Can changing schools help peer-victimized students escape their plight? A mixed-methods study

This is a sel -a chi ed e sion o an o iginal a icle. This e sion may di e om he o iginal in pagina ion and ypog aphic de ails. Au ho (s): Ti le: Yea : Ve sion: Copy igh : Righ s: Righ s u l: Please ci e he o iginal e sion: CC BY 4.0 h ps://c ea i ecommons.o g/licenses/by/4.0/ Can changing schools help pee - ic imized s uden s escape hei pligh ? A mixed- me hods s udy © The Au ho (s) 2024 Published e sion Tenhunen, Essi-Lo a; Kuusiaho, Inka-Liisa; Salmi alli, Ch is ina Tenhunen, E.-L., Kuusiaho, I.-L., & Salmi alli, C. (2024). Can changing schools help pee - ic imized s uden s escape hei pligh ? A mixed-me hods s udy. In e na ional Jou nal o Beha io al De elopmen , Ea ly online. h ps://doi.o g/10.1177/01650254241289308 2024 h ps://doi.o g/10.1177/01650254241289308 In e na ional Jou nal o Beha io al De elopmen 1 –10 © The Au ho (s) 2024 A icle euse guidelines: sagepub.com/jou nals-pe missions DOI: 10.1177/01650254241289308 jou nals.sagepub.com/home/jbd You h who a e ic imized by hei pee s a school end up chang- ing class ooms and schools mo e o en han hei non- ic imized pee s (Ca son e al., 2013), sugges ing ha such a change o con- ex is some imes mo i a ed by ic imiza ion. In o he wo ds, when o he a emp s ha e ailed, changing schools may be he las eso o end ic imiza ion. Ye , a ecen s udy sugges ed ha being a newcome in a class oom o school is associa ed wi h a heigh ened isk o ic imiza ion compa ed o es ablished s u- den s, pa icula ly when he mo e is p omp ed by p e ious ic- imiza ion expe iences (Tenhunen e al., 2024). Howe e , in hese cases, he inc eased likelihood o expe iencing ic imiza- ion in he new con ex migh be linked o indi idual cha ac e is- ics ha con ibu ed o ic imiza ion in bo h he p e ious and cu en school: he ole o school change hus emains unknown. The p esen s udy add esses he ques ion o whe he and how p e ious ic imiza ion his o y mode a es he e ec o school change on subsequen ic imiza ion, and how s uden s wi h and wi hou ic imiza ion his o y pe cei e he change. No ma i e and Non-No ma i e School Changes and Thei Associa ion Wi h Pee Vic imiza ion While his s udy ocuses on non-no ma i e school changes, i is essen ial o acknowledge he b oade con ex o school ansi- ions wi hin he educa ional se ing. No ma i e ansi ions, such as hose occu ing be ween middle and high school o due o he me ging o schools, ep esen ypical s ages in s uden s’ aca- demic jou neys. E en hese an icipa ed ansi ions can impac nega i ely he s abili y o iendships (Ng-Knigh e al., 2019) and place child en in o new pee g oups. Despi e he possible loss o iendships, a educ ion in pee ic imiza ion was epo ed among s uden s expe iencing middle school ansi ions, compa ed o hose a ending schools wi hou such ansi ions (Fa me e al., 2011). In ano he s udy, no ma i e middle school ansi ions we e ound o dec ease ic imiza ion among gi ls (Wang e al., 2016). Acco ding o s udy by Felmlee e al. (2018), s uden s unde going high school ansi ions aced addi ional challenges pos - ansi ion, including changes in iendship nom- ina ions and inc eased isola ion. Mul i-school ansi ions, whe e s uden s mo e in o one school om se e al eede schools, ini- ially dis up exis ing iendships (Temkin e al., 2018) bu o en lead o he eme gence o new ones by he end o he pos - ansi- ion yea (Newcomb & Bagwell, 1995). Can changing schools help pee - ic imized s uden s escape hei pligh ? A mixed-me hods s udy Essi-Lo a Tenhunen1, Inka-Liisa Kuusiaho2 and Ch is ina Salmi alli1 Abs ac Changing schools ends o be mo e equen among ic imized s uden s and is some imes used as a means o pu an end o pe sis en bullying. Howe e , whe he hese changes bene i ic imized s uden s emains unclea , as bullying ic imiza ion o en pe sis s ac oss di e en con ex s. This mixed-me hods s udy (n = 58,799 and n = 68 o quan i a i e and quali a i e da a, espec i ely) conduc ed in Finland examines he ole o ic imiza ion his o y in he associa ion be ween newcome s a us and ic imiza ion in he new school among elemen a y school s uden s ( ou h o six h g ade s, ha is, 10–12 yea s old) and middle school s uden s (se en h o nin h g ade s, ha is, 12–15 yea s old). The indings showed ha elemen a y school newcome s wi h mos equen ic imiza ion his o y seemed o bene i om he school change compa ed o es ablished s uden s wi h simila ic imiza ion his o ies. A co esponding, albei much weake , e ec was obse ed among middle schoole s. P e iously ic imized s uden s had ini ially been mo e willing o change schools han non- ic imized s uden s. The quali a i e con en analysis showed a mo e complex eali y—some o he p e iously ic imized newcome s had s ong aspi a ions o a be e u u e, while o he s expe ienced ea s and losses o es ablished iendships. E en ually, he school change can b ing bo h disappoin men s and ewa ds, depending on he newcome s’ abili y o make new iends and escape bullying. Keywo ds Newcome s a us, school bullying, school mobili y, pee ic imiza ion, pe sis en ic imiza ion 1Uni e si y o Tu ku, Finland 2Uni e si y o Jy äskylä, Finland Co esponding au ho : Essi-Lo a Tenhunen, INVEST Resea