The Role of Attending Center-Based Care for Kindergarten-Aged Children With Disabilities

by Michael A. Gottfried - 2017

Background/Context: Families have been increasingly utilizing center-based care both during prekindergarten as well as before/after school during kindergarten (CBC-K), and the literature has addressed the relative effectiveness of attending the former on early schooling outcomes. However, missing in the field is an analysis of the efficacy of attending the latter. To compound this, no research has considered how attending CBC-K is associated with early outcomes for children with disabilities.

Purpose/Objective/Research Question/Focus of the Study: There were four research questions: 1. Do kindergarten academic outcomes differ for children with disabilities who did and did not attend CBC-K? 2. Do kindergarten socioemotional outcomes differ for children with disabilities who did and did not attend CBC-K? 3. Do these relationships differ by disability? 4. Do these relationships differ by socioeconomic status (SES)?

Population/Participants/Subjects: This study utilizes data from the newly released Early Childhood Longitudinal Study – Kindergarten Class of 2011 (ECLS-K:2011). The ECLS-K: 2011 represents the most contemporary national-level data available to study these research questions.

Research Design: This study examined two sets of outcomes. The first set included reading and math achievement. The second set included socioemotional scales based on both teacher and parent ratings of child behavior. These outcomes were regressed on a measure for having attended CBC-K as well as a wide span of child and family characteristics. Error terms were clustered at the school level to account for nested data.

Findings: The results based on the first and second research questions indicated that for children with disabilities, there was a negative association between attending CBC-K and academic and socioemotional measures compared to not attending CBC-K. As for the third research question, children with emotional or learning/communication disabilities had the largest observed negative association with the set of outcomes from having attended CBC-K (along with children with developmental delays when considering achievement outcomes) compared to children from the same groups who did not attend CBC-K. As for the fourth research question, there was little difference of attending versus not attending CBC-K by SES.

Conclusions/Recommendations: When considering transitions into kindergarten, attending CBC-K certainly appears to have an effect, albeit negative for both children with and without disabilities. Having documented these patterns, this study helped to provide a well-rounded portrait into how center-based care affects outcomes for school-aged children.

For numerous reasonsincluding a growing maternal workforce, shifts in job prospects, increased dialogue around early childhood education, and changes to family structuremore families have been utilizing formal childcare (Blau & Currie, 2004; Burchinal, 1999; Spain & Bianchi, 1996; West, Denton, & Germino-Hausken, 2000; Yamauchi & Leigh, 2011). To be clear, formal care is defined as center-based care and Head Start; features include care outside the childs home, trained providers, extensive peer interaction, and an overt focus on development and learning (Cannon, Jacknowitz, & Karoly, 2012; Crosnoe, 2007; Takanishi, 2004). In contrast, fewer children are now found exclusively in informal care settings, which encompass parental care, relative care, and nonrelative nonparental care such as a babysitters, nannies, or caretakers; these options are characterized by care often taking place in the childs home (though these may also be out-of-home, such as at a nannys house), unstructured activities, untrained providers, no overt focus on development or learning, and potentially less peer interaction (Crosnoe, 2007).

The share of young children in formal childcare is not trivial. Blau and Currie (2004) found that the majority of prekindergarten-aged children in the U.S. were being cared for outside of the homeand most of these children were in center-based care (Loeb, Bridges, Bassok, Fuller, & Rumberger, 2007). More recent research has shown that in a national sample of children, approximately 70% were in center-based care in the year prior to starting kindergarten (Gottfried, 2015b). To put this in perspective, just over a decade earlier, national estimates were only at 50% (Claessens, 2012).

Given this increase in the utilization of center-based care in particular, a significant body of research has focused on the effectiveness of this childcare alternative in boosting outcomes. It is now fairly well established that children benefit academically from center-based care in the year prior to kindergarten (Burger, 2010; Loeb et al., 2007; Loeb, Fuller, Kagan, & Carrol, 2004; Magnuson, Meyers, Ruhm, & Waldfogel, 2004; NICHD Early Child Care Research Network, 2006). This near-consensus seems logical, given that center-based care focuses on learning and is administered by trained providers, as described above. However, research finds that attending center-based care during prekindergarten tends to weaken socioemotional development (Baker, Gruber, & Milligan, 2008; Herbst & Tekin, 2010; Loeb et al., 2007; Magnuson, Ruhm, & Waldfogel, 2007; Yamauchi & Leigh, 2011). There is no single explanation for these negative findings, though it is often asserted that adapting to new settings and separating from parents increases stress for children (Fiese et al., 2002; Morrissey, 2009; Youngblade, 2003).

A large body of research has focused on the role of center-based care during the prekindergarten year. We know much less, however, about the effectiveness of center-based childcare before/after school during the kindergarten school year itselfwhat this study will refer to as CBC-K henceforth. To be clear, CBC-K refers to care that a child receives in addition to attending kindergarten during that same year. The paucity of research in this area is surprising, given that there are a growing number of children attending these programs in addition to prekindergarten programs (Claessens, 2012). The driving forces behind the increase in CBC-K utilization are certainly the same for center-based care during prekindergarten and kindergartenthose related to changes in workforce, economy, and family.

The few studies that have examined the role of CBC-K portrayed mostly negative or null effects on kindergarten-year outcomes. Claessens (2012) examined the relationship between CBC-K and childrens academic and socioemotional outcomes at the end of kindergarten. The findings indicated that attending CBC-K was associated with increased problem behaviors and decreased social skills. Only small gains were made in math from attending CBC-K, with no gains in reading. Gottfried (2015b) found that children attending center-based care in both prekindergarten and kindergarten years had lower social skills and higher problem behaviors at the end of kindergarten compared to those who did not. Gottfried concluded that leaving the home for center-based care involves having to adapt to multiple settings and interactions. In prekindergarten, the settings may be home and center. In kindergarten, these may be home, center, and school. Therefore, multiple years of adapting to multiple settings may exacerbate stress (and hence may lower social skills and higher problem behaviors) associated with continually new setting changes and interactions. This same study found no relationship between attending CBC-K and achievement outcomes. Gottfried (2015a) found that while attending center-based care in prekindergarten reduced chronic absenteeism once in kindergarten, there was no link between CBC-K and these same absence measures.

 

Little work on CBC-K exists beyond these three studies. One missing line of research has been examining heterogeneity in the differences of the effects of CBC-K. Though these aforementioned studies examined differences by gender and socioeconomic status (SES), there has been a lack of attention paid to differences how attending CBC-K might relate to outcomes for children with significant needsand namely scant research on childcare for school-aged children with disabilities (Jinnah-Ghelani & Stoneman, 2009). This gap is critical to address, particularly given that approximately the same rising percentage of children with and without disabilities have consistently been placed in center-based care (Booth-LaForce & Kelly, 1998, 2004; Bruder, 1998; Stahmer & Carter, 2005) and mothers with and without children with disabilities have been consistently entering the workforce at the same rates (Landis, 1992). In other words, the need for and participation in childcare for school-aged children appears to be similar for families with and without children with disabilities. But without knowing how attending CBC-K links to early schooling outcomes children with disabilities, it is difficult to develop policy conclusions surrounding this care alternative for all children.

