Social Inclusion
2025 •Volume 13 •A icle 10122
h ps://doi.o g/10.17645/si.10122
ARTICLE Open Access Jou nal
Longi udinal Associa ions Be ween Pe cei ed Inclusi i y No ms
and Opinion Pola iza ion in Adolescence
Mao Shani 1,2 , Ma jo ie Be ns 1, Lucy Be gen 1, S e anie Rich e s 1, K is ina K äme 1,
Sophie de Lede 1, and Maa en an Zalk 1
1Ins i u e o Psychology, Osnab ück Uni e si y, Ge many
2Facul y o Social Sciences, A iel Uni e si y, Is ael
Co espondence: Mao Shani (mao .shani@uni‐osnab ueck.de)
Submi ed: 14 Feb ua y 2025 Accep ed: 9 May 2025 Published: 9 June 2025
Issue: This a icle is pa o he issue “The Impac o Social No ms on Cohesion and (De)Pola iza ion” edi ed
by Mi anda Lubbe s (Au onomous Uni e si y o Ba celona), Ma cin Bukowski (Jagiellonian Uni e si y), Oli e
Ch is (Fe nUni e si ä in Hagen), E a Jaspe s (Uni e si y o U ech ), and Maa en an Zalk (Uni e si y o
Osnab ück), ully open access a h ps://doi.o g/10.17645/si.i438
Abs ac
A ec i e pola iza ion, cha ac e ized by emo ional hos ili y and beha io al a oidance owa d ideological
opponen s beyond me e policy disag eemen s, can pose a signi ican h ea o social cohesion. Howe e ,
his phenomenon emains ela i ely unexplo ed in adolescence. This longi udinal s udy in es iga es whe he
pe cei ed inclusi i y no ms—emphasizing equali y‐based espec , open and cons uc i e dialogue, and
communal uni y— educe opinion‐based a ec i e pola iza ion among adolescen s. Using a sample o
839 s uden s om wo demog aphically dis inc Ge man seconda y schools (g ades 7–11), we de eloped
and alida ed measu es o pola iza ion ailo ed o adolescen s, cap u ing dialogue o ien a ion and social
dis ance owa d ideological ou g oups. Resul s e ealed s able ideological subg oup di e ences in no ms
and a i udes, wi h conse a i e s uden s exhibi ing lowe desc ip i e and p esc ip i e inclusi i y no ms and
highe a ec i e pola iza ion compa ed o hei libe al pee s. Howe e , a signi ican p opo ion o
adolescen s demons a ed luid poli ical o ien a ions o e ime, highligh ing he malleabili y o ea ly poli ical
iden i ies. C oss‐lagged analyses showed no e idence ha inclusi i y no ms di ec ly educe a ec i e
pola iza ion, al hough ea ly dialogue o ien a ion signi ican ly p edic ed g ea e social openness in di e se
se ings. Ou indings ad ance he unde s anding o adolescen poli ical iden i y de elopmen by
demons a ing ha while ideological o ien a ions emain uns able du ing his pe iod, g oup‐based
di e ences in no m pe cep ions and pola iza ion endencies a e al eady e iden . We emphasize how his
de elopmen al luidi y p esen s bo h oppo uni ies and challenges o in e en ions, sugges ing ha
e ec i e depola iza ion s a egies mus accoun o he dis inc cha ac e is ics o adolescen poli ical
socializa ion and he a ying in luence o school‐based no ma i e con ex s.
© 2025 by he au ho (s), licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License (CC BY). 1
Keywo ds
adolescen s; a ec i e pola iza ion; dialogue; inclusi i y no ms; poli ical in ole ance; poli ical pola iza ion;
school no ms; social dis ance; social no ms
1. In oduc ion
A ec i e pola iza ion, cha ac e ized by heigh ened nega i e emo ions and beha io al a oidance owa ds
ideological opponen s beyond policy disag eemen s (Bakke & Lelkes, 2024; Iyenga e al., 2019), is linked by
subs an ial esea ch o nega i e democ a ic consequences, including poli ical g idlock, e oded us , and he
sp ead o mis/disin o ma ion (Iyenga e al., 2019; Shah, 2025; Tang e al., 2022; Wagne , 2024). While
esea ch has ex ensi ely examined hese de imen al ou comes and he de elopmen o such an agonis ic
o ien a ions in adul s, unde s anding hei ini ial eme gence du ing adolescence has ecei ed su p isingly
limi ed a en ion (Boonen, 2015; Hooghe & Wilken eld, 2008; Hu chins, 2024; Lay e al., 2023; Rekke e al.,
2015; Tyle & Iyenga , 2023; Wang e al., 2017). Adolescence is a c i ical pe iod in poli ical socializa ion,
ma ked by iden i y o ma ion and heigh ened social sensi i i y (Dahl e al., 2018; Van Hoo n e al., 2017).
Du ing his s age, young indi iduals o en de elop poli ically cha ged iden i ies and pa isan‐ inged
hos ili ies, e en lacking ully a icula ed pa isan a ilia ions (Tyle & Iyenga , 2023). Recen e idence
indica es a ec i e pola iza ion can mani es as ea ly as age 11, making adolescence an essen ial window o
unde s anding poli ical animosi ies’ oo s (Hu chins, 2024).
