Skip to main content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.


The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Child Care Arrangements in Urban and Rural Areas

Publication Date

Kendall Swenson

Office of the Assistance Secretary for Planning and Evaluation (ASPE)
U.S. Department of Health and Human Services (HHS)


This paper compares non-parental care arrangements of pre-school age children in urban and rural areas of the United States using data from the 2005 National Household Education Survey (NHES), Early Childhood Program Participation Survey (ECPP).  Data from the NHES show that among preschool-age children, those in rural areas are about as likely as those in urban areas to receive care from someone other than their parents on a weekly basis.  The NHES data also show that when rural children participate in non-parental care they are more likely than urban children to receive this care from relatives and are less likely to receive care in center programs.  Additionally, rural children are in families that, on average, made fewer out-of-pocket contributions toward the cost of their care.  However, it is difficult to interpret these differences since the cost of living in these areas may differ.



The author is grateful for the thoughtful feedback on earlier drafts of this paper by Barbara Broman, Nikki Forry, Susan Hauan, Amy Madigan, Melissa Pardue, Kristin Smith, and Bobbie Webber, and would like to thank Jana Liebermann for her assistance in the library at the U.S. Department of Health and Human Services. The views expressed in this paper are solely those of the author and do not represent the official position of the U.S. Department of Health and Human Services.


Recent policy discussions involving ways to improve the well-being of young children and their families have placed increased attention on the importance of child care and early education programs.  Research suggests that, although parental influences are important, the quality of non-parental care arrangements has long lasting effects on child development (Belsky, Vandell, Burchinal, Clarke-Stewart, McCartney, & Owen, 2007; Vandell & Wolfe, 2000; Zaslow, Halle, Guzman, Lavelle, Dombrowski, Berry, & Dent, 2006). Research also suggests that access to child care may be a factor in the employment and earning patterns of families, especially those with limited incomes and those transitioning from government assistance programs into the workforce (Schaefer, Kreader, & Collins, 2006).

While the body of literature regarding the relationship between non-parental care and child development has increased considerably over recent years, it is still a growing field with remaining knowledge gaps.  One area where additional research could be beneficial is analysis on child care arrangements and early education programs in rural areas.  As discussed later in the paper, families residing in rural areas, on average, have smaller household incomes and live in less densely populated areas than urban families.  These differences could potentially influence the availability of certain types of non-parental care in rural areas if child care providers are unable to attract enough families with the desire, transportation, and financial resources to participate in them.  Lower population densities might also be related to the quality of some care arrangements if rural child care providers are unable to offer high enough salaries to attract more highly qualified caregivers.

Using data from the National Household Education Survey (NHES), this paper furthers our understanding of child care and geography by comparing and contrasting the non-parental care arrangements of children living in urban and rural areas of the United States.

Literature Review

This section reviews the results of several research studies. They were chosen because they examined differences in urban and rural child care arrangements from a national perspective, which makes their results more comparable to the findings from the NHES than studies that only examined populations from specific localities or regions.  Comparisons across the studies presented in this section, though important, should be done cautiously because they utilized different research designs.  For example, the studies examined different age groupings of children, defined urban and rural areas differently, and categorized types of care settings differently.  In addition, some of the studies presented in this section examined all pre-school age children, while others focused on children with working mothers.

Two of the most thorough studies examining child care arrangements were completed in the early 1990s; the National Child Care Survey (Hofferth, Brayfield, Deich, & Holcomb, 1991) and the Profile of Child Care Settings (Kisker, Hofferth, Phillips, & Farquher, 1991).  The National Child Care Survey (NCCS) examined a nationally representative sample of U.S. families with children and produced an extensive set of descriptive findings, including a series of tables on the child care arrangements of pre-school age children under age five.  The 1990 Profile of Child Care Settings Study examined a nationally representative sample of center directors and regulated home-based care providers.

