24 July 2021
AperTO - Archivio Istituzionale Open Access dell'Università di Torino
Original Citation:
Beliefs of social equality and attribution of traits to in-group and out-group peers: a study of Italian Children
Terms of use:
Open Access
(Article begins on next page)
Anyone can freely access the full text of works made available as "Open Access". Works made available under a
Creative Commons license can be used according to the terms and conditions of said license. Use of all other works
requires consent of the right holder (author or publisher) if not exempted from copyright protection by the applicable law.
Availability:
This is the author's manuscript
1
i
i
--· ,'-{ l l:
;
l !Cognition, Brain, Behavior. An lnlerdisciplinary Journal
Copyright ~J 2010 Romanian A'sociation for Cognitive Science. Ali rights reserved.
ISSN: 1224-8398
Volume XIV. No. l (March). 47-61
Cognition
Bra
in&Behavior
BELIEFS OF SOCIAL EQUALITY AND ATTRIBUTION
OF TRAITS TO IN-GROUP AND OUT-GROUP PEERS:
A STUDY OF ITALIAN CHILDREN
Cristina
MOSSO·.
Emanuela
RABAGLJETTJ, Giovanni BR/ANTE,
Silvia
CIAIRANO
Department of Psychology, University of Torino, Torino
,
Italy
ABSTR4CT
The study investigatedfìrst"v, gender muf age d[[ferences with respect to the beliefs
of social equality, a cognitive component of system justification; second"v, the attrihution of positive and negative traits to in-group m1d out-group peers in school
age children. Thirdfcv m1d jìnalfcv, we explored the relations between socio! equa/il}'
muf attribution o.ftraits, controllingfor gender m1d age.
163 Italim1 children (Af age = 8.3 7 years, :::,7) = 1.11; 49% girls) participated in the study. They were administered a short se(f-report questionnaire investigating social equa/il}' m1d were asked lo attribute positive mzd negative traits to the figures oftwo children (one in-group chi/d with "white" skin; one out-group chi/d with "black"
skin). We .found that: a) o/der children perceived higher social equa/il).• m1d girls were less like(v thm1 boys to attribute negative traits to the in-group peer; b) children
·w/10 had higher social equality belie.fs were less like~v to attribute negative traits m1d more likefcv to attribute positive traits to both in-group mul out-group peers, a/so controlling .for gender m1d age. Increasing the belie.fs o.f soci al equality in children
appears a useful educational interventionfor promoting both in-group mui out-group
non-discriminato!}' peer relations.
KEYWORDS: socio! equality, trait attribution, chiÌdren.
• Corresponding author: E-mail: cristina.mosso(0unito.it
48 C. Mosso, E. Rabaglietti, G. Bria11te, S. Ciaira11o
JNTRODUCTJON
Immigrant chi1dren are the fastest growing segment in ali westem chi1d popu1ation, including Italy, the country where this study was conducted (Ine, 2008: lstat 2008). Thcse chil.dren come from a varicty of nations, speak a multitudc of 1anguages, and have a range of ethnic, cultura!, religious, and socioeconomic backgrounds. Somctimes some physical characteristics. such as the color of the skin aml/or a traditional drcssing, suggest a forcign origin with respect to the country where thcy Iivc and indicate their belongìngs to thc out-group. Besides, one may be perceived and categorizcd as out-group and therefore stigmatizcd as '"diverse" cven if he/shc fccls an in-group member.
Ncvcrtheless, some episodcs showed that in many Westem countries inunigrant and/or diverse peoplc. including children, may be victims of
discrimination and even of racism (Zincone, 2001; Johnson & Lambrinos, 1985;
Licata & Klein, 2002). Episodes of discrimination and racism are often supported
by negative stereotvpes versus the out-group and in-group favoritism. Furthermore,
violence and discrimination towards peers in generai have dramatically increased at ali school levels (Menesini, 2008). Therefore, the investigation of the processes underlying discrimination among children is becoming an increasingly pressing issue for scholars, as well as educators and policymakers.
Wc do not want to enter bere in the large debate (e.g., JosL Burgess, &
Mosso, 2001), which has concemed especially adults, about the possible relations among ideology (H avei, 1991) and other belief systems that serve as excuses an d justifications for discriminatory individuai, political. soci aL and economie behavior and attitudes. We would just mcntion that according to the system justification theory some idcology and belicfs make people feel better when expressing in-group
favouritism and negative stereotypes towards the out-group (Jost & Hunyady,
2005). However, othcr previous studies on social dominance (Sidanius & Pratto,
1999), ego justification (Fein & Spencer, 1997), and social identity (T~jfel &
T umcr, 1986) contributed to an appreciation of the functional basis of stereotype content. A key assumption is that pcople in generai, and also children from the age
of tlrree (Aboud & Amato, 200 l), are motivated to evaluate their 0\\11 group
positively, enhancing or maintaining a positive scnse of their social sclf. For children, leaming an adequate m1derstanding of soeial rules, regu1ations, and
practices is cmcial in the proeess of growing up (Fischer & CmmelL 2003). The
acquisition of this knowledge is Ì!Ùlerently linked to group membership and self-categorisation (Hirschfeld, 1996). We also already know that social self-categorisation, as thc proeess for cstablishing favourable distinctiveness and in-group favouritism do not lead only to a positive identity and self-perccption (Aboud, 2003). They can -also lead to a distorted pereeption of variability, as they exaggerate differences
between and similaritics within groups (Messick & Mackie. 1989). Besides, social
categorisation an d in-group favouritism may encourage the attribution of negative Cognitio11, Brai11, Be/wvior. A11 Interdiscipli11ary Jour11al
C. Mosso, E'. Rabaglietti, G. Briante, S. Ciairano 49
traits to the out-group and the fom1ation and use of socia\ stereotypcs,
which
arebased on thc notion that ali mcmbers of a particular category are alike in some way
(Hamilton & Troiler, 1986). The exaggeration of simìlarities is cspeciall:v
pronounced when one evaluates groups to which he/she does not belong.