Ye , signi ican me hodological and heo e ical gaps cons ain ou unde s anding o adolescen a ec i e
pola iza ion, which his longi udinal s udy aims o add ess. Fi s , adi ional adul ‐ ocused pola iza ion
measu es a e o en de elopmen ally inapp op ia e o adolescen s, whose poli ical iden i ies equen ly
cen e on speci ic con en ious social issues a he han o mal pa isanship (Lay e al., 2023). We, he e o e,
de eloped and alida ed no el opinion‐based pola iza ion measu es cap u ing how young people’s
issue‐speci ic s ances mani es in social dis ancing and diminished dialogue willingness wi h ideological
opponen s. Second, o add ess he dea h o de elopmen al s udies, we examined whe he opinion‐based
g oups exhibi ed di e en ial pa e ns o e ime in openness o dialogue and social in e ac ion, key
beha io al indica o s o a ec i e pola iza ion (Kekkonen e al., 2022). Thi d, while esea ch has explo ed
how pe sonal belie s d i e in e g oup hos ili y, ewe s udies ha e sys ema ically in es iga ed he ole o
social no ms, pa icula ly wi hin schools, in exace ba ing o mi iga ing adolescen pola iza ion (Bicchie i
e al., 2023). Schools a e i al socializa ion en i onmen s whe e adolescen s in e nalize no ms abou
in e g oup beha io s ia pee in e ac ions and ins i u ional messaging (González, 2024; Tanka d & Paluck,
2016). Thus, we examined he po en ial o school‐based inclusi i y no ms, pe cep ions ha pee s endo se
and enac equali y‐based espec , open dialogue, and communal uni y, o a enua e a ec i e pola iza ion.
1.1. Pola iza ion and I s A ec i e Mani es a ion
In ecen yea s, pola iza ion has eme ged as a undamen al challenge ha is inc easingly shaping poli ical
landscapes ac oss he globe (Kingze e e al., 2021; McClosky, 1964; McCoy & Some , 2019). While
adi ional pola iza ion esea ch ocuses on ideological dis ance, a ec i e pola iza ion cen e s on he
emo ional dimension o in e g oup ela ions in he poli ical sphe e (Bakke & Lelkes, 2024; Iyenga e al.,
2019; Shah, 2025; Tang e al., 2022; Wagne , 2024). I in ol es nega i e eelings (dislike, dis us , ange ) and
Social Inclusion •2025 •Volume 13 •A icle 10122 2
beha io al a oidance di ec ed owa ds membe s o opposing poli ical o opinion‐based g oups, independen
o speci ic policy disag eemen s. This “us s. hem” men ali y aligns wi h social iden i y heo y, whe e g oup
membe ship os e s ing oup loyal y and ou g oup de oga ion (Bakke & Lelkes, 2024; Iyenga e al., 2012;
Schedle , 2023).
Pe aining o democ a ic no ms, a p ominen hypo hesis sugges s ha pa isans who ha bo mo e nega i e
eelings owa d opponen s will be mo e inclined o o e look ansg essions o democ a ic no ms by
ep esen a i es o hei camp (G aham & S olik, 2020; S olik, 2019). As he endency o dismiss o
delegi imize opposing iewpoin s inc eases, he capaci y o inclusi e go e nance and easoned comp omise
de e io a es (Le endusky, 2013). Consis en ly, a obus co ela ion be ween a ec i e pola iza ion and
democ a ic backsliding was ound ac oss 50 coun ies (O han, 2022). On he o he hand, some esea ch
sugges s ha a ec i e pola iza ion migh ha e po en ial posi i e e ec s, such as inc easing o e u nou by
heigh ening he pe cei ed s akes o elec ions (Fe ei a Da Sil a & Ga zia, 2024; Ha e eld & Wagne , 2023).
Ano he s udy demons a ed ha ac oss Ge many, he UK, and he US, mode a e le els o a ec i e
pola iza ion s eng hen democ a ic a i udes by inc easing poli ical engagemen and ep esen a ion, while
bo h e y low and ex emely high le els o pola iza ion a e associa ed wi h weake democ a ic suppo
(Janssen & Tu kenbu g, 2024). S udies also show ha ci izens may o en p io i ize mo e ele an
conside a ions o e aw pa isan sen imen when making poli ical judgmen s (B oockman e al., 2023), and
ha “pa isan animosi y is no a uni ying concep unde pinning he psychology o pola iza ion and
democ acy” (Voelkel e al., 2024, p. 9).
Howe e , a ec i e pola iza ion may sys ema ically e ode ins i u ional us (Iyenga & Wes wood, 2015;
Reiljan & Ryan, 2021), impede cons uc i e discou se, unde mine socie al cohesion by diminishing
meaning ul c oss‐g oup con ac , and po en ially ca alyze socie al di isions ha culmina e in disc imina o y
beha io s o in e g oup con lic (Be n zen e al., 2024; O ian Ha el e al., 2023). These g oup dynamics align
wi h social iden i y heo y’s co e p oposi ions, which posi ha g oup a achmen s can o e ide a ional
policy delibe a ion (G eene, 2004; Taj el & Tu ne , 2004). As iden i ica ion wi h pa icula iewpoin s
solidi ies, opposing g oups ansi ion om being pe cei ed as ellow ci izens o being iewed as an agonis ic
ou g oups ha h ea en undamen al alues (S ephan & S ephan, 2016), which also may in il a e e e yday
social decisions, om employmen p e e ences o in e pe sonal ela ionships (Gi & Gi , 2015; Hube &
Malho a, 2017). Unde s anding how hese a ec i e p ocesses become embedded in g oup iden i ies, and
iden i ying s a egies o mi iga e esul an hos ili ies, is c ucial o p ese ing democ a ic discou se and
ins i u ional legi imacy.
