|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
Measuring Social Competence in Preschool-Aged Children Through the Examination of Play Behaviors by Eun-Yeop Lee A thesis submitted in partial fulfillment of the requirements for the degree of Education Specialist Department of Psychological and Social Foundations College of Education University of South Florida Major Professor: Kat hy Bradley-Klug, Ph.D. Linda Raffaele-Mendez, Ph.D. John Ferron, Ph.D. Date of Approval: May 26, 2006 Keywords: early childhood, social comp etence, play behavior Copyright 2006, Eun-Yeop Lee
ii Table of Contents List of Tables iv Abstract v Chapter One Introduction 1 Statement of the problem 1 Developmental Framework 2 Social Skills Theory 4 Gender Stereotypes 6 Assessment of Social Skills 6 Play Behavior 8 Research questions 10 Hypotheses 11 Significance of the study 14 Chapter Two Literature Review Introduction 16 Social Competence in Early Childhood 19 Social Competence and Academic Outcomes 25 Relating Social competence and Emotional Competence 27 Social Skills Deficits and ADH D/Disruptive Behavior Disorder 35 Current Assessment Issues 40 Social Skills Acquisition through Play Behavior 41 Parental Factors and Presc hoolers Social Competence 43 Peer Play Interactions and Measurements of Child Development 47 Gender Issues in Measures of Social Competence 54 Guidelines for Early Childhood Assessment 58 Parent and Teacher Reports 60 Current Assessment Methods 63 Purpose of the Study 66 Chapter Three Method 69 Introduction 69
iii Participants 69 Instrumentation 75 Parent-Teacher Communication Form 75 Preschool and Kindergarten Behavior Scales 75 Penn Interactive Peer Play Scale (PIPPS) 80 Procedure 84 Protection of research participants 84 Recruitment of center for participation 84 Parent data collection 85 Teacher data collection 87 Scoring of Protocols 87 Chapter Four Results 88 Internal Reliability for the PKBS-Social Skills Scale and PIPPS 88 Descriptive Statistics for the PKBS-2 and PIPPS Systems 90 Concurrent Validity of the PI PPS-Parent with the PKBS-2 Social Skills Scale Parent 93 Concurrent Validity of the PI PPS-Teacher with the PKBS-2 Social Skills Scale Teacher 94 Differences in Group Means between Gender and PKBS-2 Social Skills Scale and PIPPS 95 Relationships between the PIPPSTeacher and the PIPPS--Parent 97 Predictive Validity of the Parent Teacher Communication Form with the PKBS and PIPPS 98 Summary of findings 100 Chapter Five Discussion 102 Concurrent Validity 103 Gender Differences in Pare nt and Teacher ratings 105 Relationships between Teacher and Parent Versions 108 Predictive Validity 111 Limitations and Implications for Future Research 112 Implications for Practice 117 Informing the Curriculum 118 Establishing methods of pare nt and teacher communication 119 Conclusions 119 References 120 Appendices 131 Appendix A: Parent and Teacher Consent forms 132 Appendix B: Instrumentation 135
iv List of Tables Table 1 Power Analysis 73 Table 2 Response Rate by Preschool Center 74 Table 3 Demographic Information by Preschool Center 74 Table 4 Internal Consistency (alpha) Reliability for the PKBS-2 Social Skills Scale and PIPPS 90 Table 5 Descriptive Statistics 92 Table 6 Correlations of PIPPSParent Factors with PKBS-2 Social Skills ScaleParent 94 Table 7 Correlations of PIPPSTeacher Factors with PKBS-2 Social Skills ScaleTeacher 95 Table 8 Mean Scores for Parent and Teacher Measures as a Function of Gender 97 Table 9 Correlations of PIPPS-Teacher Factors with PIPPS-Parent Factors 98 Table 10 Correlations of PIPPS-Teacher, PIPPS-Parent, PKBS-2 Social Skills Scale-Parent, PKBS-2 Social Skills Scale-Parent with the Parent-Teacher Communication form 99
v Measuring Social Competence in Preschool-Aged Children Through the Examination of Play Behaviors EUN-YEOP LEE ABSTRACT For young children, a primary component of social competence is establishing effective interactions with peers during play. To inform the development of practices that promote this competency starting in early childhood, quality asse ssment measures are needed. These instruments must have the capac ity to establish linkages between the home and school as well as utilizing multiple informants. A promising early childhood assessment measure is the Penn Interactive Pe er Play Scale (PIPPS) which is a rating scale created with parent a nd teacher versions. Previous research has established its validity for preschoolers from among various populations. The purpose of this study was to examin e the validity of the PIPPS system in a population of preschool children, by investigating: (1 ) the concurrent validity of parent and teacher versions of the PIPPS and a sta ndardized assessment measure of social competence (PKBS-2 Social Skills Scale); (2 ) the relationship between teacher/parent ratings and child gender; (3) the relationships between the teacher and parent versions of
vi the PIPPS; and (4) the predictive validity of teacher and parent ratings on the PIPPS and PKBS-2 with level of communica tion between the two parties. To meet inclusion criteria, teachers a nd parents had to have contact with preschool students ages 3-5 year s enrolled in a preschool classroom for at least 4 months, and who were proficient in ei ther English and/or Spanish. In total, across the three participating preschool centers, 50 students were found eligible to participate in this study and 32 students returned with completed p ackets parent rating scales (64%). Results indicated some relationship between the parent and teacher versions of the PIPPS and PKBS-2 Social Skills rating systems as well as the influence of communication level. However, there were no statistically significant findings for the influence of gender on these ratings. There we re several limitations to the external validity of the results of this study. Limitations included sample bias and the use of selfreport questionnaires. Implicati ons and future directions for research are discussed.
2 Chapter One Introduction Statement of the problem Alarming national figures describe th e extent to which young children are vulnerable to the experiences of facing nega tive consequences associated with academic and socially-related failure. Observational data on preschoolers indicate that between 4% and 6% of all young children have serious em otional and behavioral disorders, and between 16% and 30% pose on-going problems to classroom teachers (Raver & Knitze, 2002). Many of the problematic behaviors exhibited early on in life may be attributed to impaired social relationships, which are hi ghly correlated with academic difficulties and negative social, emotional, psychological ou tcomes for children both with and without disabilities (Merrell, 1995). In addition, poor peer relationshi ps established early on in life have been shown to be relatively stable over time and predictive of later adolescent and adult psychopathology (Shapiro & Kratchowill, 2000). These data are important to address cons idering that four million children are entering kindergarten each year. Although some of these children bring to the transition to school the vital social, emotional, language and cognitive skills that are necessary to succeed, significant groups are at risk for early school failure (Raver & Knitze, 2002). There has been a strong emphasis on the impor tance of school read iness and preparing children for literacy. Pre-existi ng skills that contribute t o, and are related to school
3 readiness are based on a foundation of strong so cio-emotional skills. Children attending preschool classrooms that manifest close teacher-student relations hips, low levels of problem behaviors, and opportunities for positive so cial interactions are noted to be more socially competent and fare be tter academically during the fi rst two years of elementary schooling than do children who graduate fr om more disruptive classrooms (Raver & Knitze, 2002). While researchers, educators, and parents may differ on their interpretations of tasks that are most indica tive of social and emotional health, there is general consensus in that ch ildren cannot thrive in isola tion (Raver & Zigler, 1997). Throughout a childs life, each stage of de velopment presents the acquisition of distinct social skills (Fantuzzo et al., 1995). Among th ese domains of development, the acquisitions of social competence results in succ essful interactions with peers and pose as a primary developmental task. Relatively l ittle is known about di mensions within the social skills domain for young children. It has only been in the last decade that largescale, national normative databases on childre ns social skills have been developed (Merrell, 1996). Additionally, as early chil dhood programs strive to ensure educational and social success for young children, quality assessment measures are necessary in guiding the development of appr opriate curricula and intervention strategies (Merrell, 1996). Developmental Framework There is a guided perspective which purports that as chil dren develop, they acquire competences across many domains, including social, emotional, linguistic, cognitive, and physical func tioning. Moreover, development in one domain has the potential to influence the development of other domains across va rious settings. Children
4 acquire these skills primarily through interactions with adults in the earlier years. It is pertinent for individuals working in and developing curricula for early childhood programs to recognize the competencies that em erge during this period in order to ensure that these skills are promoted and children r eceive appropriate assistance in these areas (Coolahan et al., 2000). The understanding of the environmental in fluences on childrens development is also informative to quality early childhood programs. According to Bronfenbrenners developmental ecological model, the individual participates in various levels of nested social systems. During the early childhood y ears, the family, school, and peers make up the innermost level, called the microsystem Within this system, the child experiences daily life and interacts with those that are cl osest to them. The next level describes the relationships that exists between individuals that are included in the childs microsystem, and is called the mesosystem The exosystem comprises the settings which have an indirect effect on the childs development. An example of such a system may include school administrators who make decisions aff ecting the childs daily school experiences. Lastly, the macrosystem describes the larger ideological and institutional patterns of a particular culture that provides a blueprint for the nature and stru cture of other social system levels. These systems may include e ducational, social, political, and economic systems that encompass distinct ideologies a nd practices that influence the micro-, meso-, and exosystems (Bronfenbrenner, 1979). The developmental ecological model depict s the information that can be gathered and emphasized through focus on the importance of social systems in which a child develops. Families and schools are the primary systems that influence children early on.
5 Although the family and school microsystems i ndividually affect childrens development, the connection that exists between these two sy stems is also important. The issues of skill development as they are influenced and enhanced by these systems is worthy of examination. Social Skills Theory Over the past two decades, there has been increased focus and attention in the area of social skills, mainly in regards to children (Hatch, 1987; Merrell, 1995). This increasing emphasis and presence is evidenced by the emergence of new social skills assessment tools, and developments in the fi eld of study as introduced in the literature and presented at professional conferences (Merrell, 1995). Social skills have been explained and defined in nume rous ways. A discussion of social skills and social competence must address a definition of the te rms. To date, a single definition of social skills does not exist. However, research seeks to link the outcomes of engaging in specific social behaviors to producing positive so cial outcomes as a general description of social competence (Merrell, 1995, 1 999; Merrell & Wolfe, 1998). The initiation of appropriate social skills in a given situation is predicted to increase the likelihood of rein forcement and decrease the probability of punishment in response to an individuals be havior (Merrell, 1999). Although social skills and social competence have commonly been used intercha ngeably in the literature to convey the same meaning, efforts have been made to differentiate the two domains. Social skills refer to discrete, learned beha viors that are exhibited by an individual for the purposes of interacting with others and fo r performing tasks (Shapiro & Kratchowill, 2000). In earlier stages of research, social competence was broadly defined to reflect individuals
6 personal and social maturity in multiple domains (Raver & Zigler, 1997). Since then, efforts to refine the definition of social competence have included the capacity to effectively manage ones emotions, as demonstrated in verbal and nonverbal communicative behavior in the context of peer interaction. Many have proposed commonly identified developmental tasks such as peer acceptance, academic achievement, and compliance with standard, soci etal rules of conduct to be included as indices of competence as well (Raver & Zigler 1997). Presented definitions are neither age, situation, nor skill specific. However, there are hallmarks of social competence for different developmental periods and the integrity of distinguishing these periods is highlighted (Bracken, 2000). The benefits of utilizing a broader definition include allowing a more holistic evaluati on of a child. This focus defe rs researchers from solely examining the cognitive outcomes of young ch ildren which are incomprehensive of a young childs development. Examiners cannot simply focus on specific factors to evaluate social competence of a child (Bracken, 2000). Concrete measures of social competence are generally composed by the evaluative judgments and descrip tions that are used by others to describe the behaviors and social interactions of a child (Shapiro & Kratchowill, 2000). Teachers and caregivers of younger children commonly assess social co mpetence through rating scales of social skills. Children who demonstrate social skills deficits either may not have the necessary skills in their repertoire, or may not have learned the steps for successfully engaging in a behavior (Shapiro & Kratchowill, 2000). Ta rgeting social skills may improve outcomes in related areas for young children.
7 Gender Stereotypes Evidence in the litera ture suggests that th e adoption of sex-stereotypic behaviors may also be related to the development of social competence. Gender differences in the styles of social interactions are striking and further assume an important role in the development of children in the playroom (Cramer & Skidd, 1992). Gender-stereotyped styles of social interactions are apparent in the preschool years and the interpretation and use of these gender-stere otyped behaviors are associated w ith the childs perceived social competence. In previous research, evidence shows that there have been consistent patterns of sex differences pa ralleling sex stereotypes in ol der children (Cramer & Skidd, 1992). Boys have higher self-concepts when assesse d in areas related to their ability in sports and athletic qualities, wh ereas girls are more confident in abilities that are related to academics and peer interactions. While gi rls are generally expected to respond more prosocially to their peers, (F arver & Branstetter, 1994) boys ar e more likely to engage in rough and tumble active play (Fabes, Sh epard, Guthrie, & Martin, 1997). These statistics are supported by the bias in the prevalence of disruptive behavior disorders diagnosed in younger children. Boys show a greater likeliness than girls to be referred for externalizing behaviors whereas girls are more likely to be referred for internalizing behaviors earlier on. Assessment of Social Skills Evaluation of preschoolers constitutes a f undamentally distinct task than assessing school-age children for a number of reasons. The behaviors of these children within testing situations can affect the accuracy of results as thei r abilities and behaviors do not
8 remain static until later developmental stag es. Because preschoolers cognitive abilities depend to such an extent on other skills, it is pertinent to examine various domains of functioning, including social competence, emotional expression, self-regulation, coping with new situations and challe nges, and play behavior, rather than focusing solely on cognitive development as may be the case with school-aged children (Gagnon & Nagle, 2004). Assessment in this way is convergent, and allows the involvement of formal and informal measurement tools to enhance so cial and treatment va lidity by providing the most valid estimate of developmental functi oning at specific ag e ranges (Fisher, 1992; Gagnon & Nagle, 2004). Researchers have attemp ted to measure social competence for a number of years through the use of rating scales completed by teachers, parents and students in addition to direct observations and sociometri c assessments. Despite the source of data, it is importan t to define the construct of social competence as reflecting the integrity of children in different de velopmental periods (Pellegrini, 2000). The Preschool and Kindergarten Behavi or Scales-Second Edition (PKBS-2) is an example of a commonly used norm-referenced, standardized behavior rating instrument that was developed for use in a variety of settings across multiple informants. Although it is useful for multiple purposes, the PKBS-2 specifically can be used as a research instrument for studying the social, and emo tional characteristics and patterns of young children (Merrell, 2002). The Social Cooperation subscale in cludes items that reflect behavioral characteristics th at are important in followi ng directions from adults, cooperating and compromising with peers, and showing appropriate self-restraint. The items in this subscale are linked to both p eer-related and adult-related forms of social adjustment and involve the assessment of appr opriate compliance with types of structure
9 and regulation that are typically imposed by parents and preschool teachers (Merrell, 2002). Play Behavior An alternative option may be assessments conducted through play behavior by which researchers are able to examine natu ral social interactio ns taking place among young children. During early childhood, the primary context for establishing positive interactions with peer s is through play. Through play interactions res earchers are able to acquire information relating to a childs la nguage development, cognitive development, socio-emotional growth, general maturation, in addition to observation of a wide variety of individual traits (Fantuzzo, Mend ez, & Tighe, 1998; Gagnon & Nagle, 2004). Teachers and caregivers alike encourage play behavior in their children to model and teach positive social behaviors through th e interactions with others. Play is an activity which provides an important window through which to view development and progress since it proceeds through a regular developmental sequence during childhood (Gagnon & Nagle, 2004). For the purposes of re ducing ambiguity in the research of play behavior, it has been broadly defined as self-generated, intrinsically driven, hedonic activity that is characterized by variable behavior and pretense (Fisher, 1992). Young children naturally enjoy play and are motivated to engage in it while expending substantial social and cognitive energy towa rd its engagement (Gagnon & Nagle, 2004). Widely known developmental theorists such as Piaget (1952, 1962), Vygotsky (1976) and Erikson (1968) supported views that play was fostered social development and provided major opportunities for interaction in the preschool years. They identified childrens peer play as a primary contex t for the acquisition of important social
10 competencies (Fantuzzo & McWayne, 2002; Fi sher, 1992). Piaget believed that play was often the medium through which children were enabled to build social collaboration skills and learn to coordinate multiple points of vi ew (perspective-taking). Through interaction with peers in play, children are encouraged to move away from egocentric perspectives and progress towards acknowledging realities that are outside of their own views (Fisher, 1992). Specifically related to this belief, play interactions within the peer group are seen to be critical in the provision of feedback that is necessary for responses to be acquired in the establishment of socialization. Peers may effectively demand that young children inhibit aggressive behaviors to avoid ostrac ism and simultaneously serve as coaches in supporting and promoting prosocial behaviors (Fisher, 1992; Raver & Zigler, 1997). Research has revealed significant correl ations among preschoolers levels of socio-dramatic play, measures of social competence, and peer acceptance (Fantuzzo, Sutton-Smith, Coolahan, Manz, Canning, & De bnam, 1995). Domains that tend to be emphasized in the preschool curricula include cognitive, language, social-emotional, and academic skills. Progresses made in these areas are readily observable and open to intervention in play behavi ors (Fantuzzo et al., 1995, Fi sher, 1992; Gagnon & Nagle, 2004). An example of a play assessment tool that continues to expand its utility in the literature is the Penn Interactive Peer Pl ay Scale (PIPPS; Fantuzzo, Sutton-Smith, Coolahan, Manz, Canning & Debnam, 1995). Th e PIPPS was designed as a behavioral rating instrument useful for understanding peer play behaviors that are evaluated by both parents and teachers across settings du ring early childhood (Fantuzz & McWayne, 2002). Results from the PIPPS may be used to: (a) as sess childrens play be haviors in free play
11 contexts in the home and school; (b) create a method of communicati on between teachers and parents; and (c) inform an early ch ildhood curriculum (McWayne, Sekino, Hampton & Fantuzzo, 2002). In conclusion, it is important to conduct furt her research in the area of social skills with preschool aged children. Currently soci al skills are perceived and interpreted by others as a way to encourage prosocial beha viors and promote skills that will assist children in developing positive interactions w ith others. It is during these years that children begin to form peer groups and stab le friendships. Although research has shown that friendship formations may readily cha nge in preschool/kindergarten years, soon after, children appear to establish more stab le reputations within peer groups (Denham & Holt, 1993). In addition, observing play behavi or may allow educators/families to obtain additional sources indicating measurements of social-emotional development in preschool-age children. Many of the social-emotional behavior rati ng scales available for use with very young children are downward extensions of rating scales originally designed for use with older children. Thus, such rating scales may not adequately address the unique developmental characteristic s of this population. There is a continuing need for the development of useful behavioral assessm ent instruments focused exclusively on the early childhood/preschool ag e range (Merrell, 1996). Research questions In order to contribute to th e research in this area, speci fic research questions were developed for this study and are presented:
12 1) Does the preschool version of the PIPPS de monstrate concurrent validity with the Social Skills scale of the PKBS-2? 2) Do teacher and parent ratings for presc hoolers' social competence vary based upon a childs gender? 3) Are PIPPS-Teacher and the PIPPS-Parent versions congruent and significantly related to one another? 4) Do parent and teacher ratings on the PIPPS and PKBS-2 vary as a function of the frequency of communications that are re ported as occurring between parties? Hypotheses Convergent validity indicates that an inst rument correlates positively with other variables with which it would be expected to correlate. It was hypothe sized in this study that convergent validity woul d be acquired between the PIPPS Play Interaction factor and the PKBS-2 Social Skills scale by produci ng positive correlations. Utilizing a play behavior assessment tool is beneficial for the purposes of deriving interventions that teachers and parents may easily implement across settings. Although not a central focus of the pres ent investigation, the obtained data provided an opportunity to eval uate sex effects in the study of social competence in children who are typically younger than previous ly studied. It was speculated that boys who show a greater use of male-stereotyped styles of intrusion, (i.e. pushing out into space, joining a group without permission) and domination (i.e. physical/verbal aggression, challenging, asserti on, interference, and criticism) would be directly related to ratings of perceived social competence. Girls on the other hand, were evaluated on
13 greater use of female-stereotyped styles of inclusion, and affiliation (i.e. questioning, requesting, inviting, and joining in play) to determine ra tings of perceived social competence. This hypothesis is based on theo ry and research stating children whose styles of interaction which are characteri stically sex-stereotyped will receive more positive feedback from their peers, and as a result will be perceived as having greater social skill capabilities than children whos e styles of interaction deviate from sexstereotyped expectations (Cra mer & Skidd, 1992). Same-sex peers are noted to reinforce each other more than they re inforce opposite-sex peers while this is also the case in reinforcing socially expected behavior that is same-sex typed, and punishing or ignoring deviant behavior. Generally, in studies that have been conducted on the younger population, boys have been shown to be rate d substantially high er on measures of anger-aggression and lower on social competence than girls. However, studies that have s hown this relationship have used traditional measures of behavior ally inferred ratings of social competence rather than observations of play behavior It was hypothesized that boys would be rated lower overall in social competence through observations of play behavior as compared to girls (Cramer & Skidd, 1992). Lastly, teacher ratings and parent ratings of a childs social competence and play behavior were hypothesized to di ffer on various aspects of the s cales. This is due in large part to the relative objectivity of teachers, as compared to parents. Other children in the classroom in addition to those children th at teachers may have seen throughout their careers act as references to norms which allow them to make objective comparisons. Furthermore, teachers have frequent contact with the child and are able to base their
14 judgments on numerous observations of the childs behavior in the natural environment on a consistent basis, once a child starts at tending preschool. Pare nts on the other hand may see a child interact in other settings including extracurricular activities and in interactions with their siblings and other adults to obtain differ ing judgments about a child (Aktins & Pelham, 1991). Parent and teacher perceive d social competencies are likely to influence their interactions with children also making it more or less likely to pr ovide opportunities to learn and stimulate children in the respective environments. Related to this hypothesis, it was speculated that the greater amounts of communication occurring between teachers and parents would increase the likelihood of obtaining higher correlations between the beliefs of social competence across informan ts. Clearly, increased access to parental input is essential to obtain an accurate picture of peer play interactions. Furthermore, joint involvement on behalf of the teacher and parents toward a common goal may be beneficial toward progress and consistently promoting specific areas for success. As children transition from preschool to kindergarten, they e xperience a sense of continuity or discontinuity. Familiarity with the expectations of the two environments may ease the adjustment to new settings. Ho wever, if children encounter unexpected demands and practices in the new environment, they may have more difficulty adjusting and making the smooth transition. Early ch ildhood programs should target ways to enhance the continuity between the home and school environments in order to facilitate successful school experiences.