ch Flagship, Depa men o Psychology, Uni e si y o Tu ku, 20500 Tu ku, Finland. Email: [email p o ec ed] 1289308JBD0010.1177/01650254241289308In e na ional Jou nal o Beha io al De elopmen X(X)Tenhunen e al. esea ch-a icle2024 Special Sec ion: Pee Vic imiza ion 2 In e na ional Jou nal o Beha io al De elopmen 00(0) The cu en s udy ocuses on non-no ma i e school changes, whe e a child changes schools and en e s as a newcome in al eady es ablished pee g oups. We examine how such school changes a e associa ed wi h pee ic imiza ion in he new school among s u- den s wi h a ying ic imiza ion his o ies. Acco ding o a s udy by Ca son and colleagues (2013), pee ic imiza ion declines a e changing schools, while wo s udies (Tenhunen e al., 2024; Ve nbe g, 1990) ound ha newcome s in class ooms end o expe- ience mo e pee ic imiza ion han es ablished s uden s. Two o he s udies (Ramba an e al., 2020; Vandell e al., 2021) did no ind a s a is ically signi ican e ec o newcome s a us on ic imiza ion, bu he di ec ion o he associa ion was owa d inc easing ic imi- za ion. Howe e , none o hese s udies conside ed s uden s’ p io ic imiza ion his o y. In es iga ing he ini ial easons o s uden s’ mobili y, Tenhunen e al. (2024) ound ha ic imiza ion in he new school was mos likely among s uden s who had swi ched schools because hey we e ic imized in he p e ious school. Ne e heless, hey did no compa e cu en ic imiza ion among p e iously ic- imized s uden s who had e sus had no changed schools and he e o e ailed o unco e whe he p e iously ic imized newcom- e s expe ienced mo e o less ic imiza ion han hey would ha e aced in hei p e ious schools. Gaps in Resea ch on School Change and Vic imiza ion Despi e he g owing body o li e a u e on school mobili y and ic- imiza ion, some no able gaps emain. Fi s , he e is a sho age o esea ch examining whe he ic imiza ion expe iences end o dec ease, inc ease, o pe sis a simila le els when p e iously ic- imized s uden s a e ans e ed ( s no ) o a new school. In es iga ing whe he school mobili y p o ides elie om ic imi- za ion o leads o con inuing mise y in new en i onmen s is essen- ial o unde s anding he dynamics o bullying and he e ec i eness o school change in pu ing an end o pee ic imiza ion. Second, no s udies ha e in es iga ed how p e iously ic imized child en pe - cei e non-no ma i e school changes. One quali a i e s udy by Messiou and Jones (2015) b ough o ligh he expe iences o chil- d en who had ecen ly changed schools by in e iewing hem a e he change. The conce ns child en b ough up du ing he in e iews we e o en linked o losing and inding iends o becoming a a ge o bullying (Messiou & Jones, 2015). While hese we e hough s aised by newcome s whose ic imiza ion his o y was mos ly unknown (excep o wo child en who men ioned hei ic imiza- ion his o y), i is no known how p e iously ic imized newcome s pe cei e school change. This is o pa icula impo ance because anecdo al e idence sugges s ha ans e ing a ic imized s uden o a new school is a ela i ely common—albei much deba ed— p ac ice in handling pe sis en bullying cases. Hea ing he pe spec- i es o p e iously ic imized newcome s can p o ide aluable insigh s in o hei psychosocial adjus men , he eby con ibu ing o he de elopmen o mo e p ecise and e ec i e s a egies o ease hei in eg a ion in he new pee g oup. The Impac o School Change on P e iously Vic imized S uden s: Po en ial Bene i s and Risks We p opose ha child en who ha e expe ienced equen ic imi- za ion in p e ious schools may bene i om school change. Speci ically, hose wi h a his o y o equen ic imiza ion migh epo lowe le els o pee ic imiza ion du ing hei i s yea in a new school compa ed o s uden s wi h simila ic imiza ion his o ies who emain wi h hei p e ious classma es. Despi e he heigh ened ini ial isk associa ed wi h newcome s a us, i is pos- sible ha ic imiza ion dec eases o becomes less in ense in he new con ex . In he p e ious class oom, g adual cogni i e changes among pee s migh ha e led o a wo sening pe cep ion o he ic imized child o e ime (Olweus, 1978). While a ic- im’s epu a ion may pe sis and con inue o nega i ely in luence pe cep ions among hese same pee s (Schus e , 2001), new pee s in he new school, unbu dened by en enched biases, migh ini- ially pe cei e he ic imized child di e en ly, po en ially esul - ing in dec eased ic imiza ion. On he o he hand, we also know ha pee ic imiza ion ends o pe sis in new se ings likely due o ulne abili ies o p e iously ic imized s uden s (e.g., being insecu e and igh ened abou wha will happen in he new class- oom), which make hem an easy a ge o pe sis en bullying (Salmi alli e al., 1998). In addi ion, newcome s gene ally expe- ience highe le els o pee ic imiza ion compa ed o es ab- lished s uden s (Tenhunen e al., 2024). Thus, i is also concei able ha p e iously ic imized child en con inue o expe ience high le els o ic imiza ion a e he school change, compa able o o e en exceeding hose expe ienced by s uden s wi h simila ic- imiza ion his o ies who emain in he same school. Mo eo e , s uden mobili y can lead o b oade in e pe sonal consequences beyond ic imiza ion. Resea ch indica es ha school changes o en esul in ewe iends, smalle