Factors influencing child care-related maternal work exits

Children and Youth Services Review 46 (2014) 168–176

Contents lists available at ScienceDirect

Children and Youth Services Review journal homepage: www.elsevier.com/locate/childyouth

Factors influencing child care-related maternal work exits Meirong Liu a,⁎, Manrong Chen b, Steven G. Anderson c,d,1 a

School of Social Work, Howard University, 601 Howard Place NW, Washington, DC 20059, United States The School of Social Welfare, SUNY at Albany, United States c School of Social Work, Arizona State University, 411 N. Central, Suite 800, Phoenix, AZ 85004-0689, United States d School of Social Work, Michigan State University, 655 Auditorium, Room 254 Baker Hall, East Lansing, MI 48824, United States b

a r t i c l e

i n f o

Article history: Received 10 March 2014 Received in revised form 19 July 2014 Accepted 22 July 2014 Available online 9 August 2014 Keywords: Working mothers Child care-related work exits Neighborhood factors

a b s t r a c t For mothers with young children, child care challenges can pose significant barriers for their labor force participation. Working mothers must arrange for someone else to care for their children when working outside the home. Previous research has shown that women with children spend less time in the labor force compared to women without children. This study used the Fragile Families and Child Wellbeing study to examine whether a range of child care characteristics, neighborhood factors, and individual factors caused mothers of young children to leave the work force. The results indicated that child care-related work exits are common occurrences for mothers in large urban areas. Of those mothers in the FFCW sample who used non-parental child care, more than one in ten mothers reported work exits due to child care-related problems. Logistic regression analysis further revealed that common risk factors for work exits included changing child care arrangements, using multiple types of child care, living in neighborhoods with a higher percentage of Hispanic population, being African American, and having household income between 50 and 99% of FPL. The findings are useful in informing social policies and interventions to help mothers better bridge the gap between adequate child care and gainful employment. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction Studies have found that women with children spend less time in the labor force compared to women without children (Pew Research: Social & Demographic Trends, 2013). Challenges related to child care often pose significant barriers that may partially explain such differences in labor force participation (Dodson & Bravo, 2005). For example, using a national sample, Usdansky and Wolf (2008) found that 31% of mothers of young children had to make special arrangements because their child care arrangements fell through at some point in the month preceding the interview; among them, 46% reported missing work or school as a consequence. Henly and Lyons (2000) utilized a sample of lowincome mothers in California and found that two-fifths of the sample experienced child care problems that interfered with their ability to retain employment. Understanding the predictors of child care-related work exits is a necessary first step toward developing programs and policies to help mothers to better bridge the domains of child care and employment (Usdansky & Wolf, 2008). In addition, breakdowns in child care are ⁎ Corresponding author. Tel.: + 1 202 806 4720; fax: + 1 202 387 4309. E-mail addresses: [email protected] (M. Liu), [email protected] (S.G. Anderson). 1 Tel.: +1 602 496 0058; fax: +1 602 496 0959.

http://dx.doi.org/10.1016/j.childyouth.2014.07.013 0190-7409/© 2014 Elsevier Ltd. All rights reserved.

estimated to cost U.S. businesses $3 billion each year in workplace disruptions (Shellenback, 2004). Nonetheless, there have been relatively few studies examining the factors that make some mothers prone to child care-related job termination. Hofferth and Collins (2000) examined how cost, quality, and availability of non-maternal child care influenced the probability of a maternal work exit using the National Child Care Survey (NCCS); Han and Waldfogel (2001) utilized Census Population Survey (CPS) and the Survey of Income and Program Participation (SIPP) data to examine the effects of child care costs, quality and availability on probability of employment for women with pre-school-aged children; Bub and McCartney (2004) used data from the National Institute for Child Health and Human Development (NICHD) to assess how hours in child care during infancy, toddlerhood, and the preschool years predicted mothers' maternal wages and hours of employment when their children reached first grade. More recently, Forry and Hofferth (2011), and Usdansky and Wolf (2008) each employed data from the Fragile Family and Child Wellbeing (FFCW) study to explore circumstances in which mothers had to make special arrangements because their usual child care arrangements fell through. None of the above studies, however, have examined whether the reasons that mothers had to quit a job, school, or training stemmed from problems in arranging child care or maintaining a child care arrangement. In addition, while studies have concentrated on mothers' individual and child care characteristics, relatively little research has

M. Liu et al. / Children and Youth Services Review 46 (2014) 168–176

examined neighborhood effects on child care-related maternal work exits. As suggested by the ecological model (Bronfenbrenner, 1994), contextual factors, such as neighborhood socioeconomic and demographic characteristics, can also influence mothers' child care-related work performance in important ways; this theoretical perspective is used to frame the hypotheses and related analyses in this study. As described by Bronfenbrenner (1994), ecological theory is grounded in a “person–process–context” paradigm in which both personal characteristics of the individual family members and broader process and contextual factors are viewed as important in understanding parental behaviors (i.e., maternal child care-related work exits). More generally, it provides interconnected, hierarchical constructs hypothesized to influence human behaviors, including: 1) the micro-system, which is the immediate environment in which individuals are embedded; 2) the meso-system, which refers to the interrelationships between the various settings of the micro-system; 3) the exo-system, which also refers to the linkages and processes between two or more settings, but in which at least one of the settings indirectly influences the individual; 4) the macro-system, which consists of “the overarching pattern of micro-, meso-, and exo-systems characteristics of a given culture or subculture, with particular references to the belief systems, bodies of knowledge, material resources, customs, life-styles, and opportunity structures and life course options that are embedded in each of these broader systems”; and 5) the chrono-system, which includes change or consistency over time not only in the characteristics of the person but also in the environment in which that person lives (Bronfenbrenner, 1994). Each of these systems shapes human development, and each system can impact the others. The general ecological model has been applied extensively to research in a number of fields to explain the development of a variety of behaviors, including individual's family and employment outcomes (Grzywacz & Butler, 2005; Wayne, Grzywacz, Carlson, & Kacmar, 2007). The ecological model is particularly well-suited to the study of child care-related work exits. The mother is embedded in the immediate social settings (micro-system, for example, mothers' race/education), which are directly affected by other settings in the community (exosystem, for example, child care factors). These two systems are in turn affected by broader economic and political structures, such as neighborhood socioeconomic characteristics, that influence how social institutions are organized in our society (macro-system). In the following section, the ecological model is used to organize and integrate research findings on child care-related work exits, so that the relevant influences can be more specifically understood and research limitations can be identified. Although the selection and use of child care is a process that affects and is affected by the entire family, the primarily focus of the review will be mothers' characteristics and environments. This decision to concentrate on mothers rather than on fathers is based on prior literature concerning child care choice behavior, which has found that mothers have the primary responsibility for arranging and maintaining nonparental care arrangements in most families (Atkinson, 1991; Hofferth, Shauman, Henke, & West, 1998). This article applies the Bronfenbrenner's theoretical model to examine maternal child care-related work exits under the influence of a range of environmental context variables, child care characteristic, and mother's individual factors. Using recently available data from the Fragile Families and Child Wellbeing (FFCW) study, we investigated how child care factors, mother's individual characteristics, and neighborhood characteristics shape mothers' experiences of child care-related work exits in a national sample of mothers of three-year old children. We begin with a literature review of the implication of child care factors, mother's individual characteristics, and neighborhood characteristics. Next, we use the FFCW study to conduct analyses of the risk of child care-related maternal work exits. We conclude with a discussion of the implications of our findings to help mothers bridge the gap between reliable child care and gainful employment. While we recognize that many fathers play an active role in managing child

