitudinal approach rarely found in research on education and labor markets in Africa. The time period covered is especially significant, as it is characterized by declining fertility and rising unemployment, trends seen in much of sub-Saharan Africa.

The study seeks to understand the effects of these trends on educational investment and childbearing decisions in households, while taking into account the local influence of family planning services, educational options, and school quality. A variety of research questions are can be considered here: Given the high rate of unemployment, do parents focus on educating sons rather than daughters? Does high unemployment influence parents’ childbearing choices (leading to smaller family sizes)? How does education influence men and women’s employment and earnings trajectories? Has access to education changed as a result of fertility declines or other factors operating at the same time? Have the labor market returns to education changed?

The earlier grant supported interviews with members of 3,369 nationally representative households in Cameroon, gathering data on schooling and labor market histories of parents and children. The current study updates those histories and provides more data on hours worked and earnings, school quality, and family investments in education. Data will be collected for all available respondents from the prior study, with particular attention to individuals who had just entered, or were preparing to enter, the labor market in the earlier study, as these individuals will have experienced labor market shifts early in their careers, and will likely be in a life stage where they will be making critical decisions about family size and investments in their children’s educations. Combining these new data with earlier data promises to provide valuable insights into the relationship of recent demographic and labor market trends in Africa to educational and family decisions.


Christopher Jepsen, University of Kentucky
Peter Mueser, University of Missouri, Columbia
Kenneth Troske, University of Kentucky

Women's Economic Returns from Obtaining a GED

Most previous research on the effects of earning a GED has focused on men. This study examines examine how earning a GED influences economic and educational outcomes for women, as well as the costs and benefits to both the individual and the government of women earning the GED. There are reasons to believe that the related processes and outcomes differ for women and men. For example, due to early pregnancy and parenting, girls may end up taking the GED later than boys, and age may affect the labor market value of the GED. Employers may also make different assumptions about personal qualities of females who take the GED versus males, and, since females may be entering different employment sectors (sales, for example), the GED may confer different advantages to them.

This study combines unique administrative data from Missouri on all women who took the GED in that state between 1997 and 2002 (more than 30,000 women) with a dataset that includes all high school dropouts between 1998 and 2001 (more than 13,000 women). In addition to information about GED test scores and high school enrollment, these datasets provide demographics information, employment histories, information on welfare receipt, and postsecondary enrollment data. Based on these data, regression discontinuity analyses (which focus on outcome differences for individuals closely above or below an arbitrary cutpoint) will be used to determine the specific effect of earning a GED on economic and post-secondary educational outcomes and how these are related to the experiences and circumstances of women to take the GED.

Gary Orfield

Suburban Change and the Schools: The Effect on the Educational Opportunities of Poor and Minority Students
University of California, Los Angeles

This study concerns recent demographic changes in suburbs, particularly increases in the proportion of poor and minority students, and how these changes influence the educational opportunities of young people. In recent decades, the education of poor and minority students has become the responsibility of a broader range of locations, as suburbs have incorporated more students who in the past were likely to be concentrated in urban areas. To understand this phenomenon and its educational implications, Orfield asks three questions: (1) how have student demographic patterns changed in the twenty-five largest metropolitan areas over the past fifty years? (2) How do educational institutions and their communities define and think about issues of diversity? (3) What strategies, both internal to educational organizations (for example, programs aimed specifically at raising the test scores of minority students) and external (such as housing policies) make it easier or more difficult for low-income and minority students to be accommodated by suburban school districts? He takes a broad approach to these questions, examining how suburban communities, including political institutions, local business and other organizations, school boards and administrators, parent groups, and other interested parties, identify and define diversity, and how they interact with each other to respond to diversity.