iendship ne wo ks, less cen al posi ions in social ne wo ks, and inc eased social isola ion (Haynie & Sou h, 2005; Sou h & Haynie, 2004; Ve nbe g e al., 2006). Thus, i is c ucial o ho oughly unde - s and he e ec s o non-p omo ional school changes on pee ic- imiza ion, pa icula ly among p e iously ic imized s uden s. Cu en S udy The exis ing li e a u e p o ides aluable insigh s in o se e al key a eas: he po en ial nega i e e ec s o being a newcome in a class oom o school (Tenhunen e al., 2024; Ve nbe g, 1990), he high equency o school changes among ic imized s uden s (Ca son e al., 2013), and he impac o he ini ial easons o mobili y on newcome s’ pee ic imiza ion (Tenhunen e al., 2024). Howe e , p e ious s udies did no conside he possibili y ha bo h p e ious ic imiza ion expe iences and changing schools a e likely o a ec subsequen ic imiza ion, and impo - an ly, school change may ha e di e en implica ions o s u- den s wi h a ying ic imiza ion his o ies. Ano he aspec la gely missing om he li e a u e is he pe spec i e o s uden s who changed schools. The p esen mixed-me hods s udy aims o ill hese gaps by add essing he ollowing ques ions: i s , we explo e how posi- i ely o nega i ely s uden s in gene al, and s uden s wi h di e - en ic imiza ion his o ies in pa icula , pe cei ed he change o schools, bo h a he ime i occu ed and a e hey s a a ending he new school. Second, we in es iga e he unique and in e ac- i e e ec s o p e ious ic imiza ion and school change on sub- sequen ic imiza ion in he new con ex . We hypo hesize ha bo h s uden s’ p e ious ic imiza ion and he change o schools a e independen ly associa ed wi h a highe isk o ic imiza ion, al hough we do no ha e a di ec ional hypo hesis ega ding he Tenhunen e al. 3 in e ac ion e ec be ween hese ac o s. Finally, we examine how newcome s wi h a cons an ic imiza ion his o y (CVH) desc ibe hei expe iences ega ding school change and iden i y he ac- o s in hei p e ious o new school ha a e ela ed o hese expe- iences. This explo a o y ques ion is add essed h ough a quali a i e con en analysis o he expe iences o p e iously ic- imized newcome s ega ding school change, including a e si e and a ac i e ac o s ela ed o hei o me and cu en schools. Ma e ials and Me hods S udy Design We u ilize mixed me hods, wi h a quan i a i e analysis comple- men ed by quali a i e con en analysis. By doing so, we can cap- u e bo h he measu able phenomenon and deepe con ex ual insigh s (C eswell & Plano Cla k, 2018). This app oach imp o es he alidi y o ou indings and enables a comp ehensi e exami- na ion o ic imized s uden s’ school change, unco e ing nuances ha may be di icul o de ec wi h one me hod alone. Pa icipan s and P ocedu es Da a conce ning ic imiza ion expe iences in p e ious and cu en school yea s we e collec ed in Ma ch 2022 in he con ex o an annual s uden su ey in Finnish schools implemen ing KiVa an i- bullying p og am. In Finland, ac i e pa en al consen is no equi ed o la ge-scale su eys ha a e in eg a ed in o he school’s egula ac i i ies. The e o e, no ins i u ional e iew boa d (IRB) app o al was needed o his s udy. Bo h pa en s and s uden s ha e been in o med ha s uden s’ anonymous esponses may be u ilized o esea ch pu poses. In 2022, ques ions ega ding s uden mobil- i y and ic imiza ion his o y we e included in he su ey o s u- den s in G ades 4–9 (ages 10–15). A o al o 58,936 s uden s ac oss 498 comp ehensi e schools pa icipa ed in he su ey. Due o missing da a ac oss all a iables, 137 cases we e omi ed, esul ing in a inal sample o 58,799 pa icipan s. Among hem, 56.91% we e in elemen a y school (G ades 4–6), 43.09% we e in middle school (G ades 7–9), and 4.59% we e newcome s o hei schools. In he quali a i e pa o he s udy, 94 newcome s wi h CVH we e su eyed wi h open ques ions. Responses ha we e ound o be ongue-in-cheek and did no answe he ques ions we e excluded, esul ing in a inal sample o 68 s uden s. Measu es G ade le el (4–9, ha is, 10–15 yea s) was epo ed by he pa - icipan s a he beginning o he su ey. Newcome s a us was measu ed wi h he ollowing ques ion: Ha e you ans e ed o his school om ano he school du ing he cu en school yea (i.e., a e las summe b eak)? (1) No. I’m in he same school as in he p e ious yea ; (2) Yes. Only I changed schools, no he o he classma es; (3) Yes, because mos (o all) o he s uden s o my age ans e ed o his school las au umn; and (4) Yes, mos (o all) o he s uden s ans e ed o a new school, bu I ans e ed o a di e en school han o he s. In his s udy, indi iduals who selec ed op ion 2 o 4 we e classi ied as newcome s. Gi en ha da a collec ion ook place du ing he sp ing semes e (in Ma ch), hese newcome s had been in hei cu en class oom o a pe iod anging om 0 o 7 mon hs. Cu en pee ic imiza ion was assessed by ou i ems co e - ing di e en o ms o pee ic imiza ion (name-calling, social exclusion, physical ic imiza ion, and umo -sp eading). The s uden s we e asked: “How o en ha e you been bullied like his du ing he las wo mon hs?” ollowed by he i ems: I was called names, mocked, o eased in a hu ul way; The o he s uden s igno ed me comple ely o excluded me om he g oup; I was hi , kicked, o pushed; and The o he s uden s sp ead mean o o