To understand the link between CBC-K and early schooling outcomes for children with disabilities requires us to consider that the kindergarten school year represents one of the most significant transitions that children face (Ladd & Price, 1987). Successful transition into kindergarten entails children addressing new academic challenges, adapting to institutional expectations, and socializing (Bogart, Jones, & Jason, 1980; Holland, Kaplan, & Davis, 1974). Unsuccessful transition into kindergarten leads to children feeling less secure about their environments and experiencing increased stress (Ladd & Price, 1987).¹

Prior research has suggested that children can often adapt to the transition of schooling through other activities occurring during that same year (Bodrova & Leong, 2005). It seems feasible, then, that CBC-K could affect the transition into school uniquely for children with disabilities. There may potentially be academic benefits for those children with disabilities attending CBC-K. As mentioned, center-based childcare is characterized as having trained providers and an explicit focus on learning (Cannon et al., 2012; Clarke-Stewart, Gruber, & Fitzgerald, 1994; Crosnoe, 2007; Scarr, 1998; Takanishi, 2004). In kindergarten, the small and/or null effects of CBC-K for children in the general population may not be surprising; because kindergarten is already a structured learning environment, typically-developing children may not find any academic gain from attending a second structured educational environment in that same year. However, for children with disabilities, these additional educational opportunities might prove to be valuable, particularly for those with learning disabilities² and others who might face unique challenges in the traditional classroom that might inhibit learning (e.g., behavioral disabilities). Through these additional academic opportunities, the academic transition into kindergarten is eased for children with disabilities (Booth-LaForce & Kelly, 2004). Though unexplored for CBC-K in particular, these structured educational experiences in prekindergarten center-based care have been shown to be particularly beneficial for easing the transition from prekindergarten into kindergarten for children with disabilities (Booth-LaForce & Kelly, 2004; Ingersoll, Schreibman, & Stahmer, 2001; McGee, Morrier, & Daly, 1999). Therefore, it might be expected that concurrent educational experiences in a childcare center might further facilitate this academic transition by directly reinforcing content learned in the kindergarten classroom in that same year.

CBC-K was linked to negative socioemotional development in the general child population, and it was hypothesized in this study that CBC-K would also be linked to lower socioemotional development for children with disabilities. However, before addressing the potentially negative consequences, it is important to consider the possibility that there might be potential for positive socioemotional gains for children with disabilities. There is the potential for positive socioemotional growth because prior research has established that in prekindergarten center-based care, children with disabilities make gains in the areas of communication, social skills, and motor skills (Brady, Martin, Williams, & Burta, 1991; Burack & Volkmar, 1992; Walton, 2001). The same might hold for children with disabilities in CBC-K, particularly if those formal programs explicitly focus on developing these types of skillsets (Pears, Kim, Healy, Yoerger, & Fisher, 2015). That is, if CBC-K provides outside-of-school yet school-like opportunities to reinforce socioemotional skills such as self-regulation and social interaction, then this could serve as a benefit when these children are in other settings, like the kindergarten classroom itself. As described previously, one theoretical benefit of CBC-K for children with disabilities is that the transition to schooling is facilitated through other concurrent activities aimed at boosting socioemotional development during that same year (Bodrova & Leong, 2005).

The literature suggests, however, that it is more likely that children with disabilities will experience negative socioemotional effects from having attended CBC-K, particularly given the findings in regards to children in the general population. These effects might arise from both the mental and physical stress associated with the transition into kindergarten and CBC-K. In terms of mental stress, attending CBC-K represents a second new environment to which children must adapt in addition to kindergarten. Therefore, in addition to spending time learning new rules and behaviors pertaining to kindergarten (Pianta & Cox, 1999), children in CBC-K must also learn to adapt to the expectations of a second environment, including a new set of adult-child and peer relationships. Changing environments and learning to develop new relationships as such have been shown to be extremely stressful on development for children during kindergarten (Fiese et al., 2002; Pianta & Cox, 1999). Building these transition skills has been shown to be particularly difficult for children with disabilities (Pears et al., 2015), including those with developmental and behavioral challenges (McIntyre, Blacher, & Baker, 2010). Therefore, it might be hypothesized that children with disabilities who attend both kindergarten and CBC-K might experience increased stress compared to children in the general population due to not only the transition from home to kindergarten but also the transition into CBC-K. Furthermore, attending both kindergarten and CBC-K increases the duration of separation from the childs home and parents. The stress of such a long daily period of separation has been surmised as being linked to mental stress (Loeb et al., 2007). This might be worsened for children with disabilities, including those with emotional disturbances such as separation anxiety (Shakuntala & Melvin, 2010), as well as those children who might have previously relied on their parents extensively for attention and care.

As for physical stress, research suggests that the rate of physical injuries in center care (albeit for prekindergarten care) is much higher for children with disabilities than for children without disabilities. Leland, Garrard, and Smith (1994) found that 20% of children with disabilities were injured in disputes with other children (compared to only 3% for children without disabilities), and 20% of children with disabilities were injured by another child (compared to 1% of children without disabilities). Though most physical injuries were not severe, it was clear that children with disabilities were involved in injuries due to interpersonal situations during care.

Children with disabilities often feel a higher degree of stress in transitioning into the kindergarten school year (Fowler, Chandler, Johnson, & Stella, 1988). Therefore, attending CBC-K might exacerbate this stress given the potentially higher degree of mental and physical issues mentioned in the above paragraphs. And this increased stress might result in negative academic and socioemotional ramifications. In terms of the latter, the additional transition into CBC-K (combined with the transition into kindergarten itself) may prove to be more challenging for children with disabilities. The previously established mental stressors for children with disabilities in center-based care (albeit for prekindergarten) combined with the stress associated with transitions might result in lower socioemotional growth for children with disabilities in CBC-K. As an indirect mechanism of this increased stress, attending CBC-K might lead to lower achievement outcomes. For instance, both the mental and physical stress associated with attending CBC-K might result in the development of negative attitudes about formal school environments. This may materialize as academic decline as attending multiple school-like environments becomes increasingly stressful.

No work has considered the role of CBC-K on the academic and socioemotional outcomes of children with disabilities. This study will be the first to ask the following four research questions:

1.

Do kindergarten academic outcomes differ for children with disabilities who did and did not attend CBC-K?

2.

Do kindergarten socioemotional outcomes differ for children with disabilities who did and did not attend CBC-K?

3.

Do these relationships differ by disability?

4.

Do these relationships differ by SES?

To assess these questions, this study relied on a newly released national dataset of kindergarten children in the 20102011 school year. These data represented the most currently available to address these research questions. Through child assessments, parent interviews, and teacher surveys, it was possible to determine whether a child had attended CBC-K and then link this information to academic and socioemotional outcomes as well as a wide of contextual measures.

The first and second research questions inquired into the direction of the effect of CBC-K for children with disabilities, relying on many measures of child success. The third and fourth questions addressed heterogeneity. Different facets of the above mechanisms might be unique for children with a specific disability. While very few studies have examined outcome differences by disability type among kindergarten-aged students, the literature has reported variation in outcomes by disability classification when evaluating other school policies and practices (see e.g., Blackorby et al., 2010; Hanushek, Kain, & Rivkin, 2002). Therefore, variation by disability was explored in this present study. Finally, prior research has found differences in the effects of center-based care by family SES (Crockenberg & Litman, 1991; Desai, Chase-Lansdale, & Michael, 1989; Yamauchi & Leigh, 2011; Youngblade, 2003). Therefore, differences by SES were examined as well.

Overall, this study was the first to position itself at the intersection of research on CBC-K and research on children with disabilities. As much more was known about center-based care in year prior to school entry, the dearth of data on care during this first year of schooling was concerning particularly given that attending CBC-K could serve as a potential policy lever to facilitate school transitions. Moreover, the lack of research on CBC-K for children with disabilities is an obstacle to developing policies for a group of children who are known to have difficulties with the transition to from home to school. Knowing whether CBC-K helps or hinders that transition enables the development of supports and target resources to best address their unique needs.