The study will begin with analyses aimed at describing demographic changes in the twenty-five largest metropolitan areas over the past fifty years using U.S. Census data. To look more specifically at student enrollment trends, Orfield will use the 1990 through 2006 NCES Common Core of Data. These data allow Orfield to compute three indexes of segregation, which he will do at the metropolitan, district, and school level. The second part of the study will use a multiple case study approach to probe questions about how different groups and individuals within suburbs conceptualize diversity, their proposed responses to increased diversity, and how efforts to adopt or change policies or practices are pursued or opposed. Four to six districts, selected to represent different patterns of residential change, will be selected. Site visits and interviews with district and school-level personnel, parents, students, policymakers, and other stakeholders will be conducted, focusing on policies, practices, attitudes, and experiences in the school context.

Sarah E. Turner, University of Virginia
Christopher N. Avery, Harvard University

Aid and Application Awareness: Reaching High Achieving Students

Turner and Avery’s study addresses a set of questions about how high-achieving, low-income students make college-application and attendance decisions. This population tends to apply to and attend less selective institutions that have much lower rates of completion than the more selective institutions they are qualified to attend. Turner and Avery focus on the contribution that information gaps may make to this pattern, including information about how to select schools that match student qualifications, the availability of financial aid, and how to apply for both admission and financial aid. Focusing on public high school students in Virginia, the study examines both family background and high school factors that might relate to information differences and subsequent decisions to apply to and matriculate in different institutions.

There are three phases to the project. A first phase uses “Search Files” from the College Board to identify low-income students who were qualified to be in the applicant pool for Harvard University or the University of Virginia, but who did not apply. These data will be used to identify high schools where students are likely to benefit from interventions aimed at increasing high-achieving, low-income students’ decisions to apply to and enroll in selective institutions. The second phase of the research tracks 400 high-achieving high school students in low-income high schools over their senior year to follow their college-application process. Surveys will be administered to these students and their parents, with the goal of understanding how much information they have about the college application and financial aid process, the likelihood of being admitted to specific institutions, and the likelihood of receiving financial aid. The final phase of the study involves a low-cost intervention to which a subset of students from the second phase of the study will be randomly assigned. In the intervention, students will be encouraged to apply to institutions matched to their qualifications and will be provided information about the various steps included in the college application and financial aid process. Overall, results of the study should improve understanding of the impediments to students being successfully matched to the institutions that might best serve them and of how these impediments may be overcome.


2009 Grant Summaries

C. Cybele Raver

Early Investments in Non-cognitive Skills: Testing the Impact of CSRP on Low-income Children’s Educational Outcomes in 3rd Grade
New York University

This study examines the long-term academic and behavioral effects of the Chicago School Readiness Project (CSRP), a pre-school intervention focusing on two non-cognitive skills: self-regulation and social relationships with teachers. Through random assignment, children received the intervention in Head Start preschool classrooms in 2004-2005 or 2005-2006. Prior research found that intervention classrooms differed from control classrooms in expected ways: they had more positive classroom climate, teachers displayed greater sensitivity to children’s needs, classroom rules and expectations were clear, and children were redirected towards more positive behaviors when they did misbehave. Furthermore, Raver and others found large effects on children’s vocabulary, letter-naming, and math skills at the end of the year of the intervention. The present study follows up with these students in the 3rd grade to determine whether these effects last beyond the year of the intervention and whether being exposed to additional treatments sustains or increases the effects. Results of the study will provide information about the role of non-cognitive factors in children’s school achievement and behavioral outcomes and will also provide guidance about program practices that most support children’s development.

Beginning in Fall 2004, two cohorts of Head Start children (N=602) participated in the study, through random assignment to a treatment or a control teacher’s classroom. The treatment included teacher training in supports believed to be important for promoting self-regulation and positive teacher-student relationships, as well as on-site social workers who provided capacity-building consultations for teachers and gave mental health support to students. The follow-up study will follow children who participated in either the control or the treatment into 3rd grade. Data to be collected from each child’s family include income, family structure, life events since the last data collection, and parents’ assessment of children’s adjustment, behavior, and academic achievement. Teachers will be asked to provide complete progress reports about each child, information about their own demographic characteristics and the characteristics of their classrooms, their use of techniques targeting social-emotional learning, and their rating of school quality. Children’s school records will be collected from 2nd and 3rd grades.

