Research Publication

Work Hours, Family Composition, and Employment Predict Use of Child Care for Low-Income Latino Infants and Toddlers

Download brief as a PDF


Households with infants and toddlers are increasingly a focus of early care and education (ECE)a policy.1 High-quality care is important for children during their early years when rapid brain development is occurring,2 so lagging availability and access to affordable ECE for these age groups, relative to preschool children, is cause for concern.3 In this brief, we examine the child, household, and community characteristics that predict ECE participation for Latinob infants and toddlers living in low-income households. These children have among the lowest utilization rates of nonparental care.4

Many low-income parents with young children face challenges as they look for a job or work while trying to secure care for their infant and toddler-age children.5 Child care may be especially difficult for Hispanic families to access because of its cost. Latino parents of young children report cost as the primary reason for difficulty finding child care, rather than other factors such as location or availability.6 Recent national data show that although some low-income Latino families have access to free ECE options (such as Head Start), about 3 in 10 families using care face high out-of-pocket costs.7 This means that for many low-income Latino parents, accessing nonparental care may be out of reach financially.5,7

In addition to cost and availability of care, several characteristics of Hispanic households are also known to be associated with whether nonparental care is used. For example, among low-income Hispanic households, national data show that immigrant households tend to have higher rates of paid provider care provided by someone not previously known to them (unfamiliar), as compared with nonimmigrant Hispanic households.4 This may signal that social networks facilitate ECE access and sources of nonparental care differently across nonimmigrant and immigrant Hispanic households.8,9 This may also signal that Latino immigrant networks may be limited in size and reach relative to those of U.S.-born Hispanic families. This, in turn, may limit immigrant Hispanic families’ awareness of what ECE options, including publicly subsidized options, are available in their communities.

Low-income Latino households also report searching differently for ECE relative to other racial/ethnic groups. For example, research shows that low-income Hispanic families with children under age 5 consider fewer ECE providers during their search for child care than their low-income Black and White counterparts.10 However, low-income Latino families report similar reasons to their peers for using care—mainly, to support their work or their children’s development.10 They also are less likely to have relatives nearby who could provide unpaid child care, compared to their low-income Black and White peers.11 Taken together, as with all families, various factors influence the selection and availability of ECE for low-income Hispanic families, including those with infants and toddlers.

Using data from the 2012 National Survey of Early Care and Education (NSECE), we examine how child, household, and community characteristics relate to low-income Hispanic families’ use of infant and toddler care (as illustrated in Figure 1). We explore a range of child-level characteristics, including number and ages of children and whether there are children with special needs in the household. At the household level, we examine family structure and household composition (including the presence of grandparents or other relatives), parents’ work status, and other sociodemographic characteristics that shape the resources they may have to secure child care arrangements (e.g., income, nativity status of the household, and the extent to which English is spoken regularly at home). For community context, we include two broad indicators of the environment in which families live—urbanicity and poverty density—because of their implications for influencing the search process and supply of care.c

Key Findings

Our findings suggest that the use of nonparental care for infants and toddlers in low-income Hispanic households depends largely on whether the child lives in a single- or two-parent household, how many parents are employed, and the number of hours parents work.

Low-income Latino infants and toddlers are more likely to experience nonparental care if both of their parents work and when their parents work a greater number of hours.

  • Infants and toddlers in households with two working parents are more than three times as likely to use ECE than households in which there is only one working parent.
  • The likelihood of using ECE also increases with parents’ number of work hours; each additional hour of work increases the odds of nonparental care by 2 percent.

Use of nonparental care among low-income Hispanic households with infants and toddlers also depends on who else lives in, or nearby, the household.

  • Infants and toddlers are more than four times as likely to be in nonparental care arrangements if they live with one parent.
  • Nonparental care is 1.75 times more likely to occur when a grandparent resides in the household or a relative lives nearby who is available to provide care. It is less likely to occur when there are preschool and school-age children living in the home.

Notably, in our study, several other characteristics of children, households, and communities (e.g., the child’s age and special needs status; limited English proficiency; nonstandard parental work hours; household income-to- poverty ratio; receipt of cash assistance; and community poverty density and urbanicity) do not significantly relate to parents’ decisions to place their infants and toddlers in nonparental care.

