Identifying Mechanisms Through Which Preschool Problem Behavior Influences Academic Outcomes

Abstract

Given theoretical and empirical support for the importance of peer play within the preschool classroom to early learning, the present study tested the hypothesis that associations between teacher-reported problem behavior and academic skills were mediated by difficulties in peer play (disruptive and disconnected play), for a representative sample (N = 507 children across 46 classrooms) enrolled in a large urban Head Start program. The study also tested whether mediation was moderated by child sex (e.g., whether the mechanism through which problem behavior was associated with learning differed for boys and girls) controlling for the multilevel nature of these data. Findings from structural equation models provided evidence for partial mediation, whereby associations between internalizing behavior and academic skills were partially explained through their effects on disconnected play; however, mediation was present for girls only. Implications for theory, research, and practice are discussed as they support low-income children’s social and academic development through the context of peer play.

Keywords

classroom behavior school readiness interactive peer play sex differences

Preschool problem behavior is negatively associated with children’s engagement in formative learning and socialization activities that support kindergarten readiness (Raver, 2002; Rimm-Kaufman, Pianta, & Cox, 2000). At greatest risk are children living in poverty exposed to early ecological stressors that increase their likelihood of experiencing problem behavior in the preschool classroom (Barbarin et al., 2006; McLoyd, 1998). In addition, low-income children face concurrent academic risks, with early reading and mathematics skills on average 1 standard deviation below their more advantaged peers at kindergarten entry (Magnuson & Duncan, 2006). Although direct associations between behavioral and academic risks are documented, research examining the mechanisms linking problem behavior to academic skills is limited for low-income children. To implement classroom interventions during early childhood when developmental skills are most amenable to change, more research is needed to identify malleable skills that could explain associations between problem behavior and academic skills for this vulnerable population.

One important mechanism that supports learning is children’s engagement with peers, often observed within the context of free play or small group cooperative activities within early childhood classrooms (Fisher, Hirsh-Pasek, Golinkoff, Singer, & Berk, 2011). A recent line of research conducted within Head Start classrooms has found that children who display productive and socially competent behavior within the context of interactive play demonstrate concurrent academic and learning-related strengths (Coolahan, Fantuzzo, Mendez, & McDermott, 2000; J. Fantuzzo, Sekino, & Cohen, 2004). Unfortunately, children who display problem behavior often exhibit difficulties engaging appropriately with peers during play, placing them at a disadvantage over time, socially and academically (Vaughn, Hogan, Lancelotta, Shapiro, & Walker, 1992). Given theoretical and empirical support for sex differences in social and emotional skills during early childhood, it is also important to consider whether associations among early problem behavior, peer play, and academic skills might differ across boys and girls.

In recent studies, externalizing and internalizing problem behaviors are associated with negative peer play interactions (such as disruptive or disconnected play) in preschool (Fantuzzo, Bulotsky, McDermott, Mosca, & Lutz, 2003; Fantuzzo, Bulotsky-Shearer, Fusco, & McWayne, 2005) and across the transition to kindergarten (Bulotsky-Shearer, Domínguez, Bell, Rouse, & Fantuzzo, 2010). Given what we understand as the role of peers in supporting learning, it is possible that difficulties engaging in peer play may account for some of the association between problem behavior and academic skills. Therefore, the purpose of the present study was to examine whether teachers’ observations of children’s difficulties engaging in peer play mediated associations between teacher-reported externalizing and internalizing problems and academic skills for a representative sample of low-income children. A secondary aim was to determine whether this mechanism differed for boys and girls. Given the important role of the preschool teacher in observing and assessing children’s social, emotional, and academic competence, we used measures (observational and rating scales) developed and validated for use by Head Start teachers (Fantuzzo, Coolahan, Mendez, McDermott, & Sutton-Smith, 1998; Fantuzzo, Hightower, Grim, & Montes, 2002; Lutz, Fantuzzo, & McDermott, 2002).

Theoretical Framework

Several theoretical models provide support for the need to examine the role of peer play interactions in children’s learning. First, a developmental–ecological model provides a broad framework for understanding the influence of proximal, dynamic processes within the classroom (such as interactions between children, peers, teachers, and instructional materials) on children’s behavioral adjustment and learning (Bronfenbrenner & Morris, 1998; Hamre & Pianta, 2007). In this model, opportunities and challenges present within the classroom setting transact with children’s developmental capacities (such as self-regulation, expressive or receptive language, and attention or persistence skills) to support or hinder children’s successful negotiation of these contexts. Peer play interactions are seen as naturally occurring opportunities in which children are intrinsically motivated to engage (Mendez, McDermott, & Fantuzzo, 2002). Through these interactions, children practice and develop skills that they can carry forward as they encounter future social and academic challenges (Cicchetti, Toth, & Maughan, 2000).

In addition, social learning and constructivist theories link peer play interactions and learning (Ladd, 2007). In both models, peer play provides children with opportunities to actively process and learn social information. Children learn about themselves and their social behavior, peers and peers’ behavior, and the situations in which they are interacting with peers through play (Ladd, 2007). Particularly through cooperative activities (e.g., block building or pretend play), constructivist theorists suggest that early skills in the cognitive domain can be practiced, modeled, reinforced, and extended by peers (for reviews, see Coplan & Arbeau, 2009; Fisher et al., 2011). In these cooperative exchanges, children acquire not only social competencies but also cognitive (such as language and mathematics skills), self-regulatory, and learning-related skills (e.g., problem solving, cognitive flexibility, turn-taking, persistence, and creativity; Fisher et al., 2011; Nicolopoulou, McDowell, & Brockmeyer, 2006; Wentzel & Watkins, 2002).

These theoretical models, however, suggest that children with difficulty engaging in positive peer play interactions within the classroom may be placed at risk. In fact, research documents that children with difficulties establishing and engaging in positive peer interactions during kindergarten display less school-liking, less engagement in classroom learning, and lower academic performance (Ladd, Kochenderfer, & Coleman, 1996). Thus in the present study, we were interested in testing a mediation model examining whether children who exhibited problem behavior early in the year demonstrated lower academic skills at the end of the year, in part because of their difficulties engaging in peer play in the classroom.

Identifying Mechanisms Linking Preschool Problem Behavior and Academic Skills

Externalizing behavior, such as aggression and inattention, is linked to reading delays (Campbell, Shaw, & Gilliom, 2000), language deficits (Qi & Kaiser, 2003), and literacy and mathematics difficulties (Domínguez Escalón & Greenfield, 2009). Internalizing behavior (such as shyness or withdrawal) is associated with lower expressive and receptive vocabulary, and lower mathematics skills (Dobbs, Doctoroff, Fisher, & Arnold, 2006; Fantuzzo et al., 2003; Fantuzzo et al., 2005). Although these findings help us understand that children displaying problem behavior exhibit lower academic competencies, less is known about the reasons why these associations are present. To inform intervention efforts, recent calls from the field highlight the need to examine mechanisms that explain these direct associations (McWayne, Hahs-Vaughn, Cheung, & Green, 2012; Snow, 2007). For example, research suggests that within early childhood classrooms, children who exhibit problem behavior receive less instruction, less positive feedback from teachers, and engage in fewer socially mediated learning activities with peers (Blair, 2002; Bulotsky-Shearer et al., 2010; Raver et al., 2008). Given the nature of learning within preschool, one such mechanism may be children’s engagement with peers during play. In addition, peer play interactions are naturally occurring opportunities easily observed by the teacher and amenable to intervention that can support children’s social, emotional, and learning needs (Zigler, Singer, & Bishop-Josef, 2004).