end- ing s o ies abou me. The esponses we e gi en on a scale om 1 o 5: 1 = no a all; 2 = once o wice; 3 = 2–3 imes a mon h; 4 = once a week; 5 = se e al imes a week. Responses we e a e - aged ac oss i ems o c ea e a sco e ep esen ing o e all pee ic- imiza ion (α = .79). Pee ic imiza ion his o y was assessed wi h wo e ospec i e ques ions. Fi s , s uden s we e asked o hink abou p e ious school yea s, and whe he hey we e ic imized hen. Those answe ing “No” ecei ed a sco e o 0 on ic imiza ion his o y. Those who answe ed “Yes” we e gi en a ollow-up ques ion: How equen ly did bullying occu ? wi h esponse op ions 1 = Single occasions; 2 = Qui e o en; 3 = Almos cons an ly; 4 = Cons an ly. Thus, pee ic imiza ion his o y anged om 0 o 4. Though s abou he school change we e assessed wi h wo ques- ions: (1) Wha did you hink abou changing schools a he ime i happened? and (2) Wha do you hink abou i now? Responses we e gi en on a scale o 1–5. Fo he i s ques ion, he op ions we e: 1 = I eally didn’ wan o; 2 = I didn’ wan o; 3 = I didn’ ma - e o me; 4 = I hough i was qui e nice; and 5 = I eally wan ed o change schools; and o he second ques ion: 1 = I was a bad hing; 2 = I was a p e y bad hing; 3 = I was nei he a good no a bad hing; 4 = I was a p e y good hing; and 5 = I was a good hing. Bo h o he abo e ques ions we e ollowed by a simple open ques- ion: Why? The answe s o hese open ques ions cons i u ed he quali a i e da a o he s udy. Despi e he i ele an esponses being excluded (see Pa icipan s and P ocedu es), he da a eached sa u a- ion poin (i.e., pa icipan s s a ed o gi e simila esponses), indi- ca ing he sample size was su icien o con en analysis. Analy ic Plan The hough s abou he school change a he ime and a e he e en we e i s examined using analysis o a iance (ANOVA), compa ing s uden s wi h a ying le els o p e ious ic imiza- ion. The ea e , mul i a ia e analyses we e conduc ed using wo- le el andom in e cep models o accoun o he nes ed da a s uc u e (s uden s nes ed wi hin schools). We used chained equa ion mul iple impu a ion wi h 10 d aws o handle missing da a and employed maximum likelihood es ima ion o he mixed e ec s models. Al hough he in aclass co ela ion coe icien (ICC) o ic imiza ion was only 0.01 be ween schools, his app oach mi iga ed he isk o inaccu a e s anda d e o s and biased es ima es. The da ase had ew missing alues (missing- ness on s udy a iables <1%), and Li le’s MCAR es indica ed ha da a we e no missing comple ely a andom (MCAR) bu may be missing a andom (MAR), which is equi ed o mul iple impu a ions. To ge close o he assump ion, we in oduced measu es o loneliness, global ic imiza ion, and global bullying as auxilia y a iables. The auxilia y a iables we e selec ed based on hei co ela ions ( ⩾ .3) wi h missingness and/o he ou come a iable. All quan i a i e analyses we e conduc ed using S a a/MP 16.1. 4 In e na ional Jou nal o Beha io al De elopmen 00(0) To del e deepe in o he CVH newcome s’ hough s abou he school change, we employed quali a i e con en analysis by cod- ing, g ouping, ca ego iza ion, and abs ac ion (Elo & Kyngäs, 2008). We w o e headings and no es o each s uden ’s esponses du ing he open coding, using he whole sen ences as analy ic uni s. Following his, we summa ized he meanings and expe i- ences (i.e., g ouping). Fu he mo e, we sea ched o simila i ies and dispa i ies, ending up wi h b oade ca ego ies. Finally, he cons i uen s we e abs ac ed o a highe logical le el, esul ing in a classi ica ion ha add esses ou esea ch ques ions. Resul s Desc ip i es Desc ip i e s a is ics and pai wise co ela ions a e displayed in Table 1. Vic imiza ion his o y was associa ed wi h cu en ic- imiza ion ( = .44, p < .001). Newcome s a us had a posi i e, albei e y weak, co ela ion wi h ic imiza ion his o y ( = .07, p < .00) as well as wi h cu en ic imiza ion ( = .06, p < .001). The mo e a child had been ic imized in hei p e ious school, he mo e posi i ely hey had hough abou he school change a he ime i happened ( = .09, p < .00). Vic imiza ion his o y was un ela ed o how child en hough abou he school change now. As indica ed by he mean sco es, s uden s we e o e all qui e posi i e abou he school change, especially now ha hey we e al eady in he new school. School Change Among S uden s Wi h Di e en Vic imiza ion His o ies We compa ed newcome s wi h di e en ic imiza ion his o ies ega ding hei hough s conce ning he school change using one- way ANOVA. We ound ha he hough s abou he ansi ion a he ime i occu ed a ied ac oss di e en ic imiza ion his o y ca ego ies, F (4, 53.84) = 13.46, p < .001, η2 = .01. In line wi h he co ela ions epo ed abo e, hose wi h no ic imiza ion his- o y pe cei ed he school change mos nega i ely a he ime i happened. In his espec , hey di e ed signi ican ly om hose wi h an almos cons an his o y o ic imiza ion, acco ding o he Tukey’s pos hoc es (mean di e ence = 0.58, p < .001). Howe e , he di e ence in nega i e hough s be ween non- ic- imized newcome s and hose wi h a cons an his o y o ic imi- za ion did no each s a is ical signi icance (mean di e ence = 0.26, p > .05). G oups wi h a ying ic imiza ion his o ies did no di - e in hei cu en hough s ega ding he school change. To