169

care, we focus on mothers because they handle a disproportionate share of this work (Pungello & Kurtz-Costes, 1999). 2. Literature review 2.1. Child care characteristics 2.1.1. Type of care The types of child care arrangements vary widely along a number of dimensions related to the reliability of care (Forry & Hofferth, 2011). Studies have shown that home-based providers were less reliable than center-based providers, and that the unavailability of small home-based providers was associated with mothers' work exits (Gordon, Kaestner, & Korenman, 2008; Knox, London, Scott, & Blank, 2003; Scott, London, & Hurst, 2005). Research also has found that mothers using informal care were more likely to miss work when their usual child care arrangements fell through than mothers who relied on center care (Boushey, 2003; Huston, Chang, & Gennetian, 2002; Usdansky & Wolf, 2008). Researchers indicated that this may be because child care centers employ multiple staff members, and thus can remain open when one employee falls ill or takes a vacation (Usdansky & Wolf, 2008). The licensure process for formal child care settings is also likely to foster a greater sense of professionalism and commitment than informal caregivers, who often perform child care to help out a relative (Anderson, Liu, & Liao, 2012; Henly & Lyons, 2000; Liu & Anderson, 2012). On the other hand, the instability of informal care may be due to providers' own employment, illness, or temporary disability (Gordon et al., 2008). Scott et al. (2005) suggested that with relative child care providers, the instability was a product of changes in the lives of the care providers: they found other jobs, recovered from illnesses, had babies themselves, or experienced other life changes that made them unavailable to continue providing the care. 2.1.2. Child care stability Previous studies have shown that child care instability affected mothers' ability to be employed, as well as their job stability (Adams & Rohacek, 2010; Bowen & Neenan, 1993; Hofferth & Collins, 2000; Mason & Kuhlthau, 1992; Presser & Baldwin, 1980). For example, Meyers (1994) found that mothers who had to give up their child care arrangements when they changed activities in California's GAIN program were more likely to drop out of the training program than those who did not have this discontinuity in child care. Hofferth and Collins (2000) also suggested that the stability of care was linked to the job stability of moderate- and high-wage mothers. Not all researchers agree, however, that child care disruptions are associated with job turnover. Miller (2005) found little evidence that child care instability causes employment instability, concluding instead that other underlying factors, such as low education and family instability, caused both child care and employment disruption. Usdansky and Wolf (2008) also suggested that changing providers was not related to the odds of work disruption or work absence. 2.1.3. Child care multiplicity Research suggests that child care multiplicity affects the parent's ability to meet family and work obligations. Parents often make multiple arrangements to cover their children's care (Eichman & Hofferth, 1993; Folk & Yi, 1994). Formal center care can exhibit relatively high price elasticity; with family care, informal care, and unpaid relative care serving as substitutes as parents move between market and nonmarket forms of care (Kimmel, 2006). Some research has suggested that the frequent use of multiple arrangements may permit more flexibility and greater ability to adapt to unexpected events, increasing one's ability to continue working. For example, Hofferth and Collins (2000) found that the presence of more child care arrangements was associated with a significantly decreased risk of leaving the present job. However,

170

M. Liu et al. / Children and Youth Services Review 46 (2014) 168–176

other research has shown that using multiple child care providers for a child was linked to an increased probability of experiencing child carerelated work disruptions (Huston et al., 2002; Scott et al., 2005; Usdansky & Wolf, 2008). For example, Usdansky and Wolf (2008) found that each additional provider increases the likelihood of care breaking down by 18%. They concluded that patchworks of care can represent either successful efforts to build a system of care provision or the tenuous nature of care arrangements, or mothers turn to patchwork system when care is unreliable. Scott et al. (2005) also found that mothers may experience confusion when they need to work with multiple providers. 2.2. Neighborhood factors Some studies have examined whether neighborhood availability of child care providers is associated with women's labor force participation. For example, Hofferth and Collins (2000) found that the availability of child care was one of the most important factors related to job stability, in that a greater number of centers and home-based programs in the same counties were associated with a lower probability of work exits for moderate-wage mothers. Han and Waldfogel (2001) also examined the effects of child care availability on women's employment decisions; they found that child care availability is correlated with employment for married mothers but not for unmarried mothers. At the same time, they concluded that this may be due to the measure they used for availability – the supply of center care and family day care – which is less relevant for unmarried mothers, who are more likely to use the kind of informal care that would not be captured by this measure. In a related vein, the supply and type of child care available to working mothers may be affected by neighborhood socioeconomic and demographic characteristics. For example, studies have found that low-income neighborhoods often have limited supplies of centerbased child care or family child-care homes (Fuller, Kagan, Caspary, & Gauthier, 2002; Huston et al., 2002); and mothers in neighborhoods with higher poverty rates and Hispanic populations were more likely to rely on informal care (Liu & Anderson, 2012). Other studies have found that neighborhood unemployment rates and median housing prices were associated with the availability of informal providers (Scott et al., 2005) and the affordability of the child care market (Weber, Grobe, Davis, Kreader, & Pratt, 2007). Forry and Hofferth (2011) further found that living in an area with a higher median housing price was positively associated with experiencing a child care-related work disruption. 2.3. Individual characteristics 2.3.1. Income Researchers suggested that while non-parental child care is generally necessary for mothers to work, it may be an even more critical factor for low-wage mothers. Usdansky and Wolf (2008) and Huston et al. (2002) found a higher household income to be a protective factor against experiencing a child care-related work disruption. Likewise, Gordon et al. (2008) found unreliable child care to be associated with work exits only among impoverished families. However, researchers also have suggested that, low-wage mothers may have access to direct subsidies through vouchers and direct provider payments that are not available to moderate- or high-wage mothers, making the cost of child care lower and the gains from employment higher. Thus, because of direct subsidies, it has been argued that many low-income families have care comparable in quality to that of higher-income families, with children from moderate-income families in lower quality care (Hofferth & Collins, 2000). 2.3.2. Work hours Previous studies have suggested that greater labor force attachment decreases the odds of an employment exit (Hofferth & Collins, 2000). In

