Socioeconomic Status, Parenting, and Child Development

Marc H. Bornstein, Robert H. Bradley

• Publisher: Routledge
• Year: 2014
• Print ISBN:
  9780805842425
• eBook ISBN:
  9781135634001
• Topic: Psicología
• Subtopic: Psicología del desarrollo

Socioeconomic Status, Parenting, and Child Development presents cutting-edge thinking and research on linkages among socioeconomic status, parenting, and child development. The contributors represent an array of different disciplines, and approach the issues from a variety of perspectives. Accordingly, their "take" on how SES matters in the lives of children varies.

This volume is divided into two parts. Part I concerns the constructs and measurement of SES and Part II discusses the functions and effects of SES. Each part presents four substantive chapters on the topic followed by an interpretive and constructively critical commentary.

The chapters--considered as a whole--attest to the value of systematically examining the components of SES and how each flows through an array of specific parenting practices and resources both within and outside the home environment to help shape the course of child development. The result is a more fully delineated picture of how SES impacts the lives of children in the 21st century--a picture that contains a road map for the next generation of studies of SES and its role in the rapidly evolving ecology of family life.

I

SES: Measurement and Ecology

1

A Decade of Measuring SES: What It Tells Us and Where to go From Here

Margaret E. Ensminger and Kate E. Fothergill

Department of Health Policy and Management, The Bloomberg School of Public Health, The Johns Hopkins University

Introduction

Numerous studies have demonstrated the relation of socioeconomic status (SES) to health, psychological well-being, and attainment of socially and culturally derived goals. Theoretical and empirical work have emphasized that families’ socioeconomic situations have an impact on how parents rear their children (Bronfenbrenner & Morris, 1997; Elder, 1996; Elder & Conger, 2000; Kohn 1977). Individual development is shaped by the processes that are influenced by SES and that occur in key social contexts, including families, schools, and neighborhoods (Alwin & Thornton, 1984; Bidwell & Friedkin, 1989; Blau & Duncan, 1967; Demo & Acock, 1996; Duncan, Brooks-Gunn, & Klebanov, 1994). Social class has been one of the most robust predictors of health status for both adults and children (Bunker, Gomby, & Kehrer, 1989; Pappas, Queen, Hadden, & Fisher, 1993; Williams, 1990).

In addition to the direct effects of SES on outcomes that influence families and children, SES may also interact with other variables, moderating or mediating the influence of other key variables (Conger, Ge, Lorenz, & Simons, 1994; McLoyd, Jayartine, Ceballo, & Borquez, 1994; Sampson & Laub, 1994). For example, in a longitudinal study examining pathways to high school dropout, Ensminger and Slusarcick (1992) found that the influence of maternal education on high school graduation was through its interaction with individual characteristics. Having a mother with at least a high school education increased the likelihood that males who performed poorly in first grade or who had low educational expectations as adolescents would graduate from high school.

Furthermore, excluding the consideration of SES in studies on children and families ignores the diversity and economic variation that exists in the United States (MacPhee, Kreutzer, & Fritz, 1994). We cannot understand the full range of normal development if ethnic or socioeconomic diverse samples are not included in the studies that provide the basis for developmental science. Yet, despite the evidence of the importance of SES on developmental and health outcomes, it is not clear that there is much research on the effects of SES on growth and development or that SES is taken into account in other than the most elementary way.

For example, in the 1970s, Kohn (1977) described how the experiences in the course of doing one’s work could influence the values related to child rearing. His study was of a nationally representative sample of men. Yet, surprisingly little has been written since that time on whether and how SES influences parenting (for an important exception see the work of Menaghan and colleagues: Menaghan, Kowaleski-Jones, & Mott, 1997; Menaghan & Parcel, 1997).

In this chapter, we investigate the ways that SES is utilized and measured in research on children and adolescents. First, we review past work that has analyzed the child development literature with regard to the inclusion of SES measures. Second, we highlight some important issues pertaining to the measurement of SES, such as the source of information and the type of measure used. Next, we review articles published in three select journals over the past decade to assess the use and measurement of SES in studies of parents, children, and adolescents. Finally, we summarize the literature review findings and make some recommendations for the field.