These data will be used to address the following three research questions, specifically concerning outcomes in the 3rd grade: (1) Does participation in programs targeting social-emotional learning affect on children’s academic and behavioral outcomes? (2) Do classroom and school quality moderate the influence of the early intervention on outcomes? (3) Do additional social-emotional services provided during early elementary school (i.e., additional programs found in some participating schools) “boost” the effects of the intervention?

James E. Rosenbaum

High School Procedures for Creating College-for-All
Northwestern University

This three year-study in Chicago’s public high schools assesses the role of guidance practices in facilitating the difficult transition from high school to college. Previous research, including well-recognized studies produced by Rosenbaum, has traced the difficulty students experience in making the transition to college to student background and behavior and high school curricula. In this study, Rosenbaum shifts attention to guidance counselor practices and organizational structures that are designed to help students in that transition.

The study will draw from and contribute to a rich body of data generated by the Consortium on Chicago School Research (CCSR). Existing data that will be used include data from teacher surveys, including responses about student advising and data on students, including transcript data, ACT/PSAE data, and CCSR student survey data, from 6th, 8th, 10th, and 12th grades, with responses to questions about topics such as motivation, expectations, and behaviors oriented towards future goals. To find out what guidance practices are used in each school, Rosenbaum will survey all guidance counselors in the 82 high schools in the Chicago Public Schools and will supplement an existing CCSR principal survey with new questions about guidance counseling. The new data will be merged with the CCSR student dataset for use by Rosenbaum, and eventually, by other researchers. Since CCSR data are longitudinal, trajectories can be modeled from 6th grade through the year following high school (using post-secondary enrollment information from the National Student Clearinghouse). Using these data, Rosenbaum be able 1) to describe the organizational structures in place and processes involved in college counseling; 2) to describe changes in guidance policies and practices over time; 3) to determine the relationship of these policies and practices to changes in students’ plans, motivations, achievement, and college enrollment; and 4) to assess the specific impacts of policies and practices for particular subgroups of students. Knowing which programs work best will provide school administrators and guidance counselors with highly relevant information about how to help their students transition to college.

Petra E. Todd
Jere Behrman
Kenneth Wolpin

Effects of Schooling Reforms on Education and Labor Markets: Lessons from Chile
University of Pennsylvania

In recent decades, Chile has embraced a series of market-oriented and related educational reforms, including student vouchers, rules allowing public and private schools to collect tuition on top of vouchers, large increases in the minimum wage for teachers, increases in school resources, performance-based school funding awards, and new teacher certification and evaluation programs. This project is an ambitious effort to learn from Chile’s unusual long-term experience with such reforms by assessing their impact on school attainment, labor market outcomes, teacher compensation and qualifications, and school quality. Economists Todd, Behrman, and Wolpin will develop dynamic structural models to investigate the effects of policies related to educational demand—that is, policies that work through students/households, such as vouchers—and those that directly affect teachers and schools to shape educational supply.

A linked dataset will be created from existing national datasets as well as data collected by the researchers relevant to the evolution of schooling in Chile, including information on voucher payments, school types by region, tuition charges, school funding received on top of vouchers, and teacher union operations and wage contracts. In addition to contributing to knowledge of the impact of market-oriented educational reforms through its findings, this project will contribute to future research by making this unusual dataset publicly available.