15 Significance of the study Relationships between childrens popularity and school adjustment with socially competent children have shown that in additi on to being well-liked by peers they will have a higher tendency to develop positive perceptions of school. Social competency is an important predictor of self-worth in young adolescents, and it is important to examine and understand how these aspects of personality develop earlier in life (Cramer & Skidd, 1992). Although the construct of social competence has been demonstrated in measures of play there is a paucity of research conducted in the younger population in social aspects in general. Recent studies indicat e that social competence may assist in decreasing the degree of exhibited negative externalizing behaviors (Ladd, Price & Hart, 1988). Identifying the degree of the relationshi p between these constructs in the younger population enables educators and pr ofessionals to target skills and behaviors in order to promote normal skill development in peer relationships. In the younger population, the observation of social competence may enhance prediction of later outcomes. Researchers have reported better prediction from indexes of social competence and ego maturity than fr om the absence of problem behaviors and symptoms (LaFreniere & Dumas, 1996). This investigation contributes useful information to the research available on soci al skills in young ch ildren, in addition to further supporting and expanding on existing literature related to the use of play behavior as an accurate method of obtaining social-e motional developmental characteristics of young children. It is also important to examine newer th ird-party rating scales which are generally demonstrated to have the best technical char acteristics and should therefore replace older
16 rating scales (Bracken, 2000). Therefore, this study further validated the use of a play behavior assessment tool derived by Fantu zzo et al., (1995) which appears to be promising in the examination of social competence in young children. The current study lastly replicated findings of obtained high correlations of parent and teacher ratings as compared to traditional social skills assessmen t tools. The examination of the effects of gender on parent and teacher ratings, and the number of communications between informants highlights the importance of asse ssing these factors and their influence on the ratings on this play behavior scale. This in formation may be considered in the future when examining the existing literature on the PIPPS.
17 Chapter Two Literature Review Introduction The development of childrens social competence is critical as maladaptive patterns of childhood are predictive of problems with peers and adults in school as well as society at large. Childrens social comp etence may be perceived as being guided by a developmental perspective (Pellegrini, 2000). Developmental models of social competence suggest that children face a set of stage-relevant tasks that are influenced through their interactions with networks of family, peers, and community systems (Bronfenbrenner, 1979). The capability to feel positively about oneself and to successfully engage in positive relationships with family and peers is defined as a component of social competence (Raver & Zigler, 1997). Among many factors that encompass social competence is the ability to manage ones emotions and the capability to engage positively through verbal and nonverb al communicative behavior in the context of peer interactions (R aver & Zigler, 1997). There is recognition that acquisition of prosocial behaviors during childhood has been found to enforce positive peer relationships and academic success. These behaviors may be influenced by social, emotional, li nguistic, cognitive, and physical domains of functioning across environments (Hampton & Fantuzzo, 2003). Failure to develop prosocial behaviors or development of antisoc ial behaviors has been associated with a
18 number of negative outcomes, including acqui sition of problematic behavioral patterns, juvenile delinquency, and re tention (Fantuzzo, Sutton-Sm ith, Coolahan, Manz, Canning, & Debnam, 1995; Merrell, 1995; Raver & Zi gler, 1997). While the majority of the existing studies on the development of these positive social behaviors have focused on the school-aged population, there is much evid ence suggesting that acquisition of positive social behaviors is an important milestone in the development of behavioral adjustment for children as young as two or three year s of age (Fantuzzo et al., 1995; Odom, McConnell, & McEnvoy; 1992). For the purpose of this study, the positive social behaviors including emotiona l regulatory skills, soci al cognition skills, and communicative behaviors that young childre n have found to demonstrate which are viewed by teachers, parents and peers as social ly competent will be referred to as social skills (Raver & Zigler, 1997). Much of the attention that has been given to the area of social skills has been invested in the provision of psychologi cal and educational services to early childhood/preschool-aged children with soci al-emotional problems (Merrell, 1996). The reasons for the increased emphasis are mainly two-fold. The original age range was expanded under federal law for the educati on of children with disabilities in 1986, as required by Public Law 97-142, toward the incl usion of children between the ages of birth to three. This act was reauthorized in 1990 (Public Law 101-476, referred to as the Individuals with Disabilities Education Act) to further highlight the importance and necessity of service provis ion to the early childhood population (Merrell, 1996). In addition, the Childrens Defense Fund continue s to highlight and present the changing social and economic conditions in publications depicting the desolate reality of
19 increasing poverty rates of young children across the nation (Childrens Defense Fund, 1997; Fantuzzo & McWayne, 2002). There are 4 million children today under the age of 6 in the country that engage in a life of poverty. Although poverty rates overall have been reduced since the 1960s, it has continued to rise solely within this age group. These children are vulnerable to depr ivation of mental health care, increased exposure to violence, abuse, and are at increased rates of being dia gnosed with a psychological disorder (Merrell, 1996). However, the curren t movement towards greater assistance and research focusing on the early childhood/pr eschool population is noteworthy in its attempt to service this neglected group in the provision of mental health services. This literature review will examine differe nt attributes of so cial skills and the evidence of development through examination of play behaviors in preschool-aged children and their implications on this popul ation. Additionally, the understanding of the importance of environmental influences on ch ildrens development in informing quality early childhood programs will be discussed. Impor tant concepts such as the definition of social skills in younger children will encompa ss how such skills are to be developed, in addition to the outcome of children who fail to acquire these skills early on in life. These children are believed to be at a much great er risk for the development of disruptive behavior disorders and dia gnoses of behavioral problem s later on in their careers. Subsequently, important predictors of social skills as examined through play behavior will be discussed by incorporati ng aspects of childrens emotionality and understanding. Topics will also cover areas examining the influence of social skills development of younger children, and the difference in ratings made by teachers
20 compared to those obtained from parents. Li mitations of the studies will be discussed in an attempt to direct future research needs. Social Competence in Early Childhood For young children, positive interactions w ith peers provide support, nurturance, and acceptance. Engaging in socially competent interactions also teach important skills such as helping and sharing which allow facilitation of late r school success and motivation toward academics and achievement (Hampton & Fantuzzo, 2003). The formation of effective peer relations in th e preschool years has been associated with positive adjustment in kindergarten as well as academic success in the elementary grades and high school and is consid ered an indicator of healt hy adjustment (Coolhan et al., 2000; Fantuzzo et al., 1995). Longitudinal research has linked children who possess social skills deficits with detrimental c onsequences during later developmental periods, including numerous academic and behavioral problems as evidenced in learning difficulties, academic underachievement, adju stment problems, conduct problems, and delinquent behavior (Fan tuzzo & McWayne, 2002; Hartup & Moore, 1990). These children are at a general increa sed risk of being diagnosed with a psychiatric disorder over the course of their lifetime (Merrell & Wolfe, 1998). Preschool-aged children begin to experien ce the peer group and its importance in their lives in beginning social interactions. Peer reputati ons and the foundations required for stable social behavior pa tterns are learned and created through interactions in the home environment and at preschools and daycar es. It is important to be cognizant of the fluctuations in social behavior according to the context in which they are observed. Peer and adult interactions elicit and support social competence on very di fferent levels. When
21 children and adults interact, adults often engage in the initiating and maintaining of social interactions. Perhaps this is why children with low levels of social competence choose to spend time with a teacher during recess rather than with their own peers (Pellegrini, 2000). Only in interactions with peers, however are children able to experiment with and practice social strategies among others of relatively equal status to themselves (Hartup & Moore; 1990, Hatch, 1987). Children are free to im itate successful social strategies used by other children and to learn from their own, and of others unsuccessful attempts (Hatch, 1987). This suggests that children take their cues from the most natural competent role models available to them (Hartup & Moore, 1990). The notion of peer similarity and mutual reinforcement appear to be a powerful determining factor in the future development of childrens behavior (Pellegrini, 2000). In peer contexts, children experience views that are discrepant from their own and often must compromise their views. Sensorimot or and preoperational stage children are socially egocentric acco rding to Piaget (1970) to the exte nt that they are neither willing nor able to consider the pers pective of others. The accommod ation of ones own point of view to anothers is said to result in conceptual growth that is necessary to move through the developmental phases of social compet ence acquisition. Although it is important for adults to be available in situations necessa ry to discourage aggre ssive behaviors, their presence may not be necessary in forming a nd directing childrens play groups in natural settings (Pellegrini, 2000). Children who develop a positive peer status may behave increasingly positive as a result of others approval and the social oppo rtunities that are afforded to them in response to their social behaviors. Conversely, children who fail to acquire positive
22 social behavior may become even more social ly deficient due to a lack of opportunity to learn positive behaviors. This cy cle may ultimately lead to frustration as a result of their low social standing (Denham & Holt, 1993; Gagnon & Nagle; 2004; Hartup & Moore, 1990; Hatch, 1987). These children are prone to continue to engage in maladaptive peer interactions, which tend to exacerbate the negativ e behaviors associated with social skills deficits (Alessandri, 1992; Hatch, 1987). The social order in presc hool classrooms has initially been perceived to be dynamic and transient. Preschool ers appear to engage in re negotiating peer status over and over again. Although it has been estab lished that children ar e capable of, and do indeed generate social strategies, some ch ildren appear to have difficulty acquiring the knowledge and developing the necessary comm unicative and social skills that are required for effective movement in and out of peer groups. Research provides evidence that children who are popular and have well developed interacti onal skills operate effectively within a self-per petuating social circle. They are reinforced through their competence thus stimulating fu rther growth (Hatch, 1987). Findings of studies in this area of research have charac terized popular children as cooperative and exhibiting prosocial behaviors toward peers while depicting unpopular children as aggressive and disruptive. Counter to this view is the proposition that prior peer reputations influence how they respond to peers and how peers perceive and respond to them (Ladd, Price, & Hart, 1988). Suppor t for these propositions can be found in studies suggesting that childre ns reputations may affect subsequent peer behavior. However, the design of previous studies ha s failed to address the question of whether childrens behavior is the cause or consequence of their peer status.
23 In an attempt to clarify the source of this controversy, Ladd and colleagues (1988) investigated the concept of alternate hypotheses concerning the antecedents of childrens peer status and playground behaviors in a school setting. Participants consisted of 28 White preschoolers (aged 3.5 to 4.5 years) from middle-class families who were attending a university-b ased preschool. Researchers hypothe sized that whereas prosocial behaviors would predict increases in peer accep tance, antisocial behaviors would forecast increases in peer rejection. In addition, peer acceptance was anticipated to inspire preschoolers to engage in fr iendly behaviors while children subject to rejection were bound to increase behaviors of withdrawal and aggression towards peers. Children interactions were observed on a common pla yground over the course of three, 6-week intervals, scheduled at the beginning, middl e, and end of the school year. Sociometric measures were derived in each classroom to obtain information about childrens peer status at each time of assessment. Codes for peer interaction and nonsocial behavior measured skills in the following areas: cooperative play, social conversation, argumentation, rough play, parallel play, so litary play, on-looking, unoccupied, and transition. Correlational analyses were used to expl ore the stability of childrens playground behaviors and peer status and to examine c oncurrent relations between the behavioral observations and status measures at each assess ment. In addition, a series of hierarchical regression analyses were employed to determine the extent to which earlier displayed behaviors could predict changes in peer status and behavior. Children who played cooperatively with peers at the outset of th e school year were found to do so at later points in time and appeared to contribute to long-term gains in peer acceptance. In
24 contrast, early arguing behavi ors exhibited by children, a lthough less stable over time, predicted increases in peer rejection throughout the year. Th is finding supports the initial hypothesis posed by researchers suggesting that negative reputations of disagreeable preschoolers persist even after behavior ma y have changed. Peers appeared to develop lasting negative attitudes to wards those who had a tendency to be argumentative despite the changing peer status composition in th e classroom. However, initially disliked children did not tend to become more withdrawn or off-task, nor did they engage in more arguing behavior over time relative to other children. Moreover, being well liked at the beginni ng of the school year did not seem to contribute to gains in childr ens prosocial behaviors. Thus, the results of this investigation failed to support the second hypothesis that childrens early peer reputations would affect subsequent behavi or toward peers. These finding s are interprete d in light of past research on childrens peer behavior and status. Future studies must employ a broader range of social and non-social behaviors to explore this c oncept with use of a larger and more diverse population sample befo re findings may be readily generalizable. Preschoolers likeability by other peers was further examined in a study conducted by Denham and Holt (1993). The st udy attempted to extend the findings from Ladd et al. (1988), suggesting th at behaviors were the drivin g force of acquired status. Teacher ratings of positive and negative social behavior, as well as peer-rated sociometric measures were used to assess 43, 3-4 year old preschoolers over a 12-month longitudinal period. This study was replicated and extende d over the next following year. Aligned with previous research, findings indicated that the level of prosocia l behavior exhibited was positively related to likeab ility, although the characteris tic of assertiveness was
25 unrelated. Being a purposeful, dominant leader was not as salient or necessary at this young age in the ability to interact while mainta ining positive affect in oneself and others, as may be the case in older age groups. Findings from the study indicated that early on in the formation of peer reputation, likeability was associ ated with child social behavi or. In Year 1, friendlier, more cooperative, tractable, less aggressive and less difficult children were more wellliked. However, Year 1 likeability, not prosoc ial behavior, predicte d Year 2 likeability ratings. Thus, findings suggest that early on, behavior is important but then reputation based on peer sociometric ratings become s more prominent in determining an individuals likeability and status among peers. The outcomes further emphasize the importance of the early development of social skills in acquiring positive peer interactions and establishing grounds for higher peer status. Children entering school with friends, are more likely to be well liked and able to make and sustain new friendships. They are also able to initiate positive rela tionships with their te achers in addition to feeling more positive about school (Denham et al., 2003). These children may be able to transition into a school setting and engage in peer interactions more easily as a result of their previous expos ure with peers and possibly siblings. It would be interesting to group the data according to children who have siblings at home compared to those who do not. Future research may also wish to expose reasons as to why certain peers are well liked compar ed to others and whether they are purely individual characteristics and temperament issues versus history of previous peer interaction exposure.
26 Along with the notion of the importance of peer likeability, friendships can also be seen as having a general role in a child s self-image. This theory is based on the concept that children develop a general idea of themselves thr ough interactions with significant others, including fr iends. A child may construct a self-image on the basis of feedback from those with whom they cons ider to identify with. The peer group and friends can provide informati on to influence how a child views themselves. For this reason emphasis is placed on the importance of fostering peer groups and positive social interactions (Pellegrini, 2000). Although there was detec tion of a causal link from sociometric status to concurre nt, stable social behavior in this study, further research must be generated with extended longitudi nal periods and a greater sample size to examine this pathway from likability to pros ocial behavior in this age range before conclusions may be validated. Social Competence and Academic Outcomes The importance of social competence outcomes extend into areas concerning academic achievement. Social competence is increasingly recognized as vital to the development of school readiness. Research has demonstrated that social and academic competences are very much inter-related (DuPaul, McGoey, Eckert, & VanBrackle, 2001; Merrell & Wolfe, 1998; Welsh, Parke, Widaman, & ONeil, 2001). Although previous research has suggest ed that children classified by their peers as popular in kindergarten received significantly higher scores in their academic achievement and teacherrated classroom adjustment in subs equent grades (Gagnon & Nagle, 2004), there has been controversy as to whether it is so cial competence that influences later academic competence or academic competence that influe nces later social competence. The nature
27 of this relation was examined in a longit udinal study conducted by Welsh et al. (2001). The primary aim of the study was to test a reciprocal model hypothesizing the relation between social and academic competence as a bidirectional influen ce, reciprocal over time. Researchers followed a moderately la rge sample over the dur ation of a 3-year period. During the first year of data colle ction, 163 kindergarteners (75 males and 88 females) and their families were targete d. Academic and sociometric information was gathered each year of the ongoing study. Meas urements of social competence included peer ratings from all of the children in the cl ass of the targeted child each year. Childrens behavioral characteristics were also assessed by having children nominate their classmates in different categories classified as those three children having the most prosocial (i.e., helping, shari ng, taking turns) and aggressi ve (i.e., fighting, saying mean things) characteristics. Teachers completed social competence and behavioral rating scales each year assessing child likeability by his/her peers, as well as behavioral characteristics. Measures of academic competence were derived from report cards depicting student achievement and effort levels. Findings indicated that academic competence consistently led to social competence over time in a bidirectional pattern of influence. Academic competence in first grade was shown to influence second-grade social competence, with academic competence being positively related to positive social competence. Lower academic competence in the first grade also influenced social competence in second grade, with academic competence being negatively related to negative social competence. This relational pattern was replicated in the following years as well. The overall findings indicate that academic competence exerts a significant
28 influence over social competence consistently over at least a 2-year period of time. Results must be interpreted with caution a nd several limitations should be noted. Due to the sample size of the study, definitive an alyses by gender and ethnicity were not possible. More participants are needed to examine the relation between social and academic competence in regard to ethnicity a nd SES. Therefore, generalizability issues arise in regard to these factors. Secondly, al though valid and reliable measures were used, the inclusion of measures which tap a wider range of social and academic functioning would be desirable in future studies in order to yield a more comprehensive approach to assessing social competence (Welsh et al., 2001 ). It may be difficult to assess whether or not self-concept and increased perception of academic competence factored into a childs social competence. However, it should be noted that these domains must be examined before concluding a direct effect between social competence and academic abilities. Additionally, future studies are warranted to focus beyond third grade to further validate the findings from this study. It would also be interesting to examine, in collaboration with current findings and suggested future research, whet her social competence could act as a protective factor for children who were achieving at a lower acad emic level. Although students may not be motivated by their performance in school, so cial support may prove to be a pertinent factor in childrens perspective of school and academics in general. Relating Social competence and Emotional Competence Another area of interest that has transpired in recent years addresses identifying the roles of emotionality and its contribution to childrens social competence (Fabes et al., 1999). Because of the ever-present nature of emotion at the core of social interaction
29 and well-being, discerning the nature of the linkages from emotional to social competence is a vital area for educators to explore (Denham et al., 2003). Socially competent children display emotions that are appropriately responsive to group norms and are in balance between their own desires and interests while keeping those of others in mind. Furthermore, childrens social co mpetence is related to their ability to distinguish and express emotions and emotiona l intentions (Fabes et al., 1999). Control and regulation of emotions are considered an important underp inning of childrens developing cognitions and internalizations of socially appropriate behaviors that are aligned with societal norms (Roberts & Strayer, 1996). Elements of emotional competence influence social competence in youn g children from 2 to 5 years of age in the development of successful initiation and promo tion of peer relationships (Denham et al., 2003). This positive affectivity is generally a ssociated with enhanced social status and popularity which render one more likeable, wh ereas emotionally negative children are faced with an increase in peer rejecti on (Fabes et al., 1999; Hatch, 1987). Children who are socially unskilled and/or socially rejected have been shown to access fewer competent solutions in the face of difficulties and tend to turn to aggressive and inept ways of solving social problems compared to children who are better accepted (Coy, Speltz, DeKlyen, & Jones, 2001). Popular children ar e adept at discerning emotional reactions of peers during interactions, and thus are better able to respond to them, while less popular children have a greater tendency to misinterpret emotional states and react inappropriately (Gagnon & Nagle, 2004). A study conducted by Fabes et al. (1999) wa s designed to explor e the relation of regulatory and emotional processes eviden ced by childrens social competence in
30 responding during interactions with peers. Ba sed on the previous literature, researchers hypothesized that regulatory control would be positively correlated with observed social competence. Participants were 135 children enrolled in six preschool or kindergarten classes. This particular age group was chosen based on previous literature indicating that the preschool period was a developmentally crit ical time in the formation of an active regulatory control system. The type of responses could further attribute to the probability of a child acquiring higher peer status and popularity through interactions among their peers. Data collection incorporated thr ee months of brief scanning observations of childrens free-play indoors and outdoors during snack period and free times. Observers rated children on the inte nsity of the peer interaction based on the amount of energy and activity displayed by the partic ipants involved in the interaction. Other general behaviors that were observed included the degree to wh ich negative emotions were displayed by the target child during an interaction and the degree to which the target childs actions contributed to positive or constructive social interactions. In addition, two teachers completed the Childrens Behavior Questionnair e in order to assess temperament. Four items from Harters Perceived Social Compet ence Scale for Children (1979) also were administered to examine acquisi tions of social competence. Hierarchical linear modeling analyses were conducted to effectively analyze both within-subject participants (observations) and betweensubject participants (effortful control) data simultaneously. Results of the study supported previous be liefs that as peer interaction increased in intensity, children with relatively high levels of effortful control (EC), or appropriate control of their emotions, were less likely to experience negative emotional arousal as compared to children w ho possessed lower levels of EC. Results of
31 the study supported previous beliefs that as peer interaction incr eased in intensity, children with relatively high levels of effortful control (EC) we re less likely to experience negative emotional arousal as compared to children who possessed lower levels of EC. Findings suggest that increas ed negative arousal appears to inhibit the ability to respond to intense interactions in socially competent ways for those individuals who are less skilled, and therefore retain lower levels of EC. High levels of negative emotions that are not appropriately controll ed can disrupt behavior incl uding empathetic responses and prosocial behaviors (Roberts & Strayer, 1996). The data compri sed in this study confirms the view that regulato ry and emotional processes make significant contributions to the quality of young childrens intera ctions with peers and influe nce their response outcomes. Although the age range of the participants was relatively narrow it may be considered in future research examining emotional expres siveness. Substantial gains in understanding emotions occur in periods of toddlerhood a nd preschool due to increasing exposure to peers and socialization. However increasing cognitive abilities may allow children to withhold more intense negative emotions thus allowing results to be consistent with socialization models in which school-age children face increasing pressure to regulate or moderate their expressions of negative aff ect, particularly anger (Roberts & Strayer, 1996). If this notion were true then the kinde rgartners may be found to have higher levels of EC. In addition, it was not stated whether there were any differences in gender which may also be useful to examine in future studies of this type. Denham et al. (2003) studied a model of the prediction of social competence by young childrens emotional competence to re plicate and extend previous research. Patterns of emotional expr essiveness, emotion situation knowledge, and emotion
32 regulatory coping were used to predict i ndices of social competence. Researchers scrutinized each of the elements composi ng overall emotional competence and defined each subset as equally pertinent and influential in further enhancing social competence. First, childrens emotional expressiveness was deemed a central aspect of their emotional competence. Positive affect is important in the initiation and regulation of social exchanges. Children adept in this area have a higher probability of responding prosocially to peers emotions and are seen as more likable. Emotional knowledge, a second key component of young childrens em otional competence was described as understanding and allowing childre n to react appropriately to others, thus bolstering their interactions and relationships. The last vital aspect of emotional competence is the ability to appropriately regulate emotion. During the preschool period emotional regulation becomes necessary when childrens emotionality and demands of their social worlds become more complex. In addition to their ab ility to further comprehend emotions there are increased opportunities that deem it neces sary to control their emotionality. Both concurrent prediction of social competence at age 3 to 4 and long itudinal prediction of social competence in kindergarten were assessed, using emotional competence components as predictors. Participants were 143 predominantly Ca ucasian, middle-income 3-4 year olds (mean age = 46 months, SD= 4.8 months). Data was collected at two points in time, during preschool and kindergarten. Children were observed in their classroom settings during free play by coders over twelve session s of two, 5-minute trials over a 6-week period. Observational methodologies were used to examine (a) emotions expressed by children during free play, and (b ) their reactions to peers emotions. Semi-structured
33 interviews were used to a ssess childrens knowledge about their own emotions. The use of a puppeteer was utilized in the enactmen t of an emotion-laden story to identify childrens understand ing of emotion. This included childrens responses to the identification of emotions that were appropriate to certain situations and also inferences of emotions in equivocal situations. Las tly, emotion regulation was examined through maternal reports of coping behavior when faced with emotionally di fficult situations with peers. A coping items scale was developed for this purpose and mothers indicated how the child was predicted to engage in each of several types of coping behavior. The present study assessed social competence vi a three methods. Sociometric assessment enabled all of the children in the class to rate their peers on a c ontinuum of likeability through the use of pictures to obtain measures of popularity. Scores for both the number of positive ratings and the number of negative ratings from their classmates were received. The Social Competence and Behavior Evaluation Short Form (SCBE) were distributed to teachers and daycare providers as methods for rating each childs social competence in the classroom. The procedure was modified slightly in kindergarten, using cards with the names of each classmate versus picture re presentations to acknowledge new gains in cognitive abilities appropriate of their age. Results showed a strong relationship between the emotional competence component of development, assessed at ages 3 to 4 years of age as contributing to both the concurrent and kindergarten social comp etence aspects. The findings bring to our attention the enhanced importance in basic understanding of the a ffective foundations of younger childrens social competence. It is important to emphasize possibilities of targeting prevention efforts with preschoolers in developing positive peer interactions and
34 relationships. The results of this study although consistent with previous research must be considered with a degree of caution. Young children who are engaging in sociometric assessments may have the tendency to choose students based on subjective reasons otherwise unknown to the researcher. In addi tion, measures relied heavily on informant reporting with maternal reports on perceived c oping behaviors, child reporting and use of semi-structured measures with young children being utilized as main assessment tools in this study. It may be beneficial to include th e reliability across informants (i.e. mothers, children, teacher reports) and examine the degr ee to which each assessment is influential to the conclusions of the study. Related to the area of em otionality, Philippot and Feldman (1990) examined the interactions taking place between social co mpetence and decoding of emotions on the faces of children. The main focus of this study wa s to take note of the fact that the face is the most visible component of a social intera ction and can be considered the central focus of attention. Both visibility and the potent ial for fine discriminating abilities make the face an important medium for communication for many social and affective processes. Facial expressions are related to empathetic processes and prosocial behavior as well as a means to facilitation of reading and successf ully directing social interactions. Being able to decode facial expressions may be considered a central component of general social skills because of the significance of faci al information in the management of social behaviors. Although there is abundant resear ch supporting the hypothesis that the ability to decode facial expression of emotion is related to various modalities of social competence, almost none of the research has been carried out in children younger than elementary-school age.