In spite of the importance of invcstigating these phenomena in childrcn, to our knowledge. no studies hm·c systematically examined whethcr and how systcm justification. and/or some ofits components like cognitive beliefs.of social equality, affcct intergroup evaluation in children. Studies h ave rather focused cither on social identit:v proccsses <md/or on cognitive dcvclopmcnt in tem1s of incrcascd ability to
interpret social rcalit:· (Biglcr. Brown, & MarkelL 2001: Nesdale. Durkin, Maass, &
Griffiths. 2004). Among others, the study by Rholes and Ruble (1984) showed that 9-l O years old children wcre relatively more consistent than young,er children in predicting thc actor's behaviour on thc bases of vignettcs that wcre dcsigned to rcvcal his/hcr abilities and/or personalit:· traits. Also. the study by Yuill and P carso n (1998) highlighted that children from 5 years apprecìated traits as psychological causes ofbehavior.
Furthcm10re, we already know that the perception of bcing similar to other peoplc (whatcver group they bclong to) is at thc basis offriendships from childhood
on (Kupcrsmidt, DeRosier, & Patterson, 1995), and it may promotc empathy and
prosocial behaviour and prevent thc usc of aggression (Eisenberg & Miller. 1987).
Previous studies illustratcd that as childrcn age and cognitive devclopment advances from concrete to formai thinking they also show an increasing abilit:· of using high-order morality and being cmphatic with other people (Hoffman, 2000).
lt is also important to consider gender differcnces whcn soci al equality and the attribution of traits !o peers are concemed. Previous studics already showed that boys and girls construct peer relations and friendships differently at !cast starting from late childhood. Girls seem more likely to appreciate the quality and the decpness of rclations, whilc boys are more likely to regard sharing actiYities with peers (Rubin, 1985: Stein, 1986). The value of diffcrent facets of social relations may affect also thc beliefs of social cquality and the use of social categorization.
However, the findings of a rccent mcta-analysis on mora! oricntation (Jafee &
Hyde, 2000) do not offcr strong support for thc claim that women use predominantly a care oricntation a.nd that men use predominantly a justice orientation.
The present study aims to investigate a) age and gender differenccs for beliefs of social equalit:·, as a cognitive component of system justification: b) thc attribution of positive and negative traits to in-group (rcpresentcd as a
"'vhitc"' child) and out-group (rcpresented as a "black" child) peers in school-age
children. Besides, it focuses an the relations betwcen the beliefs of social cquality and trai! attribution.
Cognition, Brain, Behavior. An lnterdisciplinary .Journal 14 (2010) 47-61
50 C Mosso, E Rahagfh1ti, G. Briante, S. Ciairano
More spccifically. we poscd two rcscarch qucstions:
l. Are thcrc mcan Jc,·cl diffcrcnccs with rcspcct to bcliefs of social cqualitv and
the attribution of positive and negative traits to in-group and out-group pccrs bctwccn boys an d giri s. and youngcr an d o l der children'l
2. ls thcrc a rclationship bctwccn soci al equality an d thc at1ribution of positive an d
negatiYe traits t o others. alter hm ing eontrollcd for gcnder and age?
Wc cxpectcd to find an agc-rclatcd progrcssion in thc bclicfs of social cquality in tcnns that oldcr childrcn show higher social cquality. This assumption is based on thc fact that youngcr children may be lcss able to discntanglc thcir personal intcrcsts from generai soci al considerations of other peoplc.
Similarly, wc cxpcctcd some agc-rclated progression in respcct lo trai! attribution, with vounger childrcn attributing traits to pcers more easily than older ones. In faet. al!hough trait attribution is very common in adult population. as pcoplc grow thcy secm to require a large amoun! of infonnation for trait inferencc
(Aloise. 1993).
Previous infonnation \YC had about gendcr is too scarce to allow us lo make
any strong prediction about gcndcr-rclatcd findings with rcspcct lo both social cquality an d trai! atiribution. Howcver, for trait attribution. wc h ave to con si der that thc gender of thc figure presented to thc ehildren in this study is male. On the one
side. \YC might cxpeet to find out that boys are more cxtrcme than girls in thcir
judgmenL That is. boys might atiributc in case both positive and negative traits to thc pccrs reprcscnted in thc figurcs bccausc thcir possibility of idcntification is strongcr. On thc othcr si dc, wc alrcady leamcd that the ratings of pcrsonality traits
by eight- to tcn-ycar-old childrcn rcJlcct strong biascs favoring thcir 0\\11 scx
(Pmdishta, 2005)
Finally. as adults usc negative stercotypcs towards the out-group much
more when they justif\' soeial incquality (Jost & Hunyady, 2005), we expcetcd that
!he bcliefs of social equality are ncgatively related lo thc attribution of negative traits. On the samc bascs. wc expected also that bclicfs of social equality are
positive!~ rclatcd lo the at1ribution of positive traits, espccially in the figure
rcprcsenting the out-group child. Howevcr. we did no! find any prcvious st11dy cxploring in childrcn the rolc of imbuing social incquality with legitimacy and of
sceing il as good. fai r. natura!. dcsirablc. and cvcn unavoidablc.