their study, Hofferth and Collins (2000) found that mother's work hours per week were negatively related with mother's work exits. Usdansky and Wolf (2008) also found longer parental work hours to be associated with a lower likelihood of experiencing child care-related work disruptions. They suggested that this may due to the high value parents who work long hours, ascribe to maintaining their jobs. Other researchers have explained this phenomenon by positing that adults with higher human capital may be more likely to have stable full-time jobs with regular schedules and relatively high earnings, all of which increase the likelihood of using child care, especially center care (Capizzano, Adams, & Sonenstein, 2000; Fuller et al., 2002; Huston et al., 2002). However, Forry and Hofferth (2011) found no such association between mother's work hours and child care-related work disruptions. 2.3.3. Social support Studies have found that social support serves as an important protective factor against child care-related work disruptions (Huston et al., 2002; Usdansky & Wolf, 2008). Henly and Lyons (2000) suggested that parents' informal social networks and short-term, temporary, and emergency providers helped them to successfully manage work and family demands. Hofferth and Collins (2000) also found that for the low-income women in their sample, employment stability was not associated with the stability of child care when mothers had access to another arrangement, such that the back-up provider filled in when the primary arrangement fell through. 2.3.4. Marital status Marital status provides one important source of potential resources within the family. Spouses and partners may be an important source of backup child care when regular care arrangements are unavailable (Han, 2004; Perry-Jenkins, 2005). Rather than facing pressure to find care, married mothers may be able to rely on fathers to provide temporary care while they search for more desirable permanent arrangements (Usdansky & Wolf, 2008). For example, Han and Waldfogel (2001) found that single mothers with pre-school-aged children were less likely than married mothers with pre-school-aged children to participate in the labor market. They further point out that this may be partially due to the fact that single mothers typically had low earning; moreover, their options for care were diminished by the absence of a husband as a potential caregiver. However, Usdansky and Wolf (2008) did not find that the presence of a spouse or partner reduced the risk of child care disruptions in their study. 2.3.5. Maternal education Some research has suggested that maternal education is associated with the likelihood of experiencing child care-related problems due to its association with child care preferences, income, and work characteristics (Mulligan, Brimall, West, & Chapman, 2005). For example, mothers with higher levels of education may also have better sources of information about the quality and reliability of providers, and thus may be able to find reliable care options with less disruption (Harknett, 2006). Highly educated mothers might also have better access to backup care when they do face care disruptions (Harknett, 2006). Other studies, however, have not found maternal education to be a significant predictor of child care-related work disruptions (Huston et al., 2002; Usdansky & Wolf, 2008) Previous research also has reported that child care-related work disruptions vary by race/ethnicity, but no consistent patterns have emerged (Huston et al., 2002; Usdansky & Wolf, 2008). For example, Huston et al. (2002) found that African Americans reported more child care problems than did European-Americans. However, Usdansky and Wolf (2008) found that African-American women were less likely to experience child care related work disruptions. The same study, however, found that immigrants were more likely to lose a job due to child care-related problems (Usdansky & Wolf, 2008). Other studies demonstrated that mothers with younger children were more

M. Liu et al. / Children and Youth Services Review 46 (2014) 168–176

likely to experience job termination (Hofferth & Collins, 2000; Huston et al., 2002). In summary, existing knowledge on understanding the factors that impact mothers' experiences of child care-related job termination is limited in significant ways as follows. First, few quantitative studies have been conducted in the period following the implementation of the Personal Responsibility and Work Opportunity Reconciliation Act, which significantly altered public welfare programs and related child care needs for low-income families. Second, studies conducted since PRWORA have concentrated on examining the effects of individual characteristics; there have been limited analyses of important contextual factors, such as how neighborhood socioeconomic factors influence working mothers' child care-related work exits. Third, previous studies have not examined whether the work exits are directly related with mothers' child care arrangement problems. Using a national data set, this study responds to the above limitations by examining how child care characteristics, neighborhood factors, and individual characteristics are related to mothers' child care-related work exits. 3. Method 3.1. Sample The study investigated the above research questions through secondary data analysis using the Fragile Families and Child Wellbeing (FFCW) study. The FFCW study contains a stratified random sample of 4898 primarily single mothers with children born between 1998 and 2000 in 20 cities in 15 different states (Reichman, Teitler, Garfinkel, & McLanahan, 2001). Mothers and fathers were first interviewed at the time of the focal child's birth (baseline), with follow-up telephone interviews occurring when the child reached approximately 12 (one-year follow-up), 36 (three-year follow-up), and 60 months (five-year follow-up) of age. Response rates for FFCW were high, with 4898 mothers completing the baseline interview (87% [3712] and 82% [1186] response rate for unmarried and married mothers, respectively). The Year 3 interview was completed by 4231 mothers (Center for Research on Child Wellbeing (CRCW), 2008). The study used data collected from the mother at the child's birth and Year 3 follow-up. Because of the relatively long gap of two years between data collection waves, longitudinal analyses of this data set were not included (Forry & Hofferth, 2011). The study sample included mothers who were interviewed at Year 3, living with the focal child for at least half of the time, and using non-parental child care regularly more than ten hours per week at the time of the interview. Application of these criteria resulted in a sample size of 2056. There were no missing values for the dependent variable. Among these 2056 observations, 1970 had complete neighborhood information. We then deleted cases that had missing values on other independent variables, which resulted in a final analytic sample size of 1860.