METHOD

DATASET

To assess the research questions in this study, data were sourced from the Early Childhood Longitudinal Study Kindergarten Class of 20102011 (ECLS-K:2011). The U.S. Department of Educations National Center for Education Statistics (NCES) collected information on a nationally representative sample of children and their families and schools from multiple sources including direct assessments, interviews, and surveys. To ensure national representation, NCES employed a three-stage stratified sampling design. Geographic region represented the first sampling unit, public and private school represented the second sampling unit, and race/ethnicity represented the third sampling unit. Hence, observations in the dataset are from a variety of school types, socioeconomic levels, racial, and ethnic backgrounds.

The first wave of data collection occurred in the fall of 2010, when children were in kindergarten. This study utilized data from this first wave as well as the second wave of data collectionthe spring of kindergarten. Missing values were addressed using chained multiple imputation (Royston, 2004) and were imputed back to the sample for which there were nonzero weights. In total, 10 sets of plausible values were imputed to resemble the distributions of the observed variables. Sample weights provided by NCES for the ECLS-K:2011 dataset were employed in both the imputation and in the analysis. After imputation, this sample consisted of approximately n = 18,170 child observations. Sample sizes are rounded to the nearest tens digit, per NCES rules.

OUTCOMES

Achievement

This study relied on students item response theoryscaled scores on math and reading tests. Spring math and reading scores serve as outcomes (though tests administered in the fall will serve as an independent control measure, described below). As described in more depth in the ECLS-K:2011 users guide (Tourangeau et al., 2013), these achievement tests were conducted in two stages. Stage 1 consisted of students taking a routing test to determine the test form that should be administered (i.e., low, medium, or high). Stage 2 consisted of students taking the form to which they had been assigned. At kindergarten, the reading assessment assessed basic reading skills, including print familiarity, letter recognition, and recognition of common words. The mathematics assessment consisted of questions on number sense, properties, and operations; measurement, geometry and spatial sense; data analysis, statistics, and probability; and patterns, algebra, and functions. The reported reliability alpha for reading was 0.95 and was 0.94 for math.

Teacher-Rated Socioemotional Measures

Spring measures of child socioemotional development were utilized as outcomes; fall measures were utilized as control variables. Based on the Social Skills Rating System (SSRS; Gresham & Elliott, 1990), NCES modified these scales and created its own Teacher Social Rating Scales (SRS) in ECLS-K:2011. All scales were continuous on a 4-point Likert metric, with higher scores indicating more frequent behavior.

The 7-item approaches to learning scale (α = 0.91) measured the frequency with which the child was able to keep his or her belongings organized, show eagerness to learn new things, adapt to change, persist in completing tasks, pay attention, and follow classroom rules. The 4-item self-control scale (α= 0.81) measured the extent to which the child was able to control his or her temper, respect others’ property, accept his or her peers’ ideas, and handle peer pressure. The 5-item interpersonal skills scale (α = 0.86) measured the frequency with which a child was able to get along with others, form and maintain friendships, help other children, show sensitivity to the feelings of others, and express feelings, ideas, and opinions in positive ways. The 5-item externalizing behaviors scale (α = 0.88) measured the frequency with which a child argued, fought, got angry, acted impulsively, and disturbed ongoing activities. The 4-item internalizing behaviors scale (α = 0.79) measured the extent to which the child exhibited anxiety, loneliness, low self-esteem, and sadness. As mentioned in Tourangeau et al. (2013), details about the specific items included in these scales were not provided.

Parent-Rated Socioemotional Measures.

The benefit of using a large-scale dataset like ECLS-K:2011 is that it is possible to triangulate data from multiple sources to measure child outcomes. It should be noted that teacher-reported ratings of childrens socioemotional functioning is often considered to be the most informative source of information, particularly when compared to survey responses from other sources, such as a childs parents (Konold & Pianta, 2007; Waterman, McDermott, Fantuzzo, & Gadsden, 2012). Nonetheless, having the luxury of both teacher- and parent-rated socioemotional scales enabled a greater degree of generalizability, even if the teacher-rated scales are generally most preferred.

In both the fall and the spring, parents were asked a series of questions about the frequency with which children exhibited certain behaviors. Like the teacher-rated scales, parent-rated scales were also based on the SSRS as described above. The 6-item approaches to learning scale (α = 0.72) measured the frequency with which children kept working at something until finished, showed interest in a variety of things, concentrated on a task and ignored distractions, helped with chores, were eager to learn new things, and were creative in work and play. The self-control scale (α = 0.73) included 5 items that measured the frequency with which children could control their behavior. The 3-item social interaction scale (α = 0.69) measured a child’s ease in joining in play, ability to make and keep friends, and ability to positively interact with peers. The impulsive/overactive scale contained 2 items that asked about the frequency of a child’s impulsivity and the child’s activity level. No alpha was reported in the user’s manual for a 2-item scale. Finally, the sad/lonely scale included 4 items (α = 0.62), asking about a child’s problems with being accepted and liked by others and the childs frequency of sadness, loneliness, and low self-esteem.

CBC-K

Table 1 presents all outcomes as well as independent measures. The key variable in this study was attending CBC-K. Based on the fall parent survey, a binary measure indicated if a child attended CBC-K if his or her parents indicated that he or she was in center-based care outside of kindergarten hours during the kindergarten school year. Table 1 is split between those who did and did not attend CBC-K.

DISABILITY STATUS

Children with disabilities were identified by their school records on file. In the ECLS-K:2011 dataset, a student was identified as having a disability if the student had an Individualized Education Plan (IEP). Initially, a binary measure was created to indicate if a child had an IEP in kindergarten. As described in a further section, this binary measure was expanded to include classification of disabilities.

Note that children with disabilities comprised approximately 20% of both sample populations (children attending CBC-K and children not attending CBC-K), as shown in Table 1. This was an important distinction, as this showed that children with disabilities were not disproportionately going or not going to CBC-K, which would have been the case had the percentages been drastically different between those children with disabilities in and not in CBC-K.

CONTROL MEASURES

Child Characteristics

The ECLS-K:2011 contains a wide array of sociodemographic characteristics about children in the sample as well as information on their schooling and care experiences. A common set of sociodemographic measures were included, such as gender, race (with White serving as the omitted reference category), English language learner (ELL) status (whether the childs primary home language was other than English), and kindergarten entry age. Additional child measures pertained more directly to attitudes about school, including: a binary measure indicating whether parents reported their children expressed eagerness to attend school more than once per week; and an indicator for being a chronic absentee (namely missing more than 2 weeks of school). Other control measures relating to the childs schooling/care experiences included indicators for whether the child was in full- or part-day kindergarten, whether the child had attended center-based care (including Head Start) during the year before kindergarten, and whether the child had attended center-based care in any year prior to prekindergarten. Finally, this study included a continuous measure for the number of hours of attending CBC-K.

Household Characteristics

In addition to child characteristics, the span of ECLS-K data allowed for the inclusion of a wide array of family background characteristics. These can informally be grouped into structure, home environment, and SES. For family structure, this study included measures for whether parents were married, the number of siblings, and age of mother when she first gave birth.

Home environment was captured in three capacities. The first was the number of books at home as well as home distance from kindergarten. The second used two home involvement scales, as replicated from Votruba-Drzal, Li-Grining, and Maldonado-Carreño (2008). The first scale, which was comprised of 15 dichotomously scored items, measured the number of learning activities in which children participated. This scale assessed whether the child had engaged in activities such as visiting a bookstore, taking music lessons, and attending tutoring lessons in the past month. The second scale, relating to parental involvement, was measured on a 4-point Likert scale. This 10-item scale assessed the frequency with which parents engaged the child in various activities such as playing games, singing songs, reading books, and doing arts and crafts. Third were measures of socioeconomic status. Maternal and parental education were included, with the category high school graduate serving as the omitted reference category. Household income was included, as was maternal employment status (with not working as the omitted reference category).