METHOD
Participants
Thc participants werc l 63 childrcn agcs G to l l (M age = 8.37 ycars, S'D= l. l l:
49'Yo females) attcnding the second an d the fourth gradcs (8 classes. 4 for each school) in two primary schools in Torino. ltaly. In thc following text the childrcn in thc sccond classes are classified as younger (49%) and those in the fourth classcs as oldcr. In ordcr lo cxclude the possibility that acquaintancc \YÌth immigrant matcs
Cognition, Brain, Behavior. An lnterdisciplinary .Tournaf 1-1 (2010) -17-61
C. Mosso. f:'. RahagliL11i, G. Briante. S. Ciairano 51
chnngcs thc pattcrn of findings. wc sclcctcd only classcs in \\ hich thcrc \\Crc no non-Itnlian childrcn.
Most parcnts (98'Yo of fathcrs, 891% of mothcrs) \YCrc employed. Bascd on
thc rcsponses lo questionnaircs complelcd by lhc parcnts \\C found lhal 46% of
fathers and 26'% of mothers could be classificd as lo\\· social slatus (thcy wcre Wlcmployed. or manual \mrkcrs). 54% of fnthers nnd 73% of mothcrs could be classificd at an intenncdinte leve l of social status (theY h cl d salcs. administraliYc. or inlcllcclltally-oricnlcd jobs. c.g. teachers, rcscarchcrs~ dcsigncrs) an d fina] h· l 'Yo o[
lhc mothcrs hcld managcrial positions. Thcsc socio-dcmographic figures ar~ similar lo those found in thc genera! populntion from the samc ltalian proùncc (lstnt. 20tl7).
Procedure
Parcnts proYidcd conscnt for thcir childrcn to participatc, and thc childrcn thcmsclYcs agreed lo parlicipalc in accordanec with ltalian law and the ethical code of the Association of ltalian Psychologists. The children completed questionnaires. \Yhich werc distributcd by research staff during classroom timc. They also took home a qucstionnaire rcqucsting socio-demographic infonnation to be completcd by lheir parcnts. TI1is questionnairc \Yas completcd by 94% of parents=. Wc found no rcleYant differcnces belwcen the children of parents \Yho lì l! cd in lhc questionnaire and thosc \\ho did nol with respeet to the other information considered.
TIIC childrcn· s and parents · questionnaires took approximalcly 30 and l O minutes rcspccti\·ely to complete. Both childrcn and parcnts \Ycre assured confidcntiality and anonymity. Teachers \\·ere not present in the classroom whilc the questionnaire was administered lo the ehildren. Pencils and mbbers were offered as incentiYe for participation. l 00% of the childrcn complctcd thc qucstionnaircs.
Measurcs
Perception of soci a! equality. This \Yas assesscd using fi\·c itcms - dcye]oped taking into considcration the young age of thc childrcn - aftcr haùng describcd the following situation: "[s family is poor. Z has always been hard \\orking. dcdicated an d confident in his/hcr abili
t:·
to succccd. Unfortunately. pcople aren ·t always giYen the samc opportunities to succeed in our society--. TIIC children wcre askcd n·hether or not thcy think that it is fair or tme that: a) z·s teachcr always gives him/hcr thc Jast choice of colored pcnciJs: b) z·s classma(CS llCYer pJay \YÌth1 Tbc majoritv of childrcn· parents (58. 1°ft, of L1thers and 52.7% of motbcrs) had a high
scbool diploma. 11.7% of l~Itbers and 20.9'% of mothers had a univcrsitv degree. 1\egarding famìly stmcturc, 92% ofthc parcnts lìved togcthcr, and R'X. werc separatcd or di,·orced. 'Tbc parcntal qucstìonnaìre wns tìllcd bv 69.2% ,,fmothers r_J! agc=39. 72 vears) and 43%
n
f age =43 7'1 'carsl of fnthcrs.CogniJion, Brain, Belwrior. An !nterdisciplinary .Tournal l-l (20/0) -17-61
52 C Mosso, E RahagliL'ffi, G. Briante, S. Ciairano
him/her: c) the class leader oftenmistrcats him/her: d) Z"s teacher and classmatcs
don ·t always acl the same way towards him/hcr: e) therc ·s no diffcrence betwcen
rich peoplc an d poor pcople. The answcrs ranged from l (not at ali agrce) t o 5 (complete agrec). The itcms a, b. and c wcrc recordcd for haYing highcr scores for social cquality.
Trait attrihution to in-group and out-group peers. Considering ù1c nccessity to
prcsent childrcn of diffcrent ages the samc stimulus. wc used two simple figurcs rcpresenting a whitc child (in-group) and a black (out-group) child: both lìttlc boys werc smiling and skating (see Figure l). This figure was carefully sclccted becausc i t showcd two childrcn who were identica] in every way cxccpt for thcir skin color. Wc know that the choice of rcprescnting in-group and out-group by the way of the color of the skin may be controversia!. However wc also based our choice on the study by Bigler and Liben (2006) who argued that person charactcristics that are perccptually discriminable are more likcly than other charactcristics to becomc the basis of stereotyping in childhood. l n tenns of skin color, all our participants werc likely to pcrccivc thc white boy as more similar to them than the b!ack boy. Howcver. wc cannot cxclude that thc boys \\·ould pcrceive themsclves as more similar than thc girls to both the children in the figure becausc of thc masculine gendcr.