171

day care, center care, and other); 2) the frequency of changing care arrangements since the child's first birthday (never changed, changed once or twice, and more than twice); and 3) whether using more than one type of care at the same time. 3.2.2.2. Neighborhood factors. The data for neighborhood characteristics were obtained from the FFCW study contract data. The contract data file contains tract-level 2000 census data for the tracts where the mothers and fathers lived at the time of the baseline interview, and at the one year and three year follow-up interviews. At each interview, cases were assigned a geocode and associated tract characteristics from the 2000 census, and then were linked to the individual records. For this study, the investigator used the measures of 1) tract level poverty composition, which is measured as the percentage of families that had incomes below the poverty level in 1999; 2) tract level portion of foreign born; and 3) tract level racial/ethnic composition, which includes the percentage of population who were non-Hispanic African American and the percentage of Hispanic population. The above measures were obtained by merging the core data and the contract data and were measured as continuous variables. 3.2.2.3. Individual factors. The analysis included a number of individual level's socio-economic characteristics that have been shown in previous studies to be associated with mothers' child care-related work performance. Mothers' individual level factors included mother's race/ ethnicity, household income, education level, work hours, the perceived availability of emergency child care assistance, marital status, age, immigrant status, and monthly age of the focal child. 4. Analysis The data analysis consisted of three steps. First, we calculated descriptive statistics for the sample. Second, bivariate analyses were conducted to test the associations between child care-related job termination and each of the independent variables. Chi-square tests identified the statistically significant associations, and independent samples t-tests were conducted to compare mean differences. Third, a logistic regression model was estimated to examine how child care characteristics, neighborhood socioeconomic characteristics, and individual characteristics were related to mothers' child care-related work exits. When estimating the logistic regression models, there was virtually no clustering of individual cases within neighborhoods. Therefore, there was little possibility that the standard errors would be artificially depressed due to clustering. We conducted separate analysis that adjusted for any clustering that did exist in models not shown, and it did not influence the results. Thus, for simplicity, this article presents the logistic regression model with no clustering effects. 5. Findings

3.2. Measurement

5.1. Descriptive analysis

3.2.1. Dependent variable The dependent variable in this study was mothers' child care-related work exits. At the Year 3 interviews, mothers were asked “At any time (since the last interview), have you had to quit a job, school, or training activity because you had problems arranging child care or keeping a child care arrangement?” Data from this question were translated into a dichotomous variable indicating whether a mother reported work exits because of the child care-related problems (1 = yes, 0 = no).

Table 1 presents descriptive statistics of the sample. Of particular interest in Table 1 was the extent of child care-related work exits experienced by this population. The results show that about 12.3% of the mothers had to quit a job, school, or training as a result of child carerelated problems in the two years preceding the interview (N = 1860). As for the child care characteristics, about one-half of mothers relied on child care centers for their three-year-old children; over one-third of mothers used their own or the fathers' relatives; and about one-tenth relied on non-relative family day care. More than one-half had never changed child care arrangements since the child's first birthday, while over one-third changed arrangements once or twice and about 8.5% changed arrangements more than twice. About 14% of the respondents were using more than one type of child care arrangement by the time of the interview.

3.2.2. Independent variables 3.2.2.1. Child care characteristics. The characteristics of child care arrangements are composed of three indicators: 1) the type of primary child care arrangement (including relative care, non-relative family

172

M. Liu et al. / Children and Youth Services Review 46 (2014) 168–176

Table 1 Characteristics of analysis sample (N = 1860). Sample mean (%) Child care type Center care Relative care Family day care Other Child care stability Change 1–2 times N2 times Multiple care (yes) Neighborhood Percent of African American Percent of Hispanic Percent of poverty Percent of immigrant Race White African American Hispanic Other Income b49% FPL 50–99% FPL 100–199% FPL 200–299% FPL N300% FPL Education bHigh school High school Some college College or beyond Work hour Emergency care Marital status Married Mother's age Born in the U.S. Child age in months

Table 2 Bivariate analysis: mothers' child care-related work exits (N = 1860). SE

Range

50.5 36.8 11.3 1.4 38.4 8.5 14.0 0.41 0.18 0.18 0.12 20.3 53.7 22.8 3.3

0–98% 0–97% 0–73% 0–71%

28.4 27.9 87.7 35.6

No

Yes

87.3 89.5 88.0 80.8

12.7 10.5 12.0 19.2

93.0 83.1 75.9

7.0 16.9 24.1

83.5 88.4 41.5 17.1 17.2 11.7

16.5 11.6 45.1 21.8 20.9 14.3

86.0 87.0 92.6 91.8

14.0 13.0 7.4 8.2

0–49% 50–99% 100–199% 200–299% N300% Education⁎⁎⁎

76.7 82.9 87.5 92.7 95.2

23.3 17.1 12.5 7.3 4.8

bHigh school High school Some college College or beyond Work hours⁎⁎⁎ Emergency care⁎⁎⁎ Yes No Marriage status⁎⁎

80.6 89.7 89.6 95.4 37.7

19.4 10.3 10.4 4.6 35.2

88.8 77.3

11.2 22.7

Married Not married Mother's age⁎⁎⁎ Born in the U.S. Yes No Child age in months⁎⁎⁎

91.5 86.3 28.2

8.5 13.7 26.0

88.2 84.6 35.5

11.8 15.4 36.0

Child care type Center care Family day care Relative care Other Child care stability⁎⁎⁎ Never change Change 1–2 times N2 times Multiple care⁎ Yes No Percent of African American Percent of Hispanic⁎⁎ Percent of poverty⁎⁎⁎ Percent of foreign born Race⁎⁎ African American Hispanic White Other Income⁎⁎⁎

15.5 18.0 25.8 16.2 24.6 29.0 28.3 29.9 12.8 37.4 90.8

Variable

10.1

2–90

5.9

17–50

2.3

32–46

With respect to the neighborhood variables, the respondents lived in neighborhoods with an approximate mean of 41% of African Americans and 18% of Hispanics. The average poverty rate in these neighborhoods was 18%, and the neighborhoods on average had 12% foreign born populations. More than one-half of the respondents were African American, while one-fifth was White and another one-fifth was Hispanic. More than three-quarters of the respondents had at least a high school degree. Single mothers comprised about three quarters of the sample. The respondents were 28-year-olds on average, and 90% were born in the U.S. The children's average monthly age was 35 months. On average, the respondents were working 37 h per week for their current or most recent job. About one-third of the respondents had household incomes below the Federal Poverty Line (FPL); and; 25.8% had income between 100 and 199% of the FPL. About 91% of the respondents reported having potential child care assistance from informal social support networks. 5.2. Bivariate analysis Table 2 presents the bivariate analysis for each independent variable and the dependent work exit variable. Chi-square tests found that mothers' child care-related work exits were significantly associated with child care characteristics. Mothers who changed child care arrangements since the child's first birthday were more likely to quit a job, school, or training because of the child care arrangement problems (24.1% vs. 7%). In addition, mothers who were using multiple child care were more likely to quit a job, school, or training because of child care problems (16.5% vs. 11.6%).

