University of Illinois Urbana-Champaign

Who chooses? A sociological portrait of families active in school choice in urban areas in the U.S.

Weitzel, Peter Carl

Content Files
WEITZEL-DISSERTATION-2016.pdf

Permalink

https://hdl.handle.net/2142/90523

Description

Title
Who chooses? A sociological portrait of families active in school choice in urban areas in the U.S.
Author(s)
Weitzel, Peter Carl
Issue Date
2016-04-12
Director of Research (if dissertation) or Advisor (if thesis)
Lubienski, Christopher
Doctoral Committee Chair(s)
Lubienski, Christopher
Committee Member(s)
  • Lleras, Christy
  • Lubienski, Sarah
  • Welton, Anjalé
Department of Study
Educ Policy, Orgzn & Leadrshp
Discipline
Ed Organization and Leadership
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Date of Ingest
2016-07-07T19:53:38Z
Keyword(s)
  • School choice
  • Sociology of Education
Abstract
School choice and charter school research acknowledges that choosers, families that are active in school choice, are different than non-choosers, but the nature of this difference is rarely examined directly. For years, commentators have expressed concern that chooser families will tend to have more educationally engaged parents than nonchooser families (Carnoy, 1993; Henig, 1995a). However, this hypothesis is rarely tested, in part due to data limitations in many school choice studies. Prior research has established the choosers tend to be more affluent than nonchoosers, but few attempts have been made to explain why this gap occurs (e.g. Cullen, Jacob, and Levitt, 2005; Martinez, Godwin, and Kemerer, 1996; Holme & Richards, 2009.) A more sophisticated understanding of the difference between choosers and nonchoosers is not only important for basic knowledge on school choice but also would contribute to a growing body of research on the sorting effects of school choice. There are growing concerns that school choice can function as a sorting mechanism, exacerbating segregation along racial and socioeconomic lines (e.g. Koedel, Betts, Rice & Zau, 2009; Garcia, 2008). Although some studies indicate that racial segregation may be getting worse through school choice, negative impacts on socioeconomic segregation tend to be the most common finding in studies on this topic (Holme & Wells, 2008; Garcia, 2010). However, the crudeness of socioeconomic information in school enrollment data means that these studies could be overestimating or underestimating the changes in segregation that are occurring through school choice. If choosers and nonchoosers tend to differ in ways that are not easily observable in most school data, the impacts of school choice as a sorting mechanism may not be fully understood. The gap between choosers and nonchoosers is a foundational issue for school choice research, and the question deserves to be approached with more detailed sociological analysis than has typically been conducted in school choice research thus far. Sociologists may also be interested in the relationships between parenting practices, socioeconomic status, and school choice. One of the leading theories on how parents transmit their advantages to their children emphasizes a specific, highly active type of parental involvement (Lareau, 2003). Research has found that these parenting practices are associated with higher academic achievement (Cheadle, 2009). Establishing a link between this parenting strategy and active participation in school choice would help expand the literature on how middle and upper class families utilize their social and material resources to help separate themselves from lower social strata. This study uses a recent nationally representative dataset of kindergartners and their parents to provide one of the most thorough sociological comparisons of choosers and nonchoosers to date. The 2010-11 Early Childhood Longitudinal Study (ECLS-K: 2011) has extensive surveys of parents, including questions about school choice, family structure, parenting behaviors, parent-child relationships, previous educational consumption decisions, and detailed socioeconomic data. After restricting the dataset to families in urban areas, non-choosers and three types of choosers are identified: • Residential public choosers- Families that moved to a certain location to attend a school. • Private school choosers- Families who sent their child to a private school. • Nonresidential public choosers- Families utilizing public school choice options, including those who actively selected charter schools, their “assigned” public school, magnet schools, or other traditional public schools • Nonchoosers- Families that sent their children to the assigned public school without considering other options. Three specific research questions will be examined in this study: 1) Does parental educational engagement or concerted cultivation predict choice and mediate the link between families’ socioeconomic status (SES) and their likelihood of participating in school choice? 2) Do parents’ socioeconomic statuses relative to their neighbors affect their likelihood of participating in school choice, as classic stratification theory would suggest? 3) Do the absolute SES, relative SES, and concerted cultivation effects on choice participation vary by race? Three sets of regression models will be estimated for each of these questions. The three types of active choosers will modeled against nonchoosers in logistic regressions. Regression models will also be run on a sample restricted to lower SES families in order to see if SES and concerted cultivation effects are stronger for this group. Model development for this study is informed by three major sociological perspectives and preliminary modeling of chooser status in an earlier version of this dataset, ECLS-K: 1999 (Weitzel, 2010). First, family process or family socialization models suggest that poverty creates substantial burdens for parents, making it more difficult for parents to engage in kids’ educational matters (e.g. Cooper, Crosnoe, Suizzo, and Pituch, 2010; Bodovski & Youn, 2010; Crosnoe & Cooper, 2010). Many family process models aim to define and test mediators that help explain the link between family’s SES and social or educational outcomes. Similarly, this study aims to explain the association between higher SES and a greater likelihood of participating in school choice. Family process models inform the overall mediational approach used in this study. Second, a prominent theory on parental engagement in education helps inform the mediation model for research question #1. Lareau’s (2003) concept of concerted cultivation, one of the leading theories on the transmission of socioeconomic status from parents to children, suggests that middle and upper class parents tend to take a very deliberate and hands-on approach to promoting children’s development. Working