Past Reviews of the Child Development Literature

Over the last 20 years, various content analyses have documented trends in research on SES and child health and development. These reviews have discussed the theoretical origins of SES measures, the common measures used today, the use of the measures (e.g., sample descriptors, control variables), SES’s confounding relationship with measures of race, and the precision of SES measures (Berrios & Hagen, 2000; Graham, 1992; Hagen & Conley, 1994; MacPhee, et al., 1994; McLoyd & Randolph, 1985; Smith & Graham, 1995). In addition to providing background on SES measurement in family research, these analyses highlight key areas in need of improvement.

First, in a review of articles published from 1991 through 1993 in Journal of Marriage and the Family and select psychology and sociology journals, Smith and Graham (1995) concluded that most researchers are not concerned about the theoretical and methodological issues related to using measures of SES. For example, theories of stratification, such as functionalism or Marxism, could inform research design, and Weber’s dimensions of stratification could help identify which measure will be most precise in particular circumstances. Yet, different theories of stratification and their implications are seldom referred to in the family or childrearing literature. In addition, this review concluded that there is a lack of systematic effort to identify which SES measures are most powerful in predicting family behaviors. The authors suggested that the three most common measures of SES— education, income, and occupation—may have differential effects, depending on the outcome of study (Smith & Graham, 1995; for a more detailed discussion of measurement issues, see the following section in this chapter).

Second, past content analyses also reveal a dearth of research on diverse samples. In a special 1994 issue of Child Development focused on poverty, MacPhee and colleagues (1994) reviewed nine journals, three each from child development, adolescent development, and parenting or socialization, between the years 1982 to 1991. They found that measures of family background were lacking for a significant minority of studies, and less than a third of the studies included low-income or ethnically diverse subjects. The authors noted that this absence of information on the samples significantly reduced the external validity of many of the studies.

In a review of the ethnicity of children studied in Child Development from 1980 to 1993, Hagen and Conley (1994) found that a number of studies offered no information on the backgrounds of their participants. The number of studies with unspecified samples decreased between 1990 and 1993, during which time most studies reported European American or predominantly European-American participants. According to a separate study comparing trends in Child Development to those in Developmental Psychology between 1991 and 1999, Developmental Psychology was even less likely to include studies on minority children and youth (Berrios & Hagen, 2000). Although these analyses found diversity to be lacking, Hagen et al. (2000), in a review of over 60 years of Child Development, found that since 1980, “white, middle class children continue to be very well represented, but increases in studies predominantly involving racially and ethnically diverse children are dramatic, as are increases in studies on atypical or at-risk children.”

McLoyd and Randolph (1985), in their analysis of trends in Child Development studies on African American children between 1936 to 1980, found that the specification of social class was highest between 1971–1975, but after 1975 it decreased substantially. The authors hypothesize that the increase between 1971 and 1975 was partly due to social and political changes, such as the federal antipoverty programs and federal support for research in this area, but they did not speculate as to what caused the subsequent decline. Wilson (1987) hypothesized that the controversy surrounding the Moynihan report (1965) “had the effect of curtailing serious research on minority problems in the inner city for over a decade, as liberal scholars shied away from researching behavior construed as unflattering or stigmatizing to particular racial minorities” (p. 4).

Third, in relation to failing to specifiy the class composition of African-American study samples, researchers of race also often fail to control for SES in the analyses. This is problematic because in the United States race and SES are highly correlated, and the lack of such controls can lead to misleading findings. Nonetheless, in the aforementioned analysis of more than 40 years of research on African-American children, McLoyd and Randolph (1985) found that there was a confounding of race and social class in 23% of the studies.

In a review of six psychological journals between 1970 and 1989, Graham (1992) noted that the research on African American populations was weakened by the lack of reporting on SES. In addition to finding fewer than one-half of the studies specifying the participants’ SES, she found that, “Even among those studies that did measure SES, it was not necessarily the case that this variable was appropriately examined as a factor in the analyses” (p. 634). She pointed out that researchers tended to compare low-income African Americans to middle-income European Americans without controlling for the effect of SES. She recommended that comparative racial studies should be required to report SES information and to test for interactions between race and SES.

