WELL-BEING, CHILD-CHILD CARE AND CHILD DEVELOPMENT1
Mother Well-Being, Child-Care Quality and Child Development: Do State Subsidy Policy Variations Change This Relationship?
Meryl Yoches Barofsky and Elisa L. Klein
University of Maryland, College Park
Paper Presented as Part of Panel Presentation:
Child Care Subsidies: Parental Stress, Child Care Quality, and Child Development
at Association for Public Policy Analysis and Management Annual Meeting
November, 2012
The authors thank the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) through grants R01HD36916, R01HD39135, and R01HD40421, as well as a consortium of private foundations for their support of the Fragile Families and Child Wellbeing Study.
WELL-BEING, CHILD-CHILD CARE AND CHILD DEVELOPMENT1
Abstract
Child-care subsidies (CCDF) were expanded after welfare reform in 1996 to help low-income families pay for child-care. Descriptive studies have been conducted on the relationship of subsidies to maternal work characteristics, but there is limited research on the extent to which CCDF is related to factors of maternal well-being. Further, few studies have examined how subsidy-use contributes to children’s developmental outcomes through maternal well-being and child-care experiences. Data from the Three-Year In-Home and Three-Year Child Care Study of the Fragile Families and Child Well-Being Study were used to examine these relationships in subsidy eligible families across states with different policies. Propensity score matching was used to limit the sample and multiple group comparisons assessed the invariance of the relationships in a structural equation model across families in states with divergent policy considerations. Our results indicate that subsidy-use is related to center-based care utilization but unrelated to child-care quality or maternal well-being. Quality independently is related to children’s vocabulary at three years. These relationships vary depending on choices states make about CCDF implementation, specifically the income limits set by states for eligibility and their use of a waitlist. Future research needs to unpack these findings and investigate how the choices states make about CCDF policy influence the relationship between subsidy-use, maternal well-being, child care and children’s developmental outcomes.
WELL-BEING, CHILD-CHILD CARE AND CHILD DEVELOPMENT1
Mother Well-Being, Child-Care Quality and Child Development: Do State Subsidy Policy Variations Change This Relationship?
Introduction
Although the Child Care Development Fund (CCDF, 45 CFR Parts 98 and 99), which distributes child-care subsidies to the states,was developed as a work support for low-income mothers transitioning off welfare, recent interest has focused how this program impacts children’s experiences in families receiving aid (Johnson, Martin & Brooks-Gunn, 2011). However, as the fund is a federal block grant, state level policy differences vary (Office of Child Care, 2012). Therefore the impact of subsidies on children and families may diverge depending on policy differences. Evaluations of the subsidy program have been focused on overall success with respect to the role of subsidies on employment opportunities (Forry & Hofferth, 2011), as well as recent work evaluating quality of child-care parents choose when using subsidies (Ryan, Johnson, Rigby & Brooks-Gunn, 2010). Few researchers have examined how the CCDFaffects low-income mother characteristics such as well-being and the potential impact of subsidy-use on child-care choices and children’s development (e.g. Herbst & Tekin, 2012). The purpose of this study was to examine how subsidy-useis related to mother well-being, and if mother well-being and subsidy-use jointly impact children’s development in families eligible for subsidies. To accomplish this, a sample of subsidy eligible families from the Three Year In-Home and Child Care Study of the Fragile Families and Child Well-Being Study was used. These relationships were tested across states with dissimilar policies to understand how differences instate level administration of the program differentially impact families and their children using subsidies.
Literature Review
Growing up in poverty has been linked to poor developmental outcomes in young children (Farah, et al., 2006; Noble, Shonkoff, Boyce & McEwen, 2003). A variety of factors, including less stimulating home environments, familial instability, and high rates of maternal stress in low-income families contribute to these differences in cognitive, social and emotional domains and in overall brain development (Brooks-Gunn, Klebanov & Duncan, 1996; Hart & Risley, 1995; Kneipp, Welch, Wood, Yucha & Yarandi, 2007; Linver, Brooks-Gunn & Kohen, 2002; Santiago, Wadsworth & Stump, 2011). Not surprisingly, young children growing up in poverty arrive at kindergarten behind their more advantaged peers and are often unable to catch up (Duncan & Magnuson, 2005; Janus & Duku, 2007).
Early childhood programs, specifically high-quality child-care, can have a beneficial impact on children’s development and skills when they enter formal schooling (Magnuson, Meyers, Ruhm, & Waldfogel, 2004). Children who attend high-quality child-care in the years before kindergarten enter school with higher academic skills and fewer behavior problems than similar children in low-quality settings (Burchinal, Vandergrift, Pianta, & Mashburn, 2010; Vandell, et al., 2010; Votruba-Drzal, Coley & Chase-Lansdale, 2004). For children growing up in and close to the poverty line, the effects appear to be greater (Magnuson, Ruhm, & Waldfogel, 2007; Votruba-Drzal, et al., 2004) and there is also evidence that children who attend high-quality programs carry some of these benefits into adulthood (Campbell, Pungello, Miller-Johnson, Burchinal & Ramey, 2001; Vandell, et al., 2010).