35 This study attempts to demonstrate the relationship between facial expression decoding skills and social competence in early childhood. Subjects were thirty-eight children aged 3 -5 years. Children were assi gned to either the high social skills group or the low social skills group based on scores obtained on the Social Competence portion of the Achenbach Child Behavior Checklist. Nine 10-20 second videotaped silent scenarios, with three for each category of emotion (h appiness, sadness, and fear) were designed. Pilot test questioning was conduc ted to ensure that even the youngest children would have no difficulty in understanding the pr esented situations and the emotional implications. Each subject obtained three scores, each of which corresponded to each category of emotion that was investigated. The results indicated that there was a st atistically significant difference between the performances of children that were assi gned to the high social skills group vs. those assigned to the low social skills group. In a ddition, the scores for the different categories of emotions showed that happiness was mo st accurately decoded, followed by fear, and sadness which was decoded with the lowest ac curacy across subjects. Consistent with previous observations, the most socially sk illed preschoolers outperformed their less socially skilled peers in decoding facial expressions paralle ling results in older children. Findings suggest that even prior to the acquisition of social display rules, basic processes in the decoding of fundamental emotions may be impaired in socially at-risk children. Future studies may be interested in the examination of covariation between social competence and facial expression decoding sk ills across age in the same individuals. Present results could also be replicated, and extended to include othe r types of emotions as well. Interestingly enough since sadness was decoded with the least accuracy it may be
36 beneficial to further examine this emoti on and determine reasons that may further contribute to the difficulty of decoding this emotion. The failure to properly decode sadness may be a result of underdevelope d cognitive abilities of younger children. Empathetic responses increase as children become more socialized and with increasing age. Therefore, the growing ability to exhibit empathetic responses with increasing age may have a direct relation to the inability to accurately recognize sadness at an earlier age. Social Skills Deficits and ADHD/ Disruptive Behavior Disorder Although the inability to provide appropriate responses to certain social situations may be inhibited as a result of a lack of empathetic responses across the younger population, children diagnosed with behavior problems such as Conduct Disorder or Attention-Deficit/Hyperactivity Disorder ( ADHD) are especially noted to possess social skills deficits and distinct difficulties w ith peer interactions when compared to undiagnosed children (Hughes, Dunn, & White, 1998) Negative social interactions are not confined to a single developmental period. Children with hyperactivity appear to manifest a greater amount of aggression and reso rt to more aggressive solutions to social situations than normal childre n (Stormont, 2001). As stated pr eviously, hostile or reactive aggression has been documented to be less socially acceptable am ong the peer group and affect peer reputation status. Maladjustment of early school-age peer relationships may potentially increase a childs risk for later ma ladjustment in a number of different areas (e.g. social skills, relationship s, self-esteem), even for t hose individuals who no longer meet criteria for behavioral disorders in adolescence and adulthood (DeWolfe, Byrne, & Bawden, 2000).
37 Hughes, Dunn, and White (1998) conducted a study in exploration of problematic areas and their contribution to lowered soci al competence for preschoolers that were identified as hard-to-manage. Reasons for focusing on young children in this study included the examination of a number of te stable hypotheses. One plausible hypothesis generated from the study was that disruptive children show delayed understanding of the emotional consequences of socio-moral tran sgressions and are therefore more probable than peers with higher social abilities to violate social ru les and norms. Also within the arena of social development, research was extended to consider the possible causes of individual differences in childrens devel oping understanding of beliefs and emotions. Work with children at risk of being diagnosed with behavioral disorders is important in suggesting that individual differences may extend beyond c ontrasts in the rates of development, but also affecting the kind of social understanding th at is developed. The 25-item Strengths and Difficulties Questionnai re (SDQ) was utilized to establish a sample for study. The participant group cons isted of 40 children including 24 boys and 16 girls, ages 3-5 years old. The control gr oup was recruited from a similar screening process and from the same schools as th e target group. All c ontrol children were individually matched with the target group for age, gender, and school. The two groups were compared on a various set of tasks includ ing: (1) theory of mi nd tasks (required the prediction of an emotion involving either a nice or a nasty surprise); (2) emotion understanding stories (required affective perspe ctive-taking skills as well as situational understanding); and (3) simple executive function tasks (testing inhibitory control, attentional set-shifting, and working memory).
38 The results of the study indicated po or emotion understanding among hard-tomanage preschoolers and thus were the same findings previously noted in preschoolers differentiated on level of social skills acqui sition (high vs. low). This evidence further implies that young children regardless of dia gnoses may suffer the same outcomes as a result of general social skills deficits, and at the same time may benefit from similar interventions composed for the purposes of promoting social competence. For the two groups combined, however, findings showed that happiness and sadness were understood better than fear as was previously conc luded in Philippot and Feldman (1990) with normal children. However, further research should be conducted in order to produce consistent findings across or between groups of individuals. Overall findings of this study suggest that impairments in executive functi on that are well established in school-aged children with ADHD and/or conduct disorder not only emerge early in development, but are associated with impairments in the deve lopment of social understanding in hard-tomanage children. The present study conducted by DeWolfe, Byrne, and Bawden (2000) considered several issues in order to de pict clear differences between preschool children with and without ADHD, by using parent ratings of behavior and psychosocial correlates. Participants were forty-five pr eschool children (3-5 years of age) that were assessed at a clinic by means of a 3.5 hour diagnostic prot ocol during which formal diagnoses were determined via direct observations, standardi zed questionnaires, admi nistration of tests of attention, psychological parent interview. Control individual s were twenty-f ive preschool children that were selected to match th e twenty-five clinic referred children on socioeconomic status (SES), age ( + 4), and sex. The measures employed in the study
39 were based on parental ratings of the pres chool children and child self-ratings obtained within the framework of three domains: behavi oral disturbance, social competence, and familial environment. Results indicated that preschool children with ADHD were rated by their parents as exhibiti ng significantly more aggr ession, non-compliance/conduct problems, and were more demanding of their parents attention as compared to their matched control peers. Preschool childr en with ADHD were rated as exhibiting significantly underdeveloped social skills and we re less adaptable to change in routine. Overall, parents of preschool children w ith ADHD reported a more stressful and less rewarding parent-child relationship. Parents ma y be able to cope w ith parenting-related stress by positive reframing and redirecting preschool childs challenging behaviors, and many seek solace in the expectation that the preschooler will eventually grow out of their behaviors. As a result, family functioning was not rated as being detrimentally affected. An interesting finding from this study was th e discrepancy between parental ratings of preschool children and the pres chool childs se lf-ratings. Preschool children with ADHD rated themselves as being equally competen t and as socially accepted as their matched peers. Previous research has indicated similar findings in older children. Many of these young childrens actions suggest that they may be insensitive and unaware of their impact of their behavior on others at this point in their developm ental periods. Lower self-ratings of competence and social acceptance may become more pronounced for school-age and adolescent individuals with ADHD as aligned with further maturation of cognitive capabilities, and cogni tive-social abilities.
40 In an additional study examining the differences between preschool-age children with ADHD and control children, DuPaul et al. (2001) comprehensively studied the behavior and social skills f unctioning of these children acr oss settings. Scores obtained on the Hyperactivity or Daydreams-Attention subscale of the Conners Teacher Rating Scale-Revised (CTRS-R) and the Impulsivi ty-Hyperactivity subs cale of the Conners Parent Rating Scale-Revised (C PRS-R) were used to disti nguish the comparative groups. Fifty-eight children (50 boys, 8 gi rls) were identified as having one of the three subtypes of ADHD (scores at or above the 90 th percentile), and 38 chil dren (20 boys, 16 girls) were assigned to a normal c ontrol group (scores below the 84 th percentile). Participants from both groups were primarily from middl e-class socioeconomic backgrounds. Parents and teachers completed the Preschool and Ki ndergarten Behavior Scales (PKBS) and Social Skills and Problem Behavior subscales for assessment of childrens behavior at home and school. Behavioral observations of parent-child inter actions were also conducted in a clinic playroom setting. Four different controlled situations were constructed, each of which was 10 minutes in duration. Direct observations of classroom behavior were also conducted in structured activities. Ge neral categories of variables were analyzed using Hotelling T 2 tests followed by univariate t tests for significant findings. Observations yielded results that indicat ed participants in the ADHD group were found to exhibit greater levels of negative so cial behavior in both structured classroom and unstructured situations. Children with ADHD displayed more frequent noncompliant and inappropriate behaviors than children in the control gr oup. As expected, children with ADHD were rated as less socially skil led than children in the control group. These
41 results have several implications for practit ioners who work with young children. First of all, screening among young children with behavi or difficulties regard less of a diagnosis may promote evaluation of associated probl ems involving social behavior. Also, the results highlight the need to assess pre-acad emic skills and preschool classroom behavior of young children that are at risk for behavior disorders and also t hose who are exhibiting similarly related problems. Current Assessment Issues It is important to introduce primary pr oblems and limitations that arise when assessing the social functioning of preschool children at this point. Preschoolers are unique in that they possess re stricted cognitive a nd language abilities. Most preschoolers cannot read, therefore making the use of inst ruments requiring this fundamental skill nonapplicable for describing social-emotiona l functioning (Bracken, 2000). As a result, widely used self-reports of personality or social functioning are generally eliminated from the use with preschool children. Rela ted to the notion of restricted cognitive abilities, preschoolers are also limited in thei r expressivity and have difficulty describing their full range of thoughts, feelings, or relati onships with others and may only be able to provide a general idea of what they thi nk and feel through verbal communication (Bracken, 2000). In addition, many preschoolers are limited in their understanding of emotions or feelings and questions are often misinterpreted. Anothe r limitation affecting the assessment of social competencies of pr eschoolers involves the large range of normal developmental progress. The range is typically broader for preschoolers than for older children, adolescences, and a dults. This creates a problem when attempting to
42 discriminate preschoolers who deviate from that of a standard or normal functioning individual (Bracken, 2000). Social Skills Acquisition through Play Behavior Social competence with peers encompa sses behaviors and cogn itions that reflect successful interacting with peers. A socially co mpetent child is effective in meeting his or her social goals with peers, however is able to maintain flexibility and sensitivity in responding to peer reactions. Demonstrati on of social competence with peers is observable through play. A primary context fo r preschool-aged children to acquire and express peer social competencies is through play (Fantuzzo & McWayne, 2002). Children must be able to initiate play, en ter ongoing play groups, appropriately respond to peer initiations, and resolv e conflicts with peers. It is assumed that children who engage in complex peer play forms would be socially competent in other aspects of peer relations (Howes & Matheson, 1992). The relationship between play and child development has been intensively studied in theory driven and empirical studies. Th e National Association for the Education of Young Children states, Play is an important ve hicle for childrens social, emotional, and cognitive development, as well as a reflect ion of their development. (Bredekamp & Copple, 1997, p. 6). Some common characterist ics of play include the following: (a) intrinsically motivated, not dependent on extern al rewards, (b) freely chosen whereas any coercion to engage in an activ ity may deter it from being considered play, (c) pleasurable and enjoyable eliciting positive responses, (d) involves an element of pretend or makebelieve, and (e) involves active engagement re quiring children to be more attentive than to most other stimuli (Bracken, 2000). In a ddition to the characteris tics of play already
43 mentioned, it can be considered a developmental phenomenon that follows a relatively stable sequence. Through play in teraction with peers, children are able to test out social roles and learn acceptable social rules. They are encouraged to share, take turns, cooperate, consider others perspectives and inhibit aggression through these interactions, which ultimately teaches appropr iate suppression of immediate impulses and acquiring self-control (Bracken, 2000; Gagnon & Nagle, 2004). Although most theories of play assume that it is beneficial, there is widespread debate regarding the magnitude of the benefits, and when during development thes e benefits occur. De spite considerations of the magnitude and occurrence of benefits of play it is assumed to indeed have advantages and provide areas fo r childrens social growth. Play can be viewed as an initial attempt at developing mature behaviors that must be learned, and may be used to teach skills and concepts (Pellegrini, 2000). Studies of the impact of play on ch ildhood development have addressed three main areas of development (Fisher, 1992). First, attempts have been made to demonstrate a relationship between play behavior and the emergence of cognitive development in the acquisition of crea tive problem-solving, logical thought, and perspective-taking. Another area of interest that has arisen in significance and relates to inquiry has been the contributi on of make-believe play to th e acquisition of symbols and language mastery. The last main area of res earch has come from investigating the effects of play training interventions on improving so cial interactions an d/or the building of empathic interpersonal skills, regulation of affect, peer-group popularity, and influence on childrens natural tendencies to ward egocentrism (Fisher, 1992).
44 Fisher (1992) conducted a meta-analysis combining findings from a collaboration of integrative works relating play behavior to development, dating back from the mid70s. The review presents results from 46 studi es. Statistical tests employed the use of effect sizes as the main me thod of analyzing obtained results. An overall effect size drawn from the data indicated that play re sulted in a moderately large and noteworthy improvement in childrens development that was evident in varying degrees over the three major outcome domains studied. Cohens scaling method was utilized for interpretation purposes indicating a 12% va riance that corresponded to an improvement rate, or gains in performance for 67%. The concluding remarks from the meta-analysis suggest that play enhances the progress of early developm ent from 33% to 67% through adjustments in a number of different functio nal areas (e.g., reducing failure rates in academic/adjustment functioning due to under-achievement, language problems, socioemotional difficulties, etc). The typical meta-analysis review is sus ceptible to a number of biases regarding the sets of decisions that are required to be made on behalf of the reviewer and thus such factors should be taken into consideration when reporting the results from the study. However, this meta-analysis explored the eff ects of different types of play and obtained noteworthy evidence of the infl uential qualities. Data collec ted from this study should be interpreted with caution although not disregarded. Parental Factors and Preschoolers Social Competence Preschoolers social competence may de pend on other factors, including the frequency with which informal play activiti es are initiated by pa rents, children, and playmates. Parents are in a very influentia l position in relation to many aspects of their
45 childrens social development (Hartup & Moor e, 1990; Ladd & Hart, 1992). Adults play a significant role in the socialization of young children by providing guidance as children make the necessary adjustments to life with p eers. Children expect ad ults in positions of authority to instruct them in behaviors by teaching them what is acceptable and what is not, in addition to monitoring their behavior s and protecting them from harm (Hartup & Moore, 1992). Early childhood caregivers and teachers are also functional to a considerable extent in providing a secure base of support from which the child is able to explore the out-of-home social envi ronment (Hartup & Moore, 1992). Parents are also in control of facilita ting childrens access to peers by choosing to reside in particular neigh borhoods, enrolling children in preschools, taking them to community settings (library, pool), and other places in wh ich they are likely to meet same-aged peers. Additionally, parents may create informal meeting opportunities with their peers for their children by arranging play groups, or afte r-school play activities at their homes. These informal play contacts ma y help children to meet peers, expand playmate networks, maintain existing peer relationships, and also develop good interpersonal skills (Ladd & Hart, 1992). Pa rents who arrange informal peer activities may encourage development of childrens initiation skills by involving them and modeling components of the initiation process for them (e.g., engaging children in contacting peers and extending play invitations to them) (Ladd & Hart, 1992). These acquired skills of social competence may be important contributions that are exhibited across and generalized to other settings su ch as school. Some research has found that children tend to form friendships that are cl oser, more stable, and involve less difficulties
46 when their parents have taken an active role in arranging for and stimulating peer relations and opportunities (Ladd & Hart, 1992). However, parental control and initiation is not the sole determining factor in the number and frequency of peer interactions. During preschool, children begin to sponsor their own opportunities to play informally with peers. Children initiate their own peer contacts and also receive pl ay invitations from peers (L add & Hart, 1992). The ease in which children are able to make this transi tion may justifiably be based on parental support and modeling of the proc ess of setting up these inte ractions. These efforts may ultimately lead up to the point where the ch ild individually makes a personal choice to interact. In view of these signif icant factors, a study was conducted by Ladd and Hart (1992) investigating the extent to which parents engage in this socializa tion activity, in addition to childrens own initiation attemp ts, and the potential effects on early peer relations and social competence. The samp le for the study consisted of 83 preschool children (50 boys and 74 girls) between the ag es of 3.5 and 5.6 years of age residing in a southern community. All of the participants were enrolled in one of four center-based preschool programs. A family information questionnaire was distribut ed at the preschools at the beginning of each assessment peri od. Questions were related to parents perceptions of the importance of informal peer contacts for their children, family structure and family demographics. Sociomet ric ratings and nomina tion procedures were utilized to assess childrens peer status among classmates at each time of measurement. Parent telephone logs were also utilized to assemble a collection of parents initiation practices and childrens non-sc hool peer contacts for a total of 8 days during each
47 assessment period (16 days total). Interviews were conducted in which parents were to provide an hour-by-hour account of their childs non-school so cial activities (including contacts with peers). Interviewers reco rded the duration of each reported contact (minutes), the location of the contact, and a description of the involved peer or peers. Along with the basic data, questions were used to ascertain who had in itiated the contact, the degree in which the parent involved the ch ild in initiating the contact, and also the degree to which the parent involved the child in planning the contact. Lastly observations of childrens behavior at pr eschool were documented. Each child was observed for a total of 90 scans during pre-play periods. Data was examined through analyses of variance (ANOVA) to determine whether initiation frequencies differed by sponsor (e .g., parents, childre n, or peers) and by age/gender of the child for each type of sponsor. Most of the families included in the sample, (81% of the parents) had arranged one or more peer contacts during the course of the investigation. Parents who placed greater value on informal peer-play activities had a tendency to initiate more of these experien ces and opportunities for their children. Such data lends support to the fact that parents values may motivate different socialization practices which in this case was evidenced by the number of o pportunities they had created for their children. These data obtained from the study sugge st that parents initiations may be associated with several aspects of child co mpetence. In essence, by arranging frequent peer contacts parents are provi ding a context for children to exhibit and learn prosocial behaviors with peers and promote sociabil ity. Longitudinal studies are needed to determine the age at which parents begin to in itiate peer contacts for children and when
48 during development the responsibilities are transferred to the ch ild. Beyond this, the efforts of parents and children to arrange informal play activ ities should also be examined with more diverse population samples. Peer Play Interactions and Measurements of Child Development Research currently supports that the development of social competence is a critical task for young children to accomplish in order to function successfully in school and in the social world, since socially co mpetent children are able to recognize and respond to the established norms operating w ithin both outside and in the school context (Gagnon & Nagle, 2004). Evaluation of pr eschoolers constitutes a fundamentally different task than assessing school-age children for a number of reasons. First, the behaviors of these children w ithin testing situations can affect the accuracy of results as their abilities and behaviors do not remain static until later developmental stages. Second, preschoolers cognitive abilities depend to such an extent on other skills, it is pertinen t to examine the various domains of functioning, including social competence, emotional expression, sel f-regulation, coping with new situations and challenges, and play behavior, rather than focusing solely on cognitive development (Gagnon & Nagle, 2004). Assessment in this way is convergent, and allows the involvement of formal and informal measurem ent tools to enhance social and treatment validity by providing the most valid estimate of developmental functioning at specific age ranges (Fisher, 1992; Gagnon & Nagle, 2004). Schol ars have attempted to measure social competence for a number of years through the use of rating scales completed by teachers, parents and students in addition to, direct observa tions and sociometric assessments.
49 Despite the source of data, it is important to define the construct of social competence as reflecting the integrity of child ren in different developmental periods (Pellegrini, 2000). Assessment tools must be consistent with recommendations outlined in Public Law 99-457, which state that they must have treatment validity, documented reliability and validity, and involve the active particip ation of parents in order to obtain a comprehensive evaluation. In addition, assessments should be multifaceted in that information should be gathered from multiple sources, instruments, and settings taking an ecological perspective that allows examination of the childs developmental skills within social, school, and family contexts (Gagnon & Nagle, 2004). The evaluation of play rev eals childrens patterns of interactions with both caregivers and peers. Play-based assessmen ts may be a valid substitution for other commonly used measures for school-aged children in their ability to provide information not only on developmental skills but also to reflect the importan ce of parent-child relationships, and the significance of play as a primary context for young childrens learning and development across settings (G agnon & Nagle, 2004). Third-party ratings may also be incorporated because they do not rely on the childs cooperation, which is often a noted hindrance in the assessment of preschoolers. This type of assessment also allows a broader range for acquiring information from multiple sources. In this way play-based assessments may be considered f unctional, since the re sults they produce may directly be linked to interventi on strategies for young children. There is a need for the c ontinuation of evaluating psyc hometric properties of play measures. Assessment of childrens play behavi ors are a measure design ed to be naturally consistent with recommendations for preschool assessment. Through a study conducted
50 by Gagnon and Nagle (2004) negative correla tions were found between ratings of the Play Disruption and Play Disconnection and th e social skills factors of childrens behaviors. Participants were 85, four-year old children (43 male and 42 females), of mainly Caucasian decent (80 children), enrolled in an early intervention program identified to be at risk for future problem s in school. Parent and teacher ratings of childrens play with their peers were asse ssed with respective versions of the Penn Interactive Peer Play Scale (PIPPS, Fantuzzo, Mendez, & Tighe, 1998). The PIPPS is a 32-item rating scale designed to differentia te children who display positive versus negative behaviors during play with peers. The three factors that are measured include Play Interaction (prosocial behaviors and stre ngths in play), Play Disruption (aggression and nonsocial play), and Play Disconnection (withdrawn behavi ors and lack of participation). The Vineland Social-Emotiona l Early Childhood Scales (SEEC; Sparrow, Balla, & Cicchetti, 1998) was also administered to parents and teachers of children in the study. This measure was used to examine social-emotional development. Multivariate techniques were utilized in order to analyze the obtained data. Reported findings from the study showed that preschoolers rated as exhibiting high levels of Play Interaction on the pare nt version of the PIPPS were also rated high on the Social Skills Rating System (SSRS; Gresham & E lliott, 1990). Positive correlations emerged between the Play Interaction and the Self Control, Interpersonal Skills, and Verbal Assertion scales on the SSRS. Negative correlations were found between ratings of the Play Disruption and Play Disconnection and th e social skills factors of childrens behaviors.