Several additional studies have compared child care in urban and rural areas.  Two of these used data from the Survey of Income and Program Participation (SIPP), which included a national sampling design and a series of questions regarding the child care arrangements of children.  The two studies using the SIPP cited in this paper examined the child care arrangements of children under age 5 during the time that their mothers worked (Casper, 1996; Smith, 2006).  In addition to the SIPP, a study by Grace et al. (2006) examined a national sample of children between 6 and 22 months of age using data from the Early Childhood Longitudinal Study (ECLS-B).  Unlike the two studies analyzing SIPP data, this study examined the care arrangements of all children, including those without employed mothers.  Another key difference between this study and the other studies reviewed in this section is that the population sampled for the ELCS-B included children between 6 and 22 months of age, whereas several of the other studies examined children age 0 to 4.  Finally, a study by Swenson (2007) compared the characteristics and caseload sizes of urban and rural children served by the Child Care and Development Fund (CCDF) using administrative records.

The literature has consistently shown that non-parental care is common for both urban and rural children, but it has not shown a clear pattern as to whether urban or rural children are more likely to receive this care.  Early findings from the NCCS showed that rural children under age 5 with working mothers were less likely to receive care from non-parental sources compared to those in urban areas (Hofferth et al., 1991).  However, later studies using SIPP data collected in 1993 and 2002 did not show substantial differences in participation in non-parental care when comparing children in metropolitan and non-metropolitan areas (Casper, 1996; Smith, 2006).  In contrast, the rural children sampled for the ECLS-B between 6 and 22 months of age were more likely than non-rural children to be cared for in regular non-parental arrangements (Grace et al., 2006).

Researchers of child care and human services policies often argue that center-based child care is less prevalent in rural areas.  Previous research has provided some evidence for this argument, although not all studies have shown this pattern.  Early findings from the NCCS showed that rural children under age 5 with working mothers were less likely to be cared for in center arrangements than urban and suburban children (Hofferth et al., 1991).  Swenson (2007) also showed lower participation in center-based care among rural children subsidized by the CCDF compared to similar urban children.  Additional unpublished tabulations by the author also showed that participation in center-based care among subsidized non-metropolitan children was correlated with the amount of urban influence associated with their resident counties. In other words, participation in center-based care was more common in non-metropolitan counties containing large towns and in counties adjacent to metropolitan counties than other non-metropolitan counties.

In contrast, findings from the SIPP data collected in 2002 did not show large differences between metropolitan and non-metropolitan children under age 5 in participation in similar arrangements referred to  as “organized care” (Smith, 2006).[1]  Similarly, findings from the ECLS-B did not show statistical differences in the use of center-based care between urban and rural children (Grace et al., 2006). However, it is difficult to compare the findings from the ELCS-B with the other studies reviewed above because center-based care is less common among the age group sampled by the ECLS-B (between 6 and 22 months) than among older pre-school age children. 

For children participating in non-parental care, the literature does not show a clear pattern as to whether the amount of time they spend in care each week is the same in urban and rural areas.  A study using the NCCS showed that rural pre-school age children with employed mothers were in care for less hours per week than urban children (Hofferth et al., 1991), while a study using SIPP data collected in 2002 (Smith, 2006) showed rural children being in care for slightly more hours per week than urban children. Another study that examined the ELCS-B did not find substantial urban/rural differences in weekly hours in care for pre-school age children (Grace et al., 2006).

The number of children in care per adult provider is sometimes used when describing the environments of care arrangements.  The literature is limited in its showing of urban and rural differences concerning this topic.  As showed by Kisker, Hofferth, Phillips, & Farquhar (1991), the 1990 Profile of Child Care Settings found that, compared to urban areas, the average number of children enrolled per setting in rural areas was statistically smaller for center-based programs, but not different for regulated home-based programs.[2]  Grace et al. (2006) did not find statistically significant differences in the mean number of children per adult caregiver for relative care and center-based care, but found that rural children in non-relative care had higher children-per-adult ratios than similar children in non-rural settings.

One area in which the literature has shown consistent results is in the area of child care costs; child care is less expensive in rural areas compared to urban areas.  This pattern has also been shown with studies using the NCCS and data collected for the SIPP in 2002 (Hofferth et al., 1991; Smith, 2006).

[1]  As cited in Smith (2006), organized care was defined as care that “is provided in day care centers, nursery schools, preschools, federal Head Start programs, and kindergarten.”  Informal non-relative care included “family day care providers, in-home babysitters or nannies, neighbors, friends, and other non-relatives providing care either in the child’s or provider’s home.”

[2]  Additionally, the study showed that average teacher wages in center programs were lower in rural areas compared to urban and suburban areas, although turnover in rural centers was lower than in urban and suburban centers.