Today, young children are spending considerable time in out-of-home care arrangements as more mothers are in the work force than ever before (U.S. Department of Labor, 2009). For low-income parents, this is especially true since the passage of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA, Pub .L. 104-193, 110 Stat, commonly known as welfare reform), in 1996 which stipulated that families could only receive cash assistance for a limited time, and parents had to engage in work or educational activities to qualify for aid. Child-care subsidies were expanded in the wake of welfare reform in order to help low-income families pay for child care (Office of Child Care, 2011; CCDF, 45 CFR Parts 98 and 99). Families who use child-care subsidies generally purchase more formal and higher quality care, and thus, even though child-care subsidies were developed as a work-support for low-income parents, subsidies may be related to children’s child care experiences (Ryan, et al., 2011). However, there is limited research evaluating what this program means for children’s academic skills and social development and more research is essential to understand this relationship (Brooks, 2002; Herbst & Tekin, 2010).
Even though child care can have an important influence on young children’s development, maternal influence is the strongest in predicting child outcomes (Chazan-Cohen, et al., 2009; McLoyd, 1998; Peisner-Feinberg, et al., 2001). For example, mothers who report high levels of stress and depression have children with more behavioral problems and lower academic achievement (Chazan-Cohen, et al., 2009). Mothers’ perception of social support and feelings of mastery in mothering appears to moderate the impact of stress and depression in predicting children’s behavioral outcomes and mother-child interactions (Farmer & Lee, 2011; Lee, Lee & August, 2011). Further, mothers’ feeling of control over their lives mediates the relationship between stress and incidence of abuse and neglect (Guterman, Lee, Taylor & Rathouz, 2009). Therefore, although child care can be an important predictor of children’s academic and social developmental outcomes, mothers’ feelings of well-being (such as stress and social support) are essential to consider as an influence on children’s development, especially in a low income population in which well-being may be challenged (Keating-Lefler, Hudson, Campbell-Grossman,Fleck, & Westfall,2004).
There is evidence that child-care subsidies could affect mother well-being, and these feelings may influence children’s development. Child-care subsidies may reduce maternal stress in that the program provides increased financial resources for families compared to eligible non-users. However, research on this relationship is unclear. Herbst and Tekin (2012) found that mothers using subsidies have lower overall health, and higher rates of anxiety, depression and stress than non-users. Similarly, in a study on subsidized child-care in Canada, a negative relationship was found between the introduction of a subsidy program and maternal depression and stress, which increased with subsidy-use. The authors speculated that increased demands for maternal employment could be behind this finding (Baker, et al., 2008).
Conversely, social support, another aspect of mother well-being, may increase as families use subsidies if they receive some of that support from their child-care provider (Bromer, & Henly, 2009; Kossek, Pichler, Meece & Barratt, 2008). Further, low-income families who report more social support tend to experience less stress (Greenfield, 2011). Interestingly, the net result may be that those families tend not to use subsidies because they have assistance (child care or financial) from others (Hirshberg, Huang & Fuller, 2005; Shlay, Weinraub & Harmon., 2010).
Differences in maternal well-being may contribute to the type and quality of the care that families use as well as children’s academic and social development (Lowe & Weisner, 2004) and subsidy-use could be related to this. Specifically, since families who use subsidies purchase more formal and higher quality care (Crosby, et al., 2005; Bacharach & Baumeister, 2003; Ryan, et al., 2011), an important question to ask is whether subsidies play a role in that decision, such as providing parents with the necessary financial resources to do so.
Despite the potential for child-care subsidies to help low-income families, nationally few eligible families use the program (Giannarelli, et al., 2003; Kinukawa, Guzman, & Lippman, 2004). In order to understand the potential impact of child-care subsidies on children and families, it is important to know that: (1) not all eligible families want to use subsidies, and (2) there do appear to be some barriers in using this social service (Washington & Reed, 2008). Further, there do appear to be demographic differences between subsidy-users and eligible non-users. Families living in poverty are much more likely to use child-care subsidies than families not living in poverty. However, other differences not related to income are also evident in examining family participation. African-American families are more likely to use child-care subsidies than either White or Hispanic families, even after controlling for income (Kinukawa, et al., 2004). Non-English speaking immigrant families are less likely to use child-care subsidies compared to non-immigrant, English-speaking families (Grobe, et al., 2008; Hirshberg, et al., 2005; Shlay, et al., 2010). Finally, families who use child-care subsidies differ from eligible non-recipients in terms of family structure. Single mothers are more likely to use subsidies than either married or cohabitating mothers (Kinukawa et al., 2004). Subsidy recipients also report having fewer adults living in their household, regardless of marital status, discussed with respect to social support in the previous section (Ahn, 2012).