51 These findings are relatively straightforward and consistent with previous findings from other studies. Children w ho scored high on Play Interaction convey childrens play strengths; including prosocial skills such as helping and encouraging others to participate in play and were found to rate positiv e in terms of their socialemotional skills. Significant correlations were also found between parent and teacher ratings on the PIPPS and the SEEC to refl ect the important relationships not only between the two measures, but also between th e two involved construc ts: peer interactive play and social-emotional development. Given the established importance of social emotional development to childrens succe ss in school, knowledge of childrens play behaviors in this context should be proven espe cially valuable. For children who display poor play behaviors or who fail to engage appr opriately in play with peers, specific playbased interventions should be developed to improve these behaviors. Helping children learn to interact more positively with peers during play may positively affect their success in school. The psychometric integrity of the PIPPS was further validated and extended beyond the initial norm ed sample (African-American children residing in an urban area enrolled in Headstart programs) by including a sample of primarily Caucasian children living in a more rural area Although childrens competency in interac ting with peers has been associated with long-term school success, little is known about how this competency may relate to other learning readiness domains. The findings of this research are of particular interest for lower SES children as it has been well esta blished that poor children are more likely to have difficulties in school ranging from behavioral and emotional difficulties to retention and poor school performance. C oolahan, Fantuzzo, Mendez and McDermott,
52 (2000) examined how multiple dimensions of inte ractive peer play relate to constructs of learning behaviors and problem behaviors in low-income, mi nority children specifically. The researchers hypothesized that positive pl ay interactions would enhance learning behaviors, including competence, motivation, pe rsistence, and attit ude towards school, while negatively correlating with clas sroom behavior problems. Disruptive and aggressive peer interactions would correlate positively with inattentive-passive classroom behavior and negatively with le arning behaviors, particularly in areas such as attention and task-related persistence. The participants of the study were 566 preschool children enrolled in a Head Start program in a northeastern U.S. city. Children ranged in age from 44. 8 to 71.8 months (M = 59.4). Teachers were asked to complete the Penn Interactive Peer Play Scale (PIPPS: Fa ntuzzo et al. 1995), and either the Preschool Learning Behaviors Scale (PLBS: McDermott, et al. 1996) or the Conners Teacher Rating Scale28 (CTRS28; Conners, 1990) which are measures used to assess successful and faulty learning patterns in young children. Children wh o exhibited high interactive play ratings received high social skills ratings from t eachers and were indicated as being well liked by peers, and engaged during play sessions. Ch ildren who were engaged in disruptive play received ratings of low self-control and were more likely to be engaged in solitary play. Those children who were rated as disconnected in play did not appear to be well accepted by peers and were not engaged during play sessions. Canonical and redundancy analyses were performed to investigate the relationshi p between the PIPPS factors and those of the PLBS and CTRS-28. Children who exhibited high interactive pl ay ratings received high social skills ratings from teachers and were indicated as being well liked by peers and engaged during
53 play sessions. Children who were engaged in disruptive play received ratings of low selfcontrol and were more likely to be engaged in solitary play. T hose children who were rated as disconnected in play did not appear to be well accepted by peers and were not engaged during play sessions. Greatest overlap was found with Play Disruption correlating negatively with the Attitude factor, and Play Disconnection correlating negatively with the Competence Motivation fa ctor. Overall results showed that the findings were aligned with the original hypot heses proposed by the researchers of this study. In summarization, children who demonstrated positive interactive play behaviors were actively engaged in classroom learning activities and displayed higher levels of competence motivation, attention, persistence, and withheld a positive attitude toward learning in school than did those children who were less engaged in peer play. The present study indicated that peer interact ion difficulties were tied to poor learning behaviors, including an unwillingness to accep t assistance from the teacher and lack of effort put forth when faced with new chal lenges. The findings indicate a need for longitudinal data to evaluate ways in whic h peer relationships relate to motivation towards learning and school functioning dur ing different developmental periods. Lastly, the findings of this study have implications for school psychologists and practitioners. Specifically, they indicate that childrens success at peer interaction is related to the quality and level of engagement in the cl assroom context, suggesting that greater opportunities for successful peer play interaction may enhance social competence in addition to academic adjustment. School psychologists should also attend to the assessment of peer interaction difficulties along with teachers as a marker for early school difficulties and encouraging peer inter actions amongst children with varied levels
54 of interactive play as a successful in tervention for children who may display inappropriate classroom behavi ors (Fantuzzo et al. 1996). A study conducted by Fantuzzo and Mc Wayne (2002) investigated the multivariate relationships between peer-play behaviors as evidenced in the family and neighborhood environments. More specifically, peer-pla y relationships within the family context were selected as a focus for this study to emphasize the developmental-ecological theory and research supporting the importance of primary so cialization experiences as a foundation for classroom learning. Overall, meaningful relationships were obtained across family, teacher, and independent observer reports informing the continuity of peerplay behaviors across home and school setti ngs as well as relati onships to childrens engagement in preschool classroom learning. Children exhibiting interactive play with peers at home (e.g. directing play activity, he lping other children settle conflicts, etc.) received high ratings of collaborative play by classroom teachers. These same children were also reported as having positive approaches to learning as evidenced by their increased tendencies at appropriately managi ng frustration, cooperate in learning groups, ask for and accepting help. The level of interac tive peer play that was displayed was also associated with teacher reports of childrens motivation to learn and childrens display of autonomous behavior (e.g., initiation of tasks, independent exploration). Childrens initiation of tasks and motiva tion to succeed are two important factors contributing to preschool success and later positive learning experiences. On the other hand, children who evidenced disruptive pee rplay experiences at home and in the neighborhood were reported as being disruptiv e in the classroom w ith peers and during the learning process. Such children e xhibited tendencies such as starting
55 arguments/fights, physical aggression and ve rbally offending others during play at school. The overall results are consistent in supporting evid ence that peer social and learning-related competencies are interdep endent for younger children. These research findings also indicate that parents can c ontribute quality information for facilitating communication with teachers regarding chil drens needs and competences that may impact the experiences of classroom learning. Fu ture research must al so closely consider how family variables impact and/or mediat e childrens play and school adjustment. Gender Issues in Measures of Social Competence Additional variables to consider when assessing socio-emotional characteristics in children are reported gender differences and th e biases that may drive informant ratings of social competency measur es. Issues regarding disagr eement-agreement factors among informants are also complicated by the fact that the amount of agreement will vary with the age and sex of the child (Mash & Bark ley, 2003). In a study conducted by Elliott et al. (1989) results from the teachers ratings on the SSRS indicated consistent significant mean differences between boys and girls. Gi rls appeared to exhibit more frequent prosocial behaviors and boys were more freque ntly displaying interfering or negative behaviors. Considerable evidence also sugge sts that temperamenta l characteristics are also a predicting factor of childhood adjustme nt (Bodzinksy, Elias, Steiger, Simon, Gill, & Hitt, 1992). Boys and girls are suggested to have differing physiologi cal responses to social interactions which further cont ribute differences in to adjust ment outcomes. In evocative social interactions boys are more likely to r eact aggressively to p eers whereas girls are more likely to actively and calmly defend themselves when faced with peer conflict.
56 Along the same lines, boys are also depicted as less likely to utilize problem-solving strategies to seek out assistance from others wh en put into stressful s ituations. As a result, they generally appear to score lower on measur es of regulation and delay of gratification when compared to girls (Bodzinksy et al. 1992). On the other hand, girl s play tends to be oriented more toward maintenance of c ooperation, concordant, and close relations (Fabes, Shepard, Guthrie, & Martin 1997). Gende r differences seen in arousability and regulation of arousability may be related to the high levels of physical play and dominance that contribute to the qualities linked to boys and not girls. Another component considered in the proposed study is the social groupi ng process that is assumed initially in young childre n. Boys are also more likely to play in larger groups of mainly other boys where strong ar ousal is elicited and play is generally characterized by physical contact. Additionally instances of pl ay usually take place further away from adult supervision. Moreover, girls tend to play more often in clusters of two or three and under the close supervis ion of adults (Fabes, et al. 1997). Fabes et al. (1997) discuss f actors that may contribute to the perceived differences in the behaviors of boys and girls. Resear chers proposed that gender differences were evidenced in temperament arousal and affected by their participati on in same-sex peer play interaction groups. Predic tions were made that play with same-sex peers would moderate the relation of temperamental ar ousal to childrens problem behaviors and adjustment for boys and girls. For arousable boys, it was hypothesized that behaviors of hostility and disruptiveness woul d be exacerbated the more th ey interacted with other boys. In contrast, girls who were arousable pl ay with same-sex peers were expected to
57 decrease the likelihood of the continuation of externalizing be haviors toward the generation of more gender expected behaviors. Participants were 57 children (29 boys, and 28 girls) enrolled in six preschool or kindergarten classes at a university-based day care facility. Brief behavioral observations of childrens free play were assessed everyda y over a 3 month period. Buss and Plomins (1984) Temperament Survey were administered to teachers in order to describe each childs temperament using 18 items. In addi tion, secondary teachers were requested to complete the Conduct Problems Prevention Re search Groups (1993) Problem Behavior Checklist for determining the level of exte rnalizing behaviors being exhibited by the children in the classroom. Each childs same-sex peer play score was derived by calculating the number of same-sex interactions for that child by his or her total number of observations to reflect the proportion of freeplay interactions that children spent with same-sex peers. Regression analyses were th en conducted separately for boys and girls. Generated results indicated that childre ns temperamental arousability (i.e., the degree to which children were reactive to ev ocative contexts and stimuli/degree to which they could regulate this reactivity) interacted with their tendencies to play with either same or opposite-sex peers. Importantly, these relations varied for girls and boys. On one hand, boys who were rated high in arousabilit y and who were observed to play more frequently with same-sex peers showed the greatest levels of exhibiting problem behaviors. It appeared that boys who were easily aroused were at an increased risk of becoming over-aroused and dysregulated in the context of play with other boys. This finding may indicate th at highly arousable boys ma ke seek out other highly arousable boys who also engage in rough a nd physical kinds of play that they like.
58 Younger children may be more likely to play an d relate to those that are the most similar to themselves in regards to styles of play behavior. Girls demonstr ated a very different pattern in the interaction of arousability a nd same-sex peer play. Highly arousable girls who played more with same-sex peers appeared to lower their level of rated problem behaviors. Thus, for girls high in arousability, playing with other girls seemed to inhibit their dysregulation for those who were able to move up in peer status. Those girls who were unable to lower their arous ability levels had relatively low peer status ratings. The results suggest that girls peer groups ar e effective at moderating the likelihood that originally arousable girls display problem be haviors. Girls may not accept others into their play groups if they do not adhere to th e relatively calm and friendly interactional style that is found in most gi rls play groups. Although most studies on peer interactions with younger children appear to focus on groups of same-sex peers and although developmentally children at this age do e xhibit this grouping most commonly, research may be warranted in examining the outcomes of children who engage in more oppositesex peer interactions. Future studies may seek to employ methods in which boys are mediated in their disruptive behaviors, and whether play between opposite-sex peers alleviates problem behaviors or further enhances them. Although the study of social skills across genders has promoted the expression of prosocial behaviors, it is still unclear as to what influences children to behave prosocially. Difficulties attributing to the collection of such data include the variety of studies that use diverse research methods while also involving a wide range of ages. Furthermore, factors such as childrens individual characteristics and socialization experiences have for the most part been examined independently of each other (Farver & Branstetter, 1994). In
59 response to mixed findings relating individual variation in y oung childrens social behavior, Farver and Branstette r (1994) extended prior work by examining preschoolers naturally occurring responses to their crying peers in an attempt to decipher how childrens individual ch aracteristics, and experiences w ith teachers and peers may shape their responses. Researchers hypothesized that both gender and age would contribute to increased prosocial behavior s including helping and comf orting behaviors, based on findings indicating that with maturity, child ren develop higher le vels of cognitive functioning, social skills, and moral reasoni ng than younger children. However, of more interest was the fact that gi rls were expected to respond mo re prosocially often than boys. Social competency levels were suggested to increase childrens level of prosocial behavior. Socially skilled ch ildren tend to be better at perspective taking, or surmising what a peer may be thinking or feeling, thus girls were hypothesized to hold a general higher rating of social compet ency compared to boys. Theref ore researchers beliefs were aligned with stereotypical behaviors expected of girls. Guidelines for Early Childhood Assessment In a situation where a child has not acquired adequate soci al skills it is very likely that it will be reflected in negative child-a dult relationships as well as negative childchild relationships (E lliott, Barnard, & Gresham, 1989) Thus the identification and treatment of socially delayed or at-risk preschool children warr ant the attention of teachers, parents, and psychologists alike. Up until this point, much of the review has illustrated important aspects of the normative development of social skills. However, many practical issues in the assessment and remediation of social skills deficits in preschool children have scarce ly been addressed. Such fact ors include the identification
60 of behaviors that are considered to be important to parents and teacher of young children, the use of parents and teachers in the assessm ent of preschoolers social skills and selfconcept, and also the influence of child, family and teacher background variables on social behavior and development. Young childrens social competencies have been hypothesized as key predictors for latter school adjustment, particularly when it is matched by higher teacher and parental expectations for success (Raver & Zigler, 1997). Therefore it is important to examine how parents and teachers rate such measures when evaluating young children. Similarly, research has repeatedly indicated that children who can remain emotionally positive during the course of group interactions are viewed by teachers and parents alike as more likeable and easier to get along with (Denham & Holt, 1993). These are all concerns confronting many educators and ps ychologists involved in the delivery of psychoeducational services to preschoolers, yet there is a lack of published literature which addresses these issues (El liott, Barnard, & Graham, 1989). The development of assessment measures must meet certain criteria to ensure that the instruments are based on sound scientific methodology. Practitioners are encouraged to use assessment methods th at possess qualities of reliab ility, validity, pr acticality, and social validity. However, few social skill a ssessment methods meet all these criteria, and even fewer for those tailored for preschool ers. Although many measures have been developed to measure young children at risk for behavioral issues most of them (a) focus on problem behaviors as opposed to prosocial behaviors, (b) have limited normative data, and (c) have been designed only for parent s or teachers (Elliott, Barnard & Gresham 1989). Moreover, there is a tendency for assessm ent data obtained from different sources
61 to correlate moderately at best (Ach enbach, McConaughy, & Howell, 1987) as was evidenced in the previously discussed study conducted by El liott et al. (1989), where correlations between parents and teachers on ratings of the SSRS indicated modest relationships it is still recommended to gather multiple sources of information when assessing social behaviors that occur in multiple settings. Teacher reports and parent reports each offer a unique perspective on a childs social-emotional and behavioral functioning. Ratings are most effective when taken from various informants in a comprehensive assessment. Parent and Teacher Reports Increasing numbers of children continue to be enrolled in daycare and preschool settings as parents enter the workforce, provi ding early access to peer interactions. Given this trend, it has become pertinent for educator s to develop an understa nding of the nature and value of childrens peer in teractions and the role thes e relationships play in the development of social, cognitive, and em otional development (Gagnon & Nagle, 2004). Assessment instruments should have the capac ity to establish a link between the home and school. Reports of play be havior provide the ability to track important outcome measures for teachers and pa rents to better understand their children and for interventions to be designed for young children. A multi-source, multi-setting, multi-method social-emotional assessment design for young children must focus heavily on a combination of informant-derived data since self-report measures and interviews directly with the child are typically of little use and of questionable accuracy (Merrell, 1996). Pianta and Walsh (1996) argue that a successful transition into kindergarten requires the collaborat ion of parents and teachers
62 while emphasizing the childs development. This is of particular importance for children who may evidence ethnic and cultural differences between the childrens families and their teachers since the educati on that takes place in the family is a powerful prerequisite for success in school (Ramey & Ramey, 1999). In addition it is important to be aware of teacher expectations and their perceptions of young students. Children they perceive to be academically and socially more advanced may be treated differently and tend to outperform their peers who are perceived as less competent regardless of whether this may actually be the case (Cooper, 1979; Sar acho, 1991). There are clearly differences in teachers perceptions of childrens readiness and competence ability levels. It would be much more beneficial to create such percep tions in collaboration with parents and work toward a achieving a sim ilar level of competency in young children. Furthermore it is important to view readin ess for school as related to the linkage of home-school expectations and to the social and cultural components that are considered important by both teachers and pa rents (Tudge, et al. 2003). An assessment of a childs overall functioning must include the home and school settings, because a child may demonstrate strengths in one context but experience difficulties in a different setting. Tudge et al. (2003) utilized Bronfenbr enners Process-Person-Context-Time (PPCT) ecological model to focus on the relations between school-re levant activities of preschool-aged children and teachers subsequent perception of childrens competence upon entering school. The aim of the research was to examine the relations among young childrens engagement in certain types of natu rally occurring activitie s, parents values, and teachers perceptions of competence once having entered school.
63 Participants were 20 white preschool ers drawn from two communities in a southeastern city in the United States. The children were observed wherever he or she was present (home, childcare center, friends houses, public places, etc) for 20 hours over the course of 1 week. Parents were administ ered the Kohns Q-Sort measure of parental values (Kohn, 1977) and asked to rate th e three most and three least important characteristics for their child from a provi ded list. Approximately 3-4 years after the children had been observed their teachers were asked to complete the teacher form of the Social Skills Rating Scale (SSRS-T, Gresha m, & Elliot, 1990). Findings reported that children who had initiated more academic lessons as a preschooler were more likely to be perceived as being competent in year 3, while those who had initiated more conversation with adults were significantly more likely to be perceived as competent in their first years of school (year 2). This research suggests that preschoolers who in itiate and engage in activities that involve interaction with adults are more likely to be perceived as competent by teachers in the first years of school. It is important that teachers and other educators in particular, are aware of the complex nature of child-to-child interaction a nd its direct relationshi p to childrens social development. Teachers are decision-making professionals who are able to constantly observe childrens behavior, form hypotheses from their observations and interpretations towards the implementation of interventions based on what they have learned. Teachers are able to enhance social development in their classrooms through establishing contexts, modeling behavior, coaching social strategies and teaching social awareness (Fisher, 1992). The interaction of teachers with young children affects childrens social and emotional outcomes either negatively or positively (Raver & Zigler, 2002).
64 Teachers in preschool classrooms are faced with a substantial number of students who lack important social and emotional co mpetencies and whose behavior is already problematic or are at risk for developing pr oblematic behaviors. T eachers and child care providers need assistance in promoting great er social skills in children and reducing challenging behaviors in the cla ssroom both to help individual children and to facilitate a positive learning climate. Developing accura te measures of social competence will generate areas to target and further promot e (Raver & Zigler, 2002). Renewed interest in defining and accurately measuring social comp etencies have been sparked by a growing awareness of the need to tap these multiple dimensions when assessing the costeffectiveness of intervention prog rams (Raver & Zigler, 1997). Current Assessment Methods In the past, the most frequently used technique for obtaining popularity data amongst children was the use of picture soci ometric methods. Despite the popular use as a measure of social competence in preschooler s, there are many points open to criticism (Connolly & Doyle, 1981). Low reliability estimat es are frequently reported for this age group, inconsistent relationships with validati ng criteria have been reported, and also higher rates of social inte raction and mature social cognitive functioning are not uniformly correlated with sociometric resu lts (Connolly & Doyle, 1981). Since then, many research attempts have been made to create alternate methods of measuring a childs peer group status. In the study conducted by Connolly and Doyle (1981) a comparison was made of the predictive validity of the classroom teachers rank ordering of the children in terms of relative popularity as a playma te with that of the traditional sociometric measure of popularity. Findings reported that the teacher-based popularity
65 measure possessed greater predictive valid ity than did the sociometric popularity measure. Although the implications of the study indicate that te acher ranking measures are much advantageous in regards to expense and utility, there are concerns with possible halo effects, as teachers were required to complete both social competence measures and rank order measures on the children. The st udy does however support research indicating that sociometric measures may not yield accura te results in measuring social competence in preschoolers, and that pr actitioners should be advised to use these measures in conjunction with other methods or re vert to using alternate methods. In a study conducted by Elliot et al. (1989) researchers examined the use of a multi-rater behavior scale, the Social Skill Rating System (Graham & Elliott, 1992) with a diverse sample of preschooler s to test the utility of cro ss-informant (i.e. parents and teachers) ratings of prosocial and problem behaviors. The parents and teachers of 212 preschool children, (aged 31 to 66 months) fr om four states, (Florida, Louisiana, Nebraska, and Wisconsin) were volunteers in this study. Three behavior rating scales were utilized: the Social Skills Rating Sy stemTeacher (SSRS-T), the Social Skills Rating SystemParents (SSRS-P), and the Bu rks Preschool and Kindergarten Inventory (Elliott et al. 1989). Areas such as Coopera tion, Social Initiati on, Self-Control, and Interfering Behaviors were assessed. Corre lations were computed to determine the relationships among the three rati ng scales and also to compar e the ratings of parents and teachers. At the conclusion of the study both sets of adults were reported to have ranked the importance of cooperative behaviors as bei ng the highest, followed by self-control, and
66 then social initiating behavior s as being very pertinent in the acquisition of good social skills. A clear sense of convergence for the fr equency of social skills was not evident between parents and teachers ratings. Results indicate that the valu es of these parents and teachers result in some specific behavior ex pectations that differ. However, this is not surprising when one considers the differences in settings and the number and types of social contacts that are possible in each of th e settings with varying individuals present. These findings are consistent with prev ious research on th e notion parent and teachers ratings of children further reiterates the unique information that is gathered by multiple informants toward the understanding of preschool childrens social behaviors (Elliot, et al. 1989). In particular for young children it may be even more pertinent to obtain multiple ratings across vari ous settings, as behaviors are likely to be transient and variable across different situa tions and with different people more so than older children. Young children are in the pro cess of establishing their bounda ries and exploring their environments through the inte rpretation of social cues and through their perceived reactions of others. Children may impress upon t eachers to behave in certain ways in the classroom, however the interaction and influenc e is bidirectional in the home, school and alternate settings. Measures must be sensitive to and attempt to capture the variation by accurately delineate the devel oping processes of young children. Findings of the above research studies s upport the notion that social competence is an area which widely influences va rious other domains in early childhood. Furthermore, research findings support that pl ay behavior is an im portant asset to young children in experiencing and ac quiring necessary social ski lls for future academic and social success. Given the importance of social emotional development in the early years
67 to childrens success in school and social interactions, knowledge of childs play behaviors in this context should prove especi ally valuable and telling (Gagnon & Nagle, 2004). In addition, input from teachers and pa rents are necessary in a comprehensive assessment of preschool students. Open co mmunication between parents and teachers through play-based assessment measures may furt her facilitate in targ eting areas that are necessary for improvement both in the home and school settings. Data obtained from parents and teachers may contribute to aspects that influence individual ratings and reveal factors that may contribute to the disparity that has continua lly been found in measures completed separately by teachers and pare nts. Among the many factors, gender effects may be a very relevant area that is worth further exploration. The broad area of social competence offers an avenue for educators to explore and gather information towards providing effective early interven tion services to young children. Purpose of the Study By comparing scores on a norm-referenced play scale with corresponding scores on a standardized measure of early social em otional skills, which holds high regards in the field, information on the utility of a play -based measure with a sample of preschool children may be provided. The purpose of the pr esent study also seeks to further evaluate the construct validity of the parent versi on of the Penn Interactive Peer Play Scale (PIPPS) and to compare it with the teacher version in a population that extends beyond the initial standardization sample. Previous re search has validated th is measure to include a sample of primarily Caucasian children living in a more rural area. This finding indicates that the PIPPS can be used validly with other demographic groups of preschool children.
