1.00 0.50 0.00 -0.50 -1.00 1.80 1.70 1.60 1.50 1.40 1.30 female male 7.50 5.00 2.50 0.00 -2.50 -5.00 1.70 1.60 1.50 1.40 high ses mean ses low ses 7.50 5.00 2.50 0.00 -2.50 -5.00 1.60 1.50 1.40 1.30 1.20 The BELLA Preschool Study – Exploring the Relationship Between Social Skills and Behavioral Problems With a Preschool Sample Markus Hess 1 , Dieter Kleiber 1 , Rüya Kocalevent 1 ,Herbert Scheithauer 1 ,Nora Wille 2 , Michael Erhardt 2 , & Ulrike Ravens-Sieberer 2,3 1 Freie Universität Berlin 2 Universitätsklinikum Hamburg-Eppendorf 3 Robert-Koch Institute Acting in a competent way in social situations with others means a challenging task for children. Yet it is important to acquire social skills because a large part of identity develops through interaction with others (e.g. Ladd et al. 1999; Mead, 1934; Sullivan, 1953). As a result of a huge amount of studies it can be postulated that a high level of social skills is linked with a reduced risk for behavioral problems (e.g. Deater-Deckard, 2001; Parker et al., 1995; Webster- Stratton & Hammond, 1998). However research in the field has concentrated on middle and late childhood and adolescence, on externalizing behavior and on boys only (Crick, 1997). There is a lack of studies on both sexes, on externalizing and internalizing behavior and on studies with preschool children. With the present cross-sectional study we want to support findings postulating a close relationship between social skills and behavioral problems by using a preschool sample (boys and girls), by simultaneously looking at externalizing (aggression) and internalizing (anxiety) behavior rated by parents, and by using parental ratings of different aspects of social skills. Moreover we want to test wether the relation between social skills and behavioral problems are moderated by sex and socioeconomic status (SES) (Deater-Deckard, 2001). 1)How strong are externalizing (aggression) and internalizing (anxiety) behavioral problems associated with different aspects of social skills looking at preschool children? 2)Is the relationship between behavioral problems and social skills moderated by sex and by SES of the families? Markus Hess, Freie Universität Berlin, Unit Developmental Science and Applied Developmental Psychology, Habelschwerdter Allee 45, 14195 Berlin Email: mhess@zedat.fu-berlin.de This research project was supported by grants from the Robert-Koch Institute Subjects The sample consisted of 336 German preschoolers stemming from the BELLA preschool study (162 boys, 174 girls). Mean age was 4.5 years with a range from 3 to 6 years. Mean level of SES was 11.7 (sd = 4.31) on a scale ranging from 3 to 21. Measures Social skills. The sample was administered a German translation of the social skills scale of the preschool parent form of the Social Skills Rating System (SSRS, Gresham & Elliott, 1990). Taking into account recent studies discussing the factorial structure of the SSRS (e.g. Van Horn, et al. 2007) we conducted psychometric analyses of the SSRS (Hess, et al. 2008). As a result we applied a revised version of the social skills scale of the SSRS with a reduced number of items and a separate factorial structure for younger (3 / 4 years; n=161) and for older (5/ 6 years, n=175) children. Due to this all analyses in this paper were computed separately for younger and for older children. Externalizing and internalizing behavior. The subjects answered parts of the Child Behavior Checklist (CBCL), (Arbeitsgruppe Deutsche Child Behavior Checklist, 2000) and of the Diagnostic System for Mental Disorders in Childhood and Adolescence (DISYPS), (Döpfner & Lehmkuhl, 2000) SES: The Winkler-Index (Winkler, 1998) combines level of education, job prestige and income of parents in an additive way. Table 1. internal consistencies of the relevant variables 4 Results - According to our data the degree of assertive behavior exhibited by girls seems not to be associated with aggression whereas for young boys (age 3 and 4) high assertion corresponds with low levels of aggression. This can be interpreted as a strong gender-specifc parental perception of the relation between assertive and aggressive behavior prevalent in the preschool years. - SES does only moderate the relation between cooperation and aggression (in both agegroups). Interestingly we found different interaction patterns in the two agegroups (further analyses will test this possible two-way interaction). This result can be interpreted by assuming that depending on the socioeconomic status of a family cooperation (well- behaving, helpfulness and autonomy in a normative sense) is perceived as differently related to aggressive behavior depending on the age of the children. Note that we have to bear in mind the cross sectional nature of our data. - Tests for interaction between sex/ SES and social skills predicting anxiety did not reveal any significant results. - Further analyses of the data will hopefully reveal deeper insights into the familiar processes which may explain our results. For example we will take a closer look at family structure and parenting practices as well as at other clinicially relevant behavioral problems like hyperactivity. 