Psiholo.ka obzorja / Horizons of Psychology, 9, 4, 7-23 (2000 Dru.tvo psihologov slovenije 2000, ISSN 1318-18y Znanstveni empirieno-raziskovalni prispeve The Slovenian version of the Social Competence and4Behavior Evaluation Scale . Preschool Edition (OLSP):4the second preliminary validation4 MAJA ZUPANEIE1*, ALENKA GRIL2 AND TINA KAVEIE R 1University of Ljubljana, Department of Psychology, Ljubljana, Sloveni 2Educational Research Institute, Ljubljana, Sloveni Abstract: The contribution presents the first part of the second validation study on the SloveniaU version of the Social Competence and Behavior Evaluation Scale . Preschool Edition (OLSP). ThI study was implemented with a sample of 95 Slovenian kindergarten children aged from 2;6 to 6;5 years uniformly sampled by age and gender. OLSP was filled out by the childrens. kindergarten teachers a. well as their assisstants. The obtained results show that internal consistency and interrater reliability o . the eight basic and the four composite scales of OLSP are satisfactory. No significant gender differN ences were obtained for any of the scales. The Dependent-Autonomous scale was positively related t 1 the childrens. age, while the Internalizing Problems and General Adaptation composite scales related t 1 age in the expected direction with medium effect sizes. However, the age differences on the latter diB not reach statistical significance mainly due to small sample size o Key words: preschool children, social competence, reliability, age differences, gender difference. Slovenska oblika Ocenjevalne lestvice socialna prilagojenost4pred.olskih otrok (OLSP): druga pilotska validacija4 MAJA ZUPANEIE1, ALENKA GRIL2 IN TINA KAVEIE R 1Univerza v Ljubljani, Oddelek za psihologijo, Ljubljan 2Pedago.ki in.titut, Ljubljan Povzetek: V prispevku predstavljamo prvi del druge pilotske .tudije s slovensko obliko OcenjevalnI lestvice socialna prilagojenost pred.olskih otrok (OLSP). Raziskavo smo izvedle na vzorcu 95-iO slovenskih otrok, ki obiskujejo vrtce. Stari so bili med 2;6 in 6;5 let in glede na svojo starost ter spo enakomerno zastopani v vzorcu. OLSP so izpolnile loeeno otrokove vzgojiteljice in njihove pomoenice o Rezultati ka.ejo na zadovoljivo notranjo konsistentnost osmih temeljnih in .tirih sestavljenih lestvi OLSP ter na visoko zanesljivost ocen med dvema neodvisnima ocenjevalkama. Deeki in deklice se meB seboj niso razlikovali v dose.enih rezultatih na nobeni izmed lestvic. Rezultati na lestvici OdvisenN Samostojen so bili pozitivno povezani z otrokovo starostjo, medtem ko so bili rezultati na lestvicaO *Naslov / address: izr. prof. dr. Maja Zupaneie, Univerza v Ljubljani, Oddelek za psihologijo, A.kereeva 2, 100Q Ljubljana, Slovenija, e-mail: maja.zupancic@ff.uni-lj.so 84M Zupaneie, A. Gril in T. Kavei: Te.ave ponotranjanja in Splo.no prilagajanje sicer povezani z otrokovo starostjo v prieakovani smeri a niso dosegli ravni statistiene pomembnosti. Ker je bil ueinek starosti na slednji spremenljivki srednjI visok, sklepamo, da razlike med starostnimi skupinami ne dosegajo kritienih vrednosti zaradi majhneg a .tevila otrok v posameznih skupinah o Kljuene besede: pred.olski otroci, socialna prilagojenost, zanesljivost, razlike med starostnimo skupinami, razlike med spolom a CC=2840 222 1 The Slovenian version of the Social Competence and Behavior Evaluation . PreN school Edition (OLSP) is the preliminary Slovenian translation and adaptation of thI original SCBE (LaFreniere & Dumas, 1995). The SCBE is an instrument, alread standardized in the USA (LaFreniere & Dumas, 1995), Quebec and France (Dumas LaFreniere, Capuano & Durning, 1997) and in preparation for standardization iU Slovenia, designed to assess patterns of social competence, affective expression and adjustment difficulties in kindergarten children aged 30 to 78 months. Its priN mary objective is to encompass childrens. behavioral tendencies for the purposes o . socialization and education, rather than classify them into diagnostic categories. ThI information is obtained by kindergarten teachers, who assess the childrens. behavioQ with respect to everyday life situations in kindergarten. The instrument provides a standardized description of behaviors that is contextual, reliable, valid, objective easy to administer and interpret, and is useful to early childhood specialists. It foN cuses on the presence or absence of positive as well as problem behaviors in threI areas of childrens. everyday life in a kindergarten setting: their affective characterisN tics, interaction with peers and interaction with adults. Therefore, it allows the chilB specialist to pinpoint a child.s specific competencies as well as vulnerabilities o The SCBE was translated and preliminarily adapted for the purpose of its usI with Slovenian kindergarten children. Its first preliminary validation was implementeB with a sample of 95 Slovenian kindergarten children, aged 64 to 75 months (Zupaneie Gril & Kaveie, 2000a; 2000b). The results of this study show that each of the OLSP. . basic and composite scales has a high degree of internal consistency, substantia convergent and discriminant validity as estimated through the correlations with thI Questionnaire of Behavioral Reactions (Horvat, 1985). Only a few items had to bI modified from the first to the second Slovenian version of SCBE due to their relaN tively low correlations with correspondent scales (Zupaneie, et al., 2000b) o The