The trait attribution to in-group and out-group pcers was assessed using S items per cach figure of peer. The children wcre asked \Yhethcr or not they thought the \\hitc and thc black boy were: a) good: b) clean: c) nice: d) happy: e) sad:
f) dirty: g) bad: h) unplcasant. The answers ranged from l (Not at ali) to 5 (Very).
Highcr scorcs mcan highenrait atlribution.
Out-In-group child
F;g11re l
Drawing. used in thc studv.
Cognitùm, Bra in, Beharior. An lnterdisciplinary .!oumal J.l (]{}]Oì 47-61
C Mosso, E. Rahaglh1ti, G. Briante, .'::. Ciairano 5]
RESULTS
P5yclwmetric clzaracteristics o.f soda! equalitr ami trait attribution
To assess thc dimensions undcrlincd bY the itcms aimcd al invcstigating social equality and trai! attribution lo in-group and out-group pccrs, wc uscd principal
componcnt analysis. truncatcd a t thc eigenvaluc o[ >l, and \vhcthcr ncccssary
Varimax rotation. From this analys,is wc savcd the factorial scorcs and uscd thcm as nriables in thc fo!lo,ving analysis. Furthcnnore, to investigate thc rcliability of thc scales wc uscd Cronbach · s al p ha on the originai itcms.
Wc first analyzed thc itcms rcfcrring to social cquality and found only onc undcrlying dimcnsion in which all the items saturated positivcly: cigemaluc 1.74. cxplained vari ance 35'Yo. Howcvcr, the Cronbach · s alpha is rathcr low: .50.
Thc ncxt step \Vas to analyzc thc itcms aimcd at cvaluating trai! attribution lo the whitc/black child. In both cascs wc found two dimcnsions (onc negative and onc positive). Thc itcms sad, dirty, bad, and unplcasant ali saturatcd positivcly in thc first factor. Wc callcd thc first dimcnsion: negative traits. Thc items good. clcan. nice, and happy saturated positivcly in the sccond factor. Wc callcd thc sccond
dimcnsion: positive traits. Tiw psychometric infonmttion about thc two componcnts
of trait attribution are rcasonable similar for both white and black children. Negative traits: eigenvaluc 2.46, cxplaincd vari ance 31 %; rx=.57 for thc white chi Id: cigcnvalue 2.44, cxplained variance 30%: rx=. 71 for the black chi l d. Positive traits: cigenvaluc 1.87, cxplained variance 23'%: rx=.79 for thc whitc child: eigcnnlue
2.1 8, cxplained yariancc 27%: a=. 77 for the black child.
Furthcnnore wc replicatcd these analyses in the sub-samplcs of girls and boys. youngcr and oldcr and wc did not find any rclcvant diffcrcncc in this pattcm of findings.
Gender a m/age differences in mean leve[ o.fsocial equality ami trait attribution
To assess the study variables for gendcr and age diffcrcnces, \\C used 1-tests for
indcpendent samples. Therc \Ycrc no gcndcr differenccs in mean lcYels of social cquality. attribution of positiYc traits to in-group peer, and attribution of negative
and positive traits lo out-group pcer (T ab le l). Howcvcr. boys (M= .27. SD = 1.16)
attributcd more ncgatiYc qualitics to ù1c in-group child than girls (M = -.28.
SD = 71 ). 1(147) = -3.49. p< .001. Comparcd to the youngcr childrcn. thc older
childrcn reported highcr lcYcls of soci al equality (M = .27_ SD = . 77 for o l der:
M= -.29. SD = l.l3, for younger: 1(159)= -3.70. p<.OOOI ). No other age diffcrcnces wcrc found.
Cognition, Brain, Be!tavior. An Interdisciplinary Journal
54 C. Mosso, E. Rahaglietli, G. Briante, S. Ciairano
Tablc l
Afean and slandard deviation of children 'beliefs o.fsocial equality and attribulion o.f positive and negative traits lo in-group and out-group peers ·
(!-test for independenl samples).
Group Girls Boys
t-test d. f. p M(SD) M(SD) Social equality .03 (.93) -.03 (1.1) .43 !59 .67 Neeative traits -.28 (.71) .27 (1.16) -3.49 147 .001 in-~roup peer Positive traits -.08 (1.0) .08 (.97) -.95 147 .35 iu-group peer Negative tra1ts
out-group peer -.Il (.99) .Il (1.0) -1.39 !50 .17 Positive traits
out-group peer -.09 (1.15) .09 (.09) -1.10 !50 .27 Group
Younger Older t-test d.f. p
Social equality -.29 (L 13) .27 (.77) -3.70 !59 .0001 Negative traits .()3 (1.04) -.03 (.97) .34 147 .74 iu-group peer Positive traits .01 (J.l6) -.01 (.84) .12 147 .91 iu-group peer Negative traits
out-aroup peer .13 (1.14) -.1 I (.85) 1.49
I?O
.14 Posi ti ve traitsout-group peer -.10 (1.22) .09 (.74) -1.21 150 .23
Relations hetween systemjustification and the evaluation ofpeers
T o investigate the relations between beliefs of social equality, traits attribution to in-group and out-group peers, and the moderating role of gender and age, we used the hierarchical regression approach, as suggested by Hohnbeck (1997). To be consistent with our aims, we first entered gender (l= boys) and age (l= older children), second we entered social equality, and fmally we entered the interactions between social equality and gender and age.