2010 Grant Summaries

Stephen L. DesJardins
Brian P. McCall

The Effect of High School Curriculum on College Completion
University of Michigan, Ann Arbor

DesJardins and McCall’s study explores the effects of high school curriculum on college completion, specifically by considering the effects of enrollment in college-preparatory courses in high school. Based on the conventional wisdom that taking college-prep courses increases the likelihood of college enrollment and completion, schools increasingly encourage or require students to take these courses. However, based on currently available evidence, it is difficult to say what the true effects of promoting college prep courses will be. On the one hand, the positive correlation between college enrollment and taking college prep courses may be due not to a causal relationship, but, instead, to self-selection (i.e., because students likely to enroll in college prep courses are already more likely to attend college). On the other hand, requiring enrollment in college prep courses could have the unintended consequence of encouraging some students to leave high school altogether.

This study will test these possibilities using longitudinal data on students’ high school programs and their college enrollment and completion from two nationally representative longitudinal datasets—the National Education Longitudinal Study 1988 (NELS88) and the National Longitudinal Survey of Youth 1997 (NLSY97)—and data about labor markets, census tract demographic information, and post-secondary institution data from the U.S. Bureau of the Census, the Current Population Survey, and the Integrated Postsecondary Education Data System. In order to estimate the effects of students’ curriculum on high school completion, college enrollment, and college completion while controlling for selection bias, DesJardins and McCall will use models that incorporate instrumental variables—variables that correlate with independent variables, but not with the outcome of interest—a method often employed by economists. In this case, the instrumental variables will be local labor market conditions and high school course-taking requirements will be used as instruments. Additionally, the study will use event history analysis; joint modeling of the three outcomes of high school graduation, college enrollment, and college completion; and sub-group analyses to address potential differential effects for different populations of students.

Results of this study should be a valuable aid as schools try to determine the most effective ways to increase high school graduation and college enrollment and completion rates. Although this study will not evaluate alternatives to promoting a “college prep for all” strategy, it might provide evidence on which to assess the costs and benefits of promoting that strategy across the high school population. To that end, an analytic tool will also be developed that will allow policymakers to simulate the effect of high school course-taking on the outcomes of interest.

Bruce C. Fuller
Xiaoxia Newton

Decentralizing School Governance in Los Angeles: Stronger Social Ties among Principals, Teachers, and Parents?
University of California, Berkeley

Strong professional communities, displaying rich collaboration among teachers, are consistently associated with robust student achievement. Many advocates of decentralized forms of schooling -- charters, magnet, and semi-autonomous pilot schools -- claim that their human-scale institutions will nurture tighter, more motivating forms of collaboration. Fuller and his colleagues are testing this claim across diverse school forms that are spreading rapidly in Los Angeles.

The idea that decentralized school governance might be a school improvement strategy relies in part on a set of assumptions about the structure and characteristics of relationships among adults in such schools, and the social norms that surround them. According to this logic, by granting principals more authority over funding and staffing decisions, decentralization allows local leaders to create schools staffed by teachers who share a common vision, attended by students whose parents affirm and support that vision. Theoretically, these circumstances result in the formation of strong horizontal social ties, high levels of relational trust, and, as a result, a more efficient and rich distribution of what Fuller et al. call "expressive and instrumental" resources. In these schools, teachers share responsibility more broadly with the principal and each other, and work together to provide the kinds of educational and social experiences that advance student learning.

The research team identifies another unanswered question looming in prior research: How much does student and teacher selection factor into any observed differences between decentralized and traditional public schools? Other research in Los Angeles has shown, for example, that charter schools enroll fewer special-needs children and children from Spanish-speaking homes than do other public schools. And teachers selecting decentralized schools might differ from those choosing regular public schools in ways relevant to the formation of social ties inside schools.

Melissa Roderick

Does College Match Matter? The Study of the Effects of College Match for Graduates of the Chicago Public Schools
University of Chicago

Recently, work on the transition to higher education has shifted from questions about postsecondary access to a concern with college match, enrolling in colleges with selectivity levels that best match the kinds of colleges students are qualified to attend. Earlier work, some of it conducted by Roderick with prior Spencer funding, suggests that students are more likely to persist and graduate if they attend a college that matches their qualifications than if they attend an institution that is less selective than that which they qualify for. Roderick and her colleagues document, moreover, that mismatch is the rule. The majority of high-achieving graduates in Chicago, those with access to selective and very selective colleges, enroll in colleges that are less selective.