present study was designed to validate the second Slovenian version o . SCBE (OLSP) on a sample of a broader age range of Slovenian kindergarten chilN dren with respect to its internal consistency, interrater reliability, gender and agI effects on childrens. behaviors in everyday life situations in kindergarten, as asN sessed by his/her teachers o Gender effects for the basic and composite scales of SCBE, obtained in thI Social Competence and Behavior Evaluation ScalI USA and Quebec on a broad age range of preschool children, yielded the followin . results: boys were assessed more negatively than the girls for Externalizing ProbN lems, and for each of the four scales associated with this factor (Angry-Tolerant Aggressive-Calm, Egotistical-Prosocial, Oppositional-Cooperative), Social CompeN tence and General Adaptation. The boys were scored as more anxious than the girl. in the Quebec sample (Dumas, et al., 1997). Fewer and smaller gender difference. for the first version of the OLSP scales were obtained with a Slovenian sample o . older preschool children. The boys were scored higher on anxiety and anger, whilI the girls were assessed as more socially competent (Zupaneie, et al., 2000a) o Other recent European studies on gender differences in related fields of chilB development also showed only a few and small gender effects on childrens. behaviors o Kindergarten teachers longitudinally assessed boys (age 5;8 to 7;5 years) as morI anxious, manifesting more behavioral problems and less social adjustment compareB to the girls (Puklek & Gril, 1999). Parental descriptions of their childrens. personaliN ties (age 3 to 12 years) provided no support to well-documented findings (e.g. Block 1983; Maccoby & Jacklin, 1974) of girls scoring higher on anxiety and boys higheQ on assertiveness. Girls were described as more dominant, independent and sociablI than the boys who were ascribed more activity, curiosity, less conscientiousness immaturity for age, having more school problems and expressing more gender apN propriate behaviors (DeFruyt, Van Hiel & Buyst, 1998). Using the same assessmen1 procedures, Kohnstamm, Mervielde, Besevegis and Halverson Jr. (1995) found ver few differences in parental descriptions of their 3-, 6- and 9-year-olds: girls werI only described as more sociable and boys as more active compared to the girls o Except for a few tendencies (boys were ascribed less extraversion and less diligencI than girls) the parental descriptions of their infants. and toddlers. personalities d 1 not differ with respect to the child.s gender (Zupaneie, 2000) o The gender effects on the OLSP scales were expected to be low with boy. scoring somewhat higher on scales of problem behavior and girls scoring higher oU social competence o Despite a rapid general development and growing social experiences witO peers as well as with adults in early childhood, an age effect on the SCBE scales wa. found to be low. Positive correlations of age with SCBE measures, obtained on U . samples, only indicate a trend towards increased social competence and feweQ behavioral problems with older children (LaFreniere & Dumas, 1995). The highes1 correlations were obtained for the Aggressive-Tolerant, Oppositonal-CooperativI and Dependent-Autonomous (which was the highest) scales. The age effect with thI Quebec sample was even smaller. Age explained up to 4% of the variance (LaFrenierI & Dumas, 1995). The authors attribute low and even negligible correlations to thI teachers who tend to evaluate each child with reference to his/her age group as preN viously established by Behar and Stringfield (1972). In addition, no gender differN ences in the correlations between age and SCBE measures were evident in the US Quebec and French studies (Dumas, et al., 1997; LaFreniere & Dumas, 1995). N 1 104M Zupaneie, A. Gril in T. Kavei: significant age effect on anxiety, behavioral problems and socialized behavior wa. found with three age groups of Slovenian kindergarten children, aged from 5;8 to 7; years. A year later the group who remained in kindergarten, instead of starting school had increased behavioral problems, while the children who started school had deN creased problems and increased socialized behavior (Gril & Puklek, 1999) o Low age effects were expected in the present study, and no age x gender interN action, with older children scoring somewhat higher on Social Competence, espeN cially Dependent-Autonomous, and lower on problem behaviors o Method4 Participants4 Twenty kindergarten teachers (all females) and thirteen assistant teachers (twelvI females and one male) participated in the evaluation of 95 children after a writteU consent by the childrens. parents was obtained. All of the teachers and their assistN ants were working in five different kindergartens of Ljubljana and assessed the chilN dren participating in their group for at least three months continuously. The childreU were recruited from twenty different kindergarten groups. Two of these groups werI age-heterogeneous, with the others composed of age-homogeneous groups of chilN dren o Table 1. The frequency distribution of children by their age and gende 1 AGE/ GENDER 2;6 3;5 years 3;6 4;5 years 4;6 5;5 years 5;6 6;5 years TOTAL Boys 13 15 11 11 50 Girls 14 14 11 6 45 TOTAL 27 29 22 17 95 There was not a significantly higher proportion of boys than girls (c2=0.26, df=1, p=0.608) the children were proportionally distributed by their age (c2=3.65, df=3, p=0.301) and b age x gender (c2=1.28, df=3, p=0.733) o Instrument4 The second Slovenian preliminary version of OLSP (Zupaneie, Gril & Kaveie, 2000c . the Slovenian Social Competence and Behavior Evaluation Scale . Preschool EdiN tion (SCBE). The questionnaire, designed to be evaluated by the child.s kindergarN ten teacher, presents a translated and adapted version of the original SCBE (LaFreniarI & Dumas, 1995) and was further developed upon the first Slovenian validatioU (Zupaneie, et al., 2000b) o Social Competence and Behavior Evaluation ScalI 114 The second version of OLSP offers eight basic scales and four summary scales o Each basic scale consists of 10 items . brief statements describing a child.s behavioQ on the rating form (1- almost never occurs - to 6 . almost always occurs). Five of thI items within each of the basic scales describe successful adjustment and another fivI describe adjustment difficulties. The first three of the eight basic scales describe a child.s characteristic manner of emotional expression, and are defined by negativI and positive poles: Depressive-Joyful, Anxious-Secure, and Angry-Tolerant. ThI next three scales describe social interactions with peers: Isolated-Integrated, AgN gressive-Calm, and Egotistical-Prosocial. Finally, two scales assess teacher-chilB relations: Oppositional-Cooperative and Dependent-Autonomous. The internal conN sistency of the first OLSP version of the basic scales, validated with older SloveniaU preschool children, ranges from 0.77 to 0.90, and their construct validity was estiN mated as satisfactory (Zupaneie, et al., 2000b) o The four summary (composite) scales were developed on the basis of extenN sive statistical analyses and have been proven in practice (Dumas, et al., 1997e LaFreniere & Dumas, 1995; Zupaneie, et al., 2000b). The first composite scale Social Competence, consists of 40 items which assess the positive qualities of thI child.s adaptation (good adjustment, flexibility, emotional maturity, prosocia behavior). The second scale, Internalizing Problems, comprises 20 items reflectin . anxious, depressed, isolated, and withdrawn behaviors (the higher the score the feweQ problems manifested). The third scale, Externalizing Problems, comprises 20 item. and reflects angry, aggressive, selfish, and oppositional behaviors (the scoring i. reversed) in a child. The last summary scale, General Adaptation, includes all 8 1 items and makes it possible to compare children on a single index o Procedure4 Five kindergartens in Ljubljana were selected to participate in the study on the basi. of previous intensive cooperation between them and the Division of Developmenta Psychology at the Faculty of Arts, University of Ljubljana. The children were seN lected randomly within each of the kindergartens and all of their teachers/assistan1 teachers agreed to participate after the consent of the kindergarten masters was obN tained. The parents of the selected children were given the statements of agreemen1 to be signed. 95 out of 100 written agreements were obtained and the teachers/assistN ant teachers assessed only the children with written parental consent o The teachers/assistant teachers of each of the five kindergartens were giveU group instructions about the administration of OLSP: (a) a very general descriptioU of the purpose of the study (no reliability, interrater agreement or group difference. were mentioned); (b) to assess individual children successively, not a group of chilN dren simultaneously per item; (c) to read carefully each of the items; (d) to asses. each child by the rank order of items as much as possible; (e) to assess all of thI items; (f) to assess what they think of a child without any consultations with otherse 124M Zupaneie, A. Gril in T. Kavei: (g) to keep questionnaires, either non-filled or filled out, confidential from the moN ment they receive them to the time they return them to the researchers. They had tw 1 weeks to assess their group of children o 50 of the whole sample of children were randomly selected to be assessed b two raters . the teacher and the assistant teacher independently. These teachers anB their assistants were given additional instructions to evaluate the children independN ently without any communication on the child.s behavior as described in quastionnairI items before they finish their participation in the study o The obtained data were analysed with respect to internal consistenc (Cronbach.s alpha), interrater reliability (Pearson.s r), age and gender effects (twoN way ANOVA and Scheffe.s post hoc test), and with respect to relations between agI and OLSP scales (Pearson.s r) to obtain a more direct comparison of the presen1 study with others using the SCBE o Results4 The interrater agreement between kindergarten teachers and their assistants proveB to be relatively high. The internal consistency of the four composite scales is estiN mated as high as well as the homogeneity of the seven basic scales. Alpha of thI Dependent-Autonomous scale was obtained as lower (0.66, with alpha standardN ized=0.72) than the alphas of the other scales. The distribution of this scale wa. found to be symmetrical, but leptokurtic (IK = 3.17) o The two-way ANOVA yielded no significant main effect of gender on thI basic and composite scales of OLSP. However, two tendencies appeared. The girl. tended to be assessed as more cooperative (F (1,94) = 2.52, p = 0.12, MSgender = 213.55 and socially competent (F (1, 94) = 1.91, p = 0.17, MSgender = 1831.05) than the boys o Eta squared, as recommended by Clark-Carter (1997) who converted Cohen.s (1988 measures of effect size into values for h2, was used to determine the effect size o . Table 2. Interrater Agreement and Internal Consistency of Eight Basic and Fou 1 Composite OLSP Scale. SCALES Depressive-Joyful Anxious-SecureInterrater Agreement 0.69 0.69 Cronbach's Alpha 0.80 0.84 Angry-TolerantIsolated-Integrated 0.73 0.74 0.79 0.89 Aggressive-CalmEgotistical-ProsocialOppositional-CooperativeDependent-Autonomous 0.81 0.82 0.88 0.73 0.85 0.82 0.86 0.66 Social Competence Internalizing Problems Externalizing Problems General Adaptation 0.89 0.69 0.76 0.87 0.93 0.84 0.93 0.95 Social Competence and Behavior Evaluation ScalI 134 gender. In our case, h2 are 0.03 and 0.02 respectively; thus gender explains 2.8% o . the overall variance on the Oppositional-Cooperative scale and 2.2% on the Socia Competence composite scale, which are small effect sizes. The power of the tes1 computed using alpha = 0.05 is 0.35 for Oppositional-Cooperative and 0.28 for SoN cial Competence. In other words, the probabilities of committing a Type II error arI 0.65, and 0.72 respectively. Therefore, there is a 65%, and 72% chance respectively that we reject the gender differences when they actualy exist. Referring to the poweQ tables (Clark-Carter, 1997), approximately 400 participants would be needed to givI the test the power of 0.80. Since the effect sizes here are small and the sample sizI required is vast it would not make sense to replicate the study with an enlargeB sample just to (perhaps) obtain statistical significance o The significant main effect of age was obtained on the Dependent-AutonoN mous scale. Older children were assessed as more autonomous than the youngeQ children (F (3, 91) = 6.40, p = 0.00, MSage = 541.08). Age explains 18.1% of the tota variance (h2= 0.18). This value according to Clark-Carter estimates a large effec1 size. Schaffe.s post hoc test showed that the 4;6 to 5;5 year-old group of childreU was assessed to manifest significantly more autonomous behaviors in relation t 1 their teachers compared to the two younger groups . 2;6 to 3;5 years olds (Md 9.61, p = 0.01) and 3;6 to 4;5 years olds (Md = 10.22, p = 0.00) as presented iU Figure 1 o No other significant effects of age were obtained. However, two tendencie. are worth mentioning because of their medium effect size. With increasing age, thI Mean +/- 1 SD 60 50 40 30 20 10 17222927N = 2;6 & 3;5 3;6 & 4;5 4;6 & 5;5 5;6 & 6;6 Age group Figure 1. Means scores and standard deviations on Dependent-Autonomous Scale bW childrens. agI 14M Zupaneie, A. Gril in T. Kavei: children tend to be assessed as decreasing internalizing problems (F (3, 91) = 2.31, p 0.08, MSage = 602.48) and increasing general adaptation (F (3, 91) = 2.03, p = 0.12 MSage = 5912.88). Age explains 7.4% of the overall variance (with the power o . 0.56) on the Internalizing Problems scale and 5.6% of the variance on the Genera Adaptation Index (with the power of 0.50): These effect sizes are estimated as meN dium according to Clark-Carter (1997). Thirty and fourty-five participants respecN tively in each of the age groups would need to participate in order to support thI eventual age differences with respect to these two variables, which still seems reaN sonable. Scheffe.s post hoc test revealed no significant differences between pairwisI age group comparisons. There was only a tendency of 4;6 to 5;5 year-old group to bI assessed as having fewer internalizing problems and better general adaptation thaU the other age groups o Consistent with the results of the two-way analysis of variance, only the reN sults of the Dependent-Autonomous scale are significantly positively related witO the childrens. age. The older children are estimated to be more autonomous than thI younger who show more dependence on the kindergarten teacher. There are als 1 slight tendencies of internalizing problems to decrease with age, and of cooperativeN ness with peers, social competence and general adaptation to increase. These correN lations, however, are low and do not reach statistical significance. Compared to thI US results, the correlations with a smaller Slovenian sample of kindergarten chilN dren of the same age are generally lower except for the Dependent-Autonomous anB Internalizing Problems scales o Discussion4 For the purpose of assessing the reliability of OLSP scales, interrater agreement anB internal consistency were the most important and therefore reported within the firs1 Table 3. Comparison of the Correlations (Pearson.s r) between Age and OLSP Score. with the Slovenian and Two US Samples (obtained by LaFreniere & Dumas, 1995 Scale Depressive-Joyful Anxious-SecureAngry-TolerantIsolated-Integrated Correlation (N=94) -0.02 0.09 0.05 0.09Probability level 0.838 0.375 0.650 0.369 US Correlations (2 samples - N=824; 439) 0.14; 0.20 0.12; 0.22 0.26; 0.23 0.09; 0.23 Aggressive-CalmEgotistical-Prosocial Oppositional-Cooperative Dependent-Autonomous 0.04 0.05 0.15 0.32 0.710 0.635 0.150 0.002 0.23; 0.23 0.18; 0.23 0.24; 0.23 0.20; 0.32 Social Competence Internalizing Problems Externalizing Problems General Adaptation 0.13 0.18 0.030.13 0.213 0.091 0.791 0.201 0.17; 0.31 0.20; 0.20 0.24; 0.15 0.23; 0.29 Social Competence and Behavior Evaluation ScalI 154 phase of the second preliminary