68 Another goal of this study is to accentu ate the importance of parental input for obtaining a more accurate picture of peer play interactions and acquiring of social skills. The opportunities created by an assessmen t process may foster parent-teacher communications while encouraging parental in volvement in their childrens education early on. Through forming partnerships between parents and incorporating their views into the function of the classroom, early e ducation programs can begin to bridge the discontinuities that are presen t in the home and school contex ts. Play behavior measures are convenient and accessible for individuals in obtaining data that may be acquired across settings. This common information a nd knowledge may readily start dialogue and be shared between parents and teachers about important emerging developmental competencies in preschool children on a fre quent basis without bei ng intrusive (Fantuzzo et al., 1998). Researchers and professional groups have been adamant in stating that early childhood assessment should provide informati on that is not confined to providing categorical eligibility determination but informa tion that is linked directly to intervention planning (Bracken, 2000). Information obtai ned from play-based assessments can be directly linked to and translated into reco mmendations and Individualized Education Plan (IEP) and objectives. Another strength associated with using play-based assessment is the ease with which parents can be brought into the asse ssment/intervention pro cess. Best practices advise that assessment with young children include family and the importance of involving parents in the entire assessment process should be recognized. Important information about parent-child interactions may also be obtained through such forms of play-based assessment (Bracken, 2000).
69 In addition, although the focus of transitioning into kindergarten has frequently been from the perspective of the school, fam ily involvement and informal education that takes place in the home setting is also a pow erful prerequisite for success in formal education. This further supports that a succe ssful transition requires establishing strong connections between parents and teachers in which the emphasis is on the childs development (Tudge et al. 2003). The PIPPS therefore demonstrates several advantages over other assessment instruments because of its ability to: (1) assess a primary competency or early childhood; (2) measure congruent constructs among the parent and teacher versions, and (3) attend to chil drens cultural contexts. The PIPPS has demonstrated its usefulness as an instrume nt to evaluate preschool age children as presented in the literature.
70 Chapter 3 Method Introduction The purpose of this study was to explor e the relationship between play behavior and social competence in pres chool children. In addition, th is study aimed to determine the correlation between ratings of social competence and pl ay behavior obtained from parents and preschool teachers while exam ining whether frequent communication between the parties played a factor in the degree of corre lation obtained on measures. This chapter presents information regarding pa rticipants who were involved in this study, the method through which data were collecte d, and the analyses that were conducted. Participants The participants in this study were 32 pairs of preschool teachers and parents/guardians who rated preschool student s enrolled at three area preschool centers located in and around West Central Florida. Preschool students enrolled in all of the classrooms ranged in age from 3-5 years and th ose students who were proficient in either English and/or Spanish were included in this study. The three preschool centers that particip ated in this study in cluded Preschool 1, a local university-based pres chool center, and two local preschools: Preschool 2 and Preschool 3. The teachers employed at all three preschools were recruited for participation. All of the preschool teachers i nvolved in this study were employed full time
71 in the classrooms and had equal access to parent communication as well as student interactions and opportunities for observation. In addition, pa rticipating students met the criteria of at least 4 months of participation in the pres chool classrooms. The purpose of this criterion was to ensure that teachers had adequate knowledge of the children and had sufficient social interactions and observations considered representative of the child. Additionally, this time requirement allowe d for teachers and parents to have had sufficient time to establish a pattern of communication. Students who were excluded from participating in this study had been en rolled less than 4 months in their preschool classrooms, were outside of the 3-5 year ag e range, and/or were native speakers of languages other than Spanish or English. Preschool 1 is an educational research center for child development located on a university campus in West Central Florida. Th is preschool primarily provides services for children 2 5 years of age whose parents are university students, faculty and staff, or from the community at large. The presc hool also houses a small school age program. The Center's program provides an environmen t that fosters many play opportunities. The curriculum emphasizes developmentally appropria te, active, and concre te experiences for children. Emphasis is placed on the entire proce ss of learning rather than focusing on the outcomes. Experiences are provided so ch ildren are exposed to opportunities rich in social, emotional, physical, and cognitive dom ains. The outdoor play area encourages climbing, balancing, riding wheel toys, so ciodramatic play, and gardening. This developmentally appropriate program stresse s hands-on learning. Particular emphasis is placed on: appropriate play experiences; cognitive development; emergent literacy;
72 creativity and exploration; health promotion and physical development; and social development. Two preschool teachers were employed at Preschool 1 with st udents eligible for this study. Although both had originally agreed to participate in this study, only one of the teachers was able to commit due to personal circumstances. Within this classroom there were 14 students identified as meeting all of the exclusion/in clusion criteria. Ten students (71%) returned with completed pa rent and teacher rating scales and were included in this study. Total enrollment of students at this sc hool is 68 students. Demographic data indicate that 45% of students are c onsidered White, non-Hispanic, 20% Black, 35% Hispanic, and 10% are classified as Other. Preschool 2 is an independent Montesso ri School that serves Pre-K through 8 th grade. The curriculum consists of a combina tion of Montessori principles and practices with other innovative ed ucational techniques, with part icular emphasis on independent and small, flexible learning groups which are ta ilored to the individua l needs of the child. Daily lessons include reading, math, language arts, science, soci al studies, history, Spanish, library skills, computer skills, e nvironmental studies, multicultural studies, conflict mediation, and practical life skills. Th e mission of Preschool 2 is to stimulate the natural curiosity of the child, nurture a love of learning, and build self-esteem and support through a sense of gained independence and responsibility. The ra tio of teachers to students is 5:1 and allows for vast in dividualized attention. The demographic information for Preschool 2 indicate that among a total population of 61 enrolled students, 73% are considered White, non-Hispanic, 9%Hispanic, 9% Asian/Pacific Islander, 7% African-American, non-Hispanic, and 2% Native American/Native Alaskan.
73 At this school, both preschool teachers had in dicated willingness to participate and gave consent for participation. Nineteen students we re identified as eligible participants for this study. Twelve students (63%) returned comp leted packets of parent rating scales and were included in this study. Preschool 3 is a preschool center that underscores the importance of empowering children to strengthen their talents and pursue their pass ions. This school serves students enrolled in Pre-K through 12 th grade. Their unique program is designed to challenge students while providing a safe, nurturing, a nd caring environment. The framework of this curriculum emphasizes that equal impor tance be given to four human functions: thinking/cognitive, feeling/emotional, physical/sensing, and intuition. Children are exposed to all modes of these functions and they are promoted through opportunities towards developing independence, respons ibility, consideration, and cooperation. Features of the school include: small classes with a student-tea cher ratio of 10:1 or better, free acceleration allowing students to progre ss at their own speed, broad varied rich curriculum, dual-enrollment program with local universities, full time guidance program, emphasis on self-directed learning, and forei gn language classes which begin at the age of 3 years. The early childhood program has established a responsiv e environment with the provision of advanced lear ning opportunities. The programs goal is to instill a love and excitement for learning and to devel op and optimize growth in young students. The demographic information of Preschool 3 data indicate that within a total enrollment of 73 students, 75% of the students are consid ered to be White, non-Hispanic, 9% are Hispanic, 9% Asian/Pacific Islander, and 7% Black, non-Hispanic. From this particular school, there were two preschool classrooms with 17 identified preschool students who
74 were eligible for participation. Ten students (59%) returned comple ted parent forms and were included in this study. The number of participants for this study was selected based on the sample participants accessible to the researcher. A power analysis was conducted in order to determine the N (sample size) necessary to obta in a power of 0.8 for various effect sizes. Cohen (1992) provided guidelines from which to interpret pract ical use for standardized effect sizes; a small effect size was define d as 0.10 or lower, a medium effect size was defined as 0.30, and a large effect size was defined as 0.50 or higher. The proposed sample size was therefore considered adequate if the effect size obtained was large. For example, a sample population of 30 pairs of teachers and parents were speculated to provide a statistically significant effect if an effect size of r > 0.5 was established (see Table 1). Table 1 Power Analysis N for Small, Medium, and Large Effect Size at Power = 0.8 for = 0.05 Test 0.05 Large effect r > 0.5 0.05 Medium effect r = 0.3 0.05 Small effect r < 0.1 N N N Significant r 28 85 798 Mean difference 26 64 393 In total across the three preschool centers 50 studen ts met the exclusion and inclusion criteria set forth for this stu dy. Out of the 50, a total of 39 (78%) were successful in returning consent forms by the requested date. Although all 39 students were given packets of parent rating scales, a total of 32 students (82%) returned with
75 completed packets of parent rating scales wi thin the allotted time period. In summary, a total of 50 students were found eligible to participate in th is study according to exclusion/inclusion criteria across the three pres chool centers, and 32 students returned with completed packets of parent rating sc ales (64%). There were no instances where students had returned with parent rating s cales and teachers had not completed teacher rating scales. Table 2 provides the response rate by preschool a nd Table 3 presents additional demographic information obtained for each preschool center. Therefore, the present study met the minimum requirements of sample size, which enhanced the power of the st atistical analysis. Among the children for whom the rating scales were completed, 39% were females a nd 61% were males. Pa rticipating teachers and parents completed all items on the scales so that all rating scales were included in the data analysis. Table 2 Response Rate by Preschool Center Preschool Center Total Number of Total Packets % Packets Returned Eligible Students Completed Preschool 1 14 10 71 Preschool 2 19 12 63 Preschool 3 17 10 59 Total 50 32 64 Table 3 Demographic Information by Preschool Center Preschool Center Gender % Female % Male Age Range Preschool 1 40 60 3-11 years to 4-6 years Preschool 2 33 67 3-9 years to 5-2 years Preschool 3 50 50 3-11 years to 5-7 years
76 Instrumentation Parent-Teacher Communication Form The Parent-Teacher Communication Form (See Appendix A) was developed by the researcher of the study in order to collect additional information about the participants. This form includes three questions requesting teachers to indicate their relationship to the child, the childs sex, and the number of informal and formal parentteacher communications occurring either on a daily, weekly, monthly, and/or semester basis. Preschool and Kindergarten Behavior Scales The Preschool and Kindergarten Behavior Scales-2 nd edition (PKBS-2; Merrell, 2002) is a norm-referenced, standardized behavi or rating scale that is designed for use in evaluating social skills and problem behavi ors of preschool and kindergarten children ages 3-6 years in a variety of settings w ith multiple informants. Included items are developmentally sensitive and appropriate for use with young children. The PKBS-2 is estimated to take approximately 10-12 minutes to complete and contains 72 items that are completed by the parent/guardian or teacher of a child. This scale is available in English and Spanish for both teacher and parent ve rsions. The PKBS was recently revised to include an expanded norming sample consisting of greater diversity to conform to the demographic data reported in the 2000 U. S. census. In addition, separate score conversion tables for home rater and school raters are included. T hus, behavior ratings completed by parents are compared only with ratings from other home-based raters, and behavior ratings completed by teachers are compared with ratings from other schoolbased raters.
77 Items are classified into either the Social Skills scale or Prob lem Behaviors scale. The Social Skills Scale consists of 34 items categorized into three subscales: Social Cooperation, Social Interaction, and Social Independence which de scribe adaptive or positive social skills that are characteristic of socially competent children. These items are rated on a 4-point Likert sc ale (i.e., 0 = Never, 1 = Rarely 2 = Sometimes, 3 = Often). The Social Cooperation subscale includes 12 items that reflect behavioral characteristics that are considered importan t in following instructions fr om adults, cooperating, and compromising with peers. The Social Intera ction subscale includes 11 items that reflect behaviors that are important in gaining a nd maintaining acceptance and friendship with peers. Lastly, the Social Independence subs cale includes 11 items that reflect behaviors and characteristics that are important in ach ieving social independence within the peer group. The Problem Behaviors scale includes 42 items that describe problem behaviors commonly observed in the early childhood/pr eschool population. These items are rated using the same 4-point scale described for the Social Skills Scale. The items in the Problem Behavior Scale are divi ded into two empirically deri ved subscales: Externalizing Problems and Internalizing Problems. Howeve r, for the purposes of this study, only the Social Skills Scale will be utilized ther efore further details highlighting the Problem Behavior Scale will not be included. Factor analytic studies of the PKBS-2 were conducted and provide evidence in support of the validity of the instrument based on its inte rnal structure. For the Social Skills Scales, the range of inter-correlation co efficients was reported to be moderate to strong (0.58 to 0.90). The range of correlations for the Problem Behavior scores also
78 range from moderate to strong (0.47 to 0.97) The final factor solutions used in developing the PKBS-2 subscales are psychomet rically strong, and c linically useful. In order to score the responses on the PK BS-2, the administrator calculates raw scores for the subscales and the composite scor e, which are then converted into standard scores, percentile ra nkings, and functional levels guided by a conversion table provided in the manual. Revisions to the original PKBS scoring allows each empirically derived subscale to contribute equally to the total score, regardless of the number of items in each scale which is derived by summing the standa rd scores for each subscale within each scale. Merrell (1994, 2002) found test-re test reliability estimates to range from 0.36 to 0.78. More specifically, test-retes t reliability for the Social Skills Total at 3 weeks and 3 months was 0.58 and 0.69, respectively. Test-retest reliability for the Problem Behaviors Total was 0.86 and 0.78, respectively. Alpha and split-half coefficients measuring internal reliability range from 0.90 to 0.97 for Total Scores, and ranged from 0.80 to 0.90 for the Social Skills and Problem Behaviors subscales. Inter-rater agreement coefficients for reliability within preschool settings were moderate to moderately strong, ranging from 0.36 to 0.61 for Social Skills subscales, and from 0.46 to 0.63 for Problem Behaviors. All are statistically significant at the p < 0.001 level. Inter-rater reliabilities for the PKBS-2 across raters in different sett ings range from 0.2 to 0.57 for Social Skills subscales, while the range of coefficients for in ter-rater reliability for the Pr oblem Behavior scores is 0.13 to 0.42. Overall, these results indicate that th e ratings of children are likely to vary across settings with different rate rs, indicating that preschool-a ge children may behave in different manners across the home and school settings.
79 In order to examine the crit erion-related reliability of the Spanish version, it was compared to the English version using th e Pearson bivariate product-moment method. Reliability coefficients that resulted from this procedure were 0.93 for Social Skills and 0.94 for Problem Behaviors. In addition, intern al consistency of the Spanish form was found to be high at 0.93 for Social Skills and 0.96 for Problem Behaviors (Carney & Merrell, 2002). This finding provides further support for the comparability of the two PKBS-2 forms as measuring the same construct in a similar manner. The idea supporting the establishmen t of validity between measures by investigating relationships to other measures is based on two premises. Test scores are purported to measure specific c onstructs and should theref ore correlate strongly with scores from other instruments that have dem onstrated a measure of the same construct, therefore evidencing convergent validity. S econdly, test scores purported to measure specific constructs should correlate either w eakly or inversely with scores from other instruments that have demonstrated to measure opposing constructs and exhibiting discriminant validity (Merrell 1994, 2002). Convergent and discriminant construct vali dity of the PKBS-2 have been explored with several different behavior rating scales designed to measure a variety of social, emotional, and behavioral constructs. Me rrell (1995b) investigat ed the relationship between the PKBS-2 scores and scores for the preschool form of the Social Skills Rating System (SSRS: Greshman & Elliott, 1990). In this study, ratings from both instruments were compared for a sample of 86 preschool-a ge children (3 to 5 years), who had been referred for special educati on Child Find screenings in a large urban public school district. Relationships between the raw scores obtained on the social sk ills scales for the
80 PKBS-2 and the SSRS for this sample were foun d to be correlated at a moderate to strong level, with the coefficients ranging from 0.32 to 0.76. The problem behavior scores of the two measures reported a range of coeffici ents from 0.25 to 0.83. Stronger correlations were found between internalizi ng to internalizing and exte rnalizing to externalizing problem behavior scores of the two measur es. Negative correlations were obtained between the social skills a nd problem behavior scores. An additional convergent validity study for the PKBS-2 was reported in Merrell (1995b). PKBS-2 scores and teacher ratings on the Matson Evaluation of Social Skills with Youngsters (MESSY; Matson, Esvelt-D awson & Kazdin, 1983; Matson, Rotari & Helsel, 1983, 1985) were compared. The MESSY teacher form produces two factor scores for Inappropriate Assertiveness/Impul sivity and Appropriate Social Skills. These domains appear to sample the same social skills and problem behavi ors that are purported to be measured by the PKBS-2. The results of the bivariate Pearson product-moment correlations reported most of the coefficients as statistically significant at the p < 0.001 level. Coefficients for the Social Skills Scale on the PKB-2 were moderately to very strong (0.62 to 0.85) and relatively weak to quite strong for problem behavior scores (0.22 to 0. 72). The correlations between th e PKBS-2 composite scores and the MESSY factors were substantial ( 0.84 between the Social Skills composite score and MESSY Factor II, and 0.64 between the Problem Behavi or composite score and MESSY Factor I). The majority of correlations between scores purported to measure differing constructs was negative, and was very weak to moderate in strength.
81 Penn Interactive Peer Play Scale (PIPPS) The Penn Interactive Peer Play Scale (P IPPS; Fantuzzo, Sutton-Smith, Coolahan, Manz, Canning & Debnam, 1995) is a 32-item be havioral rating instrument useful for understanding peer play beha viors and for meeting the need for congruent play assessment measures for parents and teachers across settings during early childhood (Fantuzzo, & McWayne, 2002). The PIPPS was de veloped to assess th e interactive peer play behavior of young children. It was deve loped in partnership with teachers and parents to assess play in the home and school contexts, and to gain an understanding of cultural expressions within play. The PIPPS has parallel versions for parents and teachers, and for preschool and kindergarten-age children. The parent version assesses play in the home and neighborhood, whereas th e teacher version examines at school. Each version contains 32 f our-point Likert-scale items that indicate how often the behavior has been observed during free play (i.e., never, seldom, often, or always). The items assess competencies within play to identify children who demonstrate successful peer play interact ions and those children who experience difficulties with peer play. To identify ch ildrens strengths and weaknesses, the PIPPS includes descriptions of positive and negative play interactions. Studies of the reliability and validity with preschool chil dren reveal three reliable dimensions: Play Interaction, Play Disruption, and Play Disconnection (Fantuzzo, Mendez et al., 1998; Fantuzzo, Coolahan et al., 1998; Hampton & Fantuzzo, 2003). The Play Interact ion factor includes items that describe cooperation, helpful behavi ors and creative behavi ors that contribute to successful peer play interactions. The Pl ay Disruption factor describes aggressive, antisocial behavior that inte rferes with ongoing peer play interactions, whereas Play
82 Disconnection examines nonparticipatory behavi ors in play interactions with peers (Gagnon & Nagle 2004; Hampton & Fantuzzo, 2003). Studies have evaluated the ability of the PIPPS to assess intera ctive peer play for preschool children (Coolahan, Fantuzzo, Mendez & McDermott, 2000; Fantuzzo, & McWayne, 2002; Fantuzzo, Sutton-Smith, C oolahan, Manz, Canning & Debnam, 1995; Gagnon & Nagle, 2004; Fantuzzo, Mendez & Tighe, 1998; Fantuzzo, Coolahan, Mendez McDermott & Sutton-Smith, 1998). To date, both construct and concurrent validity have been demonstrated. A series of common expl oratory factor analyses were conducted for both parent and teacher versions of the rating scale that examined both orthogonal (varimax) and oblique (promax) solutions. The three-factor or thogonal solution best satisfied criteria that were set by researchers. The constructs for each version have demonstrated reliability, with Cronbach alphas of 0.92, 0.91, and 0.89 for the respective teacher factors of Play Interaction, Play Disruption, and Play Disconnection, and Cronbach alphas of 0.84, 0.81, and 0.74, respectivel y, for the parent factors (Fantuzzo et al., 1995). In order to fulfill the need for quality p eer play assessment for Hispanic preschool children, a Spanish teacher vers ion of the PIPPS was recently developed and is currently being evaluated. Results of factor analyses of the Spanish version replicated and confirmed a three-factor soluti on similar to that of the Eng lish version. Cronbachs alphas of 0.830.88 were established (Cas tro, Mendez, & Fantuzzo, 2002). Concurrent validity of the preschool PI PPS has been examined with measures of social competence using teacher rating s cales, sociometric techniques through peer reports and direct observations of play. One study found that ch ildren who evidenced
83 high interactive play ratings also received hi gh social skills ratings from teachers, and were also well liked by peers and engaged during play sessions observed (Fantuzzo, Coolahan et al., 1998). Moreover, children who were perceived to be disruptive in their play as rated by the PIPPS received ratings of low self-control and were more likely to be engaged in solitary play. Finally, children who received high ratings of Play Disconnection were associated with low acceptability by peers and observed to engage in solitary play (Fantuzzo, Coolahan et al., 1998). Additional assessments of concurrent validity have been undertaken with measures of school functioning and dimensions of school readiness. Relationships were found between parent ratings of home-based, pe er play behaviors and four measures of childrens classroom behaviors th at contribute to effective social functioning within the classroom (Fantuzzo & McWayne, 2002). Speci fically, childrens classroom behaviors such as school-based peer play, approaches to learning, self-re gulation, and behavior problems were analyzed using bivariate correlational and multivariate methods in order to determine that play competencies exhibited in the home environment were significantly associated with prosocial behavior in the cl assroom, motivation to learn, task persistence, and autonomy. Moreover, children who were rated as exhibiting disruptive or disconnected play behaviors were significantly rela ted to demonstrating tendencies of disruptive and dys regulated experiences in the classroom with peers and with the learning process. Another study ev aluated the relationship of the PIPPS to social-emotional development as assessed by the Vineland Social-Emotional Early Childhood Scale (SEEC; Sparrow, Balla, & Ci cchetti, 1998) furt her validating the psychometric integrity of the PIPPS by exte nding beyond the initia l standardization
84 sample to include a sample of primarily Ca ucasian children living in a more rural area. Significant relationships were found between teacher and parent ratings of childrens play with their peers and their co rresponding observations of social-emotional development. Relationships additionally demonstrated the impo rtant role of play and peer relationships in the development of social competence. Mo re specifically, the significant relationships between the Parent PIPPS and the SEEC indicated that children who had acquired competent play behaviors also displayed st rong social-emotional sk ills whereas children who were negative and aggressive during play in teractions tended to display lower levels of social-emotional development. The canonica l variate that emerge d from the analyses indicated that Prosocial Involved dimensions from the SEEC reflected strong, positive relationships between Interpersonal Relations hips, Play Interacti on, Play and Leisure time and Coping skills, and a negative a ssociation with Play Disruption (Gagnon & Nagle, 2004). These studies indicate th e usefulness of the PIPPS for assessing an important aspect of childrens social competence dur ing preschool. The PIPPS has demonstrated its relation to other measures of social compet ence and classroom functioning in addition to construct validity. An explorati on of the technical properties of the two presented rating scales show that scores obtained on these instruments are adequately reliable and valid. The technical properties of these instruments, along with their lack of intrusiveness and appropriateness for the characteristics of this sample make them appropriate instruments for use in this study.