Though the percentage of children with disabilities attending and not attending CBC-K were similar, there might have been other differences within the sample of children with disabilities. Table 2 presents the set of independent variables for those children with disabilities who did and did not attend CBC-K. Here, means are presented in the table along with indicators for p-values for the test of mean differences between the two categories.

While there were some differences between children with disabilities who did and did not attend CBC-K, most measures were similar. As for the differences in child-level covariates, children with disabilities in CBC-K were more likely to be Black or to have attended center-based care in prekindergarten and were less likely to be ELLs compared to children with disabilities who did not attend CBC-K. As for differences in family covariates, children with disabilities in CBC-K were more likely to have parents with higher levels of involvement, to be from families of higher income, and to have mothers working full time versus children with disabilities who did not attend CBC-K. Between the two groups, those attending CBC-K were less likely to have siblings and to have mothers with less than a high school degree. As mentioned, most of the variables in this table did not show differences between those children with disabilities in CBC-K and those not in CBC-K. In fact, of those measures where there were statistically significant differences, most were small such as the difference in Black (4 percentage points) or number of siblings (0.44 difference in number of siblings).

ANALYSIS

With these data, the relationship between CBC-K and the set of outcomes was specified as the following:

Y_iS = β₀ + β₁CBC-K_iF*D_iF+ β₂C_iF + β₃H_iF + β₄Y_iF +ε_iS

where Y is an outcome (achievement or socioemotional scale) for child i measured in the spring wave S. All outcomes were standardized. Note that the fall measure of the outcome was included as well, represented by Y_iF. As described in Gershenson, Jacknowitz, and Brannegan (2014) and Todd and Wolpin (2003), when an outcome has a repeated measure from the prior period in the dataset, having this serve as an independent covariate can account for unobserved child (and family) characteristics that occurred prior to the current period as well as those that potentially did not change over time. Even with the wide span of variables that were already incorporated into the model, this was critical as it might not be possible to fully capture family characteristics, involvement, or motivations (especially if they were fairly unwavering over time).

Key to this analysis were the two dichotomous variables CBC-K and D, which respectively represented attending center-based care before/after the kindergarten school day and having a disability. The key coefficient in this equation was β₁, which represented an interaction between CBC-K and D; the findings indicated the extent to which attending versus not attending CBC-K was associated with outcomes specifically for children with disabilities. This partially interacted model (i.e., interaction between two key terms in the equation) is grounded in the field as an appropriate estimation technique for assessing differences in attending versus not attending early childhood care as delineated by a specific individual demographic characteristici.e., in this case, the difference between children with disabilities who did and did not attend CBC-K (e.g., Claessens, 2012; Yamauchi & Leigh, 2011). Hence, it was the approach also adopted in this current analysis, where the individual student-level characteristic of interest was D and care was CBC-K versus not attending CBC-K. Note that though not presented in this equation, the main terms CBC-K and D were also included in the regression model as is commonplace when employing an interaction.

The other terms in the model represent the control variables mentioned in the previous section: C_iF represents child characteristics and H_iF represents household characteristics. Finally, the error term was robust clusteradjusted for the fact that ECLS-K:2011 sampled children within the same school (White, 1980; Wooldridge, 2002). This adjustment corresponded to the techniques in prior research that used ECLS-K data to test for effects of center-based care (e.g., Claessens, 2012). Correcting the error for multiple children per school in conjunction with the wide span of child and household variables (in addition to the using a one-wave lagged measure of the outcome) helped to address selection issues on attending CBC-K (Claessens, 2012).

RESULTS

ACADEMIC ACHIEVEMENT

Table 3 presents coefficients and standard errors for all independent measures in the above equation. Each outcome Y is listed as a column heading, and all covariates are presented as rows. The coefficients of interest are found in β₁—i.e., the interaction between CBC-K_iF and D_iF, which compares children with disabilities who did and did not attend CBC-K. All measures pertaining to this (main effects as well as interaction) can be found in the first three rows of results. Recall that the outcomes were standardized when running the regression models.

There were three key findings, as shown in the first three rows. First, children with disabilities tended to have lower reading and math scores compared to children in the general population, as evidenced by the negative coefficient of the main disability term in the first row. This finding was consistent with prior research on achievement outcomes for children with disabilities (Snyder & Dillow, 2012). The second row of the table suggests that for the general population, there was no relationship between attending CBC-K versus not attending and achievement. The null findings here corresponded to previous research on CBC-K for the general child population, specifically for reading (Claessens, 2012), thereby providing a robustness in the empirical model selected in this present study.

The key finding in this analysis, however, was the third row of resultsthe interaction between CBC-K and disability status. The negative coefficient on the interaction term indicates that children with disabilities in CBC-K tended to have lower achievement outcomes, across both reading and math tests, compared to children with disabilities who did not attend CBC-K. Therefore, while there was no association between attending CBC-K and achievement for children in the general population (as evidenced in the main nonsignificant term of CBC-K), there was a negative association for those with disabilities who did attend CBC-K. Children with disabilities who did attend CBC-K had lower achievement scores compared to children with disabilities who did not attend CBC-K, again as depicted by the negative coefficient on the interaction term. The size of this coefficient, in fact, was as large as the size of the main term of having a disability itself.

In terms of the control variables, there was a degree of consistency between reading and math models. Several noteworthy findings are seen here. For instance, as hypothesized by prior findings, kindergarten entry age was positively associated with achievement (Datar, 2006; Elder & Lubotsky, 2009). Chronic absenteeism was negatively associated with achievement per Gottfried (2014). Attending full-day kindergarten was positively associated with achievement (Cannon, Jacknowitz, & Painter, 2006). Note that hours per week of CBC-K attendance was not statistically significant, as consistent with Claessens (2012). Along with the fact that the coefficient is approximately 0.00, there is no significant linear relationship between hours of care and student outcomes. Center-based care attendance prior to prekindergarten also did not significantly predict kindergarten outcomes, as consistent with Gottfried (2015b).

SOCIOEMOTIONAL MEASURES

Table 4 presents findings when the outcome was one of five teacher-rated socioemotional scales. As in the regression models for Table 3, the outcomes have been standardized. Note that larger positive coefficients for the first three models indicate positive behaviors (i.e., greater exhibited frequencies of approaches to learning, self-control, or interpersonal skills). Larger positive coefficients for externalizing and internalizing problems indicate worsened outcomes.

The key coefficients are presented in the first three lines of results, analogous to the achievement models. The interpretation of the findings when outcomes were teacher-rated socioemotional scales mirrored those from the achievement models. The first line of results suggests that children with disabilities have lower frequencies of approaches to learning, self-control, and interpersonal skills compared to children without disabilities. Children with disabilities were also much more likely to display higher frequencies of both externalizing and internalizing problem behaviors, as demonstrated by the positive coefficients in the latter two columns of results. These negative findings were consistent with prior research (Phelps & Hanley-Maxwell, 1997). Next, unlike in the achievement models, attending CBC-K was negatively associated with socioemotional outcomes for those children in the general child population, as depicted by the second row of findings. The size of the coefficient of CBC-K was consistent with prior research on CBC-K using secondary data (Claessens, 2012; Gottfried, 2015b).