1.2. A ec i e Pola iza ion in Adolescence
Recen esea ch challenges adi ional assump ions ha poli ical enmi y eme ges only in la e adolescence
(G eens ein, 1960; Hess & To ney‐Pu a, 2017), e ealing ha pa isan‐ inged hos ili y can eme ge as ea ly as
age 11 (Kinde & Sea s, 1985; Tyle & Iyenga , 2023). While a ec i e pola iza ion has been ex ensi ely s udied
in adul s (e.g., Bakke & Lelkes, 2024; Iyenga e al., 2019; Kinde & Sea s, 1985; Mason, 2015), i s ea ly
mani es a ions emain compa a i ely unde s udied (Boonen, 2015; Hooghe & Wilken eld, 2008; Hu chins,
2024; Lay e al., 2023; Rekke e al., 2015; Tyle & Iyenga , 2023; Wang e al., 2017). Ye , hese de elopmen al
yea s ep esen a c i ical window: child en and adolescen s o en p og ess om gene al poli ical awa eness o
emo ionally cha ged ing oup–ou g oup judgmen s (Hu chins, 2024; Lay e al., 2023). Fa om being poli ically
Social Inclusion •2025 •Volume 13 •A icle 10122 3
neu al, many you hs eadily in e nalize an agonism owa d opposing ideological g oups, e lec ing pola izing
he o ic encoun e ed in hei home en i onmen s and digi al spaces (Iyenga e al., 2019; Shah, 2025; Van De h
e al., 2011).
Adolescence ep esen s a pe iod ma ked by heigh ened ecep i i y o social cues and iden i y‐ ela ed no ms
(Dahl e al., 2018; Van Hoo n e al., 2017), wi h ea ly‐ o med opinions and g oup membe ships demons a ing
ema kable s abili y ac oss he li espan (Hooghe & Wilken eld, 2008). Th ough in e ac ions wi h pa en s, pee s,
eache s, and social media, adolescen s in e nalize which iewpoin s ga ne app o al and which spa k con lic
(Boonen, 2019; Uzun & Lüküslü, 2023).
The consequences o ea ly pola iza ion can ex end o poli ical beha io : When you hs en e adul hood wi h
es ablished pola ized a i udes, hei capaci y o comp omise and c oss‐ideological engagemen may be
comp omised (Kingze e e al., 2021; Mason, 2018b). Resea ch indica es ha adolescen s’ nega i e
o ien a ions can ma ch adul s’ pa isan animosi y in in ensi y (Boonen, 2019; La ineu e al., 2024; Raabe &
Beelmann, 2011), os e ing pa e ns o ou g oup dis us ha anscend poli ical domains. The e o e,
cul i a ing dialogue ac oss opinion‐based g oup bounda ies and de eloping c i ical ci ic skills du ing
adolescence becomes essen ial o p e en ing en enched pola iza ion in adul hood (Rei en‐Taga &
Cimpian, 2022).
1.3. Social No ms and Thei Impac on A ec i e Pola iza ion
A second c ucial gap in he li e a u e conce ns he ela i ely limi ed explo a ion o how pe cei ed social
no ms in luence a ec i e pola iza ion. While esea ch has ex ensi ely examined how pe sonal belie s d i e
nega i e pa isanship, ewe s udies ha e in es iga ed how collec i e ules and s anda ds, bo h explici and
implici , can ei he ein o ce o educe in e g oup hos ili y (Bicchie i e al., 2023; G. D. A. B own e al., 2022;
C andall e al., 2002; Feh e al., 2002; Paluck, 2009; Tanka d & Paluck, 2016). Social no ms unc ion as
in o mal egula o y mechanisms h ough wo in e ela ed o ms: p esc ip i e (injunc i e) no ms, which
cap u e g oup‐sanc ioned app op ia e beha io , and desc ip i e no ms, which e lec pe cep ions o how
g oup membe s ac ually beha e (Bicchie i e al., 2023; Cialdini e al., 1990; Mille & P en ice, 2016; Smi h
e al., 2012). Wi hin he school en i onmen , no ma i e in luences on adolescen beha io become especially
salien because hey can inad e en ly unc ion as echo chambe s ha in ensi y us‐ hem hinking (González,
2024; Nipedal e al., 2010; Vá adi e al., 2021), ye hey also hold he po en ial o os e espec ul pee
no ms ha e ame disag eemen as cons uc i e dialogue (González, 2024; T opp e al., 2016). Impo an ly,
pe cei ed no ms, e en when based on mispe cep ions o pee s’ a i udes, can shape you h beha io owa d
ei he exclusion o ole ance (Vá adi e al., 2021), ega dless o ac ual g oup no ms, cap u ing he s anda ds
ha indi iduals ecognize and in e nalize (Tanka d & Paluck, 2016).
Ul ima ely, pe cei ed no ms p esen bo h challenges and oppo uni ies o add essing pola iza ion
(González, 2024; Iyenga & Wes wood, 2015; Mullinix & Ly hgoe, 2023). While pola ized ing oup no ms can
accele a e in e g oup hos ili y (Mo yl e al., 2014; Whi e al., 2021), ecalib a ing g oup expec a ions may
os e mo e open‐minded a i udes, pa icula ly du ing adolescence (C andall e al., 2002). Educa ional
ins i u ions can es ablish ins i u ional messages o ai ness and espec , o e ing al e na i e amewo ks o
in e g oup ela ions (B acegi dle e al., 2022). In his con ex , pe cei ed inclusi i y no ms (Schä e e al.,
2024) eme ge as a signi ican o m o p esc ip i e no m—when adolescen s pe cei e pee suppo o equal
Social Inclusion •2025 •Volume 13 •A icle 10122 4
ea men , dialogue, and sha ed communi y membe ship, hese no ms may bu e agains pola iza ion by
acili a ing open discou se and educing social dis ance om ou g oup membe s. The ollowing sec ion
examines how hese inclusi i y no ms in e ac wi h you h socializa ion p ocesses o po en ially mi iga e
pola izing in luences in educa ional se ings.