Difficulties in Peer Play Within the Head Start Classroom

Two distinct types of preschool play interactions characterize children’s difficulties engaging with peers during play: disruptive play and disconnected play (Arnold, Homrok, Ortiz, & Stowe, 1999; Fantuzzo et al., 1998). Disruptive play is characterized by behaviors that actively interfere with play, including starting physical fights with peers, taking objects from peers, excluding peers from play, and tattling to the teacher (Fantuzzo et al., 1998). Disconnected play is characterized by a lack of engagement in play either due to the child being solitary and passive, or socially anxious and fearful of initiating or joining in play with peers. In the classroom, children often play alone or anxiously hover around their peers (Coplan & Arbeau, 2009; Hart et al., 2000).

Children who exhibit difficulties in peer play interactions often exhibit concurrent social, behavioral, and academic difficulties. In research conducted in Head Start classrooms, difficulties in peer play are associated with higher classroom problem behavior and lower learning outcomes (Fantuzzo et al., 2003; Fantuzzo et al., 2005; Vaughn et al., 1992). Externalizing problem behavior, such as aggressive and oppositional behavior, is associated with disruptive peer play (Coolahan et al., 2000; Fantuzzo et al., 2004) and internalizing problem behavior, such as passive, shy, or anxious behavior, is associated with disconnected peer play (Bulotsky-Shearer, Fantuzzo, & McDermott, 2008; Fantuzzo et al., 2003). In addition, disruptive and disconnected peer play are negatively associated with learning-related and self-regulatory behaviors that support academic skill development (Dobbs et al., 2006; Domínguez Escalón & Greenfield, 2009; Fantuzzo et al., 2004; Hampton & Fantuzzo, 2003; Mendez et al., 2002).

Identifying Mechanisms

A series of recent studies with low-income children have tested mediation models examining whether social and learning-related competencies might explain associations between classroom problem behavior and academic outcomes (Domínguez Escalón & Greenfield, 2009; McWayne & Cheung, 2009). Findings suggested that children who displayed problem behavior were less likely to develop competencies such as initiation, attention, persistence, and competence motivation in the context of instruction and positive social engagement in group learning activities, which in turn led to lower academic outcomes during preschool and across the transition to first grade.

In a recent study, Bulotsky-Shearer, Bell, Romero, and Carter (2012) extended this line of research and examined whether interactive peer play skills observed by the classroom teacher early in the Head Start year mediated associations between problem behavior and language, literacy, and mathematics skills. They found evidence for full mediation whereby children who displayed early problem behavior were less likely to exhibit positive interactive peer play skills that could support learning and, as a consequence, demonstrated lower academic outcomes. Although this study identified interactive peer play skills as an important mediating mechanism, to date no studies have examined whether difficulties in peer play may explain the negative association between early externalizing or internalizing problem behaviors and academic skills.

Child Sex as a Moderator

According to Eleanor Maccoby (1998), girls and boys grow up in different “cultures” (p. 32) simply due to their sex. Gender-typed behavior, behavior that is considered normative for each sex, becomes rigid and inflexible during this age. In a conceptual model proposed by Rose and Rudolph (2006), these sex differences are reflected in sex-segregated peer relationships and play interactions, which foster distinct skills and behavior over time. In girls’ peer groups, there are social expectations to engage in prosocial behavior that are reinforced by peers and adults; disruptive or aggressive behaviors are more likely to be reprimanded. In addition, even as early as preschool, girls are more likely than boys to define themselves in terms of their relationships with peers (Rose & Rudolph, 2006).

Boys typically display greater externalizing behavior problems and more disruptive and disconnected play than girls (Coolahan et al., 2000; Gmitrova, Podhajecká, & Gmitrov, 2009; Qi & Kaiser, 2003). Few differences are found between preschool girls and boys in internalizing behavior (Chen, 2010; Keenan & Shaw, 1997); however, scholars suggest that for boys displaying internalizing behavior, there may be greater negative social consequences because of the gendered expectations of teachers and parents regarding the appropriateness of social withdrawal in boys (Henderson, Fox, & Rubin, 2001).

Few early childhood studies have examined whether boys and girls differ in the pattern of associations among problem behavior, peer play, and academic skills, and findings from these studies are mixed. Several studies suggest that prosocial and problem behavior may play a more important role in supporting academic skills for girls. In their preschool sample, Zevenbergen and Ryan (2010) found a significant negative association between attention problems and academic skills for girls, but not for boys. Olson and Hoza (1993) found that preschool girls with conduct problems displayed lower cognitive-verbal skills in kindergarten; however, this association was not present for boys. A recent study conducted with Head Start children found a similar pattern, with a stronger positive relationship between prosocial behavior, literacy, and mathematics skills for girls than for boys (Bierman, Torres, Domitrovich, Welsh, & Gest, 2009). However, other studies of low-income preschool children find that externalizing behavior and negative peer relationships are more strongly associated with lower language skills for boys than for girls (Stowe, Arnold, & Ortiz, 2000). No study to date has examined whether difficulties in peer play indirectly influence associations between problem behavior and academic skills and whether this mechanism differs for boys and girls.

Study Purpose

Research examining the mechanisms through which problem behavior is associated with academic skills is limited, particularly for low-income preschool children who are at risk for problem behavior and poor academic readiness for kindergarten. Therefore, the purpose of the present study was to extend previous research by focusing on difficulties in peer play as a potential pathway through which problem behavior influenced academic skills in a sample of low-income preschool children. In addition, the study examined whether mediation was moderated by child sex.

Based on previous research and theory, we had several hypotheses. First, we expected that internalizing and externalizing behaviors would be differentially associated with disconnected and disruptive play, respectively, and that disconnected play would be associated directly with lower academic skills. Second, we hypothesized that difficulties in peer play within the preschool classroom would mediate associations between early problem behavior and academic skills, and that different pathways would emerge for externalizing and internalizing problems. Externalizing behavior (aggressive, inattentive, or oppositional behavior) would be associated with disruptive play in the classroom, and therefore interfere with children’s opportunity to practice and acquire academic skills during cooperative play-based activities (Campbell et al., 2000). For internalizing behavior, we hypothesized that children displaying shy or withdrawn behavior would exhibit greater disconnected play (less engagement, less initiation, avoidance, and difficulty initiating play-based interactions with their peers), which would result in lower learning outcomes (Hughes & Coplan, 2010). Third, we expected that difficulties in peer play would mediate associations between problem behavior and academic skills differently for boys and girls. We expected that (a) girls displaying internalizing behavior would exhibit lower academic skills as a function of disconnected play, whereas (b) boys displaying externalizing behavior would exhibit lower academic skills as a function of engaging more in disruptive play.

Method

Participants

Participants included 507 children across 46 classrooms enrolled in a large urban school district Head Start program in the northeast. Participating classrooms were randomly selected to be representative of the six geographic clusters in the school district. At the time of the study, the school district Head Start program served approximately 5,000 children (in more than 200 classrooms) in the most impoverished areas of the city. At the beginning of the year, children in the study ranged in age from 36 to 73 months (M = 53.8, SD = 6.6 months) and sex was split evenly with 48% male. Children were predominantly African American (74%) with 12% Caucasian, 8% Hispanic, 6% Asian or Other. Annual family income of 94% of families was less than US$12,000.

Classrooms had approximately 18 children, taught by a certified teacher and a teacher assistant. All lead teachers were credentialed in early childhood education with at least a bachelor’s degree. On experience, 35% of teachers reported having taught in Head Start for less than 10 years, 27% between 10 and 20 years, and 38% more than 20 years. Teachers were predominantly Caucasian (66%), with 31% African American.