The coefficients on the interaction term indicated that children with disabilities who attended CBC-K had lower socioemotional outcomes across all five measures compared to children with disabilities who did not attend CBC-K. Children with disabilities in CBC-K tended to have lower frequencies of approaches to learning, self-control, and interpersonal skills versus children with disabilities who did not attend CBC-K; they also had greater frequencies of externalizing and internalizing problem behaviors when attending.

Overall, attending CBC-K was associated with lower socioemotional outcomes for all children, as evidenced by the statistically significant coefficients on the main term CBC-K. However, those with disabilities were at a significantly greater disadvantage, as evidenced by the much larger coefficients on the interaction compared with the main term of CBC-K. In sum, children with disabilities were already at risk for having lower frequencies of social outcomes and higher frequencies of problem behaviors. For this group of children, attending CBC-K was associated with worsened socioemotional outcomes.

As mentioned, using ECLS-K:2011 allowed for the teacher-rated socioemotional scales to be tested for robustness by also examining parent-rated socioemotional scales. Table 5 presents the same models using parent-rated rather than teacher-rated measures as outcomes. Though parent-rated scales did not map directly onto teacher-rated scales and though teacher-rated scales are generally preferred, there were nonetheless similarities in the findings. As before, larger positive coefficients suggest greater frequencies of positive behaviors, such as approaches to learning, self-control, and social interaction. Larger positive coefficients on sad/lonely and impulsive/overactive suggested less favorable outcomes. All outcomes have been standardized. Finally, note that though not presented, all models included the set of control variables utilized in the prior analyses in this study.

The first row of results corresponded to Table 4: compared to children in the general population, children with disabilities tended to have lower frequencies of approaches to learning, self-control, and social interaction, similar to the teacher ratings of these children. Children with disabilities were also more likely to display sad/lonely and impulsive/overactive behaviors, as indicted in the latter two columns, and as consistent with Table 4 data showing that teachers rated children with disabilities as having higher frequencies of externalizing and internalizing problem behaviors. While not as consistent as before, there was some evidence here that attending CBC-K was associated with lower socioemotional outcomes for the general population, as indicated by the second row of results.

All coefficients were statistically significant on the interaction between CBC-K and having a disability. As in Table 4, children with disabilities who attended CBC-K had lower frequencies of positive behaviors (approaches to learning, self-control, social interaction) and higher frequencies of negative behaviors (sad/lonely, impulsive/overactive) compared to children with disabilities who did not attend CBC-K, Notably, almost all of the sizes of the interaction terms in Table 5 were similar to those in Table 4, except for impulsive/overactive which was much larger here. Therefore, this provided confidence in these findings: attending CBC-K was still shown to be negatively associated with outcomes for children with disabilities, even when socioemotional outcomes were measured by two separate data sources. Generalizing even more broadly, the negative association of CBC-K permeated both achievement and socioemotional development, exemplifying that the findings were not domain specific.

DIFFERENCES BY DISABILITY

A third research question asked whether there were differences in the outcomes by disability classification. Children with disabilities were disaggregated into more refined groupings. Though all 13 Individuals with Disabilities Education Act (IDEA) disability categories were represented in the dataset, relying on these 13 measures separately would have limited statistical power too greatly. Instead, five disability categories were constructed based on the taxonomy published by Gottfried (2014). Children with disabilities were grouped by: emotional/behavioral disabilities, learning/communication disabilities, developmental delays including mental retardation, physical impairments, and severe disabilities including autism. Note that children with category as other disability could not be classified in this portion of the analysis, as no other information was provided.

The previous models were then rerun such that the interaction term was [β₁CBC-K_iF*DisabilityCategory_iF] in order to test for differences of attending versus not attending CBC-K within each disability group. Table 6 presents these findings. Each cell represents the interaction term from a unique regression where the standardized outcome is listed down the rows and the disability category is listed as column headings. For instance, the first cell of results was derived from a regression where the outcome was reading achievement and the key interaction was [β₁CBC-K_iF*Emotional/Behavioral_iF]. Though each cell represents a unique regression, all models included both the main term CBC-K, the main term for the disability category listed in the column, as well as all control variables as utilized in all other models in this study.

Table 6. Differences by Disability

Table 6 indicates differences by disability. Therefore, having differentiated by category was a useful step in this analysis. As for achievement outcomes, there was consistent evidence that children with learning/communication disabilities or developmental delays had lower achievement outcomes when having been in CBC-K compared those from this group who did not attend CBC-K. This is portrayed by larger, statistically significant interaction coefficients for both reading and math for children in these disability groupings. There was some evidence that children with emotional/behavioral disabilities and those with severe disabilities had lower math test scores when in CBC-K compared those in this group that did not attend CBC-K. That said, Claessens (2012) found that only math achievement was negatively associated with CBC-K attendance within the general child population, and thus the findings here would correspond to that.

The results for both teacher- and parent-rated socioemotional outcomes also differed by disability. Compared to other children with disabilities, children with emotional/behavioral disabilities and those with learning/communication disabilities had lower frequencies of positive behaviors and higher frequencies of negative behaviors when in CBC-K than those from the same group who did not attend CBC-K. The interaction terms were statistically significant for almost all outcomes for children within these two disability groups. While there were some statistically significant coefficients for children in the other three disability categories, the patterns were not as consistent or striking as for those with emotional/behavioral or learning/communication disabilities.

DIFFERENCES BY SES

A final analysis examined whether differences emerged by SES, as prior research suggests that SES might moderate the relationship between attending center-based care and child outcomes (Loeb et al., 2007). To test for differential effects by SES, the models in Tables 2, 3, and 4 were rerun on two different samples, with the sample divided between low and high SES. SES was determined by whether the parents had received food stamps over the past 12 months. Note that the results were consistent when other measures SES were employed.

As in Table 6, the results presented in Table 7 represent the interaction term, though all other covariates were also included in the models though not displayed. As for achievement outcomes, the findings indicated that the subset of children with disabilities from lower SES families who attended CBC-K were at a greater disadvantage compared to the subset of children with disabilities in higher SES families who attended CBC-K. This was seen by slightly larger coefficients on the interaction terms when comparing low SES and high SES columns.

The findings for the socioemotional measures indicated much fewer differences between low- and high-SES families. There were no differences between the teacher-reported outcomes in the two columns. Some coefficients were also fairly similar in the results for parent-rated outcomes. For instance, the sizes of the coefficients for the approaches to learning scales in this table were almost identical between children with disabilities who attended CBC-K (versus not) between low- and high-SES groups. For all other outcomes, there was some evidence that children with disabilities from low-SES backgrounds had larger negative associations with these outcomes when attending CBC-K compared to children from higher SES backgrounds who attended CBC-K. However, as mentioned previously, in survey research, teacher reports on childrens socioemotional outcomes are considered to be the most relevant source for data on socioemotional functioning as opposed to parental responses (Konold & Pianta, 2007; Waterman et al., 2012). If this is assumed to be true and interpretations should rely more strongly on teacher-rated scales, then the results would suggest few differences for children with disabilities by SES.

DISCUSSION

This study examined the role of CBC-K on end-of-kindergarten academic and socioemotional outcomes for children with disabilities. Little prior research had considered the influence of attending CBC-K; none had considered its association with outcomes for children with disabilities. This was a critical lapse in the body of research, as attending CBC-K occurs during a key transitional period into formal schooling and it is fairly well established that children with disabilities might experience additional difficulties when making this transition into kindergarten (Fowler et al., 1988). Therefore, understanding how attending CBC-K either does or does not correlate to outcomes in this first, critical year of schooling can help to address the needs of those children who face additional challenges in transitioning into school.