1.4. Pe cei ed Inclusi i y No ms o Coun e A ec i e Pola iza ion
Inclusi i y no ms (Schä e e al., 2024) comp ise h ee in e ela ed componen s: (a) equali y‐based espec
no ms, (b) communal uni y no ms, and (c) cons uc i e dialogue no ms. Equali y‐based espec no ms
es ablish collec i e s anda ds ha ecognize he inhe en digni y o all g oup membe s, ega dless o hei
ideological o backg ound di e ences (Renge & Reese, 2017; Simon & Schae e , 2016). Emphasizing such
equal s a us enhances in e g oup ela ions and educes ou g oup hos ili y (Simon e al., 2019). Communal
uni y no ms, g ounded in ci ic a he han e hnic concep ions o belonging, p omo e he unde s anding ha
g oups can main ain dis inc iden i ies while pa icipa ing in a sha ed public sphe e (Ve kuy en e al., 2020).
Las ly, cons uc i e dialogue no ms os e espec ul pe spec i e‐ aking owa d ou g oup membe s, a
p ocess ha educes s e eo yping and in e g oup anxie y (Glas o d & Do idio, 2011; Gu in e al., 2013).
While pas esea ch has examined equali y‐based espec , dialogue, and uni y as indi idual belie s, ou ocus
lies on hei po en ial when pe cei ed as social no ms. These no ms ep esen a se o in e ela ed p inciples
o managing di e ences in di e se socie ies, p inciples ha can heo e ically apply beyond cul u al o e hnic
di e si y o ideological di e ences (Ve kuy en e al., 2020). Toge he , hese no ms anscend me e passi e
accep ance by p omo ing ac i e espec o ideological di e si y (Schä e e al., 2024). Such collec i e
s anda ds become pa icula ly i al in con empo a y con ex s whe e poli ical and opinion‐based an ipa hy
o en emains unchallenged (Iyenga & Wes wood, 2015; Mullinix & Ly hgoe, 2023). These no ms align wi h
b oade p inciples o in e g oup ole ance (Gibson, 1992; Vel huis e al., 2021; Ve kuy en, 2022; Ve kuy en
e al., 2019), os e ing no jus accep ance bu ac i e engagemen in sha ed social spaces wi h hose holding
di e gen belie s. This amewo k encou ages adolescen s o iew ideological di e ences as componen s o
a di e se social ab ic a he han ze o‐sum con lic s, po en ially a enua ing he emo ional ounda ions o
a ec i e pola iza ion (Le endusky & S ecula, 2023).
Ini ial esea ch p o ides encou aging, albei mixed, e idence o hese no ms’ e ec i eness. A ecen
mul ina ional in es iga ion e ealed ha pe cei ed inclusi i y no ms co ela e wi h enhanced ole ance and
coope a ion while educing a oidance o ideological opponen s, e en among indi iduals who s ongly
disag ee wi h opposing iews (Schä e e al., 2024). Addi ionally, pee ‐led equali y‐based espec p og ams
in educa ional se ings ha e demons a ed po en ial o os e posi i e ou g oup a i udes, hough wi h
ime‐bounded e ec s (Shani e al., 2023). Howe e , he ela ionship be ween inclusi i y no ms and
pola iza ion du ing adolescence has no been s udied. Ou longi udinal app oach add esses his gap by
examining po en ial bidi ec ional e ec s: whe he inclusi i y no ms p edic subsequen changes in a ec i e
pola iza ion and whe he a ec i e pola iza ion shapes he de elopmen o inclusi i y no ms.
1.5. The Cu en S udy
This a icle examines how pe cei ed inclusi i y no ms in educa ional se ings migh a enua e opinion‐based
pola iza ion among adolescen s. Focusing on ea ly— o mid‐adolescence, a o ma i e pe iod o opinion
de elopmen , we add ess “opinion‐based g oup” pola iza ion a he han o mal pa y iden i ica ion (Lay
e al., 2023). While pa isan con lic s a e isible mani es a ions, a ec i e pola iza ion also a ises wi hin
Social Inclusion •2025 •Volume 13 •A icle 10122 5

opinion‐based g oups a ound con en ious issues like clima e change o immig a ion (Bliuc e al., 2007;
Mason, 2015, 2018a; McGa y e al., 2009). This app oach aligns wi h he “policy‐o e ‐pa y” hypo hesis,
sugges ing animus p ima ily e lec s policy disag eemen s (Webs e & Ab amowi z, 2017). Though he
compe ing “pa y‐o e ‐policy” pe spec i e (pola iza ion om social iden i y a achmen s o pa ies; Mason,
2018b) has gained ac ion, an issue‐ ocus is pa icula ly app op ia e o adolescen s s ill de eloping pa isan
iden i ies and whose a i udes o en cen e on speci ic issues. Resea ch suppo s his: E en unb anded
policy p e e ences can diminish pa isanship’s e ec on in e pe sonal a ec (Dias & Lelkes, 2022),
sugges ing policy posi ions may shape in e pe sonal e alua ions mo e han pa isan iden i ies, especially o
adolescen s. In hese con ex s, indi iduals de elop obus social iden i ies a ound sha ed s ances, os e ing
ing oup loyal y and hos ili y owa d hose wi h di e en iews (Ha e eld, 2021; Hobol e al., 2021). E en
wi hou o mal pa isan iden i ica ion, s ongly held opinions can os e an us‐ hem men ali y, ein o cing
socie al di isions (Bakke & Lelkes, 2024).