Sig p = 0.573

p b 0.0001

p = 0.023

p= p= pb p= p=

0.161 0.005 0.001 0.101 0.007

p b 0.0001

p b 0.0001

p b 0.001 p b 0.0001

p = 0.002

p b 0.001 p = 0.128

p = 0.001

+

p b 0.10. ⁎ p b 0.05. ⁎⁎ p b 0.01. ⁎⁎⁎ p b 0.001.

As for neighborhood variables, t-tests found that a child care-related work exit was significantly associated with neighborhood racial composition, poverty rate, and immigrant rate. Mothers in neighborhoods with more Hispanics were more likely to quit a job because of the child carerelated problems, as were mothers in neighborhoods with higher poverty rates. In addition, mothers who lived in neighborhoods with more immigrants were more likely to quit work because of child carerelated problems. As for mother's individual level characteristics, Chi-square tests confirmed that mother's child care-related work exits were significantly associated with mother's race/ethnicity, income, educational attainment, perceived availability of emergency care, and marital status. For example, the bivariate analysis found that African American mothers and Hispanic mothers were more likely to experience child care-related work exits than Whites and others (14.0% and 13.0% vs. 7.4% and 8.2%). Mothers who had household incomes of more than 200% of the FPL were less likely to experience child care-related work exits than mothers who had household income less than 199% of the FPL (7.3% and 4.8% vs. 12.5%, 17.1%, 23.3%, respectively). In addition, mothers who had less than a high school education were more likely to

M. Liu et al. / Children and Youth Services Review 46 (2014) 168–176

experience child care-related work exits than mothers who had high school or more than high school education (19.4% vs. 10.3%, 10.4%, and 4.6%, respectively). Mothers who perceived that they would have emergency care were less likely to experience child care-related work exits (11.2% vs. 22.7%). Married mothers were less likely to loss job due to child care problems (8.5% vs. 13.7%). t-Tests found that mothers who worked more hours, who were older, and who had a younger focal child were less likely to quit their job, school, or training because of the child care arrangement problems. 5.3. Multivariate analysis Table 3 presents the results of the multivariate analysis for work exits due to child care arrangement problems. AIC and − 2 log likelihood are presented as measures of the fit statistics of the estimated statistical model. As shown in Table 3, child care characteristics were found to be significantly associated with mother's child care-related work exits. Mothers who had changed their child care arrangements once to twice, and more than twice since child's first birthday were more likely to quit their jobs due to child care disruptions (β = 0.218, p = 0.042; β = 0.659, p b 0.001, respectively). Mothers who were using multiple child care arrangements also were more likely to quit job due to child care arrangement problems than mothers who were not using multiple care (β = 0.222, p = 0.033). We did not find that the type of primary child care was associated with mother's work exits.

Table 3 Logistic regression: mothers' child care-related work exits (N = 1860). Estimate −2 log likelihood AIC Child care type (relative care) Center care Family day care Other Child care stability Change 1–2 times N2 times Multiple care (never change) Neighborhood Percent of African American Percent of Hispanic Percent of poverty Percent of immigrant Race (White) African American Hispanic Other Household income (0–49% FPL) 50–99% 100–199% 200–299% N300% Education (bhigh school) High school Some college College or beyond Work hour Emergency care Marital status Mother's age Born in U.S. Child age in months

Standard error

6. Discussions

0.177 0.241 0.424

0.530 0.878 0.737

0.218⁎ 0.658⁎⁎⁎ 0.222⁎

0.107 0.146 0.104

0.042 b0.0001 0.033

0.114 0.868+ 0.197 1.068

0.361 0.485 0.698 0.651

0.751 0.073 0.778 0.101

0.343+ −0.212 −0.148

0.189 0.209 0.391

0.098 0.311 0.705

0.366⁎ 0.041 −0.444⁎ −0.658⁎⁎

0.154 0.141 0.199 0.222

0.018 0.770 0.026 0.003

−0.169 0.031 −0.146 −0.018⁎ −0.269⁎

0.155 0.147 0.288 0.008 0.108 0.106 0.016 0.137 0.031

0.277 0.833 0.612 0.014 0.012 0.249 0.002 0.321 0.133

0.122 −0.051⁎⁎ −0.136 0.046

A likelihood ratio test between a full model and a restricted model without the neighborhood variables was conducted to identify whether, as a whole, the neighborhood factors were significantly related to the outcome variable. The results show that the − 2 log likelihood differences were χ2 (4) = 10.773 (p = 0.029), suggesting that the full model fits better than the restricted model. In other words, overall, the four neighborhood variables significantly affected the outcome variable. Specifically, according to the results of multivariate analysis, with every one-percentage increase in the Hispanic population in the census tract, the odds of quitting jobs would increase 2.38 (β = 0.868, p = 0.073). Neighborhood percentages of African American, poverty, and foreign born were not associated with mothers' work exits. We found that several individual level factors were associated with the likelihood of mother's work exit as well. African American mothers were more likely to quit their job, school, or training because of child care arrangement problems (β = 0.343, p = 0.098). Mother's household income level was also associated with the probability of child care-related work exit. Mothers with income between 50% and 99% of the FPL were more likely to quit a job because of the child care arrangement problems than mothers with income below 49% of the FPL (β = 0.366, p = 0.018); mothers with household income between 200% and 299% of the FPL and mothers with household income more than 300% of the FPL were less likely to quit job because of child care arrangement problems than mothers with household income below 49% of FPL (β = − 0.444, p = 0.026; β = − 0.658, p = 0.003). Mothers who worked longer hours were less likely to quit job because of child care arrangement problems (β = − 0.018, p = 0.014). Mothers who perceived that they would have emergency child care assistance were less likely to quit a job because of child care arrangement problems (β = − 0.270, p = 0.012). Finally, younger mothers were more likely to quit their jobs because of child care arrangement problems than older mothers (β = −0.051, p = 0.002).

p-Value

1383.972 1385.972 −0.112 0.037 0.142

173

Note: The table reports coefficient estimates, standard errors, and p-values. + p b 0.10. ⁎ p b 0.05. ⁎⁎ p b 0.01. ⁎⁎⁎ p b 0.001.