es he unique and in e ac i e e ec s o p e ious ic imi- za ion and school change on cu en ic imiza ion, we an wo- le el models (s uden s nes ed wi hin schools) (Table 2). The main e ec s model (Model 1a) showed ha newcome s a us (B = 0.05, p < .001) and ic imiza ion his o y (i.e., cons an ly ic- imized s uden s: B = 1.58, p < .001) had unique posi i e associa- ions wi h cu en ic imiza ion. In Model 2a, we es ed whe he he associa ion be ween newcome s a us and cu en ic imiza- ion was mode a ed by ic imiza ion his o y. The newcome s who had been cons an ly (B = −0.43, p < .001) o almos con- s an ly (B = −0.13, p < .01) ic imized in he pas epo ed signi i- can ly lowe pee ic imiza ion han es ablished s uden s wi h simila ic imiza ion his o ies. Newcome s wi hou ic imiza ion his o y, on he o he hand, epo ed sligh ly highe (B = 0.04, p < .05) pee ic imiza ion in new se ings compa ed o es ab- lished s uden s wi hou ic imiza ion his o y (Figu e 1). The same models we e un sepa a ely o elemen a y (Models 1b and 2b) and middle school s uden s (Models 1c and 2c) o add ess po en ial age g oup di e ences (Table 2). I u ned ou ha he school change was mo e bene icial o ele- men a y school s uden s who had been cons an ly (B = −0.46, p < .001) o almos cons an ly (B = −0.17, p < .01) ic imized in he pas (Figu e 2). In con as , he e ec was weake among middle schoole s and signi ican only among hose who had been p e iously cons an ly (B = −0.18, p < .05) ic imized. Among middle school s uden s who had been ic imized o a lesse ex en (“qui e o en”), school change inc eased he isk o ic imiza ion (B = 0.09, p < .05) (Figu e 3). Expe iences o P e iously Vic imized S uden s The quali a i e analysis p o ides a nuanced unde s anding o p e iously ic imized s uden s’ a i udes owa d changing schools, e ealing a spec um o emo ions and expe iences. This sec ion del es in o he con as ing pe spec i es o hose who we e eluc an e sus hose who we e eage o make a school change. A e si e and A ac i e Though s Abou P e ious and New School. Fi s , some s uden s we e unwilling o change schools, e en hough hey had expe ienced cons an ic imiza ion in hei p e ious school. Some o hese s uden s exp essed ea ela ed o school change, o example, S uden 1 s a ed: “[I eally didn’ wan o change schools because] I was a aid I wouldn’ make iends, bu o una ely, I was w ong.” Some imes, hey empha- sized he a ac i e aspec s o hei p e ious school while Table 1. Desc ip i e S a is ics and Co ela ion Ma ix o he S udy Va iables. n M SD 1 2 3 4 5 1. Cu en ic imiza ion ( ange, 1–5) 58,700 1.24 0.55 2. Newcome s a us 58,782 0.05 0.21 0.06*** 3. P e ious ic imiza ion ( ange, 0–4) 58,265 0.47 0.88 0.45*** 0.07*** 4. G ade Le el (g ades 4–9, 10 o 15 yea s old) 58,799 6.26 1.68 -0.01** -0.03*** 0.02*** 5. Seeing school change posi i ely when i happened ( ange, 1–5) 2,710 3.26 1.27 -0.05*** -0.01 0.09*** 0.08*** 6. Seeing school change posi i ely now ( ange, 1–5) 2,672 3.69 1.16 -0.15*** 0.06** 0.03 -0.004 0.49*** No e. M = mean, SD = s anda d de ia ion. *p < .05, **p < .01, ***p < .001. Tenhunen e al. 5 Table 2. Cu en Vic imiza ion P edic ed by Vic imiza ion His o y, Newcome S a us, and Thei In e ac ion. Pee - ic imiza ion Whole sample Elemen a y school s uden s Middle school s uden s N58,700 33,377 25,323 Model 1a Model 2a Model 1b Model 2b Model 1c Model 2c B (SE) [95% CI] B (SE) [95% CI] B (SE) [95% CI] B (SE) [95% CI] B (SE) [95% CI] B (SE) [95% CI] Newcome 0.05*** (0.01) [0.04, 0.07] 0.08*** (0.01) [0.06, 0.11] 0.04*** (0.1) [0.02, 0.07] 0.08*** (0.02) [0.05, 0.11] 0.07*** [0.04, 0.10] 0.08*** (0.02) [0.04, 0.12] G ade le el -0.01*** (0.001) [-0.01, -0.01] -0.01*** (0.001) [-0.01, -0.01] -0.02*** (0.003) [-0.03, -0.01] -0.02*** (0.003) [-0.03, -0.01] -0.02*** (0.004) [-0.02, -0.01] -0.02*** (0.004) [-0.02, -0.01] Vic imiza ion his o y 0. No ( e .) 1. Single occasions 0.20*** (0.01) [0.19, 0.21] 0.20*** (0.01) [0.19, 0.22] 0.21*** (0.01) [0.20, 0.23] 0.21*** (0.01) [0.20, 0.23] 0.19*** (0.01) [0.17, 0.21] 0.19*** (0.01) [0.17, 0.21] 2. Qui e o en 0.45*** (0.01) [0.44, 0.47] 0.45*** (0.01) [0.43, 0.46] 0.50*** (0.01) [0.48, 0.52] 0.50*** (0.01) [0.49, 0.52] 0.39*** (0.01) [0.37, 0.41] 0.38*** (0.01) [0.36, 0.41] 3. Almos cons an ly 0.76*** (0.01) [0.74, 0.78] 0.77***(0.01) [0.75, 0.80] 0.80*** (0.02) [0.76, 0.83] 0.81*** (0.02) [0.78, 0.84] 0.72*** (0.02) [0.69, 0.76] 0.73*** (0.02) [0.69, 0.77] 4. Cons an ly 1.58*** (0.02) [1.55, 1.61] 1.64*** (0.02) [1.60, 1.67] 1.29*** (0.03) [1.24, 1.34] 1.38*** (0.03) [1.32, 1.43] 1.79*** (0.02) [1.74, 1.83] 1.80*** (0.02) [1.75, 1.85] Newcome × Vic imiza ion his o y 1 × 1 -0.03 (0.03) [-0.09, 0.02] -0.008 (0.03) [-0.07, 0.06] -0.07 (0.05) [-0.15, 0.02] 1 × 2 0.01 (0.03) [-0.04, 0.08] -0.05 (0.04) [-0.13, 0.03] 0.09* (0.05) [0.00, 0.18] 1 × 3 -0.13** (0.05) [-0.23, -0.04] -0.17** (0.06) [-0.29, -0.05] -0.09 (0.07) [-0.22, 0.05] 1 × 4 -0.43*** (0.05) [-0.53, -0.33] -0.46*** (0.07) [-0.59, -0.33] -0.18* (0.08) [-0.34, -0.02] AdjR 2 ..20 .22 .19 .19 .21 .25 *p < .05, **p < .01, ***p < .001. 