Exis ing schola ship has no sys ema ically examined how pe cei ed inclusi i y no ms migh educe a ec i e
pola iza ion in educa ional con ex s, pa icula ly o you h o ming opinion‐based a he han s ic ly pa isan
iden i ies. While some in e en ions (e.g., Shani e al., 2023) show sho ‐ e m ole ance gains, he p ocesses
o no m in e naliza ion and whe he school demog aphic composi ion mode a es such e ec s emain unclea .
Mo eo e , con lic ing indings on po en ial ideological asymme ies (Kekkonen e al., 2022; Kluge e al., 2024)
highligh he need o in es iga e whe he libe al‐ o conse a i e‐leaning adolescen s di e in ecep i i y o
inclusi i y no ms.
To add ess hese gaps, we conduc ed a wo‐wa e longi udinal su ey o s uden s (g ades 7–11) in wo
comp ehensi e seconda y schools in Lowe Saxony, Ge many: one cul u ally di e se, one p edominan ly
homogeneous. Embedded wi hin he Toge he o Tole ance p ojec (Shani e al., 2023), we analyzed da a
om hese con ol schools be ween Time 1 (Janua y‐Feb ua y 2023) and Time 2 (Janua y 2024). This design
enabled he examina ion o bo h p esc ip i e and desc ip i e inclusi i y no ms and hei ela ionship o
adolescen a ec i e pola iza ion; he in e en ion’s e ec s on in e g oup ole ance a e epo ed elsewhe e
(Shani e al., 2023). Selec ing wo egional schools wi h dis inc demog aphic p o iles—one highly di e se
(u ban, high mig a ion backg ound) and one mo e homogeneous (less u ban, p edominan ly Ge man
backg ound)—allowed us o explo e how he school no ms‐pola iza ion ela ionship migh di e con ex ually.
Ge many p o ides a e ealing con ex o s udying opinion‐based a ec i e pola iza ion among adolescen s,
as i s mul ipa y sys em ex ends li e a u e beyond ypically s udied wo‐pa y sys ems (e.g., he US).
Adolescen s may de elop s ong opinions on salien social issues (like LGBTQIA+ igh s and e ugee policies,
p ominen in Ge man discou se; Schne ze e al., 2024) e en wi hou s able pa y a ilia ions cha ac e is ic
o wo‐pa y sys ems (Ban el, 2023; Schie e decke e al., 2024; Wagne , 2021). Mo eo e , Ge many’s
poli ical landscape has ans o med: adi ional Volkspa eien (mains eam pa ies) g ow weake while smalle
pa ies gain p ominence (Angenend & B ause, 2024). This shi likely in luences a ec i e pola iza ion,
pa icula ly as nega i e emo ions may a ge eme ging inge pa ies. The success o pa ies like he a ‐ igh
Al e na i e ü Deu schland among young o e s (Shukla & O o, 2025) sugges s heigh ened a ec i e
esponses. Fu he , while Ge many saw declining ideological pola iza ion (1980–2010) wi h subs an ial
policy consensus, a ec i e pola iza ion has s eadily inc eased concu en ly (Co é e al., 2025; Fe ei a
Da Sil a & Ga zia, 2024). This di e gence c ea es an ideal se ing o examine how emo ional di isions
de elop e en when ideological di e ences a e less s a k (Helbling & Jungkunz, 2020).
Social Inclusion •2025 •Volume 13 •A icle 10122 6
Impo an ly, we do no empi ically compa e a ec i e pola iza ion o i s ela ionship wi h inclusi i y no ms
be ween adolescen s and adul s. Ra he , we in es iga e how a ec i e pola iza ion ac oss opinion‐based
g oups eme ges du ing adolescence and examine whe he inclusi i y no ms can coun e ac such
pola iza ion by os e ing engagemen wi h opposing g oups ins ead o wi hd awal. Ou compa a i e analysis
is limi ed o in e p e ing ou indings agains exis ing esea ch and heo ies on adolescen poli ical
de elopmen e iewed ea lie .