This analysis of the multi-city FFCW data represents the first multivariate examination of child care-related work exits of which we are aware. It extends previous literature on child care-related job terminations among small samples of low-income mothers by examining the issue among a large sample of mothers in varying economic circumstances in multiple cities. It also builds on previous studies by examining a variety of factors including neighborhood effects that may impact child care-related work exits. The results indicate that care-related work exits are common occurrences for mothers of toddlers in large urban areas. More than one in ten mothers in the FFCW sample that used non-parental child care reported quitting a job, school, or training due to child care arrangement problems within a two-year period. We found partial support for the expected effects of child care characteristics, neighborhood factors, and individual factors on the risk of child care-related work exits. In this study, changing child care arrangements, using multiple types of child care, being African American, and having household income between 50 and 99% of FPL were all risk factors for experiencing child care-related work exits. In contrast, mothers who had emergency child care assistance, incomes of more than 200% of the FPL, longer working hours, and were older were less likely to quit a job because of child care arrangement problems. The study found that mothers who had changed their child care arrangements since their child's first birthday were more likely to quit their jobs because of problems arranging or keeping a child care arrangement. This is consistent with previous studies reporting that stability of care is linked to the job stability of mothers (e.g., Hofferth & Collins, 2000). In addition, the study found that mothers who were using multiple child care arrangements were more likely to lose their jobs due to child care disruptions than mothers who were not using multiple child care arrangements. This is consistent with prior research which found that the use of multiple child care providers for a child was

174

M. Liu et al. / Children and Youth Services Review 46 (2014) 168–176

linked to an increased probability of experiencing child care-related work disruptions (Huston et al., 2002; Scott et al., 2005; Usdansky & Wolf, 2008). As the earlier research suggested, patchworks of care may represent the tenuous nature of care arrangements, or mothers may turn to patchwork arrangements when care is unreliable. Mothers may also experience confusion when they need to work with multiple providers (Scott et al., 2005). The findings demonstrate that in Hispanic neighborhoods, mothers were more likely to quit a job, school, or training due to child care arrangement problems. Several explanations for this phenomenon are plausible. First, the relative lack of child care institutions in Hispanic neighborhoods may be at least partially responsible. This explanation is consistent with social disorganization theory developed by Shaw and McKay (1942), which states that neighborhoods having higher rates of poverty and ethnic heterogeneity are deprived of important institutional resources such as child care centers and home-based care settings. According to this theory, services and institutions whose availability and quality vary across neighborhoods may have a significant impact on individual outcomes, such as the relative presence of child care centers (Ellen & Turner, 2003). Given that earlier studies have shown availability of child care was one of the most important factors related to maternal job stability (Han & Waldfogel, 2001; Hofferth & Collins, 2000), research that more clearly established the relative supply of child care institutions in neighborhoods with varying ethnic characteristics could clarify the relationships found here between neighborhood ethnic characteristics and child care-related work performance. It may also be the case that the quality of child care settings in Hispanic neighborhoods tends to be substandard, or at least that parents perceive it as such. Parents' assessments of quality of care are important factors that affect the overall stability of their children's child care arrangements and employment stability (Hofferth & Collins, 2000). Research has shown that parents who assess the quality of care for their children to be poor are more likely to leave their jobs than are parents who perceive their children's care to be good (Meyers, 1993). Anderson, Ramsburg, and Scott (2005) have presented some findings in this respect based on focus groups with parents in low-income neighborhoods, but additional research with larger data sets would be useful to test this proposition. The study found that African American mothers were more likely to quit a job, school, or training because of child care arrangement problems. This may be because the jobs that African American mothers find frequently require them to work at inconsistent hours, part-time, evening or early morning hours, split shifts, or weekend hours, and seldom include family-supportive policies such as child care assistance; therefore, they are more likely to experience child care disruptions due to work schedule (Liu & Anderson, 2012). Mothers' household income levels also were associated with the likelihood of experiencing child care-related job termination. Mothers with household income 50% to 99% of FPL were more likely to quit a job because of child care arrangement problems than mothers with household incomes either below 49% of the FPL or above 200% of the FPL. This is consistent with previous findings that due to the availability of direct subsidies, many low-income families have child care comparable in quality to that of high-income families, with children from moderate-income families in lower quality care (Hofferth & Collins, 2000). The availability of emergency child care assistance was a protective factor for quitting jobs due to child care arrangement problems. Emergency child care support may help mothers find reliable back providers when child care problems do occur, thus reducing work absences or job terminations (Usdansky & Wolf, 2008). The findings also indicated that mothers who worked longer average hours were less likely to quit a job due to child care arrangement problems. This is consistent with previous studies in that greater labor force attachment decreases the odds of an employment exit (Hofferth & Collins, 2000; Usdansky & Wolf, 2008). Similarly, older mothers may also have greater labor force

attachment and therefore are less likely to quit a job when child care arrangements fall through. 7. Implications This study found that the instability and multiplicity of child care arrangements are associated with the instability of maternal employment. The Child Care and Development Fund (CCDF) policy has been designed to assist low-income families, families receiving temporary public assistance, and those transitioning from public assistance in obtaining child care so they can work or attend training or education (National Child Care Information and Technical Assistance Center [NCCIC], 2005). The CCDF should add a specific focus on improving child care reliability. For example, researchers have suggested tiered reimbursement systems be used to provide higher subsidy payments to formal and informal child care providers who possess indicators of high quality child care (Liu & Anderson, 2010). States could expand these indicators to include measures of child care reliability in their assessment systems. Fore example, the availability of contingency plans for provider emergencies in relative care and family day care settings, as well as nontraditionalhour-care for center care settings, could be recognized in reimbursement policies. The findings that mothers in neighborhoods with higher Hispanic populations were more likely to quit job because of child care arrangement problems have important implications for child care and workforce retention policy. Special efforts are needed to promote access to qualified child care in neighborhoods with high Hispanic populations. This may first involve the development of greater child care supply in these areas. State child care programs and other governmental agencies can play important roles in this respect. They may provide more access to formal care in such neighborhoods by partnering with the nonprofit section in establishing more Head Start programs and related child care programs. Head Start has a geographic component through which the federal government can fund centers if need can be demonstrated in the community. State and city government agencies may also respond by partnering with the nonprofit and for profit sectors to establish more child care centers in poor or ethnic neighborhoods, through mechanisms such as providing more favorable funding policies or assisting with building access and development. When creating additional child care centers in such neighborhoods, particular attention should be paid to provide culturally sensitive and bi-lingual services, as well as to alleviate parental concerns regarding child care center safety issues. Developing strategies for recruiting bi-lingual staff and for cooperating with local ethnic community leaders and institutions also may be useful directions for state and local agencies. The study found that mothers who had household income between 50 and 99% of the FPL were more likely to quit a job when child care arrangement fall through than mothers with household income below 49% of the FPL and mothers with household income higher than 200% of the FPL. Previous studies have found that programs and policies designed to increase family's access to paid child care affected the type of care used by families (Crosby, Gennetian, & Huston, 2005). The relative generosity of state child care programs, including the extent to which state funding and policies promote child care availability, affordability, and health and safety for low-income families, also is positively associated with the use of center care (Liu, 2010; Robins, 2007; Tvedit, 2009). To the extent states wish to promote child care center use as a means for ensuring better developmental outcomes and support maternal employment, higher spending levels appear to be one useful policy mechanism. Give these findings, the income thresholds at which families remain eligible for subsidies also may be very important. Given that perception of emergent backup care availability was associated with fewer job terminations, the availability of backup care appears important to avoid losing pay or even jobs. Government and private supports for the development of emergency child care centers offering drop-in services may be one responsive strategy, and