Measures

Classroom emotional and behavioral adjustment

The Adjustment Scales for Preschool Intervention (ASPI; Lutz et al., 2002) was used in this study to assess problem behavior across routine classroom situations at the beginning of the preschool year. The scale’s 144 behavioral items (122 items reflect problem behaviors, 22 reflect positive behaviors) are presented within the context of 22 classroom situations, including interactions with the teacher, relationships with peers, involvement in structured and unstructured classroom activities, and games and play. Teachers complete the scale by endorsing as many behaviors that apply to the child in each classroom situation. Construct validity studies with urban, low-income preschool populations have revealed five distinct and reliable problem behavior dimensions that assess internalizing and externalizing behaviors: Aggressive, Oppositional, Inattentive/Hyperactive, Withdrawn/Low Energy, and Socially Reticent, each with adequate internal consistency (Cronbach’s α coefficients of .92, .78, .79, .85, and .79, respectively, for the present study sample). Convergent and divergent validity was established with constructs of interactive peer play, behavior problems, temperament, emotion regulation, language skills, and learning behaviors (Bulotsky-Shearer & Fantuzzo, 2004; Lutz et al., 2002).

For the present study, two higher order latent factors were used (internalizing and externalizing behaviors), with Cronbach’s alpha coefficients of .72 and .78, respectively. Internalizing behavior included withdrawn/low-energy (e.g., “cannot work up the energy to face anything new, sits so quietly don’t know if attending”) and socially reticent behavior (e.g., “waits for teacher to greet first”). Externalizing behavior included aggressive (e.g., “tries to push in front of others in line”), oppositional (e.g., “helps with jobs unless in a bad mood” or “sometimes in an unfriendly mood with teacher”), and inattentive/hyperactive behavior (e.g., “answers questions before taking time to think” or “talks, gazes around”). Children’s raw scores were converted to standardized T-scores (M = 50, SD = 10) based on the published factor structure (Lutz et al., 2002).

Difficulties in peer play

The teacher version of the Penn Interactive Peer Play Scale (PIPPS-T; Fantuzzo et al., 1998) was used to assess children’s peer play within the classroom at the beginning and end of the preschool year. The PIPPS-T is a 32-item rating scale used to measure common play behaviors that facilitate or interfere with prosocial peer interactions in the classroom. The PIPPS-T was developed in collaboration with Head Start parents and teachers specifically for use with low-income, urban Head Start children. Construct validity studies of the PIPPS-T have revealed three dimensions: Play Interaction, Play Disruption, and Play Disconnection, each demonstrating high internal consistency (Cronbach’s α = .92, .91, and .89, respectively, for the present study sample).

For the present study, the Play Disruption and Play Disconnection scales were used as indicators of difficulties in peer play. Items on the Play Disruption scale reflect antisocial play behaviors that interfere with play interactions, such as “starts fights and arguments” and “disrupts play of others.” Items on the Play Disconnection scale reflect avoidant behaviors that are characterized by a lack of play initiation or interaction, such as “hovers outside play group,” “wanders aimlessly,” and “confused in play.” Convergent and divergent validity has been established using direct observations of play, peer sociometrics, and measures of learning behaviors, temperament, emotion regulation, psychological adjustment, and social skills (Coolahan et al., 2000; Fantuzzo et al., 1998; Mendez et al., 2002). Although correlated with the ASPI Internalizing and Externalizing scales, both PIPPS dimensions used in the present study capture unique play behavior specifically observed by the teacher within the context of play interactions which is distinct from broader problem behavior exhibited by children across several routine classroom activities over the course of the preschool day.

Academic skills

The cognitive skills dimension of the Child Observation Record (COR; High Scope Educational Research Association, 1992) was used to measure children’s spring academic skills. The COR is a 30-item observation-based tool designed for use with children ages 2½ to 6 years in early childhood settings. Teachers collect anecdotal observations of children’s developmental functioning and record children’s skill levels along a 5-point continuum. Exploratory factor analysis with low-income preschool children yielded three factors: Cognitive Skills, Social Engagement, and Coordinated Movement with strong criterion-related validity (Fantuzzo et al., 2002; Sekino & Fantuzzo, 2005). The Cognitive Skills dimension of the COR (Cronbach’s α = .95) assesses children’s emergent literacy, numeracy, and problem solving competencies. Examples of some of the 13 items on this scale include, “knowledge about books, beginning reading and beginning writing, sorting, arranging materials in order, comparing numbers of objects, and counting objects.”

Procedures

Archival data collection

This study was a secondary analysis of data collected as part of a larger collaborative university research partnership with an urban public school district Head Start program (e.g., Bulotsky-Shearer et al., 2008). Participating teachers were recruited by the program’s six educational coordinators. For the present study, data included (a) demographic information routinely collected by the program on Head Start children and staff, (b) teacher assessments of children’s emotional and behavioral adjustment (ASPI) and peer play behaviors (PIPPS), and (c) teacher observations of children’s academic skills (COR). In the fall and spring of the school year, each teacher completed ASPI and PIPPS assessments on all children in their classroom. In the spring 6 months after fall data collection, as part of a districtwide assessment initiative, teachers completed observations of children’s academic skills (COR) for a random subsample of five boys and five girls within their classroom. The ASPI, PIPPS, and COR were collected by the program as part of a university research partnership measurement development initiative to support the fulfillment of the federal Head Start assessment requirement (Performance Standard, 1304.20; U.S. Department of Health and Human Services [USDHHS], 1996).

Data Analytic Approach

Structural equation modeling (SEM) was used to test the mediation models using Mplus Version 6.0 (L. K. Muthén & Muthén, 1998–2010). SEM was chosen as the most appropriate data analytic strategy given that the models (a) can account for measurement error through the creation of latent variables, (b) include multiple observed and latent predictors and mediators in a single model simultaneously (Kline, 2005), and (c) test direct and indirect relationships (i.e., mediation) while accounting for the multilevel nature of the data (Iacobucci, Saldanha, & Deng, 2007). Missing data were accounted for in the models using Full Information Maximum Likelihood (FIML; Kline, 2005). FIML uses all available data for each case when estimating parameters and has been shown to be unbiased when data are missing completely at random (MCAR; Enders & Bandalos, 2001).

The mediation model was analyzed in two steps. First, structural models were examined to determine whether the associations between fall internalizing and externalizing problem behaviors and spring academic skills were mediated by spring disconnected and disruptive plays. Two latent variables were created for fall internalizing and fall externalizing problem behaviors based on previous factor analytic work (Lutz et al., 2002): (a) Fall externalizing problem behavior comprised aggressive, oppositional, and hyperactive/inattentive behavior and (b) fall internalizing problem behavior comprised socially reticent and withdrawn/low-energy behavior. In all models, fall disruptive and disconnected play variables were entered to account for variance attributable to fall peer play. In addition, age (in months), sex, and minority status were included as covariates of spring disruptive and disconnected plays and academic skills. Demographic variables were dummy coded as follows: sex, 0 = male, 1 = female; and minority status, 0 = White, 1 = African American, Hispanic/Latino, or Other.