The fmal models were significant with respect to the attribution of negative and positive traits to the in-group child [negative traits: If= .15, F(5, 142) = 5.16,
p<.0001; positive traits: R2= .07, F(5, 142) = 2.26] and the attribution of negative
traits to the out-group peer [ff= .IO, F(5, 145) = 3.13,p<.OI]. The fina! mode! for
the attribution of positive traits to the out-group child did not reach significance
(R2= .03,F(5. 145) = 1.86,p=.l0).
Cognition, Brain, Behavior. An lnterdisdplinary Journal
C. Mosso, E. Rahaglielii. G. Brùmte,
s·.
Cinirano .55Rcgarding thc attribution of negative traits to the in-group pecr wc found a positive rclationship bctwecn bcing male and negatìvcly cvaluating thc in-group peer (T ab le 2 ). Boys appeared t o be more likely to attributc negative characteristics
lo the in-group peer. W e al so found that the cocfficient of soci al equality and that of
the interaction betwcen age and social equality were significant and both negatively rclated to the attribution of negative quaiities to the in-group peer. In other words, children who pcrccived higher social equality were also Jess likely to attribute negaliYe traits. Furthem10re, at the same leve! of soeial cquality, older ehildren wcre cvcn less likcly than younger children lo attribute negative traits to the in-group peer.
As for the attribution ofncgative traits to the out-group peer, we found that the eoeffieient of social equality was significant and that of the interaction bctween age and soci al equality was near to significance; both were negatively related to the attribution of negative traits lo the out-group peer. At the same extent of the in-group child, children who had higher social equality were also less likely to attribute negative characteristics to the out-group peer and this too was particularly true for older children.
With respcct to the attribution of positive traits to thc in-group peer, wc found that the coefficient of soeial equality was significant and positively related to the attribution of positive characteristics lo the in-group pcer. We also found a signifìcant and negative interaction between age and socìal equality. That is at the same leve! of soeial equality, oldcr childrcn were less likely than younger children to attribute positive traits to the in-group peer.
As for lhe attribution of positive traits to the out-group peer. wc found that only the coefficient of social equality was significant. A t ù1e same extent of the in-group child, the children who had higher social cquality were also more likely to attribute positive traits to the out -group peer.
Table 2
Hierarchical regression resu/ls predicling a/tribution oftrails lo peers. Negative trait"
in-group peer
Positi\·e traits in-group pecr Negative traits out-group peer Positive traits ou1-group peer Predictors fl :1R' fJ JR' fJ JR' Stop l .07** .01 .03 Gender .27** .09 .11 Age -.04 -.02 -.12 Step 2 .04* .03* .05**
Social equality (SE) -.21 * .19* -.22**
Step 3 -lnkractions .04* .04é· .01
GendcrX SE -.18 -.09 -.09
..\ge X SE -.19* -.21 *
-.18-~-Notes: "'P< .10: *p< .05; **p< .01.
Cognition, Brain, Behavior. An Jnterdisciplinary .!ournal
1-1 (201 0) -17-61 fJ _lR' .02 .09 .09 .02* .16"' .01 .17 -.14
56 C ]!fosso, E. Rabaglietti, G. Briante, S Ciairano
DlSCUSSTON
The aims of study presented were: a) to investigate age and gender differences for beliefs of social equality, as a cognitive componcnt of system justification, b) the attribution of positive an d negative traits to in-group (represcnted as a '"whitc" child) and out-group (rcpresentcd as a "black" child) peers in school-age childrcn. Finally, it examincd whether therc is some rclatìonshìp between the bcliefs of soci al equalìty and the attribution of positive and negative charactcristics to in-group and out-in-group peers.
TI1c cxpectation that oldcr children would show hìgher levels of social equality compared to younger children secms to be confinned by our findings. !t is reasonable that the younger children have not yet developed the cognitive skills, including self-decentralization, necessary far showing more mature beliefs about social inequality, seeìng it as bad, unfair, undesirable, and even more important avoidable.
However wc also expccted a similar age-progression with respect to trait attribution with younger children more case than older in attributing traits to peers. In fact on the basis of a study that compared children to young adults (Aioise,
1993), we expected that as children grow they require a large amount of
infonnation for making trait inference. Thus, older children might be less likely to attribute traits to the other chi ldren simply o n the basis of the stimulus used in this study. Howevcr, wc di d not fin d any significant age diJferences far trait attribution. This finding certainly needs to receive further confim1ation in widcr samples and longitudinal studies. Besides, wc have to admit that the figure wc uscd as stimulus might not be thc most adcquatc far cliciting age-relatcd differences for trait attribution at ù1is stage of childrcn development.
Nevertheless, to consider age diJferences we have to also take into account the findings of the regression analyses. In fact age was showcd interacting with social equality in the attribution of negative traits to both in-group and out-group pccrs an d of positive traits to thc in-group child. Children \vho had higher soci al equality were less likely to attributc negative traits to peers in generai and more likely to attribute positive traits to the in-group peer. However, at older ages this pattem of relations seems to be reinforced in the case of the attribution of negative traits and weakened in thc case of the attribution of positive traits. Summarizing, there is at least some indication that older children may be in generai less easy than younger to judge their peers when infom1ation is scarce, as wc can expect on the basis of previous studies in adults (Aioise, !993).