6 In e na ional Jou nal o Beha io al De elopmen 00(0) highligh ing he a e si e ac o s associa ed wi h he new school. These s uden s o en had es ablished iendships in he p e ious school and losing hem in he school change added complexi y o he si ua ion. One s uden desc ibed hei posi i e hough s abou hei old school by saying (S uden 2): “[I eally didn’ wan o change schools] because I had made eally good iends o e he e and I hadn’ done ha in a while” and hen con inued wi h a desc ip ion o nega i e hough s abou hei new school: “[I was a p e y bad hing o change schools because] I ha e maybe ew iends, and a on o o he people ei he dislike me o jus ha e me.” By w i ing his, he s uden emphasizes he unde s and- able disappoin men hey aced in he school change; hey had los good iends and encoun e ed challenges in pee ela ions. On he o he hand, some s uden s we e eage o make a school change. They ended o ocus on he a ac i e elemen s o he new school and he a e si e ac o s o hei p e ious ins i u ion. Fo ins ance, his s uden exp essed hei desi e o ans e by emphasizing hei nega i e expe iences a hei o me school (S uden 3): “[I eally wan ed o change schools] because I had been bullied o h ee yea s in he old school. I was al eady so i ed o de ending mysel .” Subsequen ly, hey b ough up a posi- i e aspec o he new school: “[I was a good hing o change schools] because I’m no bullied anymo e.” A common h ead ha uns h ough in hese ypes o s uden s’ answe s is he aspi a- ion o a b igh e u u e and a since e hope o inding iends. The s uden s had expe ienced equen ic imiza ion which hey managed o escape by changing schools, allowing hem o s a anew and make iends. Ha ing e sus no ha ing iends o aspi a ion o ind new iends may be he mos c i ical ac o s di e en ia ing CVH newcome s who we e ini ially willing e - sus unwilling o change schools. Ha ing a s ong social ne wo k in he p e ious school may discou age ic imized s uden s om pu suing a school change, while eelings o isola ion and a lack o social suppo se e as mo i a ions o conside ing a ans e o a new school, as he e is no hing wo hy o lose. School Change Impac s All—Bu No All he Same. Expe i- ences in s uden s’ iendships we e di ided be ween hose who we e sa is ied, ha ing ound new iends, such as S uden 4, who w o e: “[I was a good hing o change schools because] I like his school much mo e, and I ha e iends” and hose who we e dis- sa is ied due o hei inabili y o es ablish new iendships, o example, S uden 5: “[I was nei he good no bad hing o change schools] I ha en’ eally ound sui able iends o me.” The expe iences wi h ic imiza ion a ied, wi h some s uden s epo ing a educ ion o comple e inish o bullying ic imiza ion. Fo example, S uden 6 i s explained why hey wan ed o change schools in he i s place: “[I eally wan ed o change schools because] I was hi , hu , and called names, and because o ha , I couldn’ go o school. I was sca ed he e, and I became dep essed.” Subsequen ly, he same s uden w o e abou a comple e u na ound esul ing om changing schools: “[I was a good hing o change schools because] I made iends, I’m no bullied o hu , and I’m glad.” Also S uden 7 desc ibed a posi i e u na ound by w i ing: “[I was a good hing o change schools because] I ha en’ been Figu e 1. The Mode a ing E ec o Vic imiza ion His o y on he Associa ion Be ween Newcome S a us and Cu en Vic imiza ion (Scale 1–5) Among Fou h o Nin h G ade s (n = 58,700). Figu e 2. The Mode a ing E ec o Vic imiza ion His o y on he Associa ion Be ween Newcome S a us and Cu en Vic imiza ion (Scale 1–5) Among Fou h o Six h G ade s (i.e., 10–12 Yea s Old) (n = 33,377). Tenhunen e al. 7 bullied he e, and I’ e been able o a end school no mally.” While hese s uden s expe ienced less ic imiza ion in he new school, he e we e s ill s uden s who had no been able o escape hei pligh by changing o ano he school. Fo example, S uden 8 w o e abou hei edious expe ience: “[I wan ed o change schools] because I hough I didn’ ha e o endu e bullying anymo e, bu i con inued in he new school.” Some CVH newcome s exp essed g a i ude o be e school acili ies, such as he school’s close dis ance om home o be e lea ning oppo uni ies. The la e was exp essed by S uden 9: “[I was a good hing o change schools because] he e a e a lo o new iends he e and an exp essi e a s class.” The e we e no nega i e esponses di ec ly linked o he school acili ies. Feelings o belongingness in he new school we e occasion- ally enhanced by sa is ac ion wi h he class oom (o school) cli- ma e o hinde ed by gene al challenges wi h i ing in. Fo example, one s uden desc ibed hei school and class posi i ely, al hough hey i s hesi a ed wi h he decision, S uden 10: “[I eally didn’ wan o change schools because] he new school was sca y.”A e he school change, he same s uden changed hei mind: “[I was a p e y good hing o change schools because] s uden s a e ca ed abou mo e, and eache s a e also iendlie .” The s a emen e lec s eelings o belongingness and success ul in eg a ion. In con as , hose s uden s who may ha e had di i- cul ies i ing in esponded wi h opposing iews o example, S uden 11: “[I was a bad hing o change schools because] I should ha e gone o [X school]” and S uden 12: “[I was a bad hing o change schools because] I ha e his school.” Since hese esponses did no cla i y he eason behind