85 Procedures Protection of research participants Several steps were taken to protect res earch participants. Approval was obtained from the University of South Florida Institutional Review Board (IRB), and from the directors of each of the partic ipating preschool centers. Pa rents were given an informed consent form to sign that described the purpos e of the study, rights of the participants, nature of their involvement, measures to en sure participant anonymity, methods in which data would be collected, and a description of how data would be stored during and after research completion. The explanation of pa rticipants rights included information regarding confidentiality, ability of particip ants to withdraw, refusal to answer any question, and emphasis on voluntary participatio n. Teachers were also informed of the purpose of the study, confidentiality procedures and voluntary participation in a similar manner. Rating forms completed by teacher/p arent pairs were assigned a number and all identifying information was removed from these documents. All of the forms and letters distributed to parents were thereafter organi zed according to student numbers and entered into a database. Recruitment of center for participation Teachers at the three preschool centers we re contacted by the investigator either in person, via telephone or through email and i nvited to participate after permission from the Directors of each school was obtained. Teachers were selected based on voluntary participation and also based on whether chil dren enrolled in their classrooms met inclusion/exclusion criteria a ccording to age, time of en rollment, and native languages spoken by the children. Only those students re turning with parent consent forms were
86 included in the study. Teachers were consul ted regarding the best approaches for collecting information from parents, and the ti mes most convenient for the researcher to collect data from the teachers. After the i nvestigator had consulte d with the teachers, standard procedures were developed and adhered to throughout the data collection process. Across all three sites, simila r procedures were followed. Modes of communication consisted of sending letters hom e with children as well as handing off notes to parents at the begi nning and end of the day. When letters, consent forms and rating packets were completed and returned, teachers and/or Directors from each of the sites contacted the investigator via email. The investigator then went to the locations to pick them up. Parent data collection An informal letter was distributed to all parents or guardians of children who were enrolled in each of the classrooms that had teachers serving as participants in the study. This letter briefly explained the purpose of the study, procedures for collecting data and requested their voluntary part icipation (see Appendix B). Pa rental consent forms were then administered to only those parents and caregivers who had gi ven initial consent by signing and returning the informal letter. The consent forms reiterated confidentiality of all responses and that the purpose of data co llection was strictly for research. Parents were asked to return the consent forms to their childs teacher. All consent forms and rating scales were available in both Spanis h and English in order to allow parents to complete forms in their most proficient langu age. Considerable care was taken to assure that a systematic process of translation wa s achieved in developing equivalence in all letters and consent forms so that cultural eq uivalence was achieved. Several steps were
87 taken to ensure the preservation of meaning across languages includi ng the assistance of a bilingual School Psychology Graduate student to translate all documents, and back translation by an independent bilingual indi vidual, a person not involved in the earlier translation. All students possessing surnames of Hispanic origin were given two complete sets of documents, one in each language. Rating sc ales were requested for completion by one parent, legal guardian or prim ary caregiver per child. A prim ary caregiver was defined as an adult with whom a child lives and who ha s responsibility for th e child. Individuals should have opportunities to obser ve the child in peer play interactions and have had responsibility for communicating with the classroom teacher. Consenting parents were sent home a packet through means of the classroom teacher which included the PIPPSParent and the PKBS-2 parent version forms. All rating scales were counterbalanced to decrease the possible bias resulting from or der effects. Parents of students who were assigned an even number in the database were instructed to fill out the PIPPS first while parents and teachers of students assigned an odd number were instructed to first complete the PKBS-2. The completion time was predic ted to range from 20-30 minutes. Modes of parent-teacher communication included the use of a folder that the child brought home for the parent to check, and handing documents/p apers directly to parents at the end of the day. All sets of child measures were reque sted for completion within a two week span and the investigator arranged to pick up th e completed forms at the respective schools from classroom teachers.
88 Teacher data collection Teachers received the Parent-Teacher Communication Form, PIPPS-Teacher, and the PKBS-2 teacher version to complete for those children for whom parental consent was obtained. Rating scales were counterbalanced in a similar fashion through procedures described above for parents. The completion time for teachers was predicted to range from 20-30 minutes. All sets of child measures were requested for completion within a two week span and the investig ator arranged to pick up mate rials at the respective schools when contacted by classroom teachers and/or school directors indicating that the forms had been completed. Scoring of Protocols All scores obtained on the protocols we re entered into a database according to student number. In order to ensure integrity of scoring procedures and obtain inter-rater agreement, 24% (8 out of 33 possible) of the protocols were randomly selected and scored by another School Psychology graduate student. Only scores producing at least 80% agreement according to the in ter-rater reliability formula (Agreements/Disagreements + Agreements) were included in the data analyses.
89 Chapter Four Results This chapter presents the data analyses conducted to examine parent and teacher observations of preschool chil drens play behaviors. To determine whether observations of play behavior in preschool children accurately assessed social competence as measured by traditional social skills rating scales, the scores of both parent and teacher versions obtained on the PKBS-2 and PIPPS were examined. The analyses were conducted separately for the teacher and pare nt measures to examine the relationship between the sets of variables of the PIPPS and the corresponding version of the PKBS-2. Separate analyses were conducted for the rela tionships between the PIPPS-parent and the PKBS-2 parent version and between the PI PPS-teacher and PKBS-2 teacher version. An assessment of concurrent validity was conducte d to determine how we ll the constructs of the PIPPS related to an outcome criterion select ed as a standardized measure of social competence. Internal Reliability for the PK BS-Social Skills Scale and PIPPS In order to assess the reliability for ob taining accurate scores on the PKBS-2 and PIPPS and data from further analyses, th e reliability estimates through internal consistency on both measures were calculate d. The reliability coefficient describes the degree to which the PKBS-2 Social Skills Sc ale and the PIPPS represent something other than measurement error. In essence, if two se ts of parallel measures agree perfectly then
90 the obtained coefficient should be 1.00. However, for example, if raters are unmotivated and indicate answers which may not be repr esentative of child behaviors through random selection it is likely that the results from these raters will represent only measurement error. The reliability coefficients that are provided for each of the measures estimate the correlation between the obtained scores from parents and teachers and the score on a parallel form of the measure (Glass & Hopkins, 1996). For a multi-dimensional scale like the PIPPS, reliability coefficients of .7 are considered acceptabl e. If the systems demonstrate poor reliability, then the informa tion that is produced from the scales will not be meaningful. However, if the scales produce strong reliabilities, the information is suggested to be much more meaningful (Glass & Hopkins, 1996). Results indicate that all three factors of the PIPPS were found to demonstrate high levels of internal consistency. Internal consistency coefficients for the preschool teacher version of the scale were found to be si gnificant as evidenced by high Cronbach alpha coefficients for each subdomain, with co efficients ranging from .720, .878, and .797 for the Play Interaction, Play Disruption, and Play Disconnection factors, respectively. Internal consistency coefficients for the pres chool parent version of the scale were found to be .774 for Play Interaction, .769 for Play Disruption, and .627 for the Play Disconnection factor. Internal consistency coefficients for the PKBS-2 Social skills parent scales were also found to demonstrate high reliability sc ores as indicated by a coefficient of .870, .894, and .785 for the Social Cooperation, Social Interaction, and Social Independence factors, respectively. Internal consistency coefficients for the PKBS-2 Social skills teacher scales were found to be .934 for the So cial Cooperation subscale, .860, for Social
91 Interaction subscale, and .707 for the Social Independence subscale. Overall, these data indicate that the PKBS-2 and the PIPPS meas ure social competency and play behaviors with high levels of reliability and consiste ncy. See Table 4 below for a summary of the Cronbach coefficient alphas for the two rating systems. Table 4 Internal Consistency (alpha) Reliability fo r the PKBS-2 Social Skills Scale and PIPPS Scale Coefficient alpha PIPPS-Teacher Play Interaction .720 Play Disruption .878 Play Disconnection .797 PIPPS-Parent Play Interaction .744 Play Disruption .769 Play Disconnection .627 PKBS2 Social Skills Scale Parent Social Cooperation .870 Social Interaction .894 Social Independence .785 PKBS2 Social Skills Scale Teacher Social Cooperation .934 Social Interaction Social Independence .860 .707 Note N=32 Descriptive Statistics for the PKBS-2 and PIPPS Systems Descriptive statistics were collected and provided in order to gather more information concerning overall scores on th e PKBS-2 and the PIPPS. The mean, median, standard deviation, skewness and kurtosis valu es of scores on the PKBS-2 and PIPPS are reported. Procedures to screen for outliers a nd linear relationships were instituted through preliminary data checking methods. In a ddition, data were run through a statistical
92 program both with and without the inclusion of all possible outliers. This was to ensure that outlier values were not attributable to impossible circumstances. Another School Psychology graduate student was en listed to independently ente r the data to ensure that data entry was accurate. A summarization of descriptive statis tics is provided in Table 5. Obtained skewness and kurtosis values indicate some degree of non-normality distribution for the PIPPSParent Play Disconnection (-1.497 and 3.360) and Play Disruption (-1.454 and 4.804) scales respectively. There appears to a slight degree of negative skewness to these data. The data indicate that overall a greater number of parents had a tendency to rate their children as exhibiting more negative pl ay behaviors than woul d be expected in a normal distribution of raters. A graphical representati on of the scores obtained on these subscales would present high scores on the Play Disconnection and Disr uption domains as rated by parents as clustering together at the end with higher values a nd tailing off towards the left where lower values are represented. Parent ra tings on these same subscales also produced a positively kurtotic representation. Kurtosis is examined when one wishes to determine whether distributions have more extremely high or low scores than would be expected in a normal distribution of scores. Distributions of scores such as those indicated by parents on Play Disconnection and Disr uption subscales indicate a leptokurtic distribution which signifies that there are more scores that are farther from the mean than in a normal distribution. In summarization, hi ghly skewed distributions tend to be leptokurtic because there are more scores that deviate far from the mean. Overall, the results of these descriptive statistics indicate that parents were more apt to score th eir children as having
93 more negative play behaviors than those expected for a normal distribution and also from those scores that were noted by teachers on th ese same subscales. All of the other values were noted to be < 1.0 demonstrating approximately normal distributions for the remaining subscales. Table 5 Descriptive Statistics PKBS-2 Social Skills Scale and PIPPS Measure Minimum Maximum Mean Standard Deviation PIPPS-Parent Play Interaction 29 67 50.09 7.61 Play Disconnection 10 63 47.22 10.75 Play Disruption 10 64 46.19 9.95 PIPPS-Teacher Play Interaction 34 70 51.38 8.24 Play Disconnection 28 60 46.75 8.35 Play Disruption 26 62 46.28 8.77 PKBS-2 Social Skills Parent 83 120 107.78 10.63 Teacher 80 123 104.44 12.33 In order to examine the degree to whic h observations of play behavior in preschool children accurately assess social co mpetence as measured by traditional social skills rating scales, the scores of both parent and teacher versions obtained on the PKBS2 and PIPPS were examined. However, separate analyses were conducted for the teacher and parent measures to examine the relationship between the sets of variables of the PIPPS and the corresponding ve rsion of the PKBS-2. An assessment of concurrent validity was conducted to determine how well th e constructs of the PIPPS related to an outcome criterion selected as a standardized measure of social competence.
94 Concurrent Validity of the PIPPS-Parent w ith the PKBS-2 Social Skills Scale Parent In order to assess the conc urrent validity of the PIPPS-Parent with the PKBS-2 Social Skills Scale-Parent, Pearson-product moment coefficients were computed for the three PIPPS-Parent factors and the PKBS-2 So cial Skills ScaleParent factor. This particular analysis examined the overall si gnificance of the relationship between the PIPPS and PKBS-2 and was explored in order to determine whether the overlap between the two instruments was greater than that ex pected by chance. A correlational design with provision of confidence intervals was c onducted. A 95% confidence interval was calculated to determine a range in which the population corr elation value was likely to fall. The results indicated a magnitude of correlations falling in the low to moderate range. The only correlation that was cons idered statistically significant at the p <.05 level was the PIPPSParent Play Interaction factor and the PKBS-2 Social Skills Scale-Parent ( r = .49; p < .0001). However, all of the correlati ons were in the expected direction. PKBS-2 Social Skills Scale Social competen ce scores are hypothesized to be negatively related to negative play beha viors such as those proposed on the Play Disconnection and Play Disruption domains of the PIPPS. Child ren who score high on these play domains are likely to be rated lower on overall social competence and the results of this analysis further supported this notion. Overall, these results indicate an overall low to moderate relationship between the PIPPS-Parent and PKBS-2 Social Skills Scale-Parent.
95 Table 6 Correlations of PIPPS-Parent Factors with PKBS-2 Social Skills ScaleParent PKBS-2 Social Skills ScaleParent Factor PIPPS-Parent Factors Social Competence Score 95% Confidence Interval Play Interaction .49* .17 to .72 Play Disconnection Play Disruption -.20 -.24 -.51 to .16 -.54 to .11 Note* = p< .001 Concurrent Validity of the PIPPS-Teacher w ith the PKBS-2 Social Skills Scale Teacher In order to assess the concurrent valid ity of the PIPPS-Teacher with the PKBS-2 Social Skills Scale-Teacher, Pearson-product moment coefficients were computed for the three PIPPS-Teacher factors and the PKBS-2 Social Skills ScaleTeacher factor. The results demonstrate a strong relationship be tween the two measures. As displayed in Table 7, all of the correlations were statistically significan t and were in the expected direction. The magnitude of the correlations was in the moderate to high range. The highest correlation among fact ors was between the PKBS-2 So cial Skills Scale and the Play Interaction scale of the PIPPS ( r = .74; p < .0001), followed by teacher reports on the Play Disruption, ( r = .45; p < .0001) and Play Disconnection ( r = .43; p < .0001) with the PKBS-2 Social Skills Scale Teacher. Overall, these results indicate signi ficant relationships between the PIPPSTeacher and PKBS-2 Teacher. Therefore, convergent validity was supported by the significant correlation between the PKBS 2Teacher Social Skills scale and the PIPPSTeacher factors.
96 Table 7 Correlations of PIPPS-Teacher Factors with PKBS-2 Social Skills ScaleTeacher PKBS-2 Social Skills ScaleTeacher Factor PIPPS-Teacher Factors Social Competence Score 95% Confidence Interval Play Interaction .74* .53 to .87 Play Disconnection Play Disruption -.43* -.45* -.67 to -.09 -.69 to -.12 Note = p< .001 Differences in Group Means between Gender and PKBS-2 Social Skills Scale and PIPPS The second research question was designed to study whether parent and teacher ratings of preschoolers social competence varied as a function of a childs gender. The group means and standard deviations for pare nt and teacher ratings for boys and girls were calculated and compared to one another using an independent t test. The means of the two groups of the sample were used to determine whether they varied enough to determine a statistically significant differe nce between ratings obtained for boys and girls. A 95% confidence interval was calculated around the mean differences to determine where population values were likely to fall. The difference in the group means between the childs gender and the teacher and parent ratings on the PKBS-2 and PIPPS scales were examined for statistically significant findings. Table 7 displays the descriptive statistics, independent t -test results, and 95% confidence intervals of the difference for the PKBS-2 and PIPPS on gender comparisons. Several comparisons resulted in the findings of insignificant mean differences. The mean score for parent and teacher ratings for fe males ranged from 42.77 (PIPPSParent Play Disruption subscale) to 109.62 (P KBS-2 Parent Social Skills scale). In comparison, mean scores for males indicated a range from 45.63 (PIPPSTeacher Play Disconnection
97 subscale) to 106.53 (PKBS-2 Parent Social Skills scale). Diffe rence in mean scores for females and males ranged from .23 (PIPPSPa rent Play Interaction subscale) to 5.76 (PIPPSParent Play Disrupti on subscale). These results are presented below in Table 8. Levenes Test for Equality of Variances was conducted to assess the equal variances assumption. F values ranged from 0.016 (PIPPSParent Play Disconnect) to 6.183 (PIPPSTeacher Play Disruption). Results from the t-test for equality of means indicated t values which ranged from -1.651 (PIPPSPare nt Play Disruption) to 1.092 (PKBS2 Teacher Social Skills Scale). Overall these data signified that the obtained means were not different enough to consider the parent a nd teacher ratings of males and females as distinct and separate groups as i ndicated by results of the independent t test. While the means were not distinct enough to consider the ratings of participating preschool students to be different within statistical standa rds, it should be noted that on average both teachers and parents rated boys as being more disruptive in their play as indicated on the PIPPS Play Disruption subscal e. In addition, girls on average were rated as having higher social competence scores overall by both teachers and parents as indicated by higher scores on the PKBS-2 system. However as previously indicated because the means cannot be considered differe nt, no definite conclusions may be stated based upon these data.
98 Table 8 Mean Scores for Parent and Teacher Measures as a Function of Gender Gender Female Male Measure Mean SD Mean SD Mean Differences PKBS-2 Social Skills-Parent Social Skills-Teacher PIPPS ParentPlay Interact Parent-Play Disrupt Parent-Play Disconnect TeacherPlay Interact TeacherPlay Disrupt TeacherPlay Disconnect 109.62 107.31 50.23 42.77 46.92 54.62 44.00 48.38 9.07 11.59 5.82 11.42 9.24 8.19 11.52 6.59 106.53 102.47 50.00 48.53 47.42 49.16 47.84 45.63 11.65 12.75 8.78 8.33 11.91 7.70 8.33 9.37 3.09 4.84 .23 5.76 .50 5.46 3.84 2.75 Relationships between the PIPPS-Teacher and the PIPPS--Parent The third research question was aimed to determine whether the PIPPS-Teacher and PIPPS-Parent were congruent and significantly related to one another. In order to examine the degree of relationship between the teacher and parent versions of the PIPPS system, bivariate correlational analyses were conducted (see Table 8). Significant correlations were expected between each corresponding factor for the parent and teacher versions that consider the congruence betw een the factors of the teacher and parent versions (i.e., Play Interaction-Parent correla tes with Play Interaction-Teacher). Obtained correlations were found to be moderate and ranged from .30 to .49. The only statistically significant correlation found was among the teacher and parent factors of Play Interaction ( r =.49; p < .0001). Overall, these results indicate a significant rela tionship between the parent and teacher reports on onl y the Play Interac tion factor of the PIPPS. Therefore,
99 the obtained correlations are too weak to c onclude that any statistically significant relationship was found to exist between the ove rall PIPPS parent and te acher versions of this system through this investigation. Table 9 Correlations of PIPPS-Teacher Factors with PIPPS-Parent Factors PIPPS-Parent Factor PIPPS-Teacher Factor Play Interact 95% Confidence Interval Play Disconnect 95% Confidence Interval Play Disruption 95% Confidence Interval Play Interact *.49 .17 to .72 -.31 -.59 to .04 -.16 -.48 to .20 Play Disconnect -.21 -.52 to .15 .04 -.31 to .38 -.11 -.44 to .25 Play Disruption -.15 -.48 to .21 .39* .04 to .65 .30 -.05 to .59 Note = p< .001 Predictive Validity of the Parent Teache r Communication Form with the PKBS and PIPPS A fourth analysis was conducted to determine whether the level of communication reported as occurring between th e parents and teachers was related to the individual measures of social competence. The obtained data indicated that overall levels of present parent teacher communication we re relatively high across all sample participants, with the majority of teachers communicating with parents on a daily basis. Therefore, the data were not able to be an alyzed in terms of lo w communication and high communication categorical groups. Since there was no clear indication of a natural split in the level of communication taking place be tween parents and teachers the data were examined as a continuous factor and a correlational design was conducted.
100 Bivariate correlations were computed for the three PIPPSTeacher, PIPPS-Parent, PKBS-2 Social Skills Scale Teacher, PKBS-2 Social Skills Scale Parent, and level of communication existing between parents and te achers as indicated on the PTAC Form. The levels of communication were rated on a continuum. The findings from these results are displayed in Table 9. Only the PI PPSPlay Disconnection Parent subscale ( r = 0.36; p< 0.001) obtained a statistically significant relationship with the level of communication between the parent and teacher. These results i ndicate that there is very little to virtually no predictive validity between the PTAC form and the overall ratings of social competence that were completed by parents and teachers. Table 10 Correlations of PIPPSTeacher, PIPPSPa rent, PKBS-2 Social Skills Scale-Parent, PKBS-2 Social Skills Scale-Parent with the Parent-Teacher Communication form Measure Level of Communication 95% Confidence Interval PIPPS-Teacher Factor Play Interaction .02 -.33 to .37 Play Disconnection .03 -.32 to .38 Play Disruption .29 -.06 to .58 PIPPS-Parent Factor Play Interaction .04 -.31 to .38 Play Disconnection .36* .01 to .63 Play Disruption .06 -.30 to .40 PKBS-2 Social Skills Scale-Parent Social Competence Score -.09 -.42 to .22 PKBS-2 Social Skills ScaleTeacher Social Competence Score -.12 -.45 to .24 Note = p< .001
101 Summary of findings In the present investigation, numerous analyses were conducted to assess the similarity of parent and teacher ratings on the PKBS-2 Social Skills and PIPPS rating systems. Although broad conclusions are not able to be made, specific relationships were depicted through these analyses. In examining the relationship between parent and teachers ratings on both measures, convergent validity was supported by a significant correlation between the PKBS 2-Teacher Social Skills scale and the PIPPS-Teacher factors. However, a much lower relationship was determined for the PKBS 2Parent Social Skills scale and the PIPPS-Parent factors where only one factor of the PIPPS system was found to be significantly correlated with the PKBS -2 Parent Social Skills scale. Analyses which examined the pred ictive validity through the degree of relationship between the social competence ra ting forms as a function of gender and the level of existing parent and teacher communi cations did not reveal overall significant results. Although these results we re not statistically significant, the differences that did appear can be considered of clinical impor tance as higher scores on ratings of social competence for girls and higher levels of disr uptive play for boys pr esented in the data are supported by the ex isting literature. The level of parent-teacher communicati on was not significantly related to any of the scores obtained on the individual scales or subscales of the PIPPS and PKBS system overall, however there was a statistically signi ficant relationship be tween parent ratings on the PIPPS Play Disconnection factor a nd increased level of parent teacher communication.