To examine the four research questions in this study, data from the ECLS-K:2011 were examined. The dataset represented a noteworthy source of informationit contained measures on childcare experiences, academic and socioemotional outcomes, and individual and family characteristics. Therefore, it was possible to test for a link between attending CBC-K and many child outcomes while also accounting for a wide span of control variables, thereby helping to reduce selection issues. More so, unique to this dataset, socioemotional measures were based on both teacher and parent survey items. Assessing both sets allowed for a generalizability in the findings.

The results based on the first and second research questions indicated that for children with disabilities, there was a negative association between attending CBC-K and academic and socioemotional measures compared to not attending CBC-K. As for academic outcomes, children with disabilities in CBC-K had lower achievement scores in kindergarten compared to children who did not attend CBC-K. In the sample of children without disabilities, however, there was no observed relationship between CBC-K and achievement. These results were consistent with prior research on the effects of attending CBC-K, which had found null or extremely small positive effects (Claessens, 2012; Gottfried, 2015b). As consistent with research on the effects of center-based care in prekindergarten (Baker et al., 2008; Herbst & Tekin, 2010; Loeb et al., 2007; Magnuson et al., 2007; NICHD Early Child Care Research Network, 2006; Yamauchi & Leigh, 2011), children who attended CBC-K had lower socioemotional outcomes measured in kindergarten compared to those who did not attend. However, unique in this study was the fact that the size of the coefficient was certainly largest for children with disabilities. The findings were fairly consistent between teacher- and parent-rated outcomes, thereby suggesting a robustness in the analysis.

Examining differences by disability and SES guided the third and fourth research questions, respectively. As for the third research question, this study found that children with emotional or learning/communication disabilities had the largest observed negative association with the set of outcomes from having attended CBC-K (along with children with developmental delays when considering achievement outcomes) compared to children from the same groups who did not attend CBC-K. This corresponded to prior research, which suggests that children with developmental and behavioral disabilities have extremely difficult transitions into kindergarten (McIntyre et al., 2010). Though this study was the first to examine the effects of attending CBC-K in particular, the results differentiated by disability type for outcomes in kindergarten are certainly supported in prior literature. As for the fourth research question, our study found little difference of attending versus not attending CBC-K by SES, as consistent with prior research (Claessens, 2012). Therefore, in this study, any observed heterogeneity in outcomes from attending CBC-K appear to have been driven by differences in disability and not differences in family SES.

Taken in sum, these findings would support the hypothesized negative effects of CBC-K on child outcomes, as described in the introduction of this article. While the mechanisms cannot necessarily be explored using these data, it remains hypothesized that children with disabilities face additional challenges in transitioning to school and that CBC-K might be exacerbating rather than mitigating these challenges. These obstacles might arise through mental stressors, such as having to adapt to multiple environmental transitions at oncean issue that children without disabilities also certainly face, though made more challenging when faced with specific challenges such as emotional, behavioral, or developmental disabilities. This added stressor, among others, might be reflected directly on socioemotional assessments or indirectly on achievement tests through reduced attitudes towards schooling.

Given these findings, there are several concluding implications. The first is the issue of timing of attending center-based care. This study has shown a negative correlation between attending CBC-K and the set of kindergarten outcomes for children with disabilities. However, this study adds to what is presently only a very small body of research focusing specifically on CBC-K. Though the findings of this study are not particularly positive, they do suggest that there are statistically significant relationships between center-based care and child outcomes that are largely being ignored when we as a field focus more narrowly on center-based care prior to school entry. When considering transitions into kindergarten, attending CBC-K certainly appears to have an effect, albeit negative for both children with and without disabilities. Having documented these patterns, this study helped to provide a well-rounded portrait of the ways center-based care affects outcomes for school-aged children.

A second implication builds on the first. With a growing economy (or in the face of economic pressures) in conjunction with greater maternal employment and single parenting, the utilization of CBC-K will only continue to rise in future years for children both with and without disabilities. Therefore, this study does not call for the cessation of utilizing CBC-K; rather, it calls for further investigation into what factors, contexts, and practices might particularly support those children with disabilities who attend CBC-K programs so that negative effects do not persist. In this way, policymakers, educational practitioners, and care providers can make adjustments accordingly and target resources more carefully in CBC-K programs in order to support the unique needs of children with disabilities in early education. There has been a significant focus on what factors and contexts improve outcomes for children with disabilities in prekindergarten center-based care (e.g., Bruder, 1998; Jinnah-Ghelani & Stoneman, 2009). It might be the case that many of the protective factors and practices developed to support children with disabilities in center-based prekindergarten care also translate to CBC-K settings. However, CBC-K is also unique: it represents a transition that occurs concurrently with the primary transition into formal schooling (i.e., kindergarten). Thus, further inquiry could directly address what factors might be most effective at boosting outcomes for those in CBC-K and in kindergarten, thereby protecting those children with disabilities from additional risk. This knowledge will be critical in developing policy.

Finally, the fact that the results differed by disability indicates that additional attention might be paid to addressing heterogeneity when it comes to developing supports in moving forward. For instance, this study found largest negative results for children with emotional/behavioral and learning/communication disabilities. This may indicate that there are unique processes and factors by disability type that moderate the relationship between CBC-K and academic and socioemotional development. Just as further inquiry is required to understand what overall factors might support children with disabilities in CBC-K as mentioned in the previous paragraph, it is as important to understand what factors are particular to certain disability groups. This will more effectively guide policy and practice in addressing the needs of those children in CBC-K.

Future research can be guided by the limitations in this study. First, the precise characteristics of CBC-K (including measures pertaining to quality) were not contained in the ECLS-K:2011 dataset. Therefore, in this study, it was not possible to address what factors moderated the link between CBC-K and child outcomes, nor was it possible to develop an understanding of what mechanisms might specifically underlie the patterns that arose in this study. Therefore, additional research using alternative sources of data is required to understand the role of quality in moderating the effects found in this research study. This further research would address many of the implications raised here.

A second limitation of this study was that survey-based data relies on self-reported information, including parent-reported childcare arrangements. Additionally, the socioemotional outcome scales are based on teachers and parents survey responses and as such, there may be some degree of subjectivity in the ratings (DiPerna, Lei, & Reid, 2007; Galindo & Fuller, 2010). Therefore, one must take this into account when interpreting the results.

Third, though the analysis assessed differences by disability, there was not enough statistical power to address specific disability types within categories (e.g., dyslexia within the learning disabilities category). Therefore, while this study did provide more detail than strictly samples of children with and without disabilities, it was not possible to delve further into differences by specific disability. This is a common limitation in empirical research in center care for children with disabilities, even within those studies that examined samples of one of 13 IDEA disability categories (e.g., Pears et al., 2015).

Finally, this study focused on short-term outcomes. While this was not a limitation per se, future research might develop an understanding of the long-term effects of attending CBC-K. Evaluating longer-term outcomes might show different patterns. Therefore, while the transition itself to kindergarten might prove challenging, the trajectories of children facing challenges during this transitory period may differ. Future research might build upon this study to provide an even richer portrait of center-based care and child outcomes, particularly for children facing challenges when first entering school.

Acknowledgement

Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health under Award Number R03HD071334. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Notes

1.

A key piece in understanding the link between CBC-K and outcomes for children with disabilities is that center-based childcare settings discussed here are inclusive rather than providing self-contained services. This has been the case for several decades: Wolery et al. (1993) reported that approximately 74% of childcare programs in their census-based survey included children with disabilities, with representation from all 13 disability categories designated by the Individual with Disabilities Education Act (IDEA). Similarly, this study evaluates children in general education kindergartens rather than specialized service settings. When considering the effect of CBC-K on kindergarten outcomes for children with disabilities, it is important to consider that these settings include both children with and without disabilities.

2.