In the second step, multiple group analyses were conducted to determine if mediation was dependent on children’s sex (Preacher, Rucker, & Hayes, 2007). For these analyses, the fit of two models to the data was compared. The first unrestricted model allowed parameters to be estimated freely across boys and girls testing the hypothesis that mediation varied (was different) across sex. The second model tested a restricted model in which parameters were constrained to be equal across sex testing the assumption that mediation did not vary across sex. Given the multilevel structure of the data (children nested within classrooms), a sandwich estimator was used to adjust the standard errors of the parameters as would be done within a multilevel framework using the option TYPE = COMPLEX in Mplus (B. Muthén, du Toit, & Spisic, 1997). In addition, the more conservative Satorra–Bentler scaled χ² difference test was used to compare the relative fit of structural models (Satorra & Bentler, 2001). A nonsignificant χ² difference test indicated that the two models were not significantly different in terms of model fit, and therefore, the more parsimonious model (where parameters were constrained to be equal across groups) should be retained (Steiger, Shapiro, & Browne, 1985).

For all models, the χ² test of model fit was used to assess the fit of the overall model to the data; lack of significance (p < .05) indicated acceptable model fit (Kline, 2005). However, a significant χ² test of model fit was still considered acceptable given that the χ² statistic is often statistically significant with large sample sizes (Bollen & Long, 1993). Two additional fit indices were used to assess closeness of fit: the Bentler comparative fit index (CFI; Bentler, 1990), and the root mean square error of approximation (RMSEA; Steiger & Lind, 1980). Values for CFI greater than .90 and values for the RMSEA of .08 or less were considered acceptable (Browne & Cudeck, 1992). Sample size was more than adequate (N = 507) to detect a close model fit (for all models, dfs ranged from 30–79, requiring minimal sample sizes ranging from 74–129 to detect an acceptable fit; Preacher & Coffman, 2006).

Results

Descriptive Statistics

To ensure data were normally distributed, all variables were examined for outliers, homoscedasticity, and kurtosis. No assumptions were found to be violated. Means and standard deviations of all variables for the entire sample, as well as by child sex, are provided in Table 1. Independent-samples t tests were conducted to determine whether significant sex differences were present in any of the variables of interest. Boys had significantly higher scores than girls on fall aggression, t(505) = 5.22, p < .001; fall inattentive/hyperactive behavior, t(505) = 3.81, p < .001; fall disruptive play, t(484) = 5.01, p < .001; and spring disruptive play, t(482) = 5.03, p < .001. Bivariate correlations between all variables are shown in Table 2.

Table 1.

Descriptive Statistics for Child Measures.

	Total			Boys			Girls
	n	M	SD	n	M	SD	n	M	SD
Fall
Low energy (ASPI)	507	49.02	6.94	243	48.60	6.37	264	49.40	7.41
Shy/withdrawn (ASPI)	507	48.74	8.32	243	48.97	7.85	264	48.53	8.75
Aggressive (ASPI)	507	49.67	8.07	243	51.57	8.81	264	47.92	6.89
Oppositional (ASPI)	507	49.89	8.27	243	50.34	8.23	264	49.48	8.30
Hyperactive/inattentive (ASPI)	507	49.28	8.40	243	50.74	8.98	264	47.94	7.61
Disruptive play (PIPPS)	486	46.58	12.26	236	49.38	11.34	250	43.94	12.48
Disconnected play (PIPPS)	491	45.69	14.96	236	46.66	13.93	255	44.79	15.83
Spring
Disruptive play (PIPPS)	484	47.61	10.12	231	49.97	10.31	253	45.45	9.46
Disconnected play (PIPPS)	485	45.66	15.44	229	46.10	15.74	256	45.27	15.19
Academic skills (COR)	486	49.49	10.54	231	49.52	10.59	255	49.46	10.52

Note. ASPI = Adjustment Scales for Preschool Intervention; PIPPS = Penn Interactive Peer Play Scale; COR = Child Observation Record. All scores represent standardized T-scores based on the respective standardization sample (M = 50, SD = 10).

Table 2.

Bivariate Correlations Between Child Measures.

	1	2	3	4	5	6	7	8	9	10
Fall
1. Low energy	__	.58**	−.01	−.01	.00	−.03	.41**	−.08	.35**	.37**
2. Shy/withdrawn		__	.07	.04	.05	−.00	.47**	−.03	.41**	−.33**
3. Aggressive			__	.61**	.65**	.69**	.20**	.64**	.18**	−.17**
4. Oppositional				__	.41**	.59**	.27**	.50**	.17**	−.07
5. Hyperactive/inattentive					__	.51**	.17**	.49**	.17**	−.23**
6. Disruptive play						__	.40**	.72**	.21**	−.07
7. Disconnected play							__	.26**	.59**	−.30**
Spring
8. Disruptive play								__	.30**	−.08
9. Disconnected play									__	−.39**
10. Academic skills										__

p < .05. **p < .01.

Mediation Model

An overall (boys and girls combined) structural model was tested examining spring disconnected and disruptive plays as mediators between fall problem behavior and spring academic skills, controlling for fall disconnected and disruptive plays, child age, sex, and minority status. This model fit the data well, χ²(30) = 64.118, p < .01; CFI = 0.976; RMSEA = 0.049. A significant indirect effect (B = −0.06, SE = 0.02, p < .01; see Table 3 for all indirect effects) indicated that spring disconnected play partially mediated the association between fall internalizing problem behavior and spring academic skills.

Table 3.

Specific Indirect Effects for All Models.

		Academic skills
Mediator	Predictor	B	SE	p
Overall model
Disconnected play
Internalizing behavior		−0.06*	0.02	0.007
Externalizing behavior		−0.02	0.01	0.081
Disruptive play
Internalizing behavior		0.00	0.00	0.839
Externalizing behavior		−0.01	0.03	0.836
Boys’ model
Disconnected play
Internalizing behavior		−0.04	0.02	0.132
Externalizing behavior		−0.02	0.01	0.077
Disruptive play
Internalizing behavior		0.00	0.00	0.733
Externalizing behavior		−0.01	0.03	0.705
Girls’ model
Disconnected play
Internalizing behavior		−0.07*	0.03	0.011
Externalizing behavior		−0.01	0.02	0.429
Disruptive play
Internalizing behavior		−0.00	0.01	0.761
Externalizing behavior		0.02	0.04	0.743

Note. Estimates are standardized coefficients.

p < .05. **p < .01.

Prior to interpreting the model, multiple group analyses were conducted to determine whether this model differed across child sex. First, a restricted model was analyzed in which parameter estimates were constrained equal across groups; this model fit the data well, χ²(79) = 156.931, p < .01; CFI = 0.953; RMSEA = 0.056. Second, an unrestricted model was tested in which parameters were estimated freely across sex; this model also fit the data well, χ²(60) = 118.042, p < .01; CFI = 0.965; RMSEA = 0.042. A χ² difference test of the nested models was then conducted, indicating that the mediation model best fit the data when the models were estimated freely, Δχ²(19) = 39.289, p < .01. Therefore, the results for each model are reported separately for boys and girls.

Direct effects for boys

Figure 1 displays the standardized parameter estimates for the direct effects in the boys’ model. There were a few significant direct effects. Boys’ age was positively associated with spring academic skills and boys’ minority status was negatively associated with spring disconnected play. Fall disconnected and disruptive plays were positively associated with their respective spring scores. Of most interest to this study, fall internalizing behavior was positively associated with spring disconnected play and fall externalizing behavior was positively associated with spring disruptive and disconnected plays. Finally, spring disconnected play was negatively associated with academic skills.

Figure 1.

Mediation model between fall problem behavior and spring academic skills through disconnected peer interactions and disruptive peer interactions (for boys only).