study of the instrument. As shown in Table 2, thi. phase has demonstrated high internal consistency and interrater agreement of thI basic and the composite scales with a somewhat lower, albeit still acceptable homoN geneity of the basic Dependent-Autonomous scale. The reliability estimates in thI relatively small Slovenian sample of kindergarten children are comparable to the U . (LaFreniere & Dumas, 1995), the Quebec and the French results (Dumas, et al. 1997). A comparison of the first validation of the first Slovenian OLSP versioU (Zupaneie, et. al, 2000) with older preschool children and the present study with aU independent mixed age group of preschool children reveals that the internal consistN ency of a majority of the scales remained very similar to the internal consistenc obtained in the first study. The homogeneity coefficients of the three scales (thI basic scales Aggressive-Calm, Oppositional-Cooperative and consequently, the comN posite scale Externalizing Problems) were slightly improved from the first study o However, the homogeneity of three other scales somewhat decreased (DepressiveN Joyful, Dependent-Autonomous and consequently, the composite scale Internalizin . Problems). Regardless of the latter, the reliability of the scales is estimated as satisN factory. Whether we decide to improve the Dependent-Autonomous scale depend. on the results to be yet obtained at the end of the second phase of this preliminar study. This is designed to evaluate test-retest reliability and temporal stability of thI OLSP scales as well as their concurrent criterion validity with respect to peeQ sociometrics using a typological approach o Gender differences were assessed for the eight basic and the four compositI OLSP scales. The gender effect on social behaviors was small, as expected, but nonN significant. Boys and girls did not differ with respect to any of the evaluated variN ables. However, the girls tended to be assessed as more cooperative and sociall competent than the boys with gender explaining only between 2 to 3 percent o . variance. The direction of the obtained tendencies is consistent with the U . (LaFreniere & Dumas, 1995), Quebec and French (Dumas, et al., 1997) studies, bu1 the effect sizes are smaller with the Slovenian sample. In addition, the above-menN tioned studies, using the SCBE, also obtained other consistent gender differences iU all three societies (USA, Quebec and France) with larger samples of preschool chilN dren, but with small effect sizes. The boys were scored as more angry, aggressive egotistical, oppositional and having more externalizing problems than the girls, whilI the girls were assessed higher on Social Competence scale and the General AdaptaN tion Index than the boys o Compared to recent Slovenian studies (Gril & Puklek, 1999; Zupaneie, et al. 2000a) where the older preschool children were evaluated by their kindergarten teachN ers, the results of the present study show consistency with respect to the direction o . gender differences in oppositional behaviors and social competence, but the effec1 sizes are smaller. However, in comparison to these Slovenian studies, no gendeQ differences in angry, aggressive and anxious behaviors occurred in the present study o Since the children participating in the present study covered a relatively broad agI 164M Zupaneie, A. Gril in T. Kavei: range (two and a half to six and a half years) and were divided into four small agI groups, we presume that in spite of the absence of a significant age x gender effec1 some, albeit small, differences between the genders might first become evident durN ing the late preschool period, at least in the present generation of SloveniaU preschoolers. Recent Slovenian studies on younger preschool children suggest no oQ negligible gender differences with regard to many aspects of early child developN ment. These aspects include motor, fine motor, language development, socializatioU index, sophistication of play, behaviors during toy-play, personality characteristic. as described by the child.s parents (e. g. Zupaneie, 1999; 2000) and early famil socialization measured through indicators of mother-child interaction, contents o . interactive toy-play, parental ethno-theories of child development and the use o . play objects (e. g. Kosaber, 1998; Krape., 1999; Gril, Cecia Erpie & Kaveie, 2000e Gril & Puklek, 1999; Zupaneie, 1999) o Summing up, the literature on gender differences in cognitive, social and perN sonality development suggests that boys and girls are more similar than is comN monly believed (for overview see Berk, 1997; Ruble, 1988). There is a general conN sensus among the researchers that individuals (children as well as adults) posses. both masculine and feminine characteristics. During the 70s and the 80s of the pas1 century as summarized by Ruble (1988), boys and girls began to show differen1 interests and activity preferences early in development, and such gender differentiaN tion remained throughout the life span, probably due to the shaping of gender role. through the social learning process and childrens. internal motivations to exhibi1 gender-appropriate behavior because of emerging cognitive structures or gendeQ schemes with biological bases of gender roles could not be excluded. Recently, adul1 gender roles have begun to change, at least in technologically-developed countrie. which certainly has had an impact on the development of