For students who are qualified to attend a more selective college there clearly is an association between the selectivity of the college they actually attend and educational outcomes. But what is driving these associations? Is it the match itself? Or are selective colleges, on average, better at graduating students, whatever the match? As a practical matter, if high schools are trying to maximize their graduates' outcomes, should they guide students to match schools, to more selective schools, or to schools with higher graduation rates overall? Important as it is to know the answer to these questions, even more important may be questions about what it is about the colleges themselves that leads to better outcomes. What do colleges actually do to increase the likelihood that students will succeed and graduate? Roderick's mixed-method study probes these issues.

This study represents a core component of the continuing work of the Consortium on Chicago School Research's (CCSR) Postsecondary Transition Project which is housed at the School of Social Service Administration under the direction of Principal Investigator and CCSR faculty director Melissa Roderick. The dataset includes over 36,000 students who graduated from Chicago Public Schools (CPS) between 2002 and 2009. For each student, Roderick has calculated whether their high school record qualifies them to attend very selective colleges, selective colleges, or less selective colleges. She uses the term "access" to identify the highest level of college that students qualify for. Thus, the quantitative analyses will focus on how match influences college outcomes, for example, persistence or graduation, by level of college access, and how these outcomes differ by college characteristics, including selectivity. Importantly, analyses linking college characteristics to student outcomes will be conducted for the entire cohort, as well as separately for students by access category. As a result, we will learn about what characteristics of colleges matter for CPS students who are qualified to go to very selective, selective, and somewhat selective colleges. These analyses acknowledge that what matters for student success may differ for students with different qualifications.

The second set of activities in Roderick's study aims to better understand how highly qualified students choose a college at a particular level of selectivity, as well as how they experience college once they arrive. Students in this part of the study will, by virtue of their high school academic record, have access to selective and very selective colleges and will have already shown an interest in schools that meet those selectivity thresholds. The students will be interviewed once during their senior year in high school after they have chosen a college, three times during their freshman year in college, and once during their sophomore year in college. In this qualitative study, the 50 students being interviewed attend a range of types of colleges, including elite private schools, selective flagship universities, small liberal arts colleges, and less selective public schools, thus providing an opportunity to compare similarly well-qualified students' experiences at a range of colleges with different selectivity levels.

Amy E. Schwartz
Leanna Stiefel

Migration, Immigration and Public Schools: The Consequences of Mobility for Students and their Classmates
New York University

Schwartz and Stiefel have identified an unusual opportunity to improve understanding of the complex relationship between education and student mobility. In large urban centers, students frequently change schools, sometimes in conjunction with a residential move, sometimes not. Other students, most notably immigrants, move into the school system from outside schools. Still other school changes are built into the organizational framework of the school system—for example, changing schools between elementary and middle school. Based on data from New York City (NYC) public schools from 1995-96 to 2007-08, this comprehensive study considers the impact of different types of student mobility, both for students who move between districts and schools and for the schools in which those students are enrolled. In the former case, Schwartz and Stiefel will ask how moves affect the academic achievement of native-born students, immigrant students from different countries or regions, and students from different social class backgrounds. In the latter case, they will consider how having large numbers of mobile students may affect schools’ resources and the performance of other students in the school.

The study uses data available from New York City public schools and the New York State Education Department on a wide variety of student-level and school-level characteristics from the 1995-96 school year to 2007-08. Analyses will be based on a nuanced understanding of mobility as it varies by timing and causes. Types of moves to be considered include between-year moves within and into the district, mid-year moves within and from outside the district, mandatory moves (required by grade-span scheduling), strategic moves (from lower- to higher-performing schools), reactive moves (from higher- to lower-performing schools), and cumulative between- and within-year moves. Results of the study can therefore provide a better understanding of how different types of students experience mobility and the different ways in which mobility affects these students and the schools they attend. In addition, student-level administrative data for 2002-2008 have been matched to another database to identify which students live in housing units that underwent foreclosure, allowing analysis of specific impacts of foreclosure on mobility and educational outcomes.