Ou in es iga ion had h ee p ima y objec i es. Fi s , we de eloped a me hod o ob ain opinion‐based
g oups and examine a ec i e pola iza ion speci ically adap ed o adolescen s. They o en lack s able
pa isan iden i ies, wi h a ilia ions cen e ing on opinions abou salien issues a he han o mal
pa isanship, pa icula ly in complex mul i‐pa y sys ems (Lay e al., 2023). To enhance ecological alidi y,
pa icipan s i s sha ed hei opinions on LGBTQIA+ and e ugee igh s, issues highly ele an o Ge man
adolescen s (Schne ze e al., 2024). Pa icipan s hen esponded o igne e‐based scena ios depic ing
pee s wi h opposing s ances, epo ing hei willingness o engage o dis ance hemsel es socially and
poli ically. Speci ically, social dis ance, indi iduals’ willingness o social in e ac ion wi h ou g oup membe s
(Boga dus, 1947), is a well‐es ablished a ec i e pola iza ion measu e in poli ical psychology (D uckman &
Le endusky, 2019; Iyenga e al., 2019). While a iously ope a ionalized (e.g., gene al pa isan‐di ec ed
a i udes; Iyenga e al., 2012), ou s udy adop ed assessmen s o com o wi h ou g oup membe s ac oss
speci ic social con ex s (e.g., as neighbo s, in‐laws) o cap u e conc e e beha io al mani es a ions. Social
dis ance measu es a e pa icula ly use ul o examining adolescen in e g oup ela ions (e.g., Binde e al.,
2009), as hey assess angible social p e e ences e lec ing e e yday poli ical an ipa hy. Addi ionally, we
included dialogue o ien a ion measu es (Cionea e al., 2014) o examine i a ec i e pola iza ion ex ends
beyond social a oidance o impai democ a ic discou se, a c i ical conce n, as animosi y can bias in o ma ion
p ocessing and lead o selec i e exposu e (Kaise e al., 2022; Lelkes & Wes wood, 2017; Mosleh e al.,
2025). This app oach cap u ed bo h in e pe sonal and ci ic dimensions o po en ial opinion‐based hos ili y.
Second, we u he classi ied s uden s in o opinion‐based g oups, based on hei posi ions ega ding
LGBTQIA+ and e ugee igh s, we examined whe he hese opinion‐based g oups exhibi ed di e en
pa e ns o pe cei ed no ms, a ec i e pola iza ion, and dialogue o ien a ion o e ime. While esea ch wi h
adul s consis en ly demons a es ideological asymme ies—wi h conse a i es o en exp essing s onge
ou g oup nega i i y (Iyenga e al., 2019; Kekkonen e al., 2022; Kluge e al., 2024; Pe ley e al.,
2024)—such pa e ns among adolescen s emain unexplo ed. Unde s anding hese ea ly g oup‐speci ic
endencies is c ucial o iden i ying which indi iduals migh be mos ulne able o, o p o ec ed om,
pola izing in luences in educa ional se ings (Lay e al., 2023).
Finally, we in es iga ed whe he pe cei ed inclusi i y no ms p edic ed educ ions in a ec i e pola iza ion
o e ime (Schä e e al., 2024; Ve kuy en e al., 2020). Using c oss‐lagged models, we assessed whe he
Time 1 no ms p edic ed subsequen changes in dialogue o ien a ion and social dis ance while examining wo
key mode a ing ac o s. Fi s , we analyzed whe he poli ical ideology mode a ed no m ecep i i y, and
second, whe he school di e si y in luenced no m e ec i eness. He e ogeneous en i onmen s may p o ide
mo e conduci e condi ions o inclusi i y messages (Mi chell, 2019), while homogeneous se ings could
a enua e no m salience. Addi ionally, we examined how demog aphic cha ac e is ics—including eligious
a ilia ion, mig a ion backg ound, and gende —mode a ed no ma i e in luences.
Social Inclusion •2025 •Volume 13 •A icle 10122 7
This comp ehensi e in es iga ion ad ances ou unde s anding o how opinion‐based hos ili y eme ges and
can be mode a ed wi hin adolescen communi ies, o e ing insigh s in o bo h heo e ical amewo ks o
pola iza ion and empi ically in o med in e en ions du ing his c i ical de elopmen al pe iod.
2. Me hods
2.1. Sample and P ocedu e
The s udy p o ocol ecei ed app o al om bo h he au ho s’ ins i u ional e hics commi ee and he local
school au ho i y esponsible o educa ional esea ch. S uden s p o ided ac i e op ‐in consen be o e each
da a collec ion wa e, while pa en s ecei ed op ‐ou consen o ms. Pa icipan s comple ed he su ey
du ing egula class ime using sma phones o able s, wi h ained esea ch pe sonnel p esen o p o ide
assis ance and ensu e con iden iali y. The 45‐minu e su ey assessed demog aphics, psychosocial a iables,
school ac i i ies, and social ne wo ks, wi h ou analysis ocusing on pola iza ion‐ ele an measu es
(comple e ins umen a ailable in he Supplemen a y File).
The ini ial sample comp ised 860 s uden s (329 om he di e se school, 531 om he nondi e se school)
ac oss wo wa es spaced one yea apa . A e excluding 21 pa icipan s who discon inued be o e page 20
(o 46) o exhibi ed comple ion imes unde 500 seconds, he inal analy ic sample included 839 s uden s.
The di e se school e ained 317 pa icipan s (142 a Time 1 only, 26 a Time 2 only, 149 a bo h wa es), while
he nondi e se school main ained 522 pa icipan s (168 a Time 1 only, 75 a Time 2 only, 279 a bo h wa es).
A p io i powe analysis using he semPowe package (Moshagen & Bade , 2023) indica ed ha 792 pa icipan s
would p o ide adequa e powe (80%) o de ec small c oss‐lagged e ec s in la en ‐ a iable CLPMs, con i ming
ou sample size as su icien .