androgynous gender roles especially in Sweden (Berk, 1997), and it seems to be the case in Slovenia as wel (Zupaneie & Metzing, 1994). We do not claim that traditional gender typing is eradiN cated in Slovenia, but progress has been made towards gender equality in the las1 decade. The latest Slovenian studies suggest that this might be reflected in mothersp non-differential treatment of their infants and toddlers by their gender as well as iU the mothers. attitudes towards child development and child rearing practises o In our opinion, the most plausible of the alternative explanations for the obN tained negligible gender differences in the present study might be the evaluatorsp tendency to assess the childrens. social behavior with reference to their gender groups o This possibility could be verified through direct observation of children during theiQ activities in a kindergarten setting. But, as long as all of the kindergarten teacher. assess the children with respect to their gender group, this has no implication for thI standardization procedure of OLSP o Similar to the absence of significant gender effects, only few age effects werI revealed. Age had a significant effect of a large size on the Dependent-Autonomou. scale, while two non-significant age effects of a medium size were found on thI Social Competence and Behavior Evaluation ScalI 174 Internalizing Problems scale and the General Adaptation Index. The significant maiU effect of the age which has been obtained on the scale Dependent-Autonomous reN quests some more reservations due to the relatively low internal consistency of thI scale. Correlations of the scales with chronological age were computed to comparI the results with other studies using the same instrument. The correlation estimate. on a smaller, but in its stucture similar sample of Slovenian preschool children are iU general lower than those reported in the USA (LaFreniere & Dumas, 1995) anB France (Dumas, et al., 1997), and higher than those found in Quebec (LaFreniere . Dumas, 1995). The determination indexes obtained in each of the countries, includN ing Slovenia, show that no more than 10% of the variance on the scales could bI explained by the child.s age. The relations between age and the teachers. estimaN tions of the child.s social behaviors are generally the highest, or at least among thI highest, for the Dependent-Autonomous scale. The child.s dependence on the kinN dergarten teacher decreases with age. The Slovenian correlation between age anB the Internalizing Problems scale is also similar to the one obtained on the US samN ples (the childrens. anxious, depressed, isolated and withdrawn behaviors in kinderN garten decrease with age), while others are considerably lower, resembling the corN relation estimates in the Quebec study (LaFreniere & Dumas, 1995) o Considering the rapid social development during early childhood one woulB expect much higher relations between the child.s age and his/her social behaviors iU a kindergarten setting as well as in general. Despite individual differences and a moderate stability of social behaviors in the preschool period, it was well-estabN lished that children become more tolerant and prosocial towards others, integrateB into their peer group, cooperative with adults and autonomous (for review see Berk 1997; Papalia & Olds, 1992). Compared to results obtained by direct observations o . the childrens. behaviors and their answers on interviews, the relations between thI childrens. age and their social behaviors as assessed by their teachers was found t 1 be considerably lower. Behar and Stringfield (1972) supported the hypothesis tha1 the teachers assess the children with reference to their age group which contribute. to a reasonably low differentiation of children by their age. This explanation wa. also accepted by LaFreniere and Dumas (1995) using the SCBE as a measure o . social competence. We assume that the Slovenian kindergarten children were scoreB in the framework of their chronological age as well, even in age heterogeneous groups o A closer look at the relations between age and the OLSP scales in the age homogeneN ous and the two heterogeneous groups, as well as between the gender and the OLS . scales, showed a surprising similarity of results (Zupaneie, Gril & Kaveie, 2000d which suggests that the teachers perceive individual differences in childrens. socia behaviors largely with respect to chronological age and not with reference to thI group they work with. There is of course an alternative explanation. Namely, that thI younger children in the age heterogeneous groups are led to the zone of proxima development by the older children. Nevertheless, few of the obtained age differN ences suggest that separate norms of OLSP will have to be constructed for youngeQ 184M Zupaneie, A. Gril in T. Kavei: and older children around the age of four years as indicated by the results of Scheffe. . post hoc analysis. However, it is important to notice some restrictions of the presen1 study, since they limit any extensive generalizations of the established results. The were obtained with relatively small age groups of Slovenian preschool children living in the city of Ljubljana o Conclusions4 Internal consistency and interrater agreement of the second OLSP version are satisN factory. Convergent validity was already supported (Zupaneie, et al., 2000b) whilI the procedures of test-retest reliability, temporal