Conceptual Framework

Figure 1 shows the variables that we examine in this study of low-income households with Hispanic infants and toddlers. The organizing framework depicted in Figure 1 also helps place our findings in the broader set of factors that shape families’ ECE access and utilization, as parents balance multiple priorities, needs, opportunities and constraints.8,12

This framework builds on two areas of conceptual work that have helped illuminate the complex and dynamic processes that drive ECE use. Specifically, we map the types of child, household, and community characteristics—such as who lives in the household, community poverty density, and urbanicity—identified by Weber (2011)13 and others as influential in child care decision-making. This framework reflects a comprehensive definition of ECE access, which states that “… parents, with reasonable effort and affordability, can enroll their child in an arrangement that supports the child’s development and meets the parents’ needs.”12 Notably, this definition recognizes that ECE access is multidimensional and acknowledges the role that supply, cost, preferences, and parental needs play in the use of and access to ECE.


Predictors of ECE Use for Low-Income Latino Infants and Toddlers

Using logistic regression, we examined the relative contribution of multiple child, household, and community characteristics in predicting the likelihood that a Hispanic infant or toddler will be in nonparental care. The odds ratios and confidence intervals produced by this analysis are reported in Table 1 and shown in Figure 2. Significant odds ratios less than 1.0 indicate the predictor is associated with a lower probability of the child experiencing the outcome (i.e., nonparental care); significant odds ratios greater than 1.0 indicate predictors are associated with a higher probability of the outcome.

Many of the predictors in these models are dichotomous (coded as 1 for the presence of the characteristic and as 0 for the absence of the characteristic); this means that the odds ratio reflects how much more (or less) likely a child with the characteristic is to be in care than a child who does not have the characteristic. See Data Source and Methodology for additional information on our sample, methods, and variables used in this study.

As shown in Table 1, we find that the likelihood of care varies primarily by who else is living in the home and parental employment. More specifically, the odds that a child will be in care are 4.55 times higher if they live with a single parent, versus two parents, and 1.75 times higher if they have a co-resident grandparent or a relative living nearby who can provide care, as opposed to having neither of these available. On the other hand, the odds of being in care are lower when other preschool- and school-aged children live in the home.

The likelihood of ECE use is also 3.38 times higher when a child has two working parents, versus one working parent. Two additional predictors that approached statistical significance (p < .10) are the presence of children ages 13–17, which increases the odds of using ECE, and whether the household is an immigrant family, which decreases the odds of nonparental care.

Among the child characteristics examined, neither the child’s age nor whether they had special needs, were significantly associated with the likelihood of using nonparental care. Several economic and employment variables were also not significantly associated with the use of any nonparental care for infants and toddlers, including no employed parent at home (versus one), parent working nonstandard hours, and limited household English proficiency (an indicator of access to the labor market and institutions). In addition, household income-to-poverty ratio and receipt of cash assistance were not significant contributors. Lastly, neither of the community level factors examined—community poverty density and rural/suburban—were significantly associated with the use of ECE by Hispanic households with infants and toddlers (see Table 1 and Figure 2).

How Do Different Combinations of Household Characteristics Change the Probability that Low-Income Hispanic Infants and Toddlers will be in ECE?

One strength of the current analysis is its consideration of multiple child, household, and community characteristics simultaneously, given that these factors co-occur in multiple ways in the lives of families. To further illustrate our key findings, we estimate the probability of being in child care for three hypothetical 18-month-old male, Latino toddlers with different combinations of background characteristics.

Across the three children, we vary the factors identified as statistically significant (or approaching significance) in predicting the use of child care, while holding the other nonsignificant factors constant. The one caveat to this approach is that even though household income (relative to poverty) is not a significant predictor of child care use, it is so closely tied to family structure, parental employment status and work hours, that we vary it accordingly across the three cases.

The first child, Gabriel, has all of the characteristics associated with a low likelihood of ECE use. He lives with both of his parents, along with his preschool-aged sister and school-aged brother. He has neither a co-resident grandparent nor a relative living nearby who can provide child care. At the moment, only Gabriel’s father is working (approximately 30 hours per week) and the family lives in deep poverty, with a household income level at 50 percent of the poverty threshold adjusted for household size. A toddler with Gabriel’s characteristics has an 8 percent probability of being in nonparental care.

The second child, Alejandro, shares all the same characteristics as Gabriel, except both of his parents are working a total of 60 hours per week and their household is considered low-income but hovers at 1.5 times the poverty threshold adjusted for household size. As shown, these two differences more than quadruples the probability of being in nonparental care for a toddler like Alejandro (44 percent) relative to his peer Gabriel (8 percent).