A recent report by the Centers for Disease Control (Visser et al., 2013) found the mean age of ADHD diagnosis to be 6.2 years, with a 95% confidence interval of 6.9 to 7.2 years. Therefore, it is highly likely that learning disabilities are identified either at the end of kindergarten or beginning of first grade. This implies that even if learning disabilities are only identified at mean age 6.2, they are likely to be present in kindergarten even if undiagnosed.

3.

The analyses here were also run separately for those children in full- versus part-day kindergarten. However, the main findings on CBC-K were not differentiated by children in different kindergarten settings.

References

Baker, M., Gruber, J., & Milligan, K. (2008). Universal child care, maternal labor supply, and family well being. Journal of Political Economy, 116(4), 709745. doi:10.1086/591908

Blackorby, J., Schiller, E., Mallik, S., Hebbeler, K., Huang, T., Javitz, H., & Williamson, C. (2010, January). Patterns in the identification of and outcomes for children and youth with disabilities (Publication No. NCEE 2010-4005). Washington, DC: National Center for Education Evaluation and Regional Assistance. Retrieved from http://ies.ed.gov/pubsearch/pubsinfo.asp?pubid=NCEE20104005

Blau, D., & Currie, J. (2004, August). Preschool, day care, and afterschool care: Whos minding the kids? (NBER Working No. Paper 10670). Cambridge, MA: National Bureau of Economic Research. Retrieved from http://www.nber.org/papers/w10670

Bodrova, E., & Leong, D. J. (2005). High quality preschool programs: What would Vygotsky say? Early Education and Development, 16(4), 437446.

Bogart, G. A., Jones, J. W., & Jason, L. A. (1980). School transitions: Preventive intervention following an elementary school closing. Journal of Community Psychology, 8(4), 343352.

Booth-LaForce, C., & Kelly, J. F. (1998). Child-care characteristics of infants with and without special needs. Early Childhood Research Quarterly, 603621.

Booth-LaForce, C., & Kelly, J. F. (2004). Childcare patterns and issues for families of preschool children with disabilities. Infants and Young Children, 17(1).

Brady, M. P., Martin, S., Williams, R. E., & Burta, M. (1991). The effects of fifth graders socially directed behavior on motor and social responses of children with severe multiple handicaps. Research in Developmental Disabilities, 12, 116.

Bruder, M. B. (1998). A collaborative model to increase the capacity of childcare providers to include young children with disabilities. Journal of Early Intervention, 21(2), 177186.

Burack, J. A., & Volkmar, F. R. (1992). Development of low and high functioning autistic children. Journal of Child Psychology and Psychiatry, 33, 607616.

Burchinal, M. R. (1999). Child care experiences and developmental outcomes. Annals of the American Academy of Political and Social Science, 563(1), 7397. doi:10.1177/000271629956300105

Burger, K. (2010). How does early childhood care and education affect cognitive development? An international review of the effects of early interventions for children from different social backgrounds. Early Childhood Research Quarterly, 25(2), 140165. doi:10.1016/j.ecresq.2009.11.001

Cannon, J. S., Jacknowitz, A., & Karoly, L. A. (2012, May). Preschool and school readiness: Experiences of children with non-English speaking parents. San Francisco: Public Policy Institute of California. Retrieved from http://eric.ed.gov/?id=ED531794

Cannon, J. S., Jacknowitz, A., & Painter, G. (2006). Is full better than half? Examining the longitudinal effects of full-day kindergarten attendance. Journal of Policy Analysis and Management, 25(2), 299321. doi:10.1002/pam.20174

Claessens, A. (2012). Kindergarten child care experiences and child achievement and socioemotional skills. Early Childhood Research Quarterly, 27(3), 365375. doi:10.1016/j.ecresq.2011.12.005

Clarke-Stewart, A., Gruber, C. P., & Fitzgerald, L. M. (1994). Children at home and in day care. Hillsdale, NJ: Lawrence Erlbaum Associates.

Crockenberg, S., & Litman, C. (1991). Effects of maternal employment on maternal and two-year-old child behavior. Child Development, 62(5), 930953. doi:10.1111/j.1467-8624.1991.tb01581.x

Crosnoe, R. (2007). Early child care and the school readiness of children from Mexican immigrant families. International Migration Review, 41, 152181.

Datar, A. (2006). Does delaying kindergarten entrance give children a head start? Economics of Education Review, 25(1), 4362. doi:10.1016/j.econedurev.2004.10.004

Desai, S., Chase-Lansdale, P. L., & Michael, R. T. (1989). Mother or market? Effects of maternal employment on the intellectual ability of 4-year-old children. Demography, 26(4), 545561. doi:10.2307/2061257

DiPerna, J. C., Lei, P.-W., & Reid, E. E. (2007). Kindergarten predictors of mathematical growth in the primary grades: An investigation using the Early Childhood Longitudinal Study--Kindergarten cohort. Journal of Educational Psychology, 99(2), 369379. doi:10.1037/0022-0663.99.2.369

Elder, T. E., & Lubotsky, D. H. (2009). Kindergarten entrance age and childrens achievement: Impacts of state policies, family background, and peers. Journal of Human Resources, 44(3), 641683. Retrieved from http://www.econ.wisc.edu/workshop/Fall 2008/ELApril4.pdf

Fiese, B. H., Tomcho, T. J., Douglas, M., Josephs, K., Poltrock, S., & Baker, T. (2002). A review of 50 years of research on naturally occurring family routines and rituals: Cause for celebration? Journal of Family Psychology, 16(4), 381390. doi:10.1037/0893-3200.16.4.381

Fowler, S. A., Chandler, L. K., Johnson, T. E., & Stella, E. (1988). Individualizing family involvement in school transitions: Gathering information and choosing the next program. Journal of the Division for Early Childhood, 12(3), 208216.

Galindo, C., & Fuller, B. (2010). The social competence of Latino kindergartners and growth in mathematical understanding. Developmental Psychology, 46(3), 579592. doi:10.1037/a0017821

Gershenson, S., Jacknowitz, A., & Brannegan, A. (2014). Are student absences worth the worry in U.S. primary schools? American University. Washington, DC. Retrieved from http://nebula.wsimg.com/37dfa768c0bceea23bed8cd7083da276?AccessKeyId=33C759F2990E6F78DB85&disposition=0&alloworigin=1

Gottfried, M. A. (2014). Classmates with disabilities and students noncognitive outcomes. Educational Evaluation and Policy Analysis, 36(1), 2043. doi:10.3102/0162373713493130

Gottfried, M. A. (2015a). Can center-based childcare reduce the odds of early chronic absenteeism? Early Childhood Research Quarterly, 32, 160173.

Gottfried, M. A. (2015b). Prekindergarten and kindergarten center-based child care and students early schooling outcomes. Teachers College Record, 117(11), 128.

Gresham, F. M., & Elliott, S. N. (1990). The Social Skills Rating System. Circle Pines, MN: American Guidance Service.

Hanushek, E. A., Kain, J. F., & Rivkin, S. G. (2002). Inferring program effects for special populations: Does special education raise achievement for students with disabilities? Review of Economics and Statistics, 84(4), 584599. doi:10.1162/003465302760556431

Herbst, C. M., & Tekin, E. (2010). Child care subsidies and child development. Economics of Education Review, 29(4), 618638. doi:10.1016/j.econedurev.2010.01.002

Holland, J. V., Kaplan, D. M., & Davis, S. D. (1974). Interschool transfers: A mental health challenge. Journal of School Health, 44(2), 7479.

Ingersoll, B., Schreibman, L., & Stahmer, A. C. (2001). Brief report: Differential treatment outcomes for children with autistic spectrum disorder based on level of peer social avoidance. Journal of Autism and Developmental Disorders, 31, 343349.