stability and criterion validity arI still in process o Gender effects on the scales of OLSP are small and non-significant. Takin . into account the results of the first OLSP study (Zupaneie, et al., 2000a) which inN cluded a larger sample of older preschool children and power analyses of the firs1 and the present study, we presume that some gender differences might exist in thI Slovenian preschool children, but they are probably small, explaining less than 5 of the variance. Several contemporary European studies on gender differences iU preschool children (e.g. DeFruyt, et al., 1998; Gril & Puklek, 1999; Kohnstamm, e1 al., 1995; Zupaneie, 1999; 2000) suggest that contemporary European societies inN creasingly stimulate androgynous gender-role development according to the changN ing adult gender roles which is also reflected in the early socialization. We do no1 completely exclude an alternative explanation, namely that the kindergarten teachN ers tend to evaluate boys and girls with reference to their gender groups. ThereforI the statistical effect of gender on the assessed social behaviors is minimized in spitI of the fact that some gender differences might actually exist. Regardless of thesI possibilities, the results suggest that the Slovenian version of SCBE (OLSP) migh1 not need separate norms for boys and girls o Except for the Dependent-Autonomous scale no significant age effects werI obtained. We attribute this fact to the kindergarten teachers. tendency to assess eacO child with reference to his/her age group as previously established by Behar anB Stringfield (1972) and LaFreniere and Dumas (1995). The age differences in auN tonomous behavior are probably so pronounced that the significant age effect on thi. variable was obtained in all of the countries using the SCBE. Taking into account thI effect sizes of age on other children.s social behaviors in kindergarten, the stud suggests that there might be a significant age decrease with respect to the InternalizN ing Problems and an increase in the General Adaptation, thus requiring separate agI norms for younger (2;6 to 4;5 year) and older preschool children (4;6 year and up) o The 4;6 to 5;5 year old age group, as evaluated by the kindergarten teachers appeared in some aspects to behave distinctively from all the other age groups. Thi. suggests a further in-depth study which could reveal whether there are any majoQ Social Competence and Behavior Evaluation ScalI 194 developmental changes unfolding at this age or merely the teachers. bias is for somI reason responsible for the obtained age differences o References4 Behar, L. & Stringfield, S. (1974). A behavior rating scale for the preschool child. DevelN opmental Psychology, 10(5), 601-610 o Berk, L. E. (1997). Child development. Needham Heights, MA: Allyn and Bacon o Block, J. H. (1983). Differential premises arising from differential socialization of thI sexes: Some conjectures. Child Development, 54, 1335-1354 o Clark-Carter, D. (1997). Doing quantitative psychological research. Hove: Psycholog Press o Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ Lawrence Erlbaum Associates o De Fruyt, F., Van Hiel, A. & Buyst, V. (1998). Parental personality descriptions of boys anB girls. In G. A. Kohnstamm, C. F. Jr. Halverson, I. Mervielde & V. Havill (Eds.) Parental descriptions of child personality: Developmental antecedents of the Bi . Five? (pp. 155-167). Mahwah, NJ: Erlbaum o Dumas, J. E., LaFreniere, P. J., Capuano, F. & Durning, P. (1997). Profil socio-affectifp valuation des comptences sociales et des difficults d.adaptation des enfants de . ans 1 a 6 ans. Paris: ECPA o Gril, A., Cecia Erpie, S. & Kaveie, T. (2000). Early play interaction in relation to gende 1 stereotyped toys. Paper presented at the 12th Psychology Days, Zadar o Gril, A. & Puklek, M. (1999). Mothers. beliefs about infant.s learning, playing and upN bringing. Eta evolutiva, 63, 53-60 o Horvat, L. (1985). Vpra.alnik vedenjskih reakcij [Questionnaire of Behavioral Reactions] o Ljubljana: Katedra za razvojno psihologijo o Kohnstamm, G. A., Mervielde, I., Besevegis, E. & Halverson, C. F. Jr. (1995). Tracing thI Big Five in parents. free descriptions of their children. European Journal of PersonN ality, 9, 283-304 o Kosaber, L. (1998). Igraea in struktura malekove igre z mamo [Toys and the structure o . infants. play with their mothers]. Unpublished graduate thesis, Filozofska fakulteta Univerza v Ljubljani, Ljubljana o Krape., T. (1999). Vsebina malekove igre pri interaktivni igri z igraeami [The content o . toddler.s interactive play with toys]. Unpublished graduate thesis, Filozofsk a fakulteta, Univerza v Ljubljani, Ljubljana o LaFreniere, P. J. & Dumas, J. E. (1995). Social competence and behavior evaluation Preschool edition (SCBE). Los Angeles: Western Psychological Services o Maccoby, E. E. & Jacklin, C. N. (1974). The psychology of sex differences. Stanford, CA Stanford University Press o Papalia, D. E. & Olds, S. W. (1992). Human development. New York: McGraw-Hill Inc o Puklek, M. & Gril, A. (1999). Socioemocionalne in vedenjske znaeilnosti otrok ob vstop v .olo [Socio-emotional and behavioral characteristics of children starting school] o In M. Zupaneie (Ed.), Razvojnopsiholo.ke znaeilnosti razlieno starih otrok oI 204M Zupaneie, A. Gril in T. Kavei: vstopu v .olo (str. 43-66) [Psychological characteristics of children starting schoo at different ages]. Ljubljana: i2 o Ruble, N. D. (1988). Sex-role development. In M. H. Bornstein & M. E. Lamb (Eds.) Social, emotional and personality development (pp. 411-460). Hove: LawrencI Erlbaum Associates Ltd., Publishers o Zupaneie, M. (1999). The development of object-play in the first two years of a child.s lifep Ljubljana, Lugano: i2, Centro Studi Universitari Internazionali o Zupaneie, M. (2000). Tracing the FFM in parents. free descriptions of their infants toddlers. Paper presented at the 12th Psychology Days, Zadar o Zupaneie, M., Gril, A. & Kaveie, T. (2000a). Predstavitev rezultatov prve OLSP .tudije a Sloveniji [Presentation of results of the first OLSP study in Slovenia]. Ljubljana Interno gradivo katedre za Razvojno psihologijo o Zupaneie, M., Gril, A. & Kaveie, T. (2000b). Predstavitev prve poskusne oblikI Ocenjevalne lestvice socialna prilagojenost pred.olskih otrok (OLSP) ter njen a konstruktna veljavnost [Presentation of the first pilot version of the Social CompeN tence and Behavior Evaluation Scales . Preschool Edition (OLSP) and its construc1 validity]. Psiholo.ka obzorja, 9 (3), 45-66 o Zupaneie, M., Gril, A. & Kaveie, T. (2000c). Ocena socialne prilagojenosti pred.olski . otrok (OLSP) . 2. poskusna oblika [Evaluation of social competence and behavio 1 of preschool children (OLSP) . 2nd preliminary version]. Ljubljana: Center z a psihodiagnostiena sredstva o Zupaneie, M., Gril, A. & Kaveie, T. (2000d). The scale of Social Competence anB Behavior . validation with Slovenian preschool children. Paper presented at 12tO Psychology Days, Zadar o Zupaneie, M. & Metzing, A. (1994). The concept of human nature: A comparative study iU Slovenia and Croatia. Proceedings of Alps-Adria Symposium on Psychology (pp o 262-268). Ljubljana: Filozofska fakulteta o Prispelo/Received: 05.06.200Q Sprejeto/Accepted: 05.10.200Q Social Competence and Behavior Evaluation ScalI 214 Appendix4 Table 4. Means and Standard Deviations of OLSP Scores1 by Age Groups and Gende 1 Age group Scale Gender M SD Joyful- Depressive Boys Girls 37.54 38.14 6.39 8.48 Secure Boys 34.69 8.75 Anxious Girls 36.29 11.00 Tolerant- Boys 29.31 11.52 Angry Girls 32.50 9.04 Integrated Boys 34.54 9.96 Isolated Girls 35.50 13.39 Calm- Aggressive Boys Girls 31.92 37.43 11.14 8.83 2;6 3;5 Prosocial- Egotistical Cooperative- Oppositional Boys Girls Boys Girls 27.54 32.71 33.08 40.07 10.37 9.66 13.65 6.09 Autonomous- Dependent Boys Girls 29.15 30.57 8.79 9.09 Social Competence Absence of Internalizing Absence of Externalizing General Adaptation Boys Girls Boys Girls Boys Girls Boys Girls 110.15 124.43 75.08 76.21 72.54 82.57 257.77 283.21 33.41 32.22 13.81 20.23 24.91 14.18 61.10 54.58 Joyful Depressive Boys Girls 37.73 34.00 6.24 9.18 Secure Anxious Boys Girls 35.53 33.29 7.70 8.22 Tolerant Angry Boys Girls 29.53 26.71 10.27 7.83 3;6 4;5 Integrated Isolated Calm Boys Girls Boys 37.53 35.21 34.00 6.92 9.48 9.11 Aggressive Girls 34.71 9.12 Prosocial Boys 32.47 7.30 Egotistical Girls 28.86 6.10 Cooperative Boys 36.13 10.51 Oppositional Girls 37.79 5.35 Autonomous Dependent Boys Girls 31.33 27.07 7.62 7.79 3;6 4;5 Social Competence Absence of Internalizing Absence of Externalizing General Adaptation Boys Girls Boys Girls Boys Girls Boys Girls 122.60 114.86 78.20 70.07 73.47 72.71 274.00 257.00 28.82 23.39 11.87 16.74 20.21 15.68 52.09 37.82 1The poles of the scales are here reversed for easier interpretatioU M Zupaneie, A. Gril in T. Kavei: Joyful Depressive Boys Girls 37.91 40.55 8.57 8.78 Secure - Anxious Boys Girls 38.91 40.09 7.79 10.99 Tolerant Angry Boys Girls 31.55 35.18 9.85 9.41 Integrated - Isolated Boys Girls 38.09 40.91 8.73 8.55 4;6 5;5 Calm Aggressive Prosocial Egotistical Cooperative Oppositional Autonomous Dependent Social Competence Absence of Internalizing Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls 36.09 38.45 29.18 35.82 40.09 40.27 40.91 38.09 130.09 139.91 84.91 86.00 8.88 11.18 10.38 12.13 7.22 9.47 13.29 9.60 27.46 37.91 19.78 14.71 Absence of Externalizing General Adaptation Boys Girls Boys Girls 77.73 83.45 292.73 309.36 20.93 18.97 48.31 62.07 5;6 6;5 Joyful Depressive Secure - Anxious Tolerant Angry Integrated - Isolated Calm Aggressive Prosocial Egotistical Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls 36.82 37.17 36.27 36.50 29.82 31.83 34.27 38.17 33.18 36.00 30.36 31.67 8.42 7.28 7.77 9.20 7.87 5.78 9.82 6.74 16.49 8.00 10.64 8.96 Cooperative Oppositional Boys Girls 37.55 41.17 10.31 6.94 Autonomous Boys 34.73 6.93 Dependent Girls 33.67 10.86 Social Boys 115.73 36.08 Competence Girls 135.83 23.46 5;6 6;5 Absence of Boys 81.73 14.47 Internalizing Girls 75.17 15.97 Absence of Boys 75.55 20.38 Externalizing Girls 75.17 13.48 General Boys 273.00 19.30 Adaptation Girls 285.50 19.07 Social Competence and Behavior Evaluation ScalI 2d Joyful Boys 37.52 7.12 Depressive Girls 37.31 8.72 Secure - Boys 36.22 7.92 Anxious Girls 36.31 9.96 Tolerant Boys 29.98 9.79 Angry Girls 31.27 8.78 Integrated - Boys 36.16 8.72 Isolated Girls 37.09 10.35 Calm Aggressive Boys Girls 33.74 36.64 11.31 9.25 Total Prosocial Egotistical Cooperative Oppositional Boys Girls Boys Girls 30.00 32.13 36.52 39.56 9.50 9.38 10.76 6.84 Autonomous Dependent Boys Girls 33.62 31.73 10.02 9.73 Social Boys 119.50 31.38 Competence Girls 126.76 30.96 Absence of Boys 79.64 14.93 Internalizing Girls 76.56 17.83 Absence of Boys 74.62 21.13 Externalizing Girls 78.73 16.06 General Boys 273.68 56.13 Adaptation Girls 281.76 53.05