1 Theoretical Background 2 Research Questions 3 Method 5 Discussion Selected references. Cohen, J., Cohen P., West, S.G., & Aiken, L.S. (2003). Applied multiple regression/ correlation analysis for the behavioral sciences. Mahwah: Erlbaum. Deater-Deckard, K. (2001). Annotation: Recent research examining the role of peer relationships in the development of psychopathology. Journal of Child Psychology and Psychiatry, 42, 565-579. Eagly, A.H. & Wood, W. (1991). Explaining sex differences in social behavior: A meta-analytic perspective. Personality and Social Psychology Bulletin, 17, 306-315. Fagot, B.I., Hagan, R., Leibach, M.D. & Krausberg, S. (1985). Differential reactions to assertive and communicative acts of toddler boys and girls. Child Development, 56, 1499-1505. Gresham, F. M., & Elliott, S. N. (1990). Social Skills Rating System. Circle Pines: American Guidance Service. Parker, J. G., Rubin, K. H., Price, J. M., & DeRosier, M. E. (1995). Peer relationships, child development, and adjustment: A developmental psychopathology perspective. In D. Cicchetti & D. J. Cohen (Eds.), Developmental psychopathology. Vol. 2: Risk, disorder, and adaptation (pp. 96-161). New York: Wiley., Sullivan, H.S. (1953). The interpersonal theory of psychiatry. New York: Norton. Van Horn, M.L., Atkins-Burnett,S., Karlin, E., Ramey, S.L.,& Snyder, S. (2007). Parent ratings of children’s social skills: Longitudinal psychometric analyses of the Social Skills Rating System. School Psychology Quarterly, 22, 162-199. Bivariate correlations (see table below) The social skills scales are only weakly correlated in both age groups whereas the behavioral problem measures are moderately associated. The correlations between social skills and behavioral problems show that (especially in the old group) high cooperation and high self-control correspond with lower levels of aggression and social behavioral problems but are not related to anxiety. In contrast assertion is not related to externalizing behavior but does show associations with internalizing behavior in both age groups (high assertion means low anxiety). Sex and SES are only weakly associated with social skills and behavioral problems. variable number of items Cronbachs alpha young old young old SSRScooperation 6 6 .69 .64 SSRSassertion 8 7 .72 .67 SSRSself-control 9 12 .72 .83 SSRSsocial rules 6 - .68 - DISYPSsocial behavior 4 4 .61 .64 CBCL aggression 19 19 .85 .86 DISYPSanxiety 6 6 .60 .53 CBCL anxiety 8 8 .65 .68 Table 2. Nonparametric correlations (due to the non-normal distribution of some variables), means and standard deviations for the independent and dependent variables (above the diagonal young children, below the diagonal old children) Test for interactions (see figures below) Tests for interactions (Cohen, et al. 2003) were computed to look for possible moderating functions of sex and SES where no or only weak bivariate correlations between social skills and behavioral problems were found. For example it can be hypothzised that assertion and aggressive behavior are differently related for boys and girls at least in the perception of parents and other adults (e.g. Eagly & Wood, 1991; Fagot, et al. 1985). Figure 1. relation between assertion and aggression (CBCL) moderated by sex in the young agegroup (R 2 = .09, F=5.01, p<.01) level of aggression assertion Results show that for boys only a high level of assertion is associated with a low levels of agression but not for girls. Aggression does not differ significantly between boys and girls in our sample (t(159)=1.92, p=.06). Figure 2. relation between cooperation and aggression (CBCL) moderated by SES in the younger age group (left) and in the older agegroup (right) (R 2 young = .14, F=8.22, p<.001; R 2 old = .11, F=7.14, p<.001) young old cooperation cooperation level of aggression Results show that in the younger group only in families with a high SES high cooperation is associated with lower levels of agression. A low degree of cooperation is associated with high aggression in the younger group irrespectable of SES. This pattern is reversed in the older group. In the older group low cooperation is related to higher aggression only in families with low SES. Note that all variables used in the interaction analyses above did have an acceptable level of skewness. Variable 1 2 3 4 5 6 7 8 9 10 mean sd Kendall‘s-Tau-b young old young old 1-Sex .10 .07 -.01 .11 .05 -.02 -.11 -.03 -.02 .49 .54 .50 .50 2-SES -.01 -.08 .09 .09 .14* -.07 -.14* -.11 -.07 11.68 11.66 4.22 4.40 3-SSRS- coop. .15* -.05 .31** .20** .23** -.04 -.16** -.03 -.03 .91 .97 .36 .34 4-SSRS- self-control .17** .14** .20** .21** .34** -.27** -.35** -.13* -.05 1.38 1.36 .27 .31 5-SSRS- assertion .03 -.05 .11* .15** .15** .03 -.06 -.22** -.13* 1.24 1.28 .36 .35 6-SSRS-social rules -.24** -.26** -.11 -.09 1.55 - .32 7-DISYPS-soc. Beh. -.11 -.08 -.22** -.42** -.11 .47** .21** .24** 1.47 1.51 .39 .41 8-CBCL-aggression -.20** -.10 -.16** -.39** -.09 .52** .22** .24** 1.49 1.46 .28 .28 9-DISYPS-anxiety -.01 .05 -.04 -.07 -.24** .12* .12* .44** 1.44 1.38 .41 .36 10-CBCL-anxiety -.02 .07 -.01 -.01 -.16** .15* .22** .34** 1.47 1.51 .28 .31 Note. coop. = cooperation; soc. beh.= social behavioral problems, sd = standard deviation, * p < .05, ** p < .01