Indirect effects for boys

The significant indirect effect that was present in the overall model was not present in the model for boys; no significant indirect effects were found for boys (see Table 3 displaying the specific indirect effects for the overall model, the boys-only model and the girls-only model).

Direct effects for girls

Figure 2 displays the standardized parameter estimates for the direct effects in the girls’ model. Direct effects included the following. Girls’ age was positively associated with spring disruptive play and academic skills. Fall disconnected play and disruptive play were positively associated with their respective spring scores. Of most interest to this study, fall internalizing behavior was positively associated with spring disconnected play, and fall externalizing behavior was positively associated with spring disruptive play. Internalizing and externalizing problem behaviors were negatively associated with spring academic skills. Spring disconnected play was negatively associated with spring academic skills.

Figure 2.

Mediation model between fall problem behavior and spring academic skills through disconnected peer interactions and disruptive peer interactions (for girls only).

Indirect effects for girls

In the model estimated for girls, an indirect effect (B = −0.06, SE = 0.02, p < .01) indicated that spring disconnected play partially mediated the association between fall internalizing problem behavior and spring academic skills (see Figure 2 for the model for girls). No other significant indirect effects were found.

Discussion

This study was the first to examine whether difficulties in peer play interactions mediated associations between early problem behavior and academic skills for low-income children at risk for poor kindergarten readiness. Given the importance of peer play as a context for learning in preschool, we hypothesized that children who displayed problem behavior would also display difficulties in peer play which, in turn, would lead to lower academic outcomes. We tested our hypotheses in a representative sample of children enrolled in a large, urban Head Start program using a rigorous moderated mediation SEM approach. Findings provided support for the hypothesis that the association between early internalizing behavior and academic skills would be mediated by disconnected play. However, we found that this mechanism was present for girls only, partially supporting our moderation hypothesis. Our hypothesis that disruptive play would mediate the association between externalizing behavior and academic skills was not supported in our overall model, nor was it supported for boys. Findings contribute to the knowledge base by extending our understanding of the mechanisms through which internalizing behavior is indirectly associated with learning during the preschool period, particularly for girls, and offer potential avenues to intervene to foster malleable skills within early childhood classrooms serving low-income children.

Direct Associations Consistent Across Child Sex

Several patterns of direct associations were comparable for boy and girls. For both, early internalizing behavior was positively associated with disconnected play, and externalizing behavior was positively associated with disruptive play. These direct associations between problem behavior and difficulties in peer play are supported by early childhood studies documenting differential associations between internalizing behavior such as social reticence and withdrawal in the preschool classroom, and disconnected play; as well as externalizing behavior such as aggression or opposition, and disruptive play (Fantuzzo et al., 2003; Vaughn et al., 1992). In addition, we found negative associations between disconnected play and academic skills for boys and girls at the end of the year. These findings are supported by research documenting direct associations between disconnected play and a broad range of academic skills in preschool (Coolahan et al., 2000; Fantuzzo et al., 2004), kindergarten (Hampton & Fantuzzo, 2003), and third grade (Sekino, 2006). Our study extends this literature by demonstrating that these direct associations are consistent across child sex during children’s Head Start year.

Direct Associations That Differed Across Child Sex

We found that direct associations between early problem behavior and academic outcomes differed for girls and boys. For girls, externalizing and internalizing problem behaviors were negatively associated with academic outcomes. However, for boys, no significant associations were found between early problem behavior and academic skills. Research provides support for negative associations between problem behavior and academic skills during preschool (Domínguez Escalón & Greenfield, 2009; McWayne & Cheung, 2009); however, in our sample, these associations were present for girls and not for boys.

Why would early problem behavior be associated with academic outcomes for girls and not for boys? There are some early childhood studies that find that problem behavior within the classroom is more strongly negatively associated with academic skills for girls than for boys (Olson & Hoza, 1993). This suggests that behavioral difficulties may be more academically detrimental for girls. Consistent with several early childhood studies, within our sample, boys displayed higher levels of externalizing behavior than girls. Therefore, on average, it was more common for boys to display externalizing behavior, and less common for girls. From a developmental perspective, gender differences in environmental expectations and contingencies for girls’ behavior within the classroom could place girls more at risk for academic problems because they were displaying behavior that was less commonly seen or expected for their sex. Research shows that environmental contingencies for behavior shape how problem behavior is viewed or accepted, particularly if teachers’ expectations for behavior differ for boys and girls (Fabes, Martin, Hanish, Anders, & Madden-Derdich, 2003).

With regard to internalizing behavior and academic skills, it is unclear why girls displaying internalizing behavior were more vulnerable academically than boys who displayed internalizing behavior. Early childhood studies do provide evidence for negative associations between internalizing behavior and academic skills, including learning-related behaviors that support engagement in classroom learning activities (Domínguez, Vitiello, Maier, & Greenfield, 2010; Hughes & Coplan, 2010). It could be that girls displaying internalizing behavior were less likely to be identified as needing additional assistance to engage within learning activities that supported academic skills. For example, Morgan and Dunn (1988) found that girls who were having difficulty learning tended to remain quiet, making it difficult for teachers to notice that they were struggling with learning tasks, therefore they continued to struggle. In contrast, boys who were having difficulties in learning situations displayed behaviors that attracted the teacher’s attention which then led to teachers providing more academic assistance.

Mediation That Differed Across Child Sex

In our final model, our hypothesis that disconnected play would mediate associations between early problem behavior and academic skills was supported for girls only. Disconnected play partially mediated associations between early internalizing problem behavior and academic outcomes. This suggests that girls who displayed internalizing behavior early in the school year displayed lower academic skills at the end of the year in part because they were engaging in more disconnected play during the year. The finding that early internalizing problem behavior was indirectly associated with learning outcomes through its association with peer play is supported by initial studies conducted with Head Start children (Bulotsky-Shearer, Bell, et al., 2012; McWayne & Cheung, 2009). However, the finding that disconnected play mediated these associations and that this mechanism differed across sex is a unique contribution of our study.

It is important to note that disconnected play only partially explained associations between internalizing behavior and academic skills. Internalizing behavior was also directly associated with academic skills for girls. This suggests that both are important in understanding multiple pathways to academic difficulties displayed by preschool children exhibiting internalizing behavior. Future research is needed to examine these pathways in other samples of low-income children. This is critical as children exhibiting emotional and behavioral needs in Head Start are programmatically underidentified (Lopez, Tarullo, Forness, & Boyce, 2000) and children exhibiting internalizing behavior are least likely to be identified and referred for intervention services (Fantuzzo et al., 2003).

Why did difficulties in peer play not mediate associations for boys? Although for girls internalizing behavior was indirectly associated with academic skills through disconnected play, for boys, disconnected play was only directly associated with academic skills. We had expected that for boys, disruptive play would take a more central role in school adjustment and that disruptive play would mediate associations between externalizing behavior and academic skills. It is possible that in our sample, disruptive play is associated with a different set of developmental outcomes. For example, for boys, disruptive play could be more closely associated with social maladjustment (e.g., peer rejection) rather than academic skills. Elementary school studies have found that peer rejection is a salient predictor of school adjustment, particularly for boys (Buhs & Ladd, 2001). In addition, it could be the case that children who display externalizing behavior are still engaging with peers and learning activities (albeit in a disruptive or inappropriate way); thus, they are accessing teacher or peer attention that may be engaging them in learning activities during the preschool day (Fantuzzo et al., 2003).