Joseph J. Tobin, Arizona State University
Joseph M. Valente, Florida State University
Thomas P. Horejes, Gallaudet University

Kindergartens for the Deaf in Three Countries: US, France, and Japan

Tobin, Valente, and Horejes will undertake a cross-national comparative study analyzing the acculturation of deaf children to national and Deaf cultures in early education. With distinctive national cultures, Deaf cultures, educational systems, and disability laws, the United States, France, and Japan provide an ideal basis for considering how various political and cultural pressures impact contemporary D/deaf early childhood education. Results of this study are of particular interest due to the persistent educational difficulties of deaf children, especially in the area of literacy, and emerging controversies surrounding their identity and education.

Using a video-as-cue technology employed with great success in two prior Spencer-funded projects, Tobin, Valente, and Horejes will produce a twenty-minute videotape of a kindergarten class for the Deaf in each of the three countries. Rather than using the video as primary data, they will use videos to elicit reactions from various stakeholders, and these reactions will provide the data. Videotapes from each country will be shown to the teachers, parents, and students in the videotaped school; to teachers and parents at schools elsewhere in the source country; to teachers and parents in schools in the other two countries; and to deaf education specialists and other relevant stakeholders in each country. In each case, focus groups of respondents will aim to reveal pedagogical beliefs, ideologies and political stances towards Deaf culture and education and to understand the role of the school and classroom in promoting particular cultural values about both Deaf culture and national culture.

2011 Grant Summaries

Christopher N. Avery

Randomized Controlled Evaluation of the Admission Possible Program
Harvard University

The growing emphasis on college for all has driven dramatic growth in the start-up philanthropic efforts to promote college entry among low-income students. For the most part, these organizations are happy to claim success merely on the basis that the students who participate are more likely to attend college than those who don't. This of course ignores the obvious fact that students don't sign up for these programs unless they (or their parents) are strongly pre-committed to attending college, a factor that obviously risks biasing conclusions. While non-profit projects to promote college enrollment are increasingly common, they have rarely been subject to high quality evaluation, and this study will help show the feasibility of such work, while providing insights into how programs can influence college admissions and enrollment.

This project focuses on the ability of an intensive college admissions counseling program to increase enrollment of academically successful students of lower socio-economic status (SES). College Possible (formerly known as Admission Possible) is a two-year (320 hour) after-school program that includes SAT and ACT preparation services, college application support, financial aid counseling, and other college transition support activities for high school juniors and seniors. Based on data collected by College Possible about college enrollment of students who applied to join the program in past years, past participants are almost twice as likely to enroll at a four-year college as similar students who do not participate, and more than 1.5 times more likely to enroll at a two-year college than those that do not. The proposed Randomized-Controlled Trial (RCT) seeks to confirm these results with a more powerful methodology than analysis of the programs historical records. The investigators focus on the ability of College Possible to affect the quality of the students' common application, GPA, ACT scores, and college enrollment using data provided by the National Student Clearing House.

Robert Bozick
Darwin Miller

Undocumented Immigrants in Higher Education: Assessing the Relationship between State Tuition Policies and College Enrollment
RAND Corporation

Undocumented students’ eligibility to enroll in state institutions of higher education and to receive in-state tuition is an ongoing political issue at the federal and state levels. In recent years, a number of states have passed laws that ensure undocumented students’ access to these benefits, while others have passed laws to limit such access. Bozick and Miller’s study attempts to determine the influence of these policies on undocumented immigrants, documented immigrants, and native-born students. While previous research has been conducted in this area, Bozick and Miller make original contributions to the field by (1) examining simultaneously the relationship of “accommodating” and “stringent” policies to students’ college enrollment patterns; and (2) considering the relationship of accommodating policies to students’ behaviors leading to college—specifically, taking admissions tests, applying to schools, and graduating from high school.