De ailed demog aphics appea in he Supplemen a y File, Table S1. The sample comp ised 45% emale and
51% male pa icipan s ( he emaining pa icipan s indica ed o he gende iden i ies), wi h a mean age o
14.59 yea s (𝑆𝐷 = 1.47). The schools demons a ed signi ican demog aphic di e ences: The di e se school
epo ed highe p opo ions o s uden s wi h mig a ion backg ounds (72% s. 34%, 𝑝 < .001) and dis inc
eligious composi ions (37% Muslim, 33% Ch is ian s. 11% Muslim, 59% Ch is ian in he nondi e se school,
𝑝 < .001). E hnocul u al backg ounds a ied subs an ially, wi h 60% o nondi e se school s uden s
iden i ying as o Ge man o igin, compa ed o mo e a ied egional backg ounds in he di e se school
(e.g., 24% Middle Eas /No h A ica, 4.7% Sub‐Saha an A ica). Poli ical o ien a ions (see Sec ion 2.2) u he
di e en ia ed he schools: The di e se school showed highe p opo ions o “an i‐LGBTQIA+, p o‐ e ugees”
s uden s (33% s. 8.6%), while he nondi e se school had a g ea e ep esen a ion o “Conse a i es”
(6.0% s. 2.8%) and “p o‐LGBTQIA+, an i‐ e ugees” s uden s (10% s. 2.8%, 𝑝 < .001).
2.2. Measu es
All measu es we e adminis e ed in Ge man using a ansla ion/back‐ ansla ion p ocedu e conduc ed by he
esea ch eam. All scale i ems a e p o ided in he Supplemen a y File, Table S2, ansla ed in o English.
Social Inclusion •2025 •Volume 13 •A icle 10122 8
2.2.1. Pe cei ed P esc ip i e and Desc ip i e Inclusi i y No ms
The p esc ip i e inclusi i y no ms scale comp ised six i ems, wi h pai s o i ems assessing h ee co e
componen s: equali y‐based espec , uni y, and dialogue. This measu e was adap ed om a p e iously
alida ed 10‐i em scale (Shani e al., 2023). Pa icipan s e alua ed hei en i e school communi y’s
ag eemen wi h each s a emen using a 5‐poin scale (1 =do no ag ee a all, 5 = ully ag ee). Equali y‐based
espec i ems we e adap ed om es ablished measu es (Renge & Reese, 2017; Renge & Simon, 2011),
while uni y and dialogue i ems d ew om he In e cul u alism Index (Ve kuy en e al., 2020). Explo a o y
ac o analyses (EFA) using maximum likelihood es ima ion wi h Va imax o a ion, conduc ed sepa a ely o
each school and wa e, consis en ly e ealed a one‐ ac o solu ion (see Supplemen a y File, Table S3). Fac o
e en ion was de e mined by eigen alues g ea e han 1.0 and examina ion o explained a iance
p opo ions. High ac o loadings suppo ed cons uc unidimensionali y, wi h he ac o explaining
60–72% o o al a iance. In e nal consis ency was excellen ac oss all subse s (C onbach’s 𝛼and
McDonald’s 𝜔 ≥ .90; see Supplemen a y File, Table S7).
I is impo an o no e ha while ou p esc ip i e inclusi i y no ms measu e uses examples o cul u al,
eligious, and na ional di e si y o con ex ualize he i ems o adolescen comp ehension (as es ablished
du ing p e es ing; see Supplemen a y File, Table S2, o speci ic i ems), he heo e ical cons uc ex ends
beyond e hnic/cul u al inclusi i y o encompass b oade in e g oup ela ions p inciples. Ou ac o analyses
suppo he iew ha hese i ems ep esen a cohe en unde lying cons uc a he han sepa a e
dimensions, sugges ing pa icipan s in e p e ed hem as e lec ing gene al inclusi i y p inciples a he han
solely cul u al di e si y a i udes.
The desc ip i e inclusi i y no ms measu e, de eloped h ough p elimina y ocus g oups wi h six adolescen s,
consis ed o ou i ems. Two i ems assessed posi i e inclusi i y beha io s and wo cap u ed exclusiona y
beha io s. Pa icipan s es ima ed he p opo ion o schoolma es engaging in each beha io using a 5‐poin
scale (1 =nobody, 5 =all). EFA e ealed a consis en wo‐ ac o s uc u e dis inguishing be ween posi i e
and nega i e beha io s, explaining 33–57% o o al a iance ac oss subse s (see Supplemen a y File,
Table S4). In e ‐i em co ela ions we e adequa e o nega i e desc ip i e no ms bu demons a ed lowe
consis ency o posi i e desc ip i e no ms (see Supplemen a y File, Table S7).
2.2.2. Opinion‐Based Pola iza ion Assessmen
Figu e 1 p esen s a lowcha ha desc ibes he sequence o ex s and ques ions included in he measu e.
Pa icipan s’ assignmen o a speci ic igne e p o ile (o a gene ic one) was based on hei esponses o he
ou yes/no ques ions abou LGBTQIA+ igh s (same‐sex ma iage, adop ion) and e ugee policies (admission
le els, socie al con ibu ion). An op ion o skip he ques ion was a ailable.
Pa icipan s hen encoun e ed a igne e ea u ing a same‐age, same‐gende ic ional pee (e.g., “Lisa is
14 yea s old”) holding opposing iews. Gi en ha adolescen s p ima ily o ien hemsel es owa d
same‐gende and same‐age pee s (McMillan, 2022; McPhe son e al., 2001), his ma ching enhances he
measu e’s ecological alidi y and ele ance o adolescen esponden s (P imi e al., 2016). The p o ile
assignmen logic aimed o p esen pa icipan s wi h a pee holding opposing iews on a leas one majo
issue domain, speci ically:
Social Inclusion •2025 •Volume 13 •A icle 10122 9
schools wi h dis inc mul icul u al composi ions, we in es iga ed whe he pe cei ed no ma i e clima e—
independen o obse ed beha io al no ms—could a enua e a ec i e pola iza ion o e ime. Despi e
heo e ical p oposi ions sugges ing ha no ma i e emphasis on equali y, espec , communal uni y, and
cons uc i e dialogue would educe pola iza ion, ou analyses e ealed no e idence o c oss‐lagged causal
e ec s om no ms o ei he dialogue o ien a ion o social dis ance. Below, we in e p e hese indings
h ough in e sec ing de elopmen al, con ex ual, and me hodological pe spec i es.