Limitations

This study is the first to examine whether difficulties in peer play displayed by preschoolers in the Head Start classroom mediated the association between problem behavior and academic skills, and whether this mechanism was different for boys and girls. However, there are limitations that should be acknowledged. First, our sample was limited to predominantly English-speaking, African American Head Start children in a large urban school district program. Research has highlighted the important role of children’s cultural and linguistic background in children’s socialization experiences at home and school (Bulotsky-Shearer, Manz, et al., 2012). Factors such as economic constraints, cultural norms, level of acculturation, and religious practices strongly influence parents’ and teachers’ beliefs and expectations regarding children’s behavior and subsequent encouragement of peer social interactions (Fogle & Mendez, 2006; Fuller & García Coll, 2010). It would be important for future research to examine the generalizability of our findings for other culturally and linguistically diverse groups of children.

Second, the constructs in this study were measured using teacher observations and ratings of children’s behavior. These measures were chosen intentionally because they were developed and validated for use with low-income ethnic minority preschool populations and captured behaviors readily observable within the preschool classroom (McDermott, 1986). However, it is important to note that teacher characteristics may contribute to variance in children’s scores, leading to the possibility that the associations observed may in part be due to shared source variance (Hamre, Pianta, Downer, & Mashburn, 2007; Konold & Pianta, 2007). Although potential source invariance issues were mitigated by the study’s predictive design (problem behavior assessed in the beginning of the year and academic skills observed at the end of the year) and by controlling for nesting of children within classroom (by application of multilevel correction in SEM), future studies can strengthen and replicate study findings by incorporating additional measures of children’s behavior assessed by multiple sources, such as by independent observers or by parents, and direct assessments of children’s academic skills.

Directions for Future Research on Mechanisms

It should be acknowledged that associations observed, particularly the indirect effect for girls, were relatively small. Difficulties in peer play are one of many potential mechanisms through which problem behavior can influence academic skills. Several other variables that were not available for this study should be considered, including child-level skills and contextual factors. For example, future research should examine children’s cognitive and self-regulatory approaches to learning, temperament, and language skills as potential mediators given their links to learning in preschool (Snow, 2007). In addition, contextual variables should be examined such as the teacher–child relationship and the quality of classroom instruction and emotional support (Hamre & Pianta, 2007). Interactions with the teacher may be a possible mechanism through which problem behavior influences academic skills, particularly for boys who are struggling academically in the classroom. In addition, teachers’ ability to scaffold peer-mediated learning experiences for children in the classroom may influence the quality of these learning opportunities. Many of these variables in future research should be considered because they could be important points of intervention for low-income children (McWayne & Cheung, 2009).

Because this study did not find that difficulties in peer play mediated the association between problem behavior and academic skills for boys, much more research is needed to understand other factors that influence how problem behavior affects boys’ learning experiences within the preschool classroom. More research is needed using observational or sociometric methods to examine how boys displaying problem behavior are engaging in cooperative play interactions within the classroom, which might be different from girls who display problem behavior. A closer study of these distinctions will provide researchers, teachers, and mental health clinicians with more specific information to guide the development and implementation of classroom-based interventions for boys. Additional research is greatly needed to unpack how problem behavior and disruptive play influence learning experiences particularly for boys.

Implications for Policy and Practice

We found that peer play is an important context through which early internalizing behavior may influence academic learning within the Head Start classroom, particularly for girls. Historically, there has been a tension in early childhood education between the roles of didactic (teacher-directed) versus child-directed instructional approaches, and how best to achieve a balance between the two that supports developmentally appropriate learning experiences for children (Fisher et al., 2011). Peer social interactions are important mechanisms associated with learning and should be contexts or targets for intervention within early childhood programs serving low-income children. In fact, recent preschool curricula such as Tools of the Mind (Bodrova & Leong, 2007) incorporate learning experiences through peer-mediated interactions, such as peer-assisted reading activities, dramatic play, and small group cooperative learning activities designed to promote self-regulation, literacy, and mathematics skills.

However, our study also suggests that for children exhibiting internalizing behavior, more individualized approaches may be needed by the teacher to support successful peer-mediated learning experiences. Early identification of the emotional, social, and academic needs of low-income children, particularly those exhibiting internalizing behavior is critical to inform interventions that support more adaptive behavior and successful learning experiences. Rating scales based on teacher observations such as the ASPI or the PIPPS can be used to identify children’s needs within the context of daily social and learning activities within the naturalistic classroom context, by raters familiar with the child (Neisworth & Bagnato, 2004). This information can be practically applied to teachers’ instructional approaches within the classroom. Used programmatically, these assessments can support a broader understanding of where children displaying emotional and behavioral needs are struggling socially and academically, and where additional professional development and consultancy support is needed to support teachers’ efforts. Our study is a first step toward understanding why internalizing behavior is associated with learning outcomes for girls; however, much more work is needed to understand how early problem behavior (internalizing and externalizing) is associated with boys’ preschool learning experiences at the beginning of their school experiences.

Footnotes

Acknowledgements

We gratefully acknowledge the children, teachers, education specialists, and disability coordinators from the School District of Philadelphia Head Start Program who collaborated with us on this project, and without whom this study would have not been possible.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: We gratefully acknowledge that this research project was supported by funds from an Early Career Research Award from the Society for the Study of School Psychology, awarded to the first author (2009-2011).

References

Arnold

D. H.

Homrok

Ortiz

Stowe

R. M.

(1999). Direct observation of peer rejection acts and their temporal relation with aggressive acts. Early Childhood Research Quarterly, 14, 183–196.

Barbarin

Bryant

McCandies

Burchinal

Early

Clifford

Howes

(2006). Children enrolled in public pre-k: The relation of family life, neighborhood quality, and socioeconomic resources to early competence. American Journal of Orthopsychiatry, 76, 265–276.

Bentler

P. M.

(1990). Comparative fit indices in structural models. Psychological Bulletin, 107, 238–246.

Bierman

K. L.

Torres

M. M.

Domitrovich

C. E.

Welsh

J. A.

Gest

S. D.

(2009). Behavioral and cognitive readiness for school: Cross-domain associations for children attending Head Start. Social Development, 18, 305–323.

Blair

(2002). School readiness: Integrating cognition and emotion in a neurobiological conceptualization of children’s functioning at school entry. American Psychologist, 57, 111–127.

Bodrova

Leong

D. J.

(2007). Tools of the mind: A case study of implementing the Vygotskian approach in American early childhood and primary classrooms. Geneva, Switzerland: International Bureau of Education, UNESCO.

Bollen

K. A.

Long

J. S.

(1993). Testing structural equation models. Newbury Park, CA: SAGE.

Bronfenbrenner

Morris

P. A.

(1998). The ecology of developmental processes. In Damon

(Ed.), Handbook of child psychology: Theoretical models of human development (5th ed., Vol. 1, pp. 993–1028). New York, NY: Wiley.

Browne

M. W.

Cudeck

(1992). Alternative ways of assessing model fit. Sociological Methods Research, 21, 230–258.

10.

Buhs

E. S.

Ladd

G. W.

(2001). Peer rejection as antecedent of young children’s school adjustment: An examination of mediating processes. Developmental Psychology, 37, 550–560.

11.

Bulotsky-Shearer

Fantuzzo

J. W.

(2004). Adjustment Scales for Preschool Intervention: Extending validity and relevance across multiple perspectives. Psychology in the Schools, 41, 725–736.

12.

Bulotsky-Shearer

Fantuzzo

J. W.

McDermott

P. A.

(2008). An investigation of classroom situational dimensions of emotional and behavioral adjustment and cognitive and social outcomes for Head Start children. Developmental Psychology, 44, 139–154.