Their work draws on secondary data on individuals from the ages 18-24 from the Current Population Survey (CPS) (1997-2009) and a cohort from the Education Longitudinal Study (ELS) who were high-school sophomores in 2002 and were tracked through 2006. The key outcomes they are analyzing are college enrollment, taking admission tests, completing college applications, and graduating from high school. Bozick and Miller’s plans for analysis take account of the potentially complex nature of relationships between students’ behavior and state policies in a variety of ways. For instance, they will examine student behavior across a series of timepoints—the passage of a law, its implementation, and the peak in journalistic coverage of the law—in order to take account of the different ways in which students might become aware of and respond to a given policy. In order to control for patterns of behavior that may confound the effects of educational policies with being in a particular time or state, their models will also include state and year fixed effects.

Tarika Kapur
Reagan Daly

The Influence of School Climate and Policy on Delinquency and Justice System Careers
Vera Institute of Justice

Kapur and Daly are exploring the relationship of two school-level factors, school climate and so-called "zero-tolerance" school disciplinary policies, on youths' short-term delinquency and long-term criminal behavior and experiences with the criminal justice system. According to critics, punitive school environments lead students to detach from school and to develop delinquent careers, creating a "school-to-prison pipeline." Kapur and Daly are testing this argument, but they are also interested in the primary effects of school climate on delinquency outcomes and in how effects of school disciplinary policy, school climate, students' own characteristics, and their other contexts (i.e., families, neighborhoods, and peer groups) may interact to produce different outcomes for different students in different settings. A combination of multilevel modeling of short-term and long-term effects of school conditions, with an interest in the interactions among the multiple contexts within which students are situated, makes this study a unique and sophisticated approach to important questions.

The study uses data from four waves of data spanning 13 years (1994-2007), from the National Longitudinal Study of Adolescent Health (Add Health), with a core group of about 12,000 youth in a stratified random sample of 132 US middle schools and high schools. These data include short-term measures of in-school delinquency; long-term criminal behavior and justice system involvement; school climate (based on student survey responses and aggregated to the school level); school disciplinary policy (based on administrators' survey responses about school disciplinary practices and surveillance and control measures); school-level variables, such as school size and urbanicity; and students' family, peer, and neighborhood contexts. These data will be analyzed using 2-level hierarchical linear modeling analyses, allowing the researchers to model school-level and student-level effects. Propensity scores will also be used to control for the possibility that particular types of students select into schools with certain types of school climate and policy.

Alejandro Portes

The Second Generation in Spain: A Longitudinal Study
Princeton University

Portes, along with his Spanish collaborator Rosa Aparicio, will collect a second wave of data from the students in their longitudinal study of the adaptation processes of children of immigrants in Spain. In the first round of data collection approximately 7,000 second-generation students in metropolitan Barcelona and Madrid completed surveys about their home and school experiences, attitudes, educational aspirations and expectations, etc. Parents of about one-third of the students also were interviewed. This project will resurvey the students three to four years later, when they are seventeen or eighteen. The survey will capture their experiences at the time when they will be making choices about further education or labor market entry.

The Spanish study replicates the influential Children of Immigrants Longitudinal Study (CILS), which Portes conducted in the United States with Spencer funding from 1992 through 2003. An important and enduring theoretical contribution from that work is the segmented assimilation model of second generation adaptation, which remains a dominant theory explaining the differential outcomes of various immigrant groups in the United States. The practical significance of the theory and the research that supported its development is that some immigrant populations, by virtue of certain societal barriers (racism and discrimination, a bifurcated labor market, and inadequate schools and other social problems in areas where poor immigrant families live) and personal or family qualities (parents' human capital, immigrant family structure, and the social context that they become a member of) are at great risk of downward mobility and economic stagnation. Importantly, although the theory is being increasingly invoked in Europe, prior to the Portes and Aparicio study, no research had been conducted outside of the United States testing the theory's applicability in other countries. Given the current shifts in population through much of Europe, understanding the experiences of immigrants and the children of immigrants is crucial, particularly if the goal is to produce more effective policies for helping immigrants integrate into their new country.