Urban and Rural Economics

Prior literature has compared the economies of urban and rural areas.  According to data from the U.S. Census Bureau, household incomes in non-metropolitan areas, on average, are lower than household incomes in metropolitan areas (DeNavas-Walt, Proctor, & Lee, 2006).  Additionally, wages are generally lower and poverty rates are higher in rural areas (DeNavas et al., 2006; USDA, 2004; USDA, 2007).  However, unemployment has historically been somewhat similar in urban and rural areas (USDA, 2006; Strong, Grosso, Burwick, Jethwani, & Ponza, 2005).  Another economic issue to consider when comparing urban and rural child care costs is the potential for differences in the cost of living in those areas.  Although rural areas are generally thought of as being lower-cost places to live compared to urban areas, there is an active debate among economists as to whether this is actually the case and if so, how it should be considered in measures of economic affluence (Weber, Jensen, Miller, Mosley, & Fisher, 2005).

In addition to general economic conditions, urban and rural differences in the labor force participation of parents with young children are important to note.  Early results from the 2005 NHES show that children are more likely to participate in non-parental care when their mothers are employed or are enrolled in school, and when they live in one-parent families (Iruka & Carver, 2006).  Data from the 2005 American Community Survey (ACS) provide information comparing labor force participation of the parents of urban and rural children under age 6, which is helpful in thinking about child care needs.[3]  These data show that urban and rural children generally lived in families that participated in the labor force. For example, in both urban and rural areas about 6 out of 10 children lived in families[4] where at least one parent living in the household participated in the labor force the previous year.  For children in single-parent families, only about 1 out of 4 children lived in families where the adult parent did not participate in the labor force at all the prior year.  However, when examining families with two parents, a higher percentage of rural children had both of their parents in the labor force the previous year than urban children (56 percent compared to 52 percent).[5]

[3]  Author’s tabulations from the American Community Survey, Table B23008, accessed through the American Fact Finder at The figures in this paragraph exclude children that did not live with either parent.

[4]  Includes both one and two-parent families.

[5]  Rural children were also more likely to be in two-parent families than urban children (76 percent compared to 67 percent).

Research Framework

Although the research described in the previous section provides valuable insights into the non-parental child care arrangements of rural children, there is a need for additional analysis.  One reason for additional analyses is that two of the most thorough national data collection efforts focusing on child care were conducted over 15 years ago: the National Child Care Survey (Hofferth et al., 1990) and the Profile of Child Care Settings (Kisker et al., 1990).  It is unclear whether the urban and rural differences shown in these studies are still representative of rural and urban areas because their data were collected before federal spending on child care subsidies increased significantly throughout the 1990s and before the passage of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996, welfare reform legislation that facilitated the transition of many single low-income mothers into the labor force (Besharov & Higney, 2006; U.S. Department of Health and Human Services, 2007).

A second reason to invite additional research on rural child care is that findings sometimes vary across different data collection efforts because they use dissimilar survey methodologies (Besharov, Morrow, & Fengyan, 2006) or classify urban and rural areas differently.  This paper increases our understanding of how child care arrangements in rural areas compare to those in urban areas of the United States by examining data from the 2005 National Household Education Survey (NHES), Early Childhood Program Participation Survey (ECPP).  Since some of the research findings discussed in the previous section do not show consistent patterns across different studies and data sources, this paper hopes to provide additional information that can provide new results to this discussion and clarify which results appear to be consistent across multiple studies.  It will also introduce some new topics not previously examined such as the physical locations of center programs.

Data and Methodology

The NHES used telephone interviews to collect data from the parents of a nationally representative sample of children age 6 and under that had not yet enrolled in kindergarten (Hagedorn, Montaquilla, Carver, O’Donnell, & Chapman, 2005).  The 2005 version of the ECPP was the fifth such collection effort and focused on non-parental child care arrangements including care by relatives, care by persons not related to the children, and care in day care centers and programs, including Head Start and pre-kindergarten programs.  Since 6 year old preschoolers are uncommon (NHES, 2005), they are not included in this particular analysis.  After these adjustments, records from 7,198 children were examined (6,066 urban and 1,132 rural).[6]

The public-use version of the NHES contains flags to identify the urban and rural status of the children in the sample, as defined by the urban and rural makeup of the zip codes in which they reside.  This methodology is based on classifications used by the Census Bureau; zip codes are considered urban if they are located within urban areas (UA) or clusters (UC), and zip codes that are not defined as urban under this definition are classified as rural.  When zip codes contain both urban and rural areas within their borders, the urban/rural classification is determined by the larger of the two populations (Hagedorn, et al., 2005).  Hence, zip codes that have more rural residents than urban residents are considered rural, and zip codes containing more urban residents than rural residents are considered urban.