13.

Bulotsky-Shearer

Bell

E. R.

Romero

Carter

(2012). Preschool interactive peer play mediates problem behavior and learning for low income children. Journal of Applied Developmental Psychology, 33, 53–65.

14.

Bulotsky-Shearer

Domínguez

Bell

E. R.

Rouse

H. L.

Fantuzzo

J. W.

(2010). Relations between behavior problems in classroom social and learning situations and peer social competence in Head Start and kindergarten. Journal of Emotional and Behavioral Disorders, 18, 195–210.

15.

Bulotsky-Shearer

Manz

P. H.

Mendez

J. L.

McWayne

C. M.

Sekino

Fantuzzo

J. W.

(2012). Peer play interactions and readiness to learn: A protective influence for African American preschool children from low-income households. Child Development Perspectives, 6, 225–231. doi:10.1111/j.1750-8606.2011.00221.x

16.

Campbell

S. B.

Shaw

D. S.

Gilliom

(2000). Early externalizing behavior problems: Toddlers and preschoolers at risk for later maladjustment. Development and Psychopathology, 12, 467–488.

17.

Chen

J. J.

(2010). Gender differences in externalizing problems among preschool children: Implications for early childhood educators. Early Child Development and Care, 180, 463–474.

18.

Cicchetti

Toth

S. L.

Maughan

(2000). An ecological-transactional model of child maltreatment. In Sameroff

A. J.

Lewish

Miller

S. M.

(Eds.), Handbook of developmental psychopathology (2nd ed., pp. 689–722). Dordrecht, Netherlands: Kluwer Academic.

19.

Coolahan

Fantuzzo

Mendez

McDermott

(2000). Preschool peer interactions and readiness to learn: Relationships between classroom peer play and learning behaviors and conduct. Journal of Educational Psychology, 92, 458–465.

20.

Coplan

R. J.

Arbeau

K. A.

(2009). Peer interactions and play in early childhood. In Rubin

K. H.

Bukowski

W. M.

Laursen

(Eds.), Handbook of peer interactions, relationships, and groups (pp. 143–161). New York, NY: Guilford.

21.

Dobbs

Doctoroff

G. L.

Fisher

P. H.

Arnold

D. H.

(2006). The association between preschool children’s socio-emotional functioning and their mathematical skills. Journal of Applied Developmental Psychology, 27, 97–108.

22.

Domínguez

Vitiello

V. E.

Maier

M. F.

Greenfield

D. B.

(2010). A longitudinal examination of young children’s learning behavior: Child-level and classroom-level predictors of change throughout the preschool year. School Psychology Review, 39, 29–47.

23.

Domínguez Escalón

Greenfield

D. B.

(2009). Learning behaviors mediating the relationship between behavior problems and academic outcomes. NHSA Dialog, 12, 1–17.

24.

Enders

C. K.

Bandalos

D. L.

(2001). The relative performance of full information maximum likelihood estimation for missing data in structural equation models. Structural Equation Modeling: A Multidisciplinary Journal, 8, 430–457.

25.

Fabes

R. A.

Martin

C. L.

Hanish

L. D.

Anders

M. C.

Madden-Derdich

D. A.

(2003). Early school competence: The roles of sex-segregated play and effortful control. Developmental Psychology, 39, 848–858.

26.

Fantuzzo

J. W.

Coolahan

Mendez

McDermott

Sutton-Smith

(1998). Contextually-relevant validation of peer play constructs with African-American Head Start children: Penn Interactive Peer Play Scale. Early Childhood Research Quarterly, 13, 411–431.

27.

Fantuzzo

Hightower

Grim

Montes

(2002). Generalization of the Child Observation Record: A validity study for diverse samples of urban, low-income preschool children. Early Childhood Research Quarterly, 17, 106–125.

28.

Fantuzzo

Sekino

Cohen

H. L.

(2004). An examination of the contributions of interactive peer play to salient classroom competences for urban Head Start children. Psychology in the Schools, 41, 323–336.

29.

Fantuzzo

J. W.

Bulotsky

McDermott

Mosca

Lutz

M. N.

(2003). A multivariate analysis of emotional and behavioral adjustment and preschool educational outcomes. School Psychology Review, 32, 185–203.

30.

Fantuzzo

J. W.

Bulotsky-Shearer

R. J.

Fusco

R. A.

McWayne

(2005). An investigation of preschool emotional and behavioral adjustment problems and social-emotional school readiness competencies. Early Childhood Research Quarterly, 20, 259–275.

31.

Fisher

Hirsh-Pasek

Golinkoff

R. M.

Singer

D. G.

Berk

(2011). Playing around in school: Implications for learning and educational policy. In Pellegrini

(Ed.), The Oxford handbook of play (pp. 341–363). New York, NY: Oxford University Press.

32.

Fogle

L. M.

Mendez

J. L.

(2006). Assessing the play beliefs of African American mothers with preschool children. Early Childhood Research Quarterly, 21, 507–518.

33.

Fuller

García Coll

(2010). Learning from Latinos: Contexts, families, and child development in motion. Developmental Psychology, 46, 559–565.

34.

Gmitrova

Podhajecká

Gmitrov

(2009). Children’s play preferences: Implications for the preschool education. Early Child Development and Care, 179, 339–351.

35.

Hampton

V. R.

Fantuzzo

J. W.

(2003). The validity of the Penn Interactive Peer Play Scale with urban, low-income kindergarten children. School Psychology Review, 32, 77–91.

36.

Hamre

B. K.

Pianta

R. C.

(2007). Learning opportunities in preschool and early elementary classrooms. In Pianta

R. C.

Cox

M. J.

Snow

K. L.

(Eds.), School readiness and the transition to kindergarten in the era of accountability (pp. 49–84). Baltimore, MD: Brookes Publishing.

37.

Hamre

B. K.

Pianta

R. C.

Downer

J. T.

Mashburn

A. J.

(2007). Teachers’ perceptions of conflict with young students: Looking beyond problem behaviors. Social Development, 17, 115–136.

38.

Hart

C. H.

Yang

Nelson

L. J.

Robinson

C. C.

Olsen

J. A.

Nelson

D. A.

(2000). Peer acceptance in early childhood and profiles of socially withdrawn behaviour in China, Russia, and the United States. International Journal of Behavioral Development, 24, 73–81.

39.

Henderson

H. A.

Fox

N. A.

Rubin

K. H.

(2001). Temperamental contributions to social behavior: The moderating roles of frontal EEG asymmetry and gender. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 68–74.

40.

High Scope Educational Research Association. (1992). Child Observation Record. Ypsilanti, MI: High/Scope Educational Research Foundation.

41.

Hughes

Coplan

R. J.

(2010). Exploring processes linking shyness and academic achievement in childhood. School Psychology Quarterly, 25, 213–222.

42.

Iacobucci

Saldanha

Deng

(2007). A meditation on mediation: Evidence that structural equation models perform better than regressions. Journal of Consumer Psychology, 17, 139–153.

43.

Keenan

Shaw

(1997). Developmental and social influences on young girls’ early problem behavior. Psychological Bulletin, 121, 95–113.

44.

Kline

R. B.

(Ed.). (2005). Principles and practice of structural equation modeling (2nd ed.). New York, NY: Guilford.

45.

Konold

T. R.

Pianta

R. C.

(2007). The influence of informants on ratings of children’s behavioral functioning. Journal of Psychoeducational Assessment, 25, 222–236.

46.