class parents, on the other hand, are more likely to step back and permit the “accomplishment of natural growth” in their children. The tendency of middle and upper class parents toward deliberate, positive action regarding child development could also lead to active involvement in school choice. Lastly, classic stratification theories suggest a competing notion of families’ motivations for school choice (e.g. Weber, 1947; Grusky, 2008). Classic stratification theory suggests that patterns of consumption are one of the major ways that different social classes separate themselves. For families wishing to distinguish themselves from their neighbors, school choice may be one way to accomplish that goal. In research question 2, indicators of families’ SES relative to their neighborhood are added to the models and interacted with SES to test this hypothesis. The descriptive results indicate that the private choosers were the most advantaged group on measures including socioeconomic status, family size, two- parent households, the use of English in the home, the use of center-based preschools, extracurricular activities for children, and several other measures. Non-choosers were the least affluent or advantaged of the four groups, and urban residential choosers were only slightly better off than non-choosers on most measures. Non-residential public choosers, which includes those attending magnet and charter schools, were more affluent than non-choosers and residential public choosers. The regression models include SES and concerted cultivation composites, home language indicators, preschool usage variables, a thorough set of controls for demographics, family composition, and neighborhood poverty. These models fit much better when modeling private school choice than for residential and non-residential public choice, perhaps due to the high direct costs of private school attendance. Model fit for non-residential public chooser models was better than that for residential choosers, which had rather poor fit overall. For private and non-residential public choosers, SES associations with choice were relatively strong, even after adding the full set of controls and concerted cultivation measures. The association between higher SES and the likelihood of participating in choice was even stronger when the sample was limited to families with SES below the national mean. Racial interaction effects indicated that SES effects were also stronger for black and Hispanic families, with especially strong effects for Hispanics. Interestingly, SES effects for urban residential choosers were not very strong and the direction of some other effects were the opposite of those seen for private and non-residential choosers. For example center-based preschool attendance was fairly strongly positively associated with private and non-residential public choice, but it was negatively associated with residential public choice. The positive association between residential choice and speaking a non-English language at home was also unexpected. Black and Asian families were more likely to be residential public choosers, even after all controls were added to the models. The results suggest that residential choice in urban areas may have a notably different profile than suburban residential choice. Residential school choice in urban areas may be serving as a vehicle for ethnocentric school moves, but this hypothesis will certainly need to be tested further. While SES effects varied by chooser type, the positive association between concerted cultivation and choice was moderately large and basically the same size for all three chooser types. Concerted cultivation, which was measured with a composite consisting of students’ participation in extracurricular activities and the number of books at home, did also mediate the relationship between SES and choice in a statistically significant way. As with the SES effects, concerted cultivation effects were stronger for lower SES families and for black and Hispanic families. When parental involvement in education was instead captured with measures of low-cost home-based activities like reading, making art, and telling stories at home, the composite had basically no association with school choice. High cost, transportation-intensive activities in the concerted cultivation composite, on the other hand, were rather positively associated with school choice. These results could suggest that participation in school choice is more affected by families’ economic means than by their attitudes or interest in educational engagement. This possibility would need to be investigated further in follow-up studies. Classic stratification theory suggests that families will make school selections consistent with their perceived social class. For research question #2 in this study, it was predicted that families would be more likely to select private or non-residential public schools when they were in a higher socioeconomic position than their neighbors. Results indicate that having educational attainment above the median in their neighborhood was moderately positively associated private school choice. When race interactions were added to the models, relatively strong positive associations between this measure and private choice were found for white and black families but not for other groups. For non-residential public choice, a positive association between this indicator and choice was also found for white families but not for other groups. This classic stratification hypothesis should be studied further with better measures of families’ SES positions and racial identities in relation to their neighbors. Implications of this study and additional recommendations for further research are discussed in the final chapter. Although ECLS data do not enable the direct observation of sorting effects through school choice, this exploratory study does suggest that student sorting may be occurring through school choice in ways that will be very difficult to observe through regular enrollment data. Adding better measures of family SES and parenting practices to traditional sorting studies will enable more rigorous examination of this possibility. This study also expands the literature around concerted cultivation by identifying another social outcome, school choice, with which the construct is associated. Sociologists who are particularly interested in the measurement and effects of concerted cultivation and parental engagement may wish to improve upon this exploratory study with structural equation analyses.  
Graduation Semester
2016-05
Type of Resource
text
Permalink
http://hdl.handle.net/2142/90523
Copyright and License Information
Copyright 2016 Peter Weitzel

Owning Collections

Graduate Dissertations and Theses at Illinois PRIMARY
Graduate Theses and Dissertations at Illinois