M. Liu et al. / Children and Youth Services Review 46 (2014) 168–176

improving provision of child care at the work site also merits further experimentation. In addition, local community centers and churches can design initiatives to provide working mothers with drop-in child care services. Supporting social networks for parents also has gained some attention (Warner, 2006). State child care lead agencies can also play an important role by assisting mothers in locating qualified care. Outreach programs, operated either by child care agencies or by local nonprofits, could be very valuable in this respect. These programs can be used to educate mothers about the centers available in the area, as well as any emergency child care facilities. Especially in neighborhoods with higher Hispanic populations, outreach staff also can help mothers overcome concerns about involvement with unfamiliar service agencies. 8. Directions for future research Several limitations in this research should be noted. Because this is a cross-sectional study, conclusions regarding causal effects must be cautious. Moreover, our measure of child care problems refers to a discreet time period. Thus, we can say little about the degree to which certain mothers may be prone to repeated care related job terminations. In addition, the FFCW study precluded the analysis on parental care, a significant limitation given the prevalence of parental care of young children (Hofferth, Brayfield, Deich, & Holcomb, 1991) and high levels of instability in paternal care (Hofferth & Collins, 2000). This study also could not distinguish whether the work exits were due to child illnesses or provider unavailability, or other reasons. Nevertheless, our analysis contributes to the literature on work and family in important respects. We provide the first multivariate analysis of child care related job terminations using national data. Our findings confirm the widespread nature of such problems, which, previously have been analyzed mainly in the context of small samples of economically disadvantaged mothers. This again points to the need for broader child care assistance for mothers of young children. References Adams, G., & Rohacek, M. (2010). Child care instability: Definitions, context, and policy implications. Washington, D.C.: The Urban Institute (Retrieved from http://www. urban.org/UploadedPDF/412278-child-care-instability.pdf). Anderson, S., Liu, M., & Liao, M. (2012). Subsidized child care by grandparents: Profiles of caregivers in an emerging public service context. Journal of Aging and Women, 25, 242–259. Anderson, S. G., Ramsburg, D.M., & Scott, J. (2005). Illinois study of license-exempt child care: Final report. Washington, D.C.: U.S. Department of Health and Human Services. Administration on Children, Youth, and Families, Child Care Bureau (Retrieved from www.researchconnections.org/location/ccrca7350). Atkinson, A. M. (1991). Fathers’ participation in day care. Early Child Development and Care, 66, 115–126. Boushey, H. (2003). Who cares? The child care choices of working mothers. Washington, D. C.: Center for Economic and Policy Research (Retrieved from http://www.cepr.net/ documents/publications/child_care_2003.pdf). Bowen, G. L., & Neenan, P. A. (1993). Does subsidized child-care availability promote welfare independence of mothers on AFDC: An experimental analysis. Research on Social Work Practice, 3(4), 363–384. Bronfenbrenner, U. (1994). Ecological models of human development. In T. Husen, & T. N. Postlethwaite (Eds.), International encyclopedia of education (pp. 1643–1647) (2nd ed.). New York: Elsevier Science. Bub, K. L., & McCartney, K. (2004). On child care as a support for maternal employment wages and hours. Journal of Social Issues, 60, 819–834. Capizzano, J., Adams, G., & Sonenstein, F. (2000). Child care arrangements for children under five. Washington, D.C.: The Urban Institute (Retrieved from http://www. urban.org/url.cfm?ID=309438). Center for Research on Child Wellbeing (CRCW) (2008). Introduction to the fragile families public use data. Retrieved from http://www.fragilefamilies.princeton.edu/ documentation/core/4waves_ff_public.pdf Crosby, D., Gennetian, L., & Huston, A. (2005). Child care assistance policies can affect the use of center-based care for children in low-income families. Applied Developmental Science, 9, 86–106. Dodson, L., & Bravo, E. (2005). When there is no time or money: Work, family and community lives of low-income families. In J. Heymann, & C. Beem (Eds.), Unfinished work: Building equality and democracy in an era of working families (pp. 122–155). New York, NY: New Press.