Jinnah-Ghelani, H. A., & Stoneman, Z. (2009). Elements of successful inclusion for school-age children with disabilities in childcare settings. Child Care in Practice, 15(3), 175191.

Konold, T. R., & Pianta, R. C. (2007). The influence of informants on ratings of childrens behavioral functioning: A latent variable approach. Journal of Psychoeducational Assessment, 25(3), 222236. doi:10.1177/0734282906297784

Ladd, G. W., & Price, J. M. (1987). Predicting childrens social and school adjustment following the transition from preschool to kindergarten. Child Development, 58(5), 11681189.

Landis, L. (1992). Marital, employment, and childcare status of mothers with infants and toddlers with disabilities. Topics in Early Childhood Special Education, 12, 496507.

Leland, N. L., Garrard, J., & Smith, D. K. (1994). Comparison of injuries to children with and without diabilities in a day-care center. Developmental and Behavioral Pediatrics, 15(6), 402408.

Loeb, S., Bridges, M., Bassok, D., Fuller, B., & Rumberger, R. W. (2007). How much is too much? The influence of preschool centers on childrens social and cognitive development. Economics of Education Review, 26(1), 5266. doi:10.1016/j.econedurev.2005.11.005

Loeb, S., Fuller, B., Kagan, S. L., & Carrol, B. (2004). Child care in poor communities: Early learning effects of type, quality, and stability. Child Development, 75(1), 4765. doi:10.1111/j.1467-8624.2004.00653.x

Magnuson, K. A., Meyers, M. K., Ruhm, C. J., & Waldfogel, J. (2004). Inequality in preschool education and school readiness. American Educational Research Journal, 41(1), 115157. doi:10.3102/00028312041001115

Magnuson, K. A., Ruhm, C., & Waldfogel, J. (2007). Does prekindergarten improve school preparation and performance? Economics of Education Review, 26(1), 3351. doi:10.1016/j.econedurev.2005.09.008

McGee, G. G., Morrier, M. J., & Daly, T. (1999). An incidental teaching approach to early intervention for toddlers with autism. Journal of the Association for Persons with Severe Handicaps, 24, 133146.

McIntyre, L. L., Blacher, J., & Baker, B. L. (2010). The transition to school: Adaptation in young children with and without intellectual disability. Journal of Intellectual Disability Research, 50, 259263.

Morrissey, T. W. (2009). Multiple child-care arrangements and young childrens behavioral outcomes. Child Development, 80(1), 5976. doi:10.1111/j.1467-8624.2008.01246.x

NICHD Early Child Care Research Network. (2006). Child-care effect sizes for the NICHD study of early child care and youth development. American Psychologist, 61(2), 99116. Retrieved from doi:10.1037/0003-066x.61.2.99

Pears, K. C., Kim, H. K., Healy, C. V., Yoerger, K., & Fisher, P. A. (2015). Improving child self-regulation and parenting in families of pre-kindergarten children with developmental disabilities and behavioral difficulties. Prevention Science, 16, 222232.

Phelps, L. A., & Hanley-Maxwell, C. (1997). School-to-work transitions for youth with disabilities: A review of outcomes and practices. Review of Educational Research, 67(2), 197226. Retrieved from http://www.jstor.org/stable/info/1170626

Pianta, R. C., & Cox, M. J. (1999). The transition to kindergarten. Baltimore, MD: Paul H. Brookes.

Royston, P. (2004). Multiple imputation of missing values. The Stata Journal, 4, 227241.

Scarr, S. (1998). American child care today. American Psychologist, 53(2), 95108. doi:10.1037/0003-066x.53.2.95

Shakuntala Walker, J., & Melvin, J. K. (2010). Emotional disorders (in children and adolescents). In J. H. Stone & M. Blouin (Eds.), International encyclopedia of rehabilitation. Retrieved from http://cirrie.buffalo.edu/encyclopedia/en/article/7/

Snyder, T. D., & Dillow, S. A. (2012, June). Digest of Education Statistics, 2011 (Publication No. NCES 2012-001). Washington, DC: National Center for Education Statistics. Retrieved from http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2012001

Spain, D., & Bianchi, S. (1996). Balancing act: Motherhood, marriage, and employment among American women. New York, NY: Russell Sage Foundation.

Stahmer, A. C., & Carter, C. (2005). An empirical examination of toddler development in inclusive childcare. Early Child Development and Care, 175(4), 321333.

Takanishi, R. (2004). Leveling the playing field: Supporting immigrant children from birth to eight. Future of Children, 14(2), 6180.

Todd, P. E., & Wolpin, K. I. (2003). On the specification and estimation of the production function for cognitive achievement. Economic Journal, 114(485), F3F33. Retrieved from http://www.jstor.org/stable/info/3590137

Tourangeau, K., Nord, C., Le, T., Sorongon, A. G., Hagerdorn, M. C., Daly, P., & Najarian, M. (2013, July). Users manual for the ECLS-K:2011 kindergarten data file and electronic codebook (Publication No. NCES 2013-061). Washington, DC: National Center for Education Statistics. Retrieved from http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2013061

Visser, S., Danielson, M. L., Bitsko, R. H., Holbrook, J. R., Kogan, M. D., Ghandour, R. M., . . . Blumberg, S. J. (2014). Trends in the Parent-Report of Health Care Provider-Diagnosed and Medicated Attention-Deficit/Hyperactivity Disorder: United States, 20032011. Journal of the American Academy of Child & Adolescent Psychiatry, 53(1), 3446.

Votruba-Drzal, E., Li-Grining, C. P., & Maldonado-Carreño, C. (2008). A developmental perspective on full- versus part-day kindergarten and childrens academic trajectories through fifth grade. Child Development, 79(4), 957978. Retrieved from http://www.jstor.org/stable/info/27563532

Walton, T. (2001). Why inclusion benefits everyone - insights from a parent. Child Care Information Exchange, 139, 7679.

Waterman, C., McDermott, P. A., Fantuzzo, J. W., & Gadsden, V. L. (2012). The matter of assessor variance in early childhood educationOr whose score is it anyway? Early Childhood Research Quarterly, 27(1), 4654. doi:10.1016/j.ecresq.2011.06.003

West, J., Denton, K., & Germino-Hausken, E. (2000, February). Americas kindergartners (Publication No. NCES 2000-070). Washington, DC: National Center for Education Statistics. Retrieved from http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2000070

White, H. (1980). A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica, 48(4), 817838. Retrieved from http://www.jstor.org/stable/info/1912934

Wolery, M., Holcombe, A., Venn, M. L., Brookfield, J., Huffman, K., Schroeder, C., & Fleming, L. A. (1993). Mainstreaming in early childhood programs: Current status and relevant issues. Young Children, 49(1), 7884.

Wooldridge, J. M. (2002). Econometric analysis of cross section and panel data. Cambridge, MA: MIT Press.

Yamauchi, C., & Leigh, A. (2011). Which children benefit from non-parental care? Economics of Education Review, 30(6), 14681490. doi:10.1016/j.econedurev.2011.07.012

Youngblade, L. M. (2003). Peer and teacher ratings of third- and fourth-grade childrens social behavior as a function of early maternal employment. Journal of Child Psychology and Psychiatry, 44(4), 477488. doi:10.1111/1469-7610.00138

Cite This Article as: Teachers College Record Volume 119 Number 2, 2017, p. 1-37 https://www.tcrecord.org ID Number: 21644, Date Accessed: 10/22/2021 12:16:30 AM Purchase Reprint Rights for this article or review