102 Lastly, analyses indicated that teacher re ports and parent reports on the PIPPS system were not highly correlated. These results indicated a sole significant relationship between the parent and teacher reports on only the Play Intera ction factor of the PIPPS. The correlation between the Play Disrupti on and Disconnection factors can only be considered moderate to weak at best.
103 Chapter Five Discussion The PIPPS system was developed for the purposes of meeting the needs of early childhood assessment in accordance with the best practice guidelines. Studies thus far have demonstrated that the preschool version of the PIPPS is a valid instrument that assesses peer play interactions, measures congruent constructs among the teacher and parent versions, and is sensitive to cultura l contexts. Consequently, the primary objective of this study was to replicate previous findings of validity as well as to expand the utility of this instrument by considering the e ffects of gender and communication levels on parents and teachers ratings of preschoolers play behaviors. To evaluate the validity of the PIPPS system for preschool children and its use with parents and teachers across multiple set tings, four hypotheses were tested relating to: (1) the concurrent validity of parent and teacher versions of the PIPPS and the Social Skills scale of the PKBS-2; (2) the relationshi p between teacher/parent ratings and child gender; (3) the relationship be tween the teacher and parent versions of the PIPPS; and (4) the predictive validity of teacher and parent ratings on the PIPPS and PKBS-2 with level of communication between the two parties. The fo llowing sections discuss the results of these findings in relation to the empirical res earch. The subsequent section
104 discusses the limitations of the study as well as suggestions for future research. Finally, implications of these findings for practice are presented. 1) Does the preschool version of the PIPPS de monstrate concurrent validity with the Social Skills scale of the PKBS-2? Concurrent Validity The first hypothesis asserted that the PIPPS would demonstrate concurrent validity with a standa rdized measure of social co mpetence, through divergent and convergent relationships. Biva riate analyses supported this hypothesis for both the teacher and parent versions. The Play Interact ion factor of the PIPPS correlated positively with the PKBS-2 Social Skills scale, whereas the Play Di sruption and Play Disconnection factors demonstrated negative correlations with the PKBS-2 Social Skills scale. The relationships that were derive d between the PIPPS and an instrument assessing global social skills support the us e of the PIPPS as a measure of social competence. Children who received high rati ngs for Play Interaction were viewed by teachers and parents as demonstrating soci ally skilled behavior overall. Although the magnitude of the correlations ranged from lo w associations to statistically significant results, all of the correlations were in the expected direction. Parents and Teachers indicated that children with higher ratings on either th e Play Disruption or Play Disconnection factor were more likely to ha ve lower overall Social Competence scores. Thus, the Play Interaction factor identified aspects of childrens social skills, whereas Play Disruption and Play Disconnection re flected areas of need regarding peer interactions.
105 Typically, broad social skills competence scales assess social skills and problem behaviors, with the problem behavior domain typically divided into categories composed of externalizing and internalizing problems (LaFreniere & Dumas, 1996; Merrell, 1996). Externalizing behaviors include aggressive and disruptive be haviors while internalizing behaviors typically refer to so cial withdrawal and anxious behaviors (Merrell, 1996). The PIPPS captures these categories of behaviors by considering positive social skills through the Play Interaction construct, and addressing measures of ex ternalizing and internalizing behaviors through the construc ts of Play Disruption and Play Disconnection. According to researchers, children who score high on evaluations of internaliz ing and externalizing behaviors typically demonstrate overall d eclines in peer acceptance and friendships resulting in increased self-report of loneliness (Sanderson & Siegal, 1995). Although the constructs of the PIPPS m easure the same domains of social competence as other early childhood assessm ent measures, the PIPPS also uniquely assesses social skills and problem behavior s within a particular aspect of social competence in examining peer interactions. The importance of the capacity to identify specific aspects of developmentally appropriate social competences is emphasized by the recognition that different types of peer interactions have associated consequences for childrens social and academic functioning. Prob lems with peers can lead to difficulties with socio-emotional functioning (Ladd, 1999) as well as academic functioning (Welsh et al., 2001) and can occur throughout early ch ildhood and into school age years. On the other hand, average and well-accepted peers are often characterized by the display of cooperative, friendly, and pros ocial behaviors (Denham & Holt, 1993). An understanding of the types of behaviors that ar e associated with peer rejection facilitates
106 the identification of children who may need additional supports and could benefit from specific interventions targeting prosocial skills Thus, the ability of the PIPPS to identify behaviors that are typically associated with popular, reje cted, and neglected children serves an important role in early childhood assessment. 2) Do teacher and parent ratings for preschoolers' social competence vary based upon a childs gender? Gender Differences in Parent and Teacher ratings The second hypothesis of this study stated that parent and teacher ratings would differ as a function of the rated childs gender, with boys being more likely to score lower on the overall social competence measures as compared to girls. This speculation was based on previous research indicating strong evidence to support that boys tend to score higher on classroom ratings of hyperactivity a nd aggression as compared to girls (Lutz, Fantuzzo, & McDermott, 2002). These specific factors are associated with the Play Disruption scale of the PIPPS and would be demonstrated in a higher score on this subscale as well as a lower score on the PKBS-2 Social Skills scale. The analyses that examined the overall mean scores that parents and teachers indicated on rating scores did not result in st atistically significant findings. Overall, this finding is contrary to results from previous research indicating that boys are much more likely to score significantly lower than girl s on ratings of overall so cial competence (Lutz et al., 2002). However, it is important to highl ight the fact that when differences did exist for interactions between girl s and boys play, girls were rated higher on ov erall positive play behaviors (Play Interac tion, overall Social Competence) whereas boys tended to be rated higher on negative play behaviors (Play Disruption, lo wer overall Social
107 Competence). The preschool cente rs that participated in th is study have 1:10 or better teacher to student ratios throughout their schools It is likely that the preschool classes participate in large group activities for the majority of their schedules which is not conducive for allowing smaller groups or gender based groups to form. Previous literature has indicated that boys who tend to play with same sex peers are likely to exhibit increased levels of overactive behavior which may translate to higher problem behaviors, (Fabes et al., 1997; Noone Lutz, Fantuz zo & McDermott, 2002) since oftentimes boys are likely to engage in hi gher levels of physical play and contact in groups of boys than when girls are participati ng in play. In contrast, girls play is oriented towards the maintenance of coopera tive, concordant, and close relations when involved in groups of same sex peers and therefore are more conducive to promoting positive adaptation and adjustment (Fabes et al., 1997). In this study, as previously mentioned, since the majority of the classr ooms are likely to engage in large group activities, the level of social competence may be equali zed for boys and girls since there are lesser opportunities to promot e levels of higher physical play in same sex groups of boys and therefore lead to a lack of differe ntiation in groups of boys and girls social competence and categorization of play behaviors. One of the main purposes of this res earch question was to examine the possible rating bias of preschoolers ge nder. This was to explore whet her this factor confounds the assessment of childrens social skills by parents and teachers. As supported by this particular investigation there was no evidence th at parent and teacher ratings were biased based on preschoolers gender, as these gr oups did not obtain scor es that could be considered statistically different. Gende r role socialization and development are
108 potentially shaped by a multitude of factors a nd settings including influences by parents, teachers, and peers at home, school, and play settings. In order to determine whether contradict ory results of this study to previous findings in the literature were associated with distinct characteristic s of this particular sample population above and beyond the low teac her to student ratio in the classrooms, mean scores derived from boys and girls we re compared to the normative sample means provided in the examiners manual of the PKBS -2. All of the PKBS -2 standard scores are based on a distribution w ith a mean of 100 and a standard deviation of 15. It is interpreted that a standard score of 100 is at the mean score level of the national normative sample. However, it should also be noted that according to the PKBS-2 examiners manual score distribu tions for the Social Skills Scale are somewhat skewed in the direction of more desirable behavior ra tings and fewer undesirable behavior ratings and this is simply a reflection of normative sa mples for child behavior ratings (Shapiro & Kratchowill, 2000). It is typical for a greater majority of the children to have adequate or good social adjustment than present with so cial deficits or problems. Therefore, according to the information provided in the manual on mean scores for the normative sample, mean scores for boys and girls in th is particular study re flect average scores indicating that they may be justifiably compared to results from previous studies examining boys and girls social skills scores on the PKBS-2. The teacher and parent ratings did not i ndicate any apparent influences of gender biases and expectations that would have been reflected in a statistically significantly lower overall social competence mean score for boys as compared to girls. This finding would lead one to believe that the raters in this study may be more encouraging of
109 children to explore a wider variety of activit ies without restricting females and males to engage in activities that are typically asso ciated with their respective genders. With parents, teachers, and peers encouraging expl oration of various aspects of physical and imaginative play, children may be motivated to seek out friends with varying interests as well as be involved in a wider variety of sc hool activities (Birch & Ladd, 1996; Lutz et al., 2002). 3) Are PIPPS-Teacher and the PIPPS-Parent versions congruent and significantly related to one another? Relationships between Teacher and Parent Versions Another hypothesis of this study indicated that the re sults obtained from the teacher and parent versions would demons trate congruence, and that statistically significant relationships woul d be found. Correlational anal yses provided a moderate degree of relationship between teacher and pare nt versions which confirmed their ability to provide for some agreement of findings between teacher and parent reports. Overall results from these analyses indicate that the two versions have some meaningful relationships. The capacity of the PIPPS system to meas ure the same constructs in teacher and parent versions represents an advantage over existing early chil dhood social competence measures, which lack the ability for parent s and teachers to comm unicate about the same constructs in different cont exts (Bracken, 2000; Merrell, 199 6). The development of the PIPPS was based on observations made from peer play and thereby enhances the likelihood that the ratings are based on behavi ors that teachers and parents are actually observing in the respective contexts. Additiona lly, parents and teachers assisted with the
110 construction of this instrument, increasing the likelihood of a comm on interpretation and use of a common language (Fantuzzo et al., 1995). One of the purposes of the PIPPS is to differentiate children who display positive play behaviors from those who are less successf ul in these interactions and identify the strengths of resilient children. The signifi cant but relatively low degree of overlap between parent and teacher versions suggests that there is some similarity between the reports of peer play interactions within the home and school settings. The different perspectives among parents a nd teachers may reflect the contexts in which parents and teachers observe childrens peer play interactions. Teachers may be more likely to focus on and be cognizant of pos itive play behaviors. This may be due to their teaching experience as well as their understanding of childrens social behaviors in the classroom context. In contrast, parents may be drawn to and more likely to recognize disruptive behaviors are compared to prosocial behaviors. These findings suggest that teachers may have a more objective perspective than parents in identifying positive and negative behaviors in play interactions e xhibited among preschool children (Milfort & Greenfield, 2002). Other influences on the problem-focused orientation of the PIPPS ratings is the extent and length of time in which parents and teachers are able to observe play. Teachers have numerous children to monitor in the classroom, which may allow them to be more attuned to easily observable pr osocial and problematic interac tions during peer play. In addition, the length of time in which the play occurs may influence the type of interactions observed. More complex forms of play tend to emerge over longer durations of play, which is typically observed in the home environment. Parents have the advantage
111 of observing their children over time in the home and community. In doing so, they provide information that is otherwise unava ilable to teachers. This may explain the greater differentiation that parents make in characterizing play interactions. Another hypothesis is th at teachers spend time praising positive prosocial behaviors in an effort to en courage and reinforce those type s of behaviors that serve to facilitate the school experience. In contrast, parents may be drawn to and more likely to recognize disruptive behaviors as compared to prosocial behaviors whic h are likely to be expected in the home environment rather than directly taught (Milfort & Greenfield, 2002). However, it is important to consider the way in which teacher perceptions of social and academic competence may be influen ced. There are some data to indicate links between specific skills such as childrens ab ility to initiate and engage in conversation and teachers subsequent perceptions of competence. If these skills are not encouraged and/or opportunities to engage in these situations are not valued or emphasized in the home environment, preschool children ma y likely be rated lower on overall social competence. Such findings have important im plications in that teachers can assist children who have had fewer opportunities to engage in conversation with adults and provide them with those opportu nities (Tudge et al., 2003). The differences in the types of be haviors reported by pa rents and teachers correspond may also reflect the amount of cont act between these individuals. Research has indicated the importance of informants having a common frame of reference when assessing young children (Tudge et al., 2003). Activities such as volunteering in the classroom, planning classroom activities with the teacher, and participating in school-
112 based events with the teac her all increase th e amount of communication. In addition, daily, biweekly, or multiple modes of communication between teacher and parents increase the opportunities to interact and to share perspectives about childrens behavior. Thus, increased parental involve ment may also facilitate pos itive behaviors to generalize to the school setting and likewise result in the decrease of negative behaviors to generalize across settings as well. 4) Do parent and teacher ratings on the PIPPS and PKBS-2 vary as a function of the frequency of communications that are reported as occurring between parties? Predictive Validity Lastly, this study hypothesi zed that the level of co mmunication presently ongoing between parents and teachers would be associ ated with the degree of agreement between parent and teacher measures of social co mpetency ratings. The results of bivariate analyses did not yield overall significant relationships between the Parent Teacher Communication Form and the PIPPS. Howeve r, parent ratings on the PIPPS Play Disconnection factor did yield a small, but significant associat ion with increased levels of parent teacher communication. In other wo rds, higher levels of parent-teacher communication led to higher associated ratings only on the Play Disconnection subscale. This may reflect the fact that there was ve ry little variation found in the level of communication between teachers and parents. The majority of parent teacher groups indicated that communication was made daily through verbal exchange at the end of the day or through letters sent hom e. Therefore, analyses were not able to be made on the basis of distinct groups defined by low co mmunication and high co mmunication groups.
113 However, of interest is the sole si gnificant correlation that was found. Play Disconnection reflects withdraw n behavior and nonparticipati on in peer play. Unlike children who are overtly rejected by their peers, those who app ear to be neglected are not likely to be aggressive and therefore may be harder to detect than those who display more obvious behaviors (Sanderson & Siegal, 1995). Thus, there may be a subgroup of neglected children obtaining high scores on the Play Disconnection subscale who choose to play alone and refrain from participating in social interactions Although this group of children is less likely to report feelings or pe rceive themselves as lonely, research has indicated that withdrawn behaviors as rated by teachers contributed to elementary school childrens reports of lone liness in later years. Social isolation may have its antecedents in early childhood. Therefore, it is important to address these is sues early on in a childs education (Sanderson & Siegel, 1995). Children who are identified on the Play Disconnection scale as neglected and who ar e least likely to participate in social interactions are expected to benefit fro m organized play groups and social skills instruction by increasing their desire to enga ge in social interactions and thus being taught to form friendships (Sanderson & Siegel, 1995). Limitations and Implications for Future Research Although the overall results of this study support the use of the PIPPS by parents and teachers of preschool children, the resu lts are qualified by both the sample and the methods of this assessment. A discussion of some limitations inherent in this study is in order so that the findings are inte rpreted with appropriate caution. In recognition of the importance of early childh ood assessment, this study selected a diverse sample population with respec t to the region of the country as well as
114 across individual preschool classrooms and th eir philosophies of teaching. This sample was, however, predominantly White with a grea ter distribution of males as compared to females. Additionally, the disproportionate representation of one ethnic group as well as gender within this sample limits an understand ing of the validity of these constructs for other populations of preschool children. Further research is warranted to assess the construct validity of th is measure with a more diverse sa mple of children to determine the ability of the PIPPS to assess interactive peer play among all populations of children residing in various areas. For the purposes of deriving assessment measures to guide the development of appropriate curricula and inte rvention strategies, it is important to be cognizant of the appli cability as well as appropriateness for all children within the particular early childhood pr ogram. The current evaluation of the PIPPS indicates its validity for a predominantly White sample and therefore its use for other ethnic groups merits further study. Another consideration is in regards to the sample population of preschool teachers who completed rating scales and observati ons for preschool children within the classroom setting. All preschool teacher pa rticipants in this study were female. Additional research should explore the use of the PIPPS with the inclusion of a more diverse gender population of teachers and paraprofessionals in the classroom setting. The research has indicated that there are clea r differences in teac hers perceptions of childrens readiness as applicable to academ ic and social standards. Although this concept may be somewhat ambiguous, teachers perceptions and expectations are likely to influence their attitudes and interactions with their students. Children that they perceive to be socially more advanced ma y be treated differently and therefore given
115 more opportunities to outperform their peers who are perceived as le ss competent, both in academic as well as social domains. Alt hough men are highly underrepresented in early childhood education, results from recent studie s have shown differences between female and male preschool teachers play will ingness. Sandberg and Pramling-Samuelsson (2005), found that female preschool teachers tend to value calm play and emphasize the importance of social development while male preschool teac hers accentuate the significance of physical developm ent. Although it is believed that less than 6% of all preschool teachers nationwide are males, emerging data indicate the importance of assessing male and female interactions with young children in encour aging different play interactions. The methods of assessment used in th is study also limit the results. For the concurrent validity assessment, the results ar e qualified by the use of the same rater to complete the PIPPS and the PKBS-2, which can ultimately lead to source variance. The subjectivity of the rate r and the idiosyncratic ways in which the rater completes the scales can contribute to error (Me rrell, 1999). Both the PIPPS as well as PKBS-2 versions identify homeand school-based behaviors, th erefore this study relied on these selected measures of assessment of social competen ce within these contex ts. Thus, parent and teacher rating scales provided the sole mean s of examining concurrent validity. Including additional methods of evaluating behavior with in these contexts woul d contribute to and enhance the ecological framework in understand ing the concurrent validity of the PIPPS. By employing a multi-method as well as multi-source assessment plan, source variance is reduced and a more comprehensive understand ing of a childs social functioning may be obtained (Merrell, 1999). Future studies s hould therefore employ multiple methods of
116 examining the concurrent validity. Such studies may utilize additional sources of information such as observations, peer reports, and narrative recordings. The method that was used for assessing the predictive validit y of the PIPPS must also be considered when interpreting the results of this study. The current study used the Parent Teacher Communication Form developed by the research er. This particular form requires judgments to be based on teachers s ubjective perceptions a nd referral to recent interactions with parents in determining the level of parent-teacher communication. Therefore, the Parent Teacher Communicati on Form is not an objective measure of present communication. In particular, these fo rms were only distributed to teachers and findings may have differed had they been administered to both teachers as well as parents. There are many factor s that may likely influence teacher ratings on this form. Teachers may average the time that is typica lly spent with parents in their classroom, thereby reporting the average time spent w ith parents in general without regard to individual differences across parents. In pa rticular, if the amount of communication is similar across families with the exception of those families that stand out merely because they either have much lower levels of communication or much higher levels of communication, distinctions are likely to be overlooked. Future studies should examine the predictive validity of the PIPPS to more objective measures of communication by utilizing daily reports or documentati on indicating the forms and amount of communication with individual parents. In addi tion, teachers may also have referred to the most recent week or the current week in which rating scales we re completed without considering more typical levels of communication.
117 There is research to s upport the changing amounts in parent-teacher interaction across the transition from preschool into ki ndergarten. One study f ound that the average amount of weekly contact betw een parents and teachers in Head Start was 33 minutes per child as compared to 9.2 minutes per child in kindergarten (Fantuzzo et al., 1998). It is imperative that teacher-parent communication le vels are highly encouraged to increase the likelihood that these interactions will be sustained over the transition. Teachers and parents can only benefit from opportunities to learn from one anot her about childrens functioning and develop a shared perspectiv e for understanding an individual childs strengths and deficits. An understanding of childrens strengths and needs upon entry to kindergarten can assist both teachers and pa rents in setting up a ppropriate educational experiences for each child. To maximize the effectiveness of this communication, tools are needed for providing systematic means of communication. Should a similar investigation be co nducted in the future, researchers are encouraged to obtain a larger sample size, and sample from a population whose findings would generalize to a larger group of presc hool as well as kindergarten children. This study was conducted within a single large county. Future research shou ld explore the use of the PIPPS in urban settings to determine its appropriateness with a more generalizable sample. There is wide recognition of the ri sks faced by ethnic minority children living in low-income urban areas. Therefore, it is imperative that future studi es examine groups of high-risk young children so that early intervention and pr evention efforts may increase the likelihood of their success and decrease the probability for future difficulties.
118 Implications for Practice Developmental research repor ts that it is essential fo r children to develop social competence in order to achieve success, particularly in the transition from preschool to kindergarten. For young children, a primary component of social competence is composed of their ability to establish and e ngage in effective interactions with peers during play (Raver & Zigler 1997). Acquisition of these sk ills has been shown to facilitate positive perceptions of school and motivation to attend, increased school involvement, and enhanced academic achieve ment (Birch & Ladd, 1996). This research highlights emerging evidence that emotional development and academic learning are far more closely intertwined in the early years than may have been previously understood (Raver & Knitze, 2002). As evidenced in past research as well as the findings from this particular study, the capability of the PIPPS to a ddress important aspects of so cial competence assessment has been demonstrated. This instrument has proven its ability to target specific interactions and experiences of young children who may be at risk for social skill deficits. Because there is an emphasis on the need for social skills assessment upon entry into early childhood programs to identify areas of need, guidelines have proposed that information be derived from both parents a nd teachers. Observations should elicit behaviors that occur in natura l contexts (Fantuzzo et al., 199 8). The ability of the PIPPS to satisfy and address the need for developm entally appropriate assessment of interactive peer play across home and school settings with young children attests to its usefulness as a tool for screening and assessment (Fantuzzo & McWayne, 2002). The PIPPS taps into dimensions that address success in activel y engaging with peers, social withdraw
119 symptoms, aggressive/acting out behaviors, and dimensions of problem behaviors (Raver & Zigler, 1997). Informing the curriculum The increased demand for early childhood e ducation services is partly due to the recognition of the crucial importance of e xperiences during the ear liest years of life (Fantuzzo & McWayne, 2002). Results obtained from the PIPPS can contribute to the development of educational experiences that enhance social skills development by indicating areas of individual childrens strengths as well as difficulties in interactive peer play situations. Teachers can use this instrument to determine specific aspects of peer play interactions that must be addressed in order to meet the needs of an individual child and to modify a childs curriculum based on empirically validated constructs. Establishing methods of par ent and teacher communication According to the guidelines suggested by the National Association for the Education of Young Children (NAEYC), estab lishing partnerships be tween teachers and parents and obtaining information from multip le sources are important components of early childhood education (Raver & Zigl er, 1997). The PIPPS provides a means of communication between the home and school in regards to a childs strengths and needs in the areas of problem behaviors and social competency skills. Because the parent and teacher measures assess the same constructs the same terminology may be used when discussing a childs interactive peer play and social performance. The ability to communicate using the same terms that have shared meaning strengthens the partnership and allows parents and teachers to work together in providi ng additional social opportunities for their children across contexts.