The third child, Mateo, has all of the characteristics associated with a high likelihood of ECE use. He lives with his mother, who is U.S.-born and a single-parent. Mateo’s grandmother and teen brother also live with them. His mother is currently working approximately 40 hours per week and their household income is at the federal poverty threshold, adjusted for household size. The probability that a young child like Mateo is in nonparental care is 84 percent.

Note: For all three children, the following values were assigned for variables not significantly associated with utilization – sample mean child age of 18 months, household where English is spoken, at least one parent works some nonstandard hours, the household does not receive public assistance, and the household is in an urban, high poverty community. The assigned values were selected because they represented the largest groups within the sample. The one exception to this approach is that, even though it was not a significant predictor, the household income variable was adjusted across the three sample cases to coincide with the number of working parents and total work hours.


In this study, we found that the primary drivers of low-income Hispanic families’ use of ECE for infants and toddlers are who lives in the home (or nearby) and parents’ level of work activity (i.e., number of parents employed and total work hours). Specifically, we find that family structure is the strongest individual predictor of nonparental care, with care being at least four times more likely if infants and toddlers live with only one parent. In two-parent homes, the odds of using ECE are more than three times higher if both parents are working than if only one parent works. And more generally across families, each hour of parental work increases the odds that low-income Latino infants and toddlers will be in some form of regular child care.

We found that other household factors matter as well. Nonparental care is more likely for low-income Latino infants and toddlers who have a co-resident grandparent or relative living nearby who is available to provide care. On the other hand, care is less likely when there are other preschool- or school-age children (younger than 13) living in the home.

Together, this set of findings is consistent with theory on ECE access and our conceptual framework, and shows how children’s care experiences are shaped by the needs of their household. Given the high cost of infant and toddler care, parents may engage in a financial tradeoff regarding decisions about parental employment and the need to arrange care for their young children. Support for this explanation can be seen in the lower likelihood of ECE use when a second parent is available (especially if that parent is not working) and when there are multiple young children in the home who require care. In addition, although prior research finds that low-income Hispanic parents are less likely than their White and Black peers to have a relative nearby to provide low- or no-cost care,11 our study shows that when relatives or grandparents are available to Hispanic families, the likelihood of their infants and toddlers being in nonparental care increases.

Notably, once the household and employment factors mentioned above are taken into account, several other characteristics of children, households, and communities presumed to influence families’ ECE access do not significantly predict low-income Hispanic parents’ decisions to place infants and toddlers in ECE arrangements. For example, the likelihood of being in care did not vary by child age or special needs status. Sample sizes limited our ability to examine whether these characteristics interact in more complex ways with other factors (e.g., disability and immigrant status) to affect ECE use. These types of questions could be explored further with larger samples of Hispanic infants and toddlers in low-income households.

Although prior work has identified parents’ limited English abilities as a challenge for navigating certain aspects of ECE access—obtaining child care subsidies, for example14—we did not find that English use was associated with whether families used any nonparental care. It may be that English proficiency is more consequential for the ECE search process and the particular types of care options parents can access. Similarly, we did not find that community characteristics, such as urbanicity and poverty density, affect the likelihood of using nonparental care for low-income Hispanic infants and toddlers. Yet, these factors could potentially affect ECE supply in other ways. For example, community factors may matter more for the types of care arrangements families select (e.g., center vs. home-based; public vs. private) than for whether they use any care at all (which was the focus of the current study). Additional research on how child, household, community, and policy factors intersect to shape multiple aspects of ECE use for very young children in low-income Latino communities will help inform methods by which to ensure this group gains access to high-quality, affordable child care that meets families’ needs.

Our finding that income levels and use of public assistance are unrelated to use of nonparental care suggests there may be opportunities to expand access to child care subsidies among low-income working Hispanic families with infants and toddlers. Doing so may require reducing administrative burden for seeking subsidies, which has been identified as a barrier for low income Hispanic households.14 Looking ahead, the precarious task of locating affordable care for infants and toddlers is likely to be further exacerbated by the economic and personal toll of the COVID-19 pandemic on low-income populations. As we have shown, parental employment and work hours are key drivers of the use of nonparental care. Any reductions in the supply of affordable care for infants and toddlers will be a significant burden for low-income Hispanic households seeking to continue or return to work.