As far as gender is concemed, we only found that boys were more likely than girls to attribute negative traits to the in-group child. Therefore, they appeared a bit more extrcme, and probably also more superficial than girls in their attribution of traits to ù1e other pcople. TI1is finding may be collocatcd within the generai framcwork of different facets of relations boys and girls pay attention to (Rubin,
Cogni1i.on, Brain, Behavwr. An Jnterdiscip/bf(lry .Tournal 14 (2010) 47-61
C. Mosso, E. Rabaglielti, G. Briante, S. Ciairano 57
1985: Stein, 1986): usually girls pay great attention at the qualitative aspccts of relations while boys at tl1e "quantitative" aspects as number of friends and activities shared together. The attention to different aspects of relations may also influence the processes of social categorization leading boys to be more case than girls in attributing negative traits to peers. However, our fmding might have been influenced by tl1e fact that the figure we presented to children represents a male. Wc certainly need to confrrm this fmding in future studies where tl1e figure will be a female. Though wc must first explore what happens when tl1e figure presented shows a giri, future research should al so investigate whetl1er boys and girls develop
differently in temts of their generai tendency to attribute negative or positive traits
to peers. T o our knowledge some studies already investigated gender differences in trait attribution but in adults and ·witl1 respect to specific kinds of attribution like
politica! ones (see, Alexander & Andersen, 1993).
Two main findings regard the relations between social equality and trait attribution. Firstly, the beliefs of soci al equality were found always significant an d in the expected direction. That is, thcy were negatively related to the attribution of negative traits an d positively related to tl1e attribution of positive traits. Secondly, i t is noteworthy that children · beliefs of social equality seem to work exactly in tlle
same way in tl1e case of botl1 in-group and out-group peers. In both cases higher
beliefs of social equality were showed to combat the tendency to easily attribute traits to peers.
In this we think that our fmdings go a step further what has been already underlined in adult population. People who recur less to the justification and the rationalization of social inequality are also less likely to use negative stereotypes
towards tl1e out-group (Jost
&
Hunyady, 2005). In our study wc highlighted tllatchildren who bave high beliefs of social equality may be more likely to construet not discriminatory peer relations in generai, since they are Jess easy to attribute traits to peers on tl1e basis of scarce information.
A previous study already showed that the beliefs of children that traits are stable predicted a greater tendency to make trait judgments and tltat these beliefs are associated with an emphasis on tl1e evaluative meanings of behaviors (e. g., whether the person is good or bad). That is, high beliefs tl1at traits are stable are related to
focusing on outcomes and behaviors through which traits can be judged (Heyman &
Dweck, 1998). We also found that the differences in beliefs of social equality contribute to explain why some school-aged children are Jess Hkely than others to begin the process of constructing stereotypes independently from tl1e group membership. Our findings represent an indirect confirmation that in-group identification may be independent of negative attitudes toward out-groups (e.g., Brewer, 2002, and previously Allport, 1954).
Furthermore, beliefs in social equality are something wc may address at an educational leve!, although taking into account tlle difficulty of this kind of intervention (Bigler, 1999), in order to decrease the perception of difference and
Cognition, Brain, Behavior. An Interdisciplinar.r Joumal 14 (2010) 47-61
58 C. Mosso, E. Rahaglietti, G. Brionte, S. Ciairano
deter problems related to prejudice. Educating a sense of social equality may bave an important role in preventing a negative evaluation and promoting a positive evaluation of peers, wbetber peers belong to the in-group and/or to the out-group. Educating social equality might tum out to be ev.en more important in a society that is rapidly becoming multicultural. This is due tt) the fact that the perception of the other is the foundation of ·ali out-group processes including the attribution of negative characteristics to the out-group and the over-estimation of the positive
cbaracteristics of the in-group (Tajfel & Tumer, 1986). Although the path from
implicit bias to discriminatory action is not inevitable (Dasgupta, 2004). these complex processes are the basis for both the formation of social stereotypes and the justification of aggression towards others.
Nevertheless, tltese findings clearly require further confrrmation in different and wider samples, especially with longitudinal desigus. A longitudinal research design would bave allowed us to follow the developmental patii of social equality an d trai t attribution an d tbe possible intert"'<inement between tbem. LIMITA TIONS AND STRENGTHS
Two key limitations of tbe present study are the cross-sectional design and tbe relative weakness of tlte stimulus and measures used. While the young age of the participants prevented us from using a longer series of items and more complex stimuli, tbe lack of a longitudinal contro! make it impossible for us to investigate the direction of the relations found and to apply more complex strategies of analysis, wbich could bave shed more Iigbt on the phenomenon under study. Tbe next step in this Iine of researcl1 is to investigate these associations over time, as well as to vary tlte stimuli and introduce different measures of soci al equality. The relatively small sample size and the fact that ali participants were from one Italian region al so make it difficult to generalise results to different populations.
Besides; to avoid potential interference from· acquaintance witl1 out-group cbildren we selected classrooms witb no immigrant children but this migbt bave introduced another kind of bias. We probably 1tave to Iook at the pattems of relations between social equality ·and trait attribution in classes with different proportion of immigrant cbildren. For instance, a previous study showed that in-group favouritism and out-in-group prejudice were reciprocally correlated in one sample from a racially homogeneous school but not in another santple :from a mixed-race scbool (Aboud, 2003).