175

Eichman, C., & Hofferth, S. (1993). Family strategies for managing work and family life. Paper presented at the meeting of the Population Association of America, Cincinnati, OH. Ellen, I. G., & Turner, M.A. (2003). Do neighborhoods matter and why? In J. Goering, & J.D. Feins (Eds.), Choosing a better life? Evaluating the moving to opportunity social experiment (pp. 313–318). Washington, DC: Urban Institute Press. Folk, K. F., & Yi, Y. (1994). Piecing together child care with multiple arrangements: Crazy quilt or preferred pattern for employed parents of preschool children? Journal of Marriage and the Family, 56, 669–680. Forry, N. D., & Hofferth, S. L. (2011). Maintaining work: The influence of child care subsidies on child-care related work disruptions. Journal of Family Issues, 32(3), 346–368. Fuller, B., Kagan, S. L., Caspary, G. L., & Gauthier, C. A. (2002). Welfare reform and child care options for low-income families. The Future of Children, 12, 96–119. Gordon, R. A., Kaestner, R., & Korenman, S. (2008). Child care and work absences: Tradeoffs by type of care. Journal of Marriage and the Family, 70, 239–254. Grzywacz, J. G., & Butler, A. B. (2005). The impact of job characteristics on work-to-family facilitation: Testing a theory and distinguishing a construct. Journal of Occupational Health Psychology, 10, 97–109. Han, W. (2004). Nonstandard work schedules and child care decisions: Evidence from the NICHD study of early child care. Early Childhood Research Quarterly, 19, 231–256. Han, W., & Waldfogel, J. (2001). Parental leave: The impact of recent legislation on parents' leave-taking. Demography, 40(1), 191–200. Harknett, K. (2006). The relationship between private safety nets and economic outcomes among single mothers. Journal of Marriage and Family, 68, 172–191. Henly, J. R., & Lyons, S. (2000). The negotiation of child care and employment demands among low-income parents. Journal of Social Issues, 56, 683–706. Hofferth, S., Brayfield, A., Deich, S., & Holcomb, P. (1991). National Child Care Survey 1990. Washington D.C..: The Urban Institute. Hofferth, S., & Collins, N. (2000). Child care and employment turnover. Population Research and Policy Review, 19(4), 357–395. Hofferth, S. L., Shauman, K. A., Henke, R. R., & West, J. (1998). Characteristics of children's early care and education programs: Data from the 1995 National Household Education Survey. National Center for Education Statistics. Washington, DC: U.S. Department of Education (NCES 98-128). Huston, A.C., Chang, Y. E., & Gennetian, L. (2002). Family and individual predictors of child care use by low-income families in different policy contexts. Early Childhood Research Quarterly, 17, 441–469. Kimmel, J. (2006). Child care, female employment and economic growth, community development. Journal of the Community Development Society, 37(2), 71–85. Knox, V., London, A. S., Scott, E. K., & Blank, S. (2003). Welfare reform, work, and child care: The role of informal care in the lives of low-income women and children. New York: MDRC Retrieved from http://www.mdrc.org/sites/default/files/ policybrief_40.pdf. Liu, M. (2010). Factors influencing low-income working mother’s child care arrangements under different neighborhood and state policy contexts. Urbana-Champaign: Dissertation, University of Illinois. Liu, M., & Anderson, S. (2010). Understanding caregiving patterns, motivations, and resource needs of subsidized FFN child care providers. Child Welfare, 89, 99–119. Liu, M., & Anderson, S. (2012). Neighborhood effects on working mothers' child care arrangements. Children and Youth Services Review, 34(4), 740–747. Mason, K. O., & Kuhlthau, K. (1992). The perceived impact of child care costs on women's labor supply and fertility. Demography, 29(4), 523–543. Meyers, M. (1993). Child care in JOBS employment and training programs: What difference does quality make? Journal of Marriage and the Family, 51, 593–603. Meyers, M. (1994). Cracks in the seams: Durability of child care in JOBS welfare to work programs. Syracuse, NY: University of Syracuse. Miller, C. (2005). Stability and change in child care and employment: Evidence from three states. MDRC working paper no. 20. New York: MDRC Retrieved from http://files. eric.ed.gov/fulltext/ED484620.pdf. Mulligan, G. M., Brimall, D., West, J., & Chapman, C. (2005). Child care and early education arrangements of infants, toddlers and preschoolers: 2001. Washington, DC: U.S. Government Printing Office. National Child Care Information and Technical Assistance Center (2005). Federal and state funding for early care and education. Retrieved from http://nccic.acf.hhs.gov/ categories/index.cfm?categoryId=3 Perry-Jenkins, M. (2005). Work in the working class: Challenges facing families. In S. M. Bianchi, L. M. Casper, & R. B. King (Eds.), Work, family, health, and well-being (pp. 453–472). Mahwah, NJ: Lawrence Erlbaum. Pew Research: Social & Demographic Trends (2013). On pay gap, millennial women near parity — For now. Retrieved from http://www.pewsocialtrends.org/2013/12/11/onpay-gap-millennial-women-near-parity-for-now/2/ Presser, H. B., & Baldwin, W. (1980). Child care as a constraint on employment: Prevalence, correlates, and bearing on the work and fertility nexus. American Journal of Sociology, 85(5), 1202–1213. Pungello, E. P., & Kurtz-Costes, B. (1999). Why and how working women choose child care: A review with a focus on infancy. Developmental Review, 19, 31–96. Reichman, N. E., Teitler, J. O., Garfinkel, I., & McLanahan, S. S. (2001). Fragile families: Sample and design. Children and Youth Services Review, 23(4/5), 303–326. Robins, P. (2007). Welfare reform and child care: Evidence from ten experimental welfare-to-work programs. Evaluation Review, 31, 440–468. Scott, E. K., London, A. S., & Hurst, A. (2005). Instability in patchworks of child care when moving from welfare to work. Journal of Marriage and Family, 67, 370–386. Shaw, C. R., & McKay, H. D. (1942). Juvenile delinquency and urban areas. Chicago, IL: University of Chicago Press. Shellenback, K. (2004). Child care and parent productivity: Making the business case. Ithaca, NY: Cornell University Department of City and Regional Planning, Linking Economic

176

M. Liu et al. / Children and Youth Services Review 46 (2014) 168–176

Development and Child Care Research Project (Retrieved from http://government. cce.cornell.edu/doc/pdf/ChildCareParentProductivity.pdf). Tvedit, K. (2009). The child care self-sufficiency scale: Measuring child care funding and policy generosity across states. Unpublished dissertation, Portland State University. Usdansky, M. L., & Wolf, D. A. (2008). When child care breaks down: Mother's experiences with child care problems and resulting missed work. Journal of Family Issues, 29(9), 1185–1210. Warner, E. (2006). Overview: Articulating the economic importance of child care for community development. Journal of the Community Development Society, 37(2), 1–6.

Wayne, J. H., Grzywacz, J. G., Carlson, D. S., & Kacmar, M. K. (2007). Work–family facilitation: A theoretical explanation and model of primary antecedents and consequences. Human Resource Management Review, 17, 63–76. Weber, R. B., Grobe, D., Davis, E. E., Kreader, J. L., & Pratt, H. C. (2007). Child care market rate survey practices of states, territories, and tribes. Research connections: Child care and early education. Retrieved from http://www.researchconnections.org/ childcare/resources/12266