Given relatively low rates of ECE use by low-income Latino infants and toddlers, and the fact that care options for this age group are less available and more expensive than for older children, continued focus on this population is needed in research and policy discussions. This emphasis on ECE access to support young Latino children’s development and school readiness, and for facilitating parental employment, is essential.


a In this brief, we use the terms early care and education (ECE), child care, and nonparental care interchangeably to capture the wide range of home- and center- based settings where young children spend time with caregivers when parents are unavailable. Although ECE is sometimes used in the literature to refer specifically to preschool-aged, formal arrangements, we use the term more broadly given that children’s experiences of care and educational opportunities are not limited to a particular age group or type of setting.

b In this brief, we use the terms “Hispanic” and “Latino” interchangeably.

c We do not directly examine child care market factors, including the supply of care in local communities, which can shape the availability and cost of care.

d For the purposes of this brief, we focused on the experiences of low-income Hispanic households; multi-ethnic households which contain both Hispanic and non-Hispanic members are excluded from the analysis.

e Tests evaluating multicollinearity among the predictors did not reveal notable issues. In addition, sensitivity analyses that included potentially overlapping indicators (e.g., immigrant status and limited English skills) one at a time (rather than including both in the model simultaneously) did not yield different results.

Suggested citation: Mendez, J., Crosby, D., & Siskind, D. (2020). Work Hours, Family Composition, and Employment Predict Use of Child Care for Low-Income Latino Infants and Toddlers. Report 2020-04. Bethesda, MD: National Research Center on Hispanic Children & Families. Retrieved from

Data Source and Methodology

The 2012 National Survey of Early Care and Education (NSECE) is a set of four nationally representative surveys designed to describe the early care and education landscape of the United States from the supply and demand perspectives. The data presented in this brief are drawn from the NSECE public use household survey, which collects parent-reported information about child care utilization and costs for a nationally representative sample of households with children under age 13. The NSECE oversampled in economically disadvantaged communities, resulting in large numbers of low-income families.

This brief focuses on children younger than age 3 living in low-income households (defined as those with annual income below 200 percent of the federal poverty threshold) where all members are identified by the respondent as being Hispanic (regardless of racial classification).d In the NSECE, these selection criteria apply to 1,122 children, 1,037 children (or 92 percent) of whom had complete data on the demographic and child care utilization variables included in the analysis.

Most analytic variables had less than 1 percent of missing data, with the exception of single-parent status, presence of a teenager in the household, receipt of public assistance, and parent nonstandard work hours, each of which had less than 4 percent missingness. Tests of mean differences between those included and excluded from the analytic sample indicated that children with missing data were more likely to be from an immigrant household and to live in an urban community, and less likely to have two working parents in the home (with fewer total parental work hours), to be in a household receiving public assistance, and to live in a community with moderate poverty density. Table 2 provides descriptive information for this sample. A logistic regression model is used to examine which child, household, and community characteristics are associated with a higher or lower likelihood that a low-income Hispanic infant and toddler child will be in any nonparental care. All analyses were conducted in STATA 16.0 and were weighted to be nationally representative of children living in U.S. households in 2012.e


Nonparental care. Parents reported on all of the individuals or organizations that care for the child on a regular basis (i.e., five or more hours per week), both in the week prior to the survey, as well as any regular providers not used in the past week. Children were identified as being in nonparental care if they had at least one nonparental care arrangement for five or more hours per week. This brief uses nonparental care, child care and ECE as interchangeable terms inclusive of a range of nonparental providers, including home- and center-based care providers.

Child disability/special needs status indicates children identified by parents as having a physical, emotional, developmental, or behavioral condition that has an impact on the care they need.

Household composition. Several variables were used to capture the types of individuals who co-reside with the child in the home. These included indicators for the presence of one (as opposed to two) biological/step/adoptive parent of the child, and the presence of other children younger than 5, children ages 5 to 12, and/or teenagers ages 13 to 17. In addition, we created an indicator to identify households with a grandparent in the home and/or a relative living nearby who could provide child care.

Immigrant household identifies households where at least one member is foreign-born (versus the comparison group of nonimmigrant households, where all members are U.S.-born).

Limited English household identifies those households in which English is not regularly spoken.

Parent work status and number of hours. To capture aspects of parents’ employment, we include indicators of whether no parent, one parent, or two or more parents in the household reported work hours in the week prior to the survey. We also include a continuous variable of the total hours of work reported for the child’s parents living in the household.

Nonstandard work hours are identified by an indicator of whether any of the child’s parents worked during times beyond 8 a.m. to 6 p.m., Monday through Friday.

Household income-to-poverty ratio represents the ratio of the household’s annual income to the federal poverty threshold for 2011.

Household receipt of public assistance identifies households that report receiving public assistance or welfare payments.