Finally, regulations for implementation of the federal subsidy program vary across states. While some criteria are federalized, states retain administrative control over the program, including certain eligibility components and other key conditions. There have been analyses of state level policy differences (Grobe, Weber, & Davis, 2008), but no direct comparison of child developmental outcomes or mother well-being measures across different state policies has been conducted. Studies examining subsidy-use and children’s academic and social developmental outcomes usually control for state policy differences without state-level policy comparisons (e.g., Herbst & Tekin, 2010a). These differences in policies may dictate whether families will use subsidies (Herbst, 2008; Joo, 2008; Pearlmutter & Bartle, 2003). For example, parents may not apply for or use subsidies if the reimbursement rate (amount the subsidy covers) is low and the co-payment that parents must pay out of pocket is high. Further, states’ decisions about income eligibility requirements may be related to families’ use of subsidies as well. Although federal guidelines recommend that a family income not exceed 85% of the state’s median household income (SMI); states vary widely on eligibility (Minton, et al., 2011). These differences in policies are complex. Comparing all subsidy eligible families without considering policy differences would lead to an incomplete picture of the influence of the subsidy program on children and families.
Given that over five billion dollars were appropriated for CCDF during 2011, understanding how subsidies impact children and families is essential (Office of Child Care, 2011). Little research has been conducted to examine how child-care subsidies are related to mothers’ well-being and if this program influences children’s development through mothers’ well-being (Baker, Gruber & Mulligan, 2008). This paper is a first step in answering some of these lingering questions. Specifically, we examined: 1) how use of subsidies and mother well-being affects children’s pre-academic skills and social development when they are about three years old; 2) how the use of subsidies and mother well-being are related to the type and quality of child-care that families purchased; 3) the extent to which state policy differences are related to mother well-being, child-care type and quality and children’s pre-academic skills and social development; and 4) to what extent state variation in child-care subsidy policy differentially explains these relationships.
Method
Data were analyzed from the Three-Year In-Home and Three-Year Child Care Studywaves of the Fragile Families and Child Well-Being Study, a large-scale longitudinal researchproject targeting at-risk families in large U.S. cities with children born in 1998 and 1999. The initial sampling frame included an over representation of unwed parents and their children. The baseline sample is representative of non-marital births in large U.S. cities in the years of data collection (Reichman, Teitler, Garfinkel, & McLanahan, 2001).
Procedures
The Three-Year In-Home Survey wave of the Fragile Families and Child Well-Being Study is a subsample of the Three-Year Core Survey. The In-Home Survey consisted of an in-depth, in-person mother interview about her child’s health, family routines, home environment and activities, maternal well-being, discipline, and children’s behavior problems. Direct assessments were administered to assess the child and mother’s vocabulary in English and Spanish when necessary (User’s Manual, 2008). Both the In-Home Study and Child Care Study took place in2002 and 2003. For the Child Care Study, trained researchers visited families’ child-care providers and completed quality assessments of settingsand interviewed directors.
Measures
For this study, individual variables and scales were chosen based on what was available in the Fragile Families and Child Well-Being interview. For the maternal well-being constructs, items were identified or grouped into specific categories on the interview questionnaire (e.g. Stress, Social Support, Mastery) by the study authors. We calculated scale scores for each of the following constructs: Maternal Stress, Maternal Mastery, Social Support and Maternal Self-Efficacy. The calculation of scale scores has been used previously with these data (Church, et al., 2012) and was used as opposed creating factors so each item was equally represented in the final score for the construct.
Child Care Subsidy Receipt. Based on mothers’ response to a question about whether they received government assistance paying for child care, families were classified as receiving subsidies or eligible non-users.
Maternal Stress. Twelve items taken and/or modified from the Early Head Start Study or developed for the Fragile Families and Child Well-Being Study were used to calculate this construct. Items were scored so that high scores indicate high stress.
Maternal Mastery. Five items taken from the Parental Mastery Scale (Pearlin & Schooler, 1978) were used to calculate this construct. High scores on these items indicate high feelings of mastery.
Social Support. Six items created for the Fragile Families Study were used to calculate this construct. Items were scored so high ratings indicate high levels of social support.
Maternal Self Efficacy. Four items were used to calculate this construct. The source for these questions was not provided, but these items have been used in other large-scale studies (i.e. ECLS-B). High ratings on these items indicate high feelings of self-efficacy.
Maternal Depression. Maternal Depression was measured using the Composite International Diagnostic Interview-Short Form (CIDI-SF), Section A (Kessler, et al., 1998) which consisted of 15 items assessing mothers’ mood in the past 12 months. Mothers were classified as to whether or not they were currently experiencing a major depressive episode.