Since zip codes are much smaller than counties, this classification system is able to classify neighborhoods as being urban or rural status more accurately than many other categorization systems.  However, this sorting system differs from many county-based systems, including the metropolitan/non-metropolitan system used by the Office of Management and Budget (OMB), in that it does not consider employment commuting patterns for its classifications.  The end result is that the NHES classifies as rural some children living in sparsely populated counties relatively near urban centers that would be classified as urban or metropolitan in other data sets.  It is also important to note that the urban/rural classification is based on the zip code in which the children reside, which may differ from the urban/rural status of the zip code in which the care is provided.

[6]  Only 413 records for 5 year old children were in the sample because children of this age are often in kindergarten and thus excluded from the NHES, ECPP.


This section presents findings from the 2005 NHES, ECPP by comparing the characteristics of child care and early education arrangements of children under age 6 that have not yet enrolled in kindergarten in urban and rural areas.  Care arrangements that did not occur at least once a week were excluded.[7]  Many of the figures in this section present separate calculations by type of non-parental care arrangement as defined below.  These types of arrangements may or may not require fees from the family receiving the care.

  • Relative care is provided by a relative of the child and may take place in the child’s home or in another location.  Relatives include siblings, grandparents, aunts and uncles, and other non-parental family members.
  • Non-Relative care includes care by non-relatives in the child’s home or in another location.  Arrangements sometimes called “family care” fit into this category.
  • Center Programs include Head Start, pre-kindergarten, day care centers, and other early education programs.  They may be located in places such as schools, public buildings, and buildings designed primarily for child care purposes.

[7]  Urban/rural differences are considered statistically significant in the text if the row percentages are p < .05 based on T-tests.  The standard errors were calculated using the “Jackknife” method with WestVar software in order to adjust for the design effects of the NHES.  For more information on the sampling design of the NHES, see Hagedorn, M., Montaquilla, J., Carver, P., & O’Donnell, K. (2006) National Household Education Surveys Program of 2005:  Public-Use Data File User’s Manual, Volume II.  U.S. Department of Education, National Center for Education Statistics.  The standard errors for the analysis can be found at (Reuben, put the Web link here)

Participation in Non-Parental Care

The NHES shows that rural children age 0 to 5 were about as likely as urban children to receive care from someone other than their parents at least once a week, as shown in Figure 1.[8]  Examples of non-parental providers include:  grandparents, older siblings, day care centers, pre-kindergarten programs, and care from friends, neighbors, and other non-relatives.  In both urban and rural areas, about 6 in 10 pre-school age children received non-parental care at least once a week.  About 8 in 10 children with employed mothers were cared for in non-parental care settings on a weekly basis.[9]

Figure 1:
Percent of Children Age 0 to 5 Not Yet in Kindergarten
in Non-Parental Care at Least Once a Week

Figure 1

  ** Urban/rural difference = p < 0.05
*** Urban/rural difference = p < 0.01

[8]  Additional tables on child care arrangements are shown in the Appendix.

[9]  Employed mothers in this paper include biological mothers, step-mothers, foster mothers, and guardians.

Child Care Arrangements

Figure 2a and Figure 2b present the distribution of children across three types of primary non-parental care arrangements for all children and for those with mothers employed at least part-time.  As stated earlier, the definition of center programs used by the NHES included Head Start, pre-kindergarten, day care centers, and other early education programs. Compared to children residing in urban areas, the NHES showed that rural children were more likely to be cared for by relatives and less likely to participate in center programs.  However, despite these differences, more children were cared for in center programs than in any other type of care in both urban and rural areas.  The NHES did not find statistical differences in the use of non-relative care for urban and rural children.

Figure 2a:
Percent Distribution of Primary Care Arrangements for Children Age 0 to 5 and
Not Yet in Kindergarten Participating in Weekly Non-Parental Care

Figure 2a

Child Care
Location- & Geography-Based Data
Urban Communities | Rural Communities