he exp essed eelings o aliena ion, we also checked hese s uden s’ cu en ic imiza- ion sco es. I e ealed ha o e all (wi h one excep ion), he CVH newcome s who epo ed challenges in i ing in also sco ed high on cu en pee ic imiza ion, al hough hey did no men ion i in hei open answe s. Based on he con en analysis, we can conclude ha CVH newcome s’ hough s abou he p e ious and new schools we e posi ioned on a spec um anging om a e si eness o a ac ion as ollows: he a e si e ac o s o he p e ious school, he a ac- i e ac o s o he p e ious school, he a e si e ac o s o he new school, and he a ac i e ac o s o he new school. A ac i e ac o s owa d he new school included he hope o he be e u u e, including new iends, esh s a wi hou pee ic imiza- ion and willingness o join in o weigh ed cu iculum eaching. A e si e hough s abou he new school some imes s em om he ea o he school change. S uden s men ioned ea s abou no inding a iend, he pe sis ence o ic imiza ion, no liking he school as well as mo ing o a new place. Simila ly, a ac i e ac- o s ela ed o he old school emphasized he amilia i y o he old school as well as he people and he memo ies made he e. A e si e ac o s ela ed o he old school we e mos ly ela ed o pee ic imiza ion expe iences, loneliness, and men al heal h p oblems, which we e linked o ad e se expe iences wi h pee s. Newcome s desc ibed hei expe iences ega ding he school change, de ailing i s impac on iendships, ic imiza ion expe i- ences, school acili ies, and belongingness. Some newcome s missed hei old iends while some had ound new iends. A ew newcome s epo ed inding iends which hey had no had be o e. Newcome s’ expe iences wi h pee ic imiza ion also a ied: some newcome s epo ed ha ic imiza ion con inued in he new school, while o he s expe ienced a dec ease o comple e inishing o ic imiza ion. School acili ies we e men ioned a o ably, such ha he newcome s we e happy wi h he new school’s loca ion and educa ional oppo uni ies. Belongingness was exp essed by being g a e ul o he new classma es and nice eache s. Nega i e expe i- ences o no belonging we e di icul o de ine, bu some s uden s we e gene ally unhappy wi h he new school which may indica e hei eeling o aliena ion, ha is, no belonging. Discussion This s udy i s examined newcome s’ pe spec i es on school change, bo h a he ime i happened and a e he change, and compa ed hese iews ac oss i e di e en ic imiza ion his o y ca ego ies (i.e., no ic imized, single occasions o ic imiza ion, qui e o en ic imized, almos cons an ly ic imized, and con- s an ly ic imized in he pas ). Second, we es ed whe he he ela ionship be ween newcome s a us and cu en ic imiza ion was mode a ed by ic imiza ion his o y. These quan i a i e anal- yses we e complemen ed by a quali a i e analysis explo ing he expe iences o newcome s wi h a his o y o cons an ic imiza- ion (CVH). O e all, newcome s had ela i ely posi i e hough s abou hei school change. S uden s wi hou ic imiza ion his o y an ic- ipa ed he school change mo e nega i ely (p obably ha ing mo e o lose), while hose who had been ic imized in he pas ended o iew he change mo e posi i ely a he ime i occu ed. The quali a i e analysis e ealed he a ying expe iences o CVH Figu e 3. The Mode a ing E ec o Vic imiza ion His o y on he Associa ion Be ween Newcome S a us and Cu en Vic imiza ion (Scale 1–5) Among Se en h o Nin h G ade s (i.e., 13–15 Yea s Old) (n = 25,323). 8 In e na ional Jou nal o Beha io al De elopmen 00(0) newcome s. Some s uden s had pu sued school change and iden- i ied a e si e hough s abou hei o me school (i.e., pee ic- imiza ion and loneliness) and a ac i e hough s (i.e., hope o he be e u u e, new iends, inishing o pee ic imiza ion) abou he new school. O he s, howe e , epo ed he opposi e. Since all o hese s uden s had a his o y o cons an ic imiza ion, he e mus be o he ac o s explaining he he e ogenei y in hei esponses. Ins ead, he con en analysis iden i ied h ee highe - le el classes aside om pee ic imiza ion: iendships, school acili ies, and belongingness. Some o he CVH s uden s had iends in hei o me school, while he o he s desc ibed hei pas as iendless. Thus, ha ing iends adds complexi y o he school change and makes i less desi able, e en hough a s uden is cons an ly ic imized. Feelings o belongingness o aliena ion, home- o-school dis ance, and new lea ning oppo uni ies we e also men ioned among hough s ega ding he school change. The mul ile el analysis showed ha CVH newcome s may bene i om school change when compa ed o equally ic imized s uden s con inuing wi hin he same schools. This in e ac i e e ec be ween school change and p e ious ic imiza ion was signi ican among he whole sample, bu when he analysis was done sepa a ely o elemen a y and middle schools, i was ound o be s onge among elemen a y school s uden s. In middle schools, he bene i o hose who said hey had been cons an ly ic imized in hei p e ious school was only ma ginal. Thus, elemen a y school s uden s wi h ic imiza ion his o y can bene i om changing schools, whe eas such a bene i is negligible among middle school s uden s. The lowe likelihood o ic imiza ion among newcome s who epo ed cons an (in bo h age g oups) o almos cons an (in