These analyses summarize general trends identified thus far in the study of SES within family research. Most important, they highlight necessary improvements in SES research: the consideration of theory, methods, and type of measurement. The following section provides additional details on critical measurement issues. Special attention was given to these issues because the subsequent literature review focuses on SES measurement.

SES Measurement Issues

Several articles have reviewed both the concepts and the measures of SES as they exist in the social science literature. Smith and Graham (1995) gave theoretical and historical background to SES measurement, including a discussion of shifts in gender roles that impacts whose status should be measured. They also gave examples of how different theoretical positions of social stratification might influence the hypotheses and interpretations of results of family researchers.

In articles in Child Development focused specifically on the measurement of SES and ethnicity, Entwisle and Astone (1994) and Hauser (1994) posed guidelines for researchers. Entwisle and Astone recommended measures of income, mothers’ education, and household and family structure to be asked of parents. Hauser (1994) suggested gathering information about the major adult earner in the household, in order to ascertain the education, labor force status, and occupational position of that person. Hernandez (1997) discussed the utility of using standard census questions on household composition, family income, educational attainment, race and ethnicity, and employment so that information can be compared with national population data.

In an excellent review of measurement of social class, Liberatos, Link, and Kelsey (1988) concluded that there is no one best SES measure. They suggest nine criteria for selecting measures: conceptual relevance, the possible role of social class in the study, the applicability of the measures to the specific populations being studied, the relevance of a measure to the time period of study, the reliability and validity of the measures, the number of indicators included, the level of measurement, the simplicity of the measure, and comparability with other studies.

An issue in the measurement of SES is whether scales or separate indices are preferable. The consensus seems to be that multiple components should be measured, but that these should be used in analyses separately rather than combined into one scale (Bornstein, Hahn, Suwalsky, & Haynes, chap. 2, this volume; Duncan & Magnuson, chap. 3, this volume; Entwisle & Astone, 1994; Krieger, Williams, & Moss, 1997; Liberatos, et al., 1988; Mechanic, 1989). For example, Liberatos et al. (1988) concluded that the three indicators of social class most commonly used—occupation, education, and income—appear to be related to health, yet are not themselves highly intercorrelated. Each indicator is often associated with health outcomes independent of the other two. Similarly, Smith and Graham (1995) cited findings from several studies that suggest that in family research one SES variable may have considerably more power than another to account for a specific outcome.

An implicit assumption in the literature is that SES measures are based on individual characteristics. Increasingly, this assumption is being challenged. Krieger et al. (1997) argued that both households and neighborhoods influence the social resources available to individuals and often characterize aspects of living conditions that are not captured by individual measures. The impact of neighborhood on developmental outcomes of children and adolescents has been of growing interest in the child and adolescent literature (Duncan et al., 1994; Levanthal & Brooks-Gunn, 2000).

We now turn to review how SES is used in research on parents and children. Specifically, we examine whether SES is used by investigators as an independent or control variable that potentially influences the questions under study, a dependent variable in the research, or as a descriptor of the study population. When SES is measured, we also note how it is measured, the level of the measurement, and the source of the SES information. For illustrative purposes, we compare and contrast three journals from different disciplines and targeting different audiences.

Review of SES use in Select Journals

Three journals were selected for review based on their representation of different perspectives in the field of child health and development: Child Development, American Journal of Public Health, and Journal of Health and Social Behavior. Child Development was chosen as the premier U.S. publication focusing on children and development. While it is not discipline specific, many of the journal’s articles are written from a developmental psychology perspective, and most of the articles concern children or adolescents. The American Journal of Public Health was chosen for its focus on health and its multidisciplinary approach. Many of the articles are based on populations of children, adolescents, or both. The Journal of Health and Social Behavior is one of the journals sponsored by the American Sociological Association. It examines health and behavior issues from a sociological perspective, and it reports findings from all age groups, including children.

This study analyzed 10 years of literature from 1991 to 2000. For ana...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Dedication
• Table of Contents
• Series Foreword Monographs in Parenting
• Socioeconomic Status, Parenting, and Child Development: An Introduction
• Part I: SES: Measurement and Ecology
• Part II: SES: Parenting and Child Development
• About the Authors
• Author Index
• Subject Index