M. Kathleen Thomas

High School Arts Education and Academic Success: The Access of At-Risk Students and the Impact of Participation
Mississippi State University

Thomas plans to study the role of arts education in high school, particularly for those young people who are at risk of dropping out of school. Using administrative individual-level data available for over four million Texas high school students, Thomas will investigate questions about the distribution of arts classes across schools attended by different populations of students, and the causal role taking arts classes plays for academic outcomes like persistence to the 10th grade, high school graduation, test scores, disciplinary action, and college attendance. No previous study of this topic has employed the econometric techniques used here, which allow the investigator to control for student and school characteristics to more effectively isolate the influence of involvement in the arts.

Although No Child Left Behind (NCLB) identifies the arts as a core academic subject, it is not one of the subjects included in mandated NCLB testing. Thus, despite the general endorsement, states report diminishing opportunities for students to participate in arts education as more time is devoted to subjects that are part of NCLB testing. This trend may be exacerbated in schools in which students are already failing to meet the adequate yearly progress requirements of NCLB. Thomas’s questions seem particularly important in this policy context. To the extent that arts education may improve the chances for struggling students to remain in school and be successful, the move away from offering these opportunities may have the unintended consequence of further diminishing students’ academic outcomes.

To study these issues, Thomas will use data from The University of Texas at Dallas Education Research Center. These data span grades pre-K through 16 for all students attending public schools in Texas, beginning in the early 1990s. Data include student demographic information, standardized test scores, disciplinary action, and course completion, as well as school characteristics like finance data, teacher quality measures, and the number and type of courses offered. More than 75 different arts courses are identified in these data, including those in visual arts, music, theater, and dance. This level of identification will allow Thomas to conduct analyses of different types of participation separately, as well as investigate whether taking arts courses early in high school has a more positive impact than later arts participation.

2012 Grant Summaries

Dr. Josipa Roksa, University of Virginia
Dr. Charles Blaich, Wabash College
Dr. Ernest Pascarella, University of Iowa

Race, Class, and Academic Outcomes: How Student Experiences and Institutional Contexts Shape Inequality on College Campuses

Inequality is a pervasive feature of higher education. Much research has investigated how students from less advantaged family backgrounds and racial/ethnic minority groups are less likely to enter higher education-particularly highly selective institutions-and less likely to persist and complete their degrees. However, far less attention has been paid to possible gaps in what different groups of students learn in college. This project asks how gaps in academic skills and attitudes between students from different backgrounds might be increased or diminished by their undergraduate experiences.

PIs Roksa, Blaich, and Pascarella hypothesize that different groups of students are exposed to different educational practices, even within institutions. These differences, as well as institution-level factors affecting instructional practices, student behavior, and other learning opportunities, significantly contribute to unequal academic outcomes. The goal of the project, then, is to quantify and track student exposure to different educational practices over time to see whether this contributes to gaps in academic skills and attitudes between different groups of students.

Utilizing data from three cohorts of students in the Wabash National Study of Liberal Arts Education (WNS), the PIs will identify factors that explain gaps in academic skills and attitudes between students from different racial/ethnic and socioeconomic groups. In particular, they will focus on variation in students' educational experiences -specifically, good teaching and high quality interaction with faculty, academic challenge and high expectations, interactional diversity, deep learning, and undergraduate research experiences. In so doing, the PIs will consider how exposure to and benefits of good educational practices vary across and within colleges and universities in the study.