elemen a y school) ic imiza ion his o y, in compa ison o es ab- lished s uden s wi h simila ic imiza ion his o ies, may be explained by hei escape om he con ex whe e classma es ha e de eloped a sha ed nega i e pe cep ion o hem (Olweus, 1978; Schus e , 2001) and whe e hey ha e a epu a ion as ic- ims. Such a epu a ion can esul in ewe iendship oppo uni- ies (Boul on, 2013), u he exace ba ing ic imiza ion. This ic im epu a ion is no as s ong among hose less equen ly ic imized, so changing schools p ima ily bene i s hose who a e mos equen ly ic imized. Howe e , changing schools dec eases ic imiza ion only ma - ginally among middle schoole s. The o e all e ec among CVH newcome s was −0.43, wi h a spli e ec o −0.46 o elemen a y school s uden s and −0.18 (41%) o middle school s uden s. This di e ence is likely due o he main e ec s o ic imiza ion his o y and newcome s a us on cu en ic imiza ion. These p edic ions a e s onge among middle schoole s compa ed o elemen a y schoole s, explaining why he in e ac ion e ec is lowe . Long- las ing ic imiza ion may esul in beha io s o cha ac e is ics ha expose he s uden o ic imiza ion e en in a new con ex (Salmi alli e al., 1998). Such beha io s could include ac ing agains g oup no ms o s uggling o i in wi h he new pee g oup (e.g., Hui sing e al., 2012; W igh e al., 1986). Due o hei de elopmen al s age, middle schoole s may be mo e adep a ecognizing non-no ma i e beha io s due o he s onge in luence o pee s and he g ea e emphasis on pee g oup compa ibili y a ha age (Lau sen & Veens a, 2021), which could explain why hey mo e eadily a ge CVH newcome s compa ed o elemen a y schoole s. The quan i a i e and quali a i e esul s p o ide a complemen- a y pic u e o school change. In he quan i a i e analysis, we ound ha CVH newcome s epo ed less pee ic imiza ion in new schools compa ed o s uden s wi h simila ic imiza ion his- o ies who s ayed in he same school. In line wi h his, none o he CVH newcome s epo ed an inc ease in ic imiza ion expe i- ences in hei new school in he open-ended answe s analyzed quali a i ely. They did ell abou pe sis ing, dec easing, and com- ple e ending o ic imiza ion, as well as changes in iendships, school acili ies, and belongingness. Many CVH newcome s old abou ha ing ound iends in he new school, which is also known o p o ec om pee ic imiza ion in a new con ex (Tenhunen e al., 2024). Howe e , in eg a ing esul s based on bo h me hods challenges a simplis ic in e p e a ion o he da a. The quan i a i e analysis sugges s s uden mobili y is a complex e en in which CVH newcome s may bene i , bu he less o en ic imized o non- ic imized newcome s may no expe ience he same bene i s— hey could e en encoun e nega i e conse- quences. In addi ion, he quali a i e esponses un eil e en a mo e nuanced eali y. I becomes e iden ha , despi e he bene- i s obse ed among CVH newcome s, some s uden s do no expe ience he an icipa ed ad an ages o school change. Thus, based on he esul s, we can conclude ha changing schools does no seem o exace ba e pee ic imiza ion among CVH s uden s; howe e , he possibili y o i s pe sis ence a he same le el mus be acknowledged. Also, some hing o conside is he po en ial consequences i CVH s uden s a e exposed o ic imiza ion again in he new class oom—would i be e en wo se o one’s men al heal h i hey a e no accep ed in he new g oup ei he ? Ou indings pa ly align wi h hose o Ca son e al. (2013), who obse ed a decline in pee ic imiza ion ollowing a school change. Howe e , ou esul s also e eal ha newcome s who we e no equen ly ic imized in hei p e ious school do no always expe ience a educ ion in ic imiza ion a e he school change. This con adic s he no ion ha a school change is uni- e sally bene icial and sugges s ha he impac may depend on he indi idual’s ic imiza ion his o y—a a iable no conside ed by Ca son e al. Mo eo e , ou s udy adds dep h o he indings o Ve nbe g (1990) and Tenhunen e al. (2024) by demons a ing ha while some newcome s ace inc eased ic imiza ion, o he s, pa icula ly hose equen ly ic imized in he pas , may bene i om school change. Ou esea ch also expands on he indings o Sou h and Haynie (2004) and Haynie and Sou h (2005), who highligh ed he b oade social impac s o school change, such as smalle iendship ne wo ks and inc eased isola ion. We obse ed ha while some CVH newcome s ound new iends, o he s con in- ued o expe ience social challenges, indica ing ha he impac o school change on social ne wo ks is complex and may be in luenced by indi idual his o ies o ic imiza ion. We also ound ha he s uden ’s hough s abou he school change we e simila o hose desc ibed in a s udy by Messiou and Jones (2015). Fo example, conce ns ela ed o inding iends, being ic imized, and i ing in we e highly emphasized in bo h quali- a i e samples. Fu u e esea ch should con inue o explo e he nuanced expe- iences o CVH newcome s, wi h a ocus on long- e m ou comes beyond he ini ial school change pe iod. In addi ion, in es iga - ing he ole o school en i onmen , eache suppo , and pee ela ionships in shaping he expe iences o mobile s uden s would p o ide a mo e comp ehensi e unde s anding o he ac- o s con ibu ing o success ul school change. Resea ch should