Furtbermore, we certainly need to take into account otber components of
system justification, such as tlte motivational ones (Jost, Burgess, & Mosso, 2001),
and tlte possible sources of information of cbildren 's beliefs and easiness to attributing traits. I t seems reasonable that parents and other significant adults fulfil a relevant role in tlte construction of children' beliefs of social equality and trait
Cognition, Brain, Behavior. An lnterdisciplinary Journal 14 (2010) 47-61
C. Mosso, E. Rabagfietti, G. Brionte, S. Ciairano 59
attribution through the attitude and behaviour they show more or lcss ovcrtly
(Biglcr & Libcn, 2006).
In spite of its limitations, this study al so has some merits.
Firstly. it highlights the importancc of investigating both social equality and trait attribution as potential precursors of discrimination and stereotypes in childhood, although they can assume differcnt fonns in subsequent phases of development.
Secondly. this study underlined the importancc of introducing specific curricula to promote the beliefs of social equality and, therefore, to prevent some of the negative consequences of social stcreotypcs in childhood and latcr on.
Finally, it highlightcd the potential rclevance of social cquality in tenns of prcscnt and future adjustmenL not only on an individuai leve] but also for social groups. Wc feci that the introduction in schools of activities designed to dcvelop a scnsc of equality and legitimacy could be useful for the promotion of positive soci al developmcnt in childrcn and may, in tum, even have potcntial conscquenccs for the future adjustmcnt of thc socicty in which thesc childrcn willlive
ACKNOWLEHGEMENTS
W e warmly thank Laura Negri and Renato Miceli (Department ofPsychology. Torino, Italy) for their support and sugg:estions in the realisation of this research. W e also thank the MA students in the Dcpartment of Psycholog:y, Torino, ltaly, who administered the questionnaires. Finally, w e thank the children, parents, an d teachers w ho were involved in this studv for their enthusiaslic participation.
REFERENCES
Aboud, F. E. (2003). The formation of favouritism and out-g:roup prejudice in voung: children: are they distinct attitudes? Developmental Psychology, 39. 48-60. Aboud, F. E., & Alnato, M (2001). Developmental Socialization influences on Interg:roup
bias. In R. Brow11 & S. Gartner (Eds.), Blackwell 's Handbook ofsocial psychology: lntergroup processes (pp. 65-85). Malden, MA: Blacbvell.
Alexander, D., & Andersen, K. O 993). Gender as a Factor in the Attribution of Leadership Trails. Politica/ Research Ouarterlv, 46. 527-545.
Aloise, P. A ( 1993). Trait Confrrmatio~ and Disconfirmation: The Development of Attribution Biases. Joumal of Experimental Child Psychology. 55, 177-193.
Allport, G. W. (l 954). The Nattlre of Prejudice. Boston: Beacon Press.
Big:ler. R S. (l 999). The Use of Multicultural Curricula an d Materials t o Counter Racism in Children. Joumal ofSociallssues, 55, 687-705.
Big:ler, R. S., Bro,m, C. S., & Marke!L M (2001). Vvl1en g:roups àre not create-d equa]: E!Iects of g:roup status on the lòrmation of interg:roup attitudes in children. Chi/d
Development, 4, 1151-1162.
Cognition, Brain, Behm•ior. An lnterdisciplinary .!ournal /4 (2010) 47-61
60 C. Mosso, E. Rabaglietti, G. Britmte, S. Ciairano
Bigler, R. S., & Liben, L. S. (2006). A developmental intergroup theory of soeial stereotypes and prejudiee. Advrmces in Chi/d Deve/opmenl and Behavior, 34, 39-89.
Brcwer. M. B. (2002). The Psycholog:v of Prejudice: Ingroup Love and Outgroup Hat. .loumaf ofSocial Jssues, 55, 429-444.
Eisenhcrg, N., & Miller, P. A ( 1987). The relation of empathy t o prosocial and relatcd hehaviors. Psvcholoe:icaf Bulletin, 101.91-119.
Fcin. S.,& Spcncer,
S.
J o'997). Prejudicea;
se1f-image maintenance: affirming the self lhrough negative eva1uation of others. Jouma/ of Personality and Socia/ Psychology, 73,31-44.Fischer, K. W., & Connell, M. W. (2003). Two motivational system that shape deve1opment: epistemic and self-organizing. British .louma/ of Educationa/ Psychology, Monograph Series, 2, 103-123.
Hamilton, D. L, & Trolier, T. K. (1986). Stereotypes and stereotyping: An overview of the cognitive approach. In J. F DO\·idio & S. L Gaertner (Eds.), Prejudice, discriminalion, m1d racism (pp. 127-163). San Diego. CA US: Academic Press. Haines, E. L., & Jost, J. T. (2000). Placating the powerless: Effects of legitimate and
illegitimate explanation on afTect, memory, and stereotyping. Socio/ Justice Research, 13,219-236.
Havel, V. (1991). The power of the powerless. In V. Havcl (Ed.), Open letters (pp. 125-214). London: Faber & Fa ber.
Heyman, G. D., & Dweck C. S. (1998). Children's Thinking about Traits: Imp1ications for Judgments ofthe Selfand Others. ChildDevelopment, 69, 391-403.
1-lirschfeld, L. (1996). Race in the making: Cognition, culture, mzd the child's construction o[humm1 kind~. Cambridge, MA: MIT Press.