Community poverty density is represented by three indicators of low, moderate and high poverty density that identify whether the household is located in a community where less than 13.9 percent (low), 13.9 percent to 20 percent (moderate), or more than 20 percent (high) of the population has incomes below the federal poverty threshold.

Community urbanicity is represented by three indicators of whether the household is in a community classified as: 1) urban, with high urban density (.85 to 1.00 ratio of urban to total population), 2) suburban, with moderate urban density (.30 to .84 ratio), or 3) rural, with low urban density (.29 or less ratio).

Tables 1 and 2


1 Henley, J., & Adams, G. (2018). Increasing Access to Quality Child Care for Four Priority Populations: Challenges and Opportunities with CCDBG Reauthorization. Washington, DC: Urban Institute. Retrieved from

2 National Research Council. (2015). Transforming the Workforce for Children Birth through Age 8: A Unifying Foundation. Washington, DC: The National Academies Press. Retrieved from

3 Workman, S., & Jessen-Howard, S. (2018). Understanding the True Cost of Child Care for Infants and Toddlers. Washington, DC: Center for American Progress. Retrieved from

4 Crosby, D., Mendez, J., Guzman, L., & López, M. (2016). Hispanic Children’s Participation in Early Care and Education: Type of Care by Household Nativity Status, Race/Ethnicity, and Child Age. Bethesda, MD: National Research Center on Hispanic Children & Families. Retrieved from

5 Schochet, L. (2019). The Child Care Crisis is Keeping Women Out of the Workforce. Washington, DC: Center for American Progress. Retrieved from

6 Corcoran, L., & Steinley, K. (2017). Early Childhood Program Participation, Results from the National Household Education Surveys Program of 2016 (NCES 2017-101 (NCES 2017-101). Washington, DC: National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Education. Retrieved from

7 Crosby, D., Mendez, J., & Barnes, A. (2019). Child Care Affordability is Out of Reach for Many Low-Income Hispanic Households. Bethesda, MD: National Research Center on Hispanic Children & Families. Retrieved from

8 Mendez, J. L., Crosby, D. A., & Siskind, D. (2018). Access to Early Care and Education for Low-Income Hispanic Children and Families: A Research Synthesis. Bethesda, MD: National Research Center on Hispanic Children & Families. Retrieved from

9 Vesely, C. K., Ewaida, M., & Kearney, K. B. (2013). Capitalizing on early childhood education: Low-income immigrant mothers’ use of early childhood education to build human, social, and navigational capital. Early Education & Development, 24(5), 744-765.

10 Mendez, J. L., & Crosby, D. A. (2018). Why and How do Low-Income Hispanic Families Search for Early Care and Education (ECE)? Bethesda, MD: National Center for Research on Hispanic Children & Families. Retrieved from

11 Guzman, L., Hickman, S., Turner, K., & Gennetian, L. (2016). Hispanic Children’s Participation in Early Care and Education: Parents’ Perceptions of Care Arrangements, and Relatives’ Availability to Provide Care. Bethesda, MD: National Research Center on Hispanic Children & Families. Retrieved from

12 Friese, S., Lin, V., Forry, N., & Tout, K. (2017). Defining and Measuring Access to High Quality Early Care and Education: A Guidebook for Policymakers and Researchers. Washington, DC: Office of Planning, Research & Evaluation. Retrieved from

13 Weber, R. B. (2011). Understanding Parents’ Child Care Decision-Making: A Foundation for Child Care Policy Making. Washington, DC: Office of Planning, Research and Evaluation. Retrieved from

14 Hill, Z., Gennetian, L.A., & Mendez, J. (2019). A descriptive profile of state Child Care and Development Fund policies in states with high populations of low-income Hispanic children. Early Childhood Research Quarterly, 47, 111-123. Retrieved from


The authors would like to thank the Steering Committee of the National Research Center on Hispanic Children & Families, Pamela Joshi, and staff from OPRE for their feedback on earlier drafts of this brief, as well as Sara Dean for her research assistance at multiple stages of this project.

Editor: Rhonda Smith and Brent Franklin

Designer: Catherine Nichols

About this series

This brief is part of an ongoing series aimed at better understanding the early care and education experiences of Latino children. It uses data from the National Survey of Early Care and Education (NSECE)—a set of four integrated, nationally representative surveys that describe the ECE landscape in the United States. Other briefs in this series include:

These publications and forthcoming briefs in the series
can be found in our Research & Resources library.

Related Research