120 Additionally, the PIPPS can f acilitate the importance of teaching peer play and educate parents about social competence. As parents are able to gain a better understanding of the value of peer play, they are more likely to promote opportunities for peer play interactions in the home context. Parents can further learn to identify the behaviors that support Play In teraction and behaviors that should be discouraged and relate to Play Disconnection and Play Disruption factors. Conclusions The primary purpose of this study was to provide additional evidence for the importance of considering social competen ce in young children in relation to their development and long-term outcomes. Overall, the results of this study revealed several capacities for the use of the PIPPS as a valid assessment instrument for preschool children. First, the PIPPS measures peer play interactions, which have proven its relevance to early childhood social compet ency. The importance of attaining this competency is evidenced by a growing body of research indi cating the consequences of poor peer relationships, future academic f unctioning, and overall well being. Second, this instrument has shown significant congruence in the measure of constructs among the parent and teacher versions. Establishing e ffective partnerships between parents and teachers early on in academic careers is crit ical for successful school adjustment. By allowing parents and teachers to converse over a relatively simplistic scale, the PIPPS encourages essential linkages in early childhood assessment.
121 References Achenbach, T.M., McConaughy, S.H., & Howell, C.T. (1987). Child/adolescent behavioral and emotional problems: Imp lications of cross-informant correlations for situational specificity. Psychological Bulletin, 101 213-232. Atkins, M.S., & Pelham, W.E. (1991). School-based assessment of attention deficithyperactivity disorder. Journal of Learning Disabilities, 24, 197-203. Alessandri, S.M. (1992). Atte ntion, play, and social behavior in ADHD preschoolers. Journal of Abnormal Child Psychology, 20, 289-302. Berkanovic, E. (1980). The effect of inadequate language translation on Hispanics responses to health surveys. American Journal of Public Health, 70 561-626. Bracken, B.A. (2000). The psychoeducational assessm ent of preschool children third edition Allyn and Bacon, Boston, p. 369-428. Bredekamp, S., & Copple, C. (1997). Developmentally appropriate practice in early childhood programs. Washington, DC: National Associ ation for the Education of Young Children. Brodzinsky, D.M., Elias, M.J., Steiger, C., Simon, J., Gill, M., & Hitt, J.C. (1992). Coping scale for children and youth: scale development and validation. Journal of Applied Developmental Psychology, 13 195-214. Bronfenbrenner, U. (1979). The ecology of human development Cambridge, MA: Harvard University Press.
122 Burks, H.F. (1977). Burks Preschool and Kindergar ten Behavior Rating Scale. Los Angeles: Western Psychological Services. Carney, A.G., & Merrell, K.W. (2002). Re liability and comparability of a spanishlanguage form of the preschool and kindergarten behavior scales. Psychology in the Schools Castro, M., Mendez, J., & Fantuzzo, J. (2002). The generalizability of the Penn Interactive Peer Play Scale: An empirical study of the Spanish version. Journal of School Psychology, 17 109-127. Childrens Defense Fund (1997). The State of Americas Children: Yearbook 1997 Washington, DC: Childrens Defense Fund. Childrens Defense Fund (2001). The state of Americas children, 2001 Washington, DC: Childrens Defense Fund. Cohen, J. (1992). Quantitative methods in psychology a power primer. Psychological Bulletin, 112 155-159. Conners, C.K. (1990). Conners Rating Scales Manual. North Tonowanda, NY: MultiHealth Systems. Connolly, J., & Doyle, A. (1981). Assessmen t of social competence in preschoolers: Teachers versus peers. Developmental Psychology, 17 454-462. Coolahan, K., Fantuzzo, J., Mendez, J., & McDermott, P. (2000). Preschool peer interactions and readiness to learn: rela tionships between classroom peer play and learning behaviors and conduct. Journal of Educational Psychology, 92 458-465.
123 Cooper, H.M. (1979). Pygmalion grows up: A model of teacher expectation communication and performance influence. Review of Educational Psychology, 49, 577-579. Coy, K., Speltz, M.L., DeKlyen, M., & Jones, K. (2001). Social-cognitive processes in preschool boys with and without oppositional defiant disorder. Journal of Abnormal Child Psychology, 29 107-124. Cramer, P., & Skidd, J.E. ( 1992). Correlates of self-worth in: The role of genderstereotyped styles of behavior. Sex Roles, 26 369-389. Denham, S.A., & Holt, R.W. (1993). Preschoolers likability as cause or consequence of their social behavior. Developmental Psychology, 29 271-275. Denham, S.A., Blair, K.A., DeMulder, E., Le vitas, J., Sawyer, K., Major-Auerbach, S., & Queenan, P. (2003). Preschool emotional competence: pathway to social competence? Child Development, 74 238-265. DeWolfe, N.A., Byrne, J.M., & Bawden, H.N. (2000). ADHD in preschool children: parent-rated psychosocial correlates. Developmental Medicine & Child Neurology, 42 825-830. DuPaul, G.J., McGoey, K.E., Eckert, T.L., & VanBrackle, J. (2001). Preschool children with attention-deficit/hyperactivity diso rder: Impairments in behavioral, social, and school functioning. Journal of the American Acade my of Child & Adolescent Psychiatry, 41 508-515. Elliott, S.N., Barnard, J., & Gresham, F.M. (1989). Preschoolers social behavior: Teachers and parents assessments. Journal of Psychoeducational Assessment, 7 223-234.
124 Fabes, R.A., Eisenberg, N., Jones, S., Smith, M., Guthrie, I., Poulin, R., Shepard, S., & Friedman, J. (1999). Regulation, emotionality, and preschoolers socially competent peer interactions. Child Development, 70 432-442. Fabes, R.A., Shepard, S.A., Guthrie, I.K., & Martin, C.L. (1997). Roles of temperamental arousal and gender-segre gated play in young childrens social adjustment. Developmental Psychology, 33 693-702. Fantuzzo, J., & McWayne, C. (2002). The rela tionship between peer-play interactions in the family context and dimensions of school readiness for low-income preschool children. Journal of Educational Psychology, 94 79-87. Fantuzzo, J., Coolahan, K., Mendez, J., McDermott, P., & Sutton-Smith, B. (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. Fantuzzo, J., Mendez, J., & Tighe, E. ( 1998). Parental assessment of peer play: Development and validation of the parent version of the Penn Interactive Peer Play Scale. Early Childhood Research Quarterly, 13 659-676. Fantuzzo, J., Sutton-Smith, B., Coolahan, K. C., Manz, P.H., Canning, S., & Debnam, D. (1995). Assessment of preschool play in teraction behaviors in young low-income children: Penn Interact ive Peer Play Scale. Early Childhood Research Quarterly, 10, 105-120. Farver, J.M., & Branstetter, W.H. (1994). Preschoolers prosocial responses to their peers distress. Developmental Psychology, 30 334-341.
125 Fisher, E.F. (1992). The impact of pl ay on development: A meta-analysis. Play & Culture, 5 159-181. Gagnon, S.G., & Nagle, R. J. (2004). Relations hips between peer interactive play and social competence in at-risk preschool children. Psychology in the Schools, 41 173-189. Glass, G.V., & Hopkinds, K.D. (1996). Statistical Methods in Education and Psychology Boston, MA: Allyn and Bacon. Gresham, F.M., & Elliott, S.N. (1990). The Social Skills Rating System. Circle Pines, MN: American Guidance. Hampton, V.R., & Fantuzzo, J.W. (2003). The va lidity of the Penn Interactive Peer Play Scale with urban, low-in come kindergarten children. School Psychology Review, 32, 77-91. Hartup, W.W., Moore, S.G. (1990). Early peer relations: Developmental significance and prognostic implications. Early Childhood Research Quarterly, 5 1-17. Hatch, J.A. (1987). Peer interaction and the development of social competence. Child Study Journal, 17 169-183. Howes, C., & Matheson, C.C. (1992). Sequences in the development of competent play with peers: Social and social pretend play. Developmental Psychology, 28 961974. Hughes, C., Dunn, J., & White, A. (1998). Tric k or treat? Uneven understanding of mind and emotion and executive dysfunction in hard-to-manage preschoolers. Journal of Child Psychology and Psychiatry, 7 981-994.
126 Hughes, H.M. (1984). Measures of self-esteem for preschool and kindergarten age children: Teacher report instruments. Child Study Journal, 14, 157-621. Ladd, G.W. (1999). Peer relationships and social competence during early and middle childhood. Annual Review of Psychology, 50 333-359. Ladd, G.W., & Hart, C.H. (1992). Creating info rmal play opportunities: Are parents and preschoolers initiations related to childrens competence with peers? Developmental Psychology, 6 1179-1187. Ladd, G.W., Price, J.M., Hart, C.H. (1988). Predicting preschoolers peer status from their playground behaviors. Child Development, 59 986-992. LaFreniere, P.J., & Dumas, J.E. (1996). So cial competence and be havior evaluation in children ages 3 to 6 years: The short form (SCBE30). Psychological Assessment, 8 369-377. Leinbach, M.D., & OBoyle, C. (1992). Gender labeling, gender stereotyping, and parenting behaviors. Developmental Psychology, 28 225-230. Lutz, M.N., Fantuzzo, J., & McDermott, P. (2002). Multidimen sional assessment of emotional and behavioral adjustment pr oblems of low-income preschool children: development and initial validation. Early Childhood Research Quarterly, 17 338355. Mash, E.J., & Barkley, R.A. (2003). Child Psychopathology Second Edition The Guilford Press, New York, pg. 26. Matson, J.L., Rotari, A.F., & Helsel, W.J. (1983). Development of a rating scale to measure social skills in young children: The Matson Evaluation of Social Skills with Youngsters (MESSY). Behaviour Research and Therapy, 21 335-340.
127 Matson, J.L., Rotari, A.F., & Helsel, W.J. (1985). Development of a rating scale to measure social skills in children: Th e Matson Evaluation of Social Skills with Youngsters (MESSY). Psychopharmacology Bulletin, 21 855-867. McDermott, P.A., Green, L.F., Francis, J.M., & Stout, D.H. (1996). Preschool Learning Behaviors Scale. Philadelphia: Edumetric and Clinical Science. McWayne, C., Sekino, Y., Hampton, G., & Fantu zzo, J. (2002). Penn Interactive Peer Play Scale (PIPPS) Manual; Teacher & Parent Rating Scales for Preschool & Kindergarten Children. Philadelphia, PA. Merrell, K.W. (1993). Using behavi or rating scales to assess so cial skills and antisocial behavior in school settings: Developmen t of the School Social Behavior Scales. School Psychology Review, 22 115-133. Merrell, K.W. (1994). The alliance of adap tive behavior and social competence: An examination of relationships between the Scales of Independent Behavior and the Social Skills Rating System. Research in Developmental Disabilities, 15 39-47. Merrell, K.W. (1994, 2002). Preschool and Kindergarten Behavior Scales (PKBS) Second Edition Examiners Manual. Austin, TX: Pro-Ed. Merrell, K.W. (1995). An inves tigation of the relationship between social skills and internalizing problems in early childhood: Construct validity of the Preschool and Kindergarten Behavior Scales. Journal of Psychoeducational Assessment, 13 230-240. Merrell, K.W. (1996). Social-emotional asse ssment in early childhood: The Preschool and Kindergarten Behavior Scales. Journal of Early Intervention, 20 132-145.
128 Merrell, K.W. (1996). Social-emotional proble ms in early childhood: New directions in conceptualization, assessment, and treatment. Education and Treatment of Children, 19 458-473. Merrell, K.W. (1998). The relationship of teach er-rated social ski lls deficits and ADHD characteristics among kindergarten-age children. Psychology in the Schools, 35 101-109. Merrell, K.W. (1999). Behavioral, Social, and Emotional Assessment of Children & Adolescents New Jersey: Lawrence Erlbaum Associates Publishers. Merrell, K.W., & Wolfe, T.M. (1998). The rela tionship of the teacher-rated social skills deficits and ADHD characteristic s among kindergarten-age children. Psychology in the Schools, 35 101-110. Milfort, R., & Greenfield, D.B. (2002). Teacher and observer ratings of head start childrens social skills. Early Childhood Research Quarterly, 17 581-595. National Association of Early Childhood Specialists in State Departments of Education (2003). Early childhood curriculum, assessment, and program evaluation: Building an effective, accountable syst em in programs for children birth through age 8. Joint position statement. Washington, DC: NAEYC. Online: www.naeyc.org/resources/position_statements/CAPEexpand.pdf. Noone Lutz, M., Fantuzzo, J., & McDermott, P. (2002). Multidimensional assessment of emotional and behavioral adjustment pr oblems of low-income preschool children: development and initial validation. Early Childhood Research Quarterly, 17 338355.
129 Odom, S.R., McConnell, S.R., & McEnvoy, M.A (1992), Social competence of young children with disabilities Baltimore, M.D: Paul H. Brookes. Onwuegbuzie, A. J. & Teddlie, C. (2003). A framework for analyzing data in mixed methods research. In A. Tashakkori & C. Teddlie (Eds.), Handbook of mixed methods in social & behavioral resear ch. Thousand Oaks, CA: Sage Publications. Onwuegbuzie, A.J. (2003). Expanding the framew ork of internal and external validity in quantitative research. Research in the Schools, 10 71-89. Pellegrini, A.D. & Blatchford, P. (2000). The Child at School Interactions with Peers and Teachers. New York: Oxford University Press Inc. Phillippot, P., & Feldman, R.S. (1990). Age and social competence in preschoolers decoding of facial expression. British Journal of Social Psychology, 29 43-54. Pianta, R.C., & Walsh, D.J. (1996). High-risk children in schools: Constructing sustaining relationships. New York: Routledge. Raver, C.C., & Knitze, J. (2002). Promoting the emotional we ll-being of children and families policy paper no. 3 National Center for Ch ildren in Poverty, 3-24. Raver, C.C., & Zigler, E.F. (1997). Soci al competence: An untapped dimension in evaluating Head Starts success. Early Childhood Research Quarterly, 12 363385. Roberts, R., & Strayer, J. (1996). Empat hy, emotional expressiveness, and prosocial behavior. Child Development, 67 449-470. Sanderson, J.A., & Siegal, M. (1995). Lonelines s and stable friendship in rejected and nonrejected preschoolers. Journal of Applied Developmental Psychology, 16 555567.
130 Saracho, O.N. (1991). Teacher expectations of students performance: A review of the literature. Early Child Development and Care, 76, 27-41. Schneider, M.J., & Leitenberg, H. (1989). A comparison of aggressive and withdrawn childrens self-esteem, optimism and pessi mis, and causal attributions for success and failure. Journal of Abnormal Child Psychology, 17 133-143. Shapiro, E.S., & T.R. Kratchowill, (2000). Behavioral Assessment in Schools Second Edition. Theory, Research, and Clinical Foundations New York: Guilford Press. Sparrow, S.S., Balla, D.A., & Cicchetti, D. V. (1998). Vineland social-emotional early childhood scales (SEEC). Circle Pines, MN: American Guidance Service. Sandberg, A., & Pramling-Samuelsson, I. (2005). An interview study of gender difference in preschool teachers attitudes toward childrens play. Early Childhood Education Journal, 32, 297-305. Stormont, M. (2001). Social outcomes of ch ildren with AD/HD: contributing factors and implications for practice. Psychology in the Schools, 38 521-531. Tudge, J.R.H., Odero, D., Hogan, D.M., & Et z, K.E. (2003). Relations between the everyday activities of preschoolers an d their teachers perceptions of their competence in the first years of school. Early Childhood Research Quarterly, 18 42-64. Vygotsky, L.S. (1978). The role of play in development. Cambridge, MA: Harvard University Press. Welsh, M., Parke, R.D., Widaman, K., & ONe il, R. (2001). Linkages between childrens social and academic competence: A longitudinal analysis. Journal of School Psychology, 39, 463-481.
131 Winsler, A., & Wallace, G. (2002). Behavior problems and social skills in preschool children: Parent-teacher agreement a nd relations with classroom observations. Contemporary Educational Psychology, 13, 41-58.
133 Appendix A: Parent and Teacher Consent forms Institutional Review Board Early Childhood Cover Letter Dear Parent/Caregiver: The following information is being presented to help you decide whether or not you would like to take part in a minimal risk research study. Pl ease read this carefully. If you do not understand anything, please feel free to contact the person in charge of the study. The goal in conducting this study is to learn about the experiences of young children and the way they engage in play behavior with their peers and with individuals in the home environment. English and Spanish speaking parents and caregivers of children ages 3-5 years, currently enrolled at ________ for a minimum of 4 months are being included in this study. Why Weve Sent This Letter to You : You are being asked to participate in this study because you are a parent or caregiver of a child (ages 3-5 years) who is enrolled in a preschool classroom at ________. The researcher has obtained consent from the Director, and the classroom teachers have agreed to take part in this study as well. The study is entitled: Measuring the Social Competence in Preschool-Aged School Children through the Examination of Play Behavior. The researcher would like to find out more about how young children play with their friends and their siblings at home and at school and how this is interpreted by parents and teachers alike. Why You Should Participate: By taking part in this research study, you may increase your overall knowledge of the importance of your child s play behaviors and how well they show these behaviors in school as compared to home. You may also find that factors of play behavior and social skills are easy concepts to speak to your childs classroom teacher about. In addition, you may find it helpful to speak to them in regards to your childs social progress in school by referring to specific skills mentioned on the rating scales. Completing the Survey : You will be asked to fill out two ra ting forms for this study during times that are most convenient for you. It is estimated that it will take 10-15 minutes for completion. You will need to hand in your forms to the cl assroom teacher when you are finished filling them out. We would like for you to hand them into the classroom teacher within two weeks of receiving them. Please Note : Your participation is completely voluntary. By returning the survey to the classroom teacher at ________, you are agreeing that you consent to participate in this research. If you choose not to participate, or if you withdraw this will in no way affect your relationship with _______, USF, or any other party. Confidentiality of Your Responses : There is minimal risk to you for participating in this research. Your privacy and research records will be kept confidential to the extent of the law. Authorized research personnel, employees of the Department of Health and Human Services, and the USF Institutional Review Board, and others wo rking on their behalf may inspect the records from this research project. Your individual responses will not be shared with any other school system personnel or anyone other than Eun-Yeop Lee after they are collected from your teachers. All names will be removed and replaced with a number to protect the confidentiality of your responses.
134 Appendix A: (Continued) What Well Do With Your Responses : The purpose of this research study is to find out if play behaviors are able to determine the level of social skills that young children have acquired. It is expected that this study will add useful info rmation to the research available on social and emotional factors with young children. This study is also expected to add to the literature which has shown that examining play behavior may be an additional method of accurately measuring social skills with young children. The results of this study may be published. However, the data obtained from you will be combined with data from other people in the publication. The published results will not include your name or any other information that would in any way personally identify you. Questions? If you have any questions about this research study, please contact Eun-Yeop Lee, MA. at (XXX) XXX-XXXX. If you have questions about your rights as a person who is taking part in a research study, you may contact a member of the Division of Research Compliance of the University of South Florida at 813-974-5638. Want to See the Results ? The researcher will be more than happy to share results of this study. In addition, the completed masters thesis will be kept on reserve at the USF Library. Thank you for taking the time to participate! Sincerely, Eun-Yeop Lee, MA USF Graduate Student Consent to Take Part in This Research Study By signing this form I agree that: I have fully read or have had read and e xplained to me this informed consent form describing this research project. I have had the opportunity to question one of the persons in charge of this research and have received satisfactory answers. I understand that I am being asked to particip ate in research. I understand the risks and benefits, and I freely give my consent to partic ipate in the research project outlined in this form, under the conditions indicated in it. I have been given a signed copy of this info rmed consent form, which is mine to keep. _______________________________ Signature of Participant ________________________________ Printed Name of Participant ______________________ Date
135 Appendix A: (Continued) Letter of Support To Whom It May Concern: This letter is to provide s upport for and allow my school to participate in the study entitled Measuring social competence in pr eschool-aged children through the examination of play behaviors , which will be conducted by Eun-Yeop Lee, MA (USF School Psychology Graduate student) for her mast ers thesis. I can verify that I have received materials describing the study as well as the procedures that will be conducted. In addition, I have the understanding that pare nt and teacher rating forms will be used in this study without further intrusion on, or di rect interaction with the students in the classroom. Thank you,
136 Appendix B: Instrumentation Parent-Teacher Communication Form Please indicate your answers below by placing a check mark next to the response that applies: 1. Gender of the child 2. What is your relation to this child? A. Male A. Pre-K classroom teacher B. Female B. Paraprofessional in Pre-K classroom For the following question please read the instructions below: 1. Please indicate forms of parent-teacher communication (i.e. letters home, face-to-face meetings, phone calls home, etc.) that have ta ken place in the past school year by placing a check next to the following options that apply. 2. Please rank order the top two forms of communica tion that occur the most frequently by placing a 1 and a 2 next to the responses that apply. 3. Please indicate the frequency in which any of these communications have occurred in the past school year by placing a check next to the following options that apply. 3. Type of Communication: ( Rank order top two) Frequency of Communication: ___ Letters home ___ 2 or more times a week ___ Phone calls home ___ Once a week ___ Parent-teacher conference ___ Once a month ___ Speak to parent/guardian at start/end of school day ___ Once a semester ___ Other: (please indicate) ____ ______________ ___ Once a year ___ Never Thank you for your time. Your assistance is greatly appreciated! Please feel free to call if you have any further questions (XXX) XXX-XXXX
Appendix B: (Continued) 137
Appendix B: (Continued) 138
Appendix B: (Continued) 139
Appendix B: (Continued) 140
Appendix B: (Continued) 141
Appendix B: (Continued) 142
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001796731
007 cr mnu|||uuuuu
008 070718s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001597
Measuring social competence in preschool-aged children through the examination of play behaviors
h [electronic resource] /
by Eun-Yeop Lee.
[Tampa, Fla] :
b University of South Florida,
ABSTRACT: For young children, a primary component of social competence is establishing effective interactions with peers during play. To inform the development of practices that promote this competency starting in early childhood, quality assessment measures are needed. These instruments must have the capacity to establish linkages between the home and school as well as utilizing multiple informants. A promising early childhood assessment measure is the Penn Interactive Peer Play Scale (PIPPS), which is a rating scale created with parent and teacher versions. Previous research has established its validity for preschoolers from among various populations. The purpose of this study was to examine the validity of the PIPPS system in a population of preschool children, by investigating: (1) the concurrent validity of parent and teacher versions of the PIPPS and a standardized assessment measure of social competence (PKBS-2 Social Skills Scale); (2) the relationship between teacher/parent ^ratings and child gender; (3) the relationships between the teacher and parent versions of the PIPPS; and (4) the predictive validity of teacher and parent ratings on the PIPPS and PKBS-2 with level of communication between the two parties.To meet inclusion criteria, teachers and parents had to have contact with preschool students ages 3-5 years enrolled in a preschool classroom for at least 4 months, and who were proficient in either English and/or Spanish. In total, across the three participating preschool centers, 50 students were found eligible to participate in this study and 32 students returned with completed packets parent rating scales (64%).Results indicated some relationship between the parent and teacher versions of the PIPPS and PKBS-2 Social Skills rating systems as well as the influence of communication level. However, there were no statistically significant findings for the influence of gender on these ratings. There were several limitations to the external validity^ of the results of this study. Limitations included sample bias and the use of self-report questionnaires. Implications and future directions for research are discussed.
Thesis (Ed.S.)--University of South Florida, 2006.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 142 pages.
Adviser: Kathy Bradley-Klug, Ph.D.
x School Psychology
t USF Electronic Theses and Dissertations.