Ladd

G. W.

(2007). Social learning in the peer context. In Saracho

O. N.

Spodek

(Eds.), Contemporary perspectives on socialization and social development in early childhood education (pp. 133–164). Charlotte, NC: Information Age Publishing.

47.

Ladd

G. W.

Kochenderfer

B. J.

Coleman

C. C.

(1996). Friendship quality as a predictor of young children’s early school adjustment. Child Development, 67, 1103–1118.

48.

Lopez

M. L.

Tarullo

L. B.

Forness

S. R.

Boyce

C. A.

(2000). Early identification and intervention: Head Start’s response to mental health challenges. Early Education and Development, 11, 265–282.

49.

Lutz

M. N.

Fantuzzo

McDermott

(2002). Multidimensional assessment of emotional and behavioral adjustment problems of low-income preschool children: Development and initial validation. Early Childhood Research Quarterly, 17, 338–355.

50.

Maccoby

E. E.

(1998). The two sexes: Growing up apart, coming together. Cambridge, MA: Belknap.

51.

Magnuson

K. A.

Duncan

G. J.

(2006). The role of family socioeconomic resources in the black–white test score gap among young children. Developmental Review, 26, 365–399.

52.

McDermott.

P. A.

(1986). The observation and classification of exceptional child behavior. In Brown

R. T.

Reynolds

C. R.

(Eds.), Psychological perspectives on childhood exceptionality: A handbook (pp. 136–180). New York, NY: Wiley-Interscience.

53.

McLoyd

V. C.

(1998). Socioeconomic disadvantage and child development. American Psychologist, 53, 185–204.

54.

McWayne

C. M.

Cheung

(2009). A picture of strength: Preschool competencies mediate the effects of early behavior problems on later academic and social adjustment for Head Start children. Journal of Applied Developmental Psychology, 30, 273–285.

55.

McWayne

C. M.

Hahs-Vaughn

D. L.

Cheung

Green

L. E.

(2012). National profiles of school readiness skills for Head Start children: An investigation of stability and change. Early Childhood Research Quarterly, 27, 668–683. doi:10.1016/j.ecresq.2011.10.002

56.

Mendez

J. L.

McDermott

Fantuzzo

J. W.

(2002). Identifying and promoting social competence with African-American preschool children: Developmental and contextual considerations. Psychology in the Schools, 39, 111–123.

57.

Morgan

Dunn

(1988). Chameleons in the classroom: Visible and invisible children in nursery and infant classrooms. Educational Review, 40, 3–12.

58.

Muthén

du Toit

S. H. C.

Spisic

(1997). Robust inference using weighted least squares and quadratic estimating equations in latent variable modeling with categorical and continuous outcomes. Unpublished manuscript. Retrieved from http://gseis.ucla.edu/faculty/muthen/articles/Article_075.pdf

59.

Muthén

L. K.

Muthén

B. O.

(1998–2010). Mplus user’s guide (6th ed.). Los Angeles, CA: Author.

60.

Neisworth

J. T.

Bagnato

S. J.

(2004). The mismeasure of young children: The authentic assessment alternative. Infants & Young Children, 17, 198–212.

61.

Nicolopoulou

McDowell

Brockmeyer

(2006). Narrative play and emergent literacy: Storytelling and story-acting meet journal writing. In Singer

Golinkoff

Hirsh-Pasek

(Eds.), Play = learning: How play motivates and enhances children’s cognitive and social-emotional growth (pp. 124–144). New York, NY: Oxford University Press.

62.

Olson

S. L.

Hoza

(1993). Preschool developmental antecedents of conduct problems in children beginning school. Journal of Clinical Child Psychology, 22, 60–67.

63.

Preacher

K. J.

Coffman

D. L.

(2006, May). Computing power and minimum sample size for RMSEA [Computer software]. Available from http://www.quantpsy.org/

64.

Preacher

K. J.

Rucker

D. D.

Hayes

A. F.

(2007). Addressing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research, 42, 185–227.

65.

C. H.

Kaiser

A. P.

(2003). Behavior problems of preschool children from low-income families: Review of the literature. Topics in Early Childhood Special Education, 23, 188–216.

66.

Raver

C. C.

(2002). Emotions matter: Making the case for the role of young children’s emotional development for early school readiness. Social Policy Report: Giving Child and Youth Development Knowledge Away, 16, 1–20.

67.

Raver

C. C.

Jones

S. M.

Li-Grining

C. P.

Metzger

Champion

K. M.

Sardin

(2008). Improving preschool classroom processes: Preliminary findings from a randomized trial implemented in Head Start settings. Early Childhood Research Quarterly, 23, 10–26.

68.

Rimm-Kaufman

S. E.

Pianta

R. C.

Cox

M. J.

(2000). Teachers’ judgments of problems in the transition to kindergarten. Early Childhood Research Quarterly, 15, 147–166.

69.

Rose

A. J.

Rudolph

K. D.

(2006). A review of sex differences in peer relationship processes: Potential trade-offs for the emotional and behavioral development of girls and boys. Psychological Bulletin, 132, 98–131.

70.

Satorra

Bentler

P. M.

(2001). A scaled difference chi-square test statistic for moment structure analysis. Psychometrika, 66, 507–514.

71.

Sekino

(2006). Investigation of the relationship between preschool peer play and third-grade outcomes for low-income urban students (Doctoral dissertation, University of Pennsylvania). Retrieved from http://repository.upenn.edu/dissertations/AAI3246236/

72.

Sekino

Fantuzzo

J. W.

(2005). Validity of the Child Observation Record: An investigation of the relationship between COR dimensions and social-emotional and cognitive outcomes for Head Start children. Journal of Psychoeducational Assessment, 23, 242–261.

73.

Snow

K. L.

(2007). Integrative views of the domains of child function: Unifying school readiness. In Pianta

R. C.

Cox

M. J.

Snow

K. L.

(Eds.), School readiness and the transition to kindergarten in the era of accountability. Baltimore, MD: Brookes Publishing.

74.

Steiger

J. H.

Lind

J. C.

(1980, May). Statistically based tests for the number of common factors. Paper presented at the Annual Meeting of the Psychometric Society, Iowa City, IA.

75.

Steiger

J. H.

Shapiro

Browne

M. W.

(1985). On the multivariate asymptotic distribution of sequential chi-square statistics. Psychometrika, 50, 253–264.

76.

Stowe

R. M.

Arnold

D. H.

Ortiz

(2000). Gender differences in the relationship of language development to disruptive behavior and peer relationships in preschoolers. Journal of Applied Developmental Psychology, 20, 521–536.

77.

U.S. Department of Health and Human Services. (1996). Final rule—Program performance standards for the operation of Head Start programs by grantee and delegate agencies (45 CFR Part 1304, Federal Register, 61, 57186–57227). Washington, DC: Government Printing Office.

78.

Vaughn

Hogan

Lancelotta

Shapiro

Walker

(1992). Subgroups of children with severe and mild behavior problems: Social competence and reading achievement. Journal of Clinical Child Psychology, 21, 98–106.

79.

Wentzel

K. R.

Watkins

D. E.

(2002). Peer relationships and collaborative learning as contexts for academic enablers. School Psychology Review, 31, 366–377.

80.

Zevenbergen

A. A.

Ryan

M. M.

(2010). Gender differences in the relationship between attention problems and expressive language and emerging academic skills in preschool-aged children. Early Child Development and Care, 180, 1337–1348.

81.

Zigler

E. F.

Singer

D. G.

Bishop-Josef

S. J.

(2004). Children’s play: The roots of reading. Washington, DC: Zero to Three Press.