1-lolmhcck, G. N. (1997). Toward temlino1ogical, conceptual, and statistica! clari~· in the study of mediators and moderators: Examp1es from the child-clinical and pediatric psychology literatures. Joumal ()[ Comu/ting m1d Clinica/ Psychology, 65, 599-610.
Hoffman, M. L. (2000). Empathy and mora/ development. Jmplications jàr cm·ing mul justice. Cambridge: Cambridge Universi!)• Press
Ine (2008). The 2007 Nationaf lmmigrant Si1rvey: extending the knowfedge on immigrm1ts heyond registers, Unece-Eurostat Work session on migration statistics (Geneva, 3-5 March 2008).
Avai1ahle online at www.unece.org/stats/documents/2008.03.migration.htm [Accessed August 3 2009].
Istat (2008). Statistics on migralion in lta~v. a reassessmenl of sources m1d methods, Confercnce of Europea n Statisticians, 56th plenary session (Paris, l 0-12 June 2008).
Availa ble online al http://www. unece.org/stats/documents/ece/ces/2008/43 .e. pdf [Accessed August 3 2009].
lstat (2007\. Annuario statistico italimw 2007 [ltalim1 Statistica! Yearbook 2007]. Rome, IT: ISTAT. Avai1able online at http://www.istat.it [Accessed July 15 2008).
Jaffee, S., & Hyde, J S. (2000). Gender Differences in Mora] Orientation: A Mcta-Analvsis. Psychological Bulletin, l 26, 703-726.
Johnson, W. G., & Lambrinos, J. (1985). Wage Discrimination against Handicapped Men and Women. The Jouma/ of!Jumrm Resources, 20, 264-277.
Cognition, Brain, Be!ta~·ior. Anlnterdisciplinary .Tournal
C Mosso, E Rabaglietti, G. Briante, S'. Ciairano 61
.Tost, .T. T., Burg.ess. D.. & Mosso, C. O (2001). Conllicts of leg.itimation among. sdf. group. and svstem: The integrative potcntial of svstem justification theorv. In .T. T. .lost &
B. M~ior (Eds.). The p.~vchology of legitimacy: Eml!lging perspectives on ideolog;•. juslic~, m1d intergroup relations (pp. 363-388). New York. NY, US: Cambridge
Universitv Press.
Jost, .l. T., & Hu~vadv, O. (2005). Antecedents and consequcnces of svstem-justii':>ing
ideologics.· Ctin-ent Directions in Psychologicaì Science, 14. 260-265.
Kupersmidt, .T. B., Melissa. E, DeRosier, M. E., & Patlerson, C. P (1995). Similaritv as the Basis for Children's Fricndships: Thc Roles of SociDmetric Status, Ag.gressiw and Withdra\\11 Behavior, Academic Acbievemenl and Demog.rapbic Characteristics.
Joumal o.fSocial and Personal Re/ations, 12, 439-452.
Licata. L.. & Klcin, O. (2002). Does European citizcnship brecd xenophobia0 European
identification as a predictor of intolerance towards immigrants. .Joumal oj" Community & Applied Socio/ Psycholog;•, 12,323-337.
Mcnesini, E. (2008). School bullving in Jtaly: Nature, functions from childhood to adolescence and related interventions. In D. Pepler & W. Craig (Eds.). Understanding and addressing bullying: An inlemational perspcctive, PREVnet Series, V. l (pp. 253-270). Blooming, US: Au1hor House.
Messick, D. M., & Mackie, D. M. (1989). lntcrgroup Relations. Annua/ Re1·iew o.f
Psycholog;·, 40. 45-81.
Ncsdale, D., Durkin, K., Maass, A., & Griffiths, J. (2004). Group status. out-group ethnicitv and children· s ethnic attitudes. Applied Developmental P.~vchology, 25, 237-251.
Powlishta, K. K. ( 1995). Gender bias in children's perceptions of personalitv traits. Sex Roles, 32, 17-28.
Rholes W. S., & Ruble. D. N. (1984). Children's Understanding: of Dispositional Characteristics of Others. Chi1d Development, 55, 550-560.
Rubin, L. B. (1985). Just.fi'iends: The ro/e ofji·iendship in our lives. Ncw York: Harpcr and
Row.
Sidanius, L & Fratto, F. (1999). Soci al dominm1ce: An intergroup theory of".wcial hierarchy
aml oppression. New Y ork: Cambriclg.e TJniversitv Press.
Stein, P . .T. (1986). J\fen and theirfi"iendships. In R. A. Lewis & R. E. Salt (Eds.). Men in !àmilies (pp 261-269). Newburry Park, CA, US: Sag:e.
Tajfel, H., & Tumcr, .T. C. (1986). The social identitv themv of interg.roup behaviour. In S. Worchel & W. G. Austin (F'Ais.), The psyclwlog;· of"intergroup relations. V. 2. New York: Nelson Hall.
Yuill N., & Pearson A. (1998). The dcvclopment of bases for trait attribution: children's understanding of traits as causa] mcchanisms based on desire. Deve/opmenral
P.'i)·cholo?J'. 3./, 574-86.
Zincone, G. (2001 ). Secondo rapporto sull'integrazione degli immigrati in Italia. Bologna: Il Mulino.
Cognition, Brain, Belun·ior. An lnterdiscipli11ary Joumal 1./ (2010) ./7-61