4.1. Di e ences Ac oss Poli ical O ien a ions
Ou mixed linea models e ealed obus and consis en e ec s o poli ical o ien a ion on no ma i e
pe cep ions and in e g oup engagemen pa e ns, wi h ideological subg oups main aining dis inc i e
a i udinal p o iles ac oss wa es. Conse a i e adolescen s exhibi ed a consis en pa e n cha ac e ized by
lowe pe cep ion o p esc ip i e no ms, heigh ened pe cep ion o nega i e desc ip i e no ms, and educed
p opensi y o c oss‐g oup dialogue. In con as , libe al‐iden i ying adolescen s demons a ed he s onges
inclusi i y‐o ien ed disposi ions, consis en ly pe cei ing bo h p esc ip i e and posi i e desc ip i e no ms
while exhibi ing enhanced dialogue eadiness. Hyb id g oups e ealed mo e nuanced a i udinal
con igu a ions: he an i‐LGBTQIA+/p o‐ e ugees g oup pe cei ed s onge p esc ip i e no ms despi e
showing educed dialogue eadiness, while he p o‐LGBTQIA+/an i‐ e ugees g oup demons a ed no ably
lowe social dis ance om ideological opponen s.
These pa e ns align wi h esea ch on ideological asymme ies showing conse a i e o ien a ions associa ed
wi h heigh ened ou g oup dis us and educed in e g oup con ac willingness (B and & C aw o d, 2020;
Ganzach & Schul, 2021). Conse a i e endencies owa d ing oup loyal y and esis ance o social change
(e.g., Feldman & Johns on, 2014; Jos , 2017) con as wi h libe al cha ac e is ics o enhanced empa hy and
educed ing oup emphasis (Jos e al., 2009; Mo is, 2020). Ou indings ex end es ablished wo k linking
s onge ideological iden i ies o inc eased poli ical animosi y (J. K. B own & Hohman, 2022; Iyenga e al.,
2019; Iyenga & Wes wood, 2015; Mason, 2018a). Impo an ly, we ound ha many adolescen s hold hyb id
posi ions ha anscend he adi ional libe al‐conse a i e di ide. These mixed s ances—such as suppo ing
e ugee policies while opposing LGBTQIA+ igh s—can gene a e s ong emo ional a achmen s and
issue‐speci ic esis ance o dialogue. While conside able ins abili y was obse ed o e one yea , some
s abili y pa e ns eme ged, sugges ing ha o ce ain adolescen s, ideological iden i ies may c ys allize
ea lie han p e iously ecognized (Hess & To ney‐Pu a, 2017; La ineu e al., 2024; Oos e ho e al.,
2021; Rei en‐Taga & Cimpian, 2022).
4.2. Me hodological Insigh s and De elopmen al Pa e ns in Adolescen Poli ical Iden i y Fo ma ion
Ou igne e‐based app oach o measu ing opinion‐based a ec i e pola iza ion o e ed bo h
me hodological insigh s and heo e ical e ela ions abou adolescen poli ical de elopmen . I p esen ed
conc e e, de elopmen ally app op ia e scena ios a he han abs ac poli ical concep s; cap u ed
con ex ually ele an social issues; and measu ed bo h beha io al in en ions and communica i e
o ien a ions in na u alis ic pee con ex s. By ocusing on conc e e social scena ios a he han abs ac
poli ical e alua ions, ou me hodology cap u ed beha io al mani es a ions o a ec i e pola iza ion h ough
social dis ance (D uckman & Le endusky, 2019) and dialogue o ien a ion measu es (Cionea e al., 2014).
Howe e , he subs an ial p opo ion o pa icipan s showing wi hin‐indi idual luc ua ions and
Social Inclusion •2025 •Volume 13 •A icle 10122 16

wi hin‐domain inconsis en esponses e ealed bo h me hodological challenges and heo e ically meaning ul
pa e ns in adolescen poli ical iden i y o ma ion.
The obse ed malleabili y in poli ical o ien a ions may e lec genuine opinions ha a e seemingly
incohe en (i.e., hyb id on he conse a i e‐libe al dicho omy) o inconsis en (i.e., mixed wi hin domains),
bu may also be held by he wide public. Al e na i ely, i may e lec b oade de elopmen al p ocesses,
whe e you h ac i ely explo e and e ise hei belie s h ough encoun e s wi h new in o ma ion and social
in luences. This ins abili y may s em om mul iple ac o s: adolescen s may s ill be de eloping he cogni i e
amewo ks needed o o ganize poli ical belie s sys ema ically, hei opinions may be genuinely in lux as
hey engage wi h di e se pe spec i es, o hey may lack su icien exposu e o poli ical discou se o o m
s able iewpoin s on complex social issues. Unlike adul popula ions, whe e ideological consis ency o en
indica es poli ical sophis ica ion, adolescen hyb id g oups (e.g., p o‐LGBTQIA+, an i‐ e ugees, o ice e sa)
and inconsis encies (e.g., suppo ing same‐sex ma iage bu opposing adop ion igh s) may ep esen ac i e
engagemen wi h complex poli ical issues a he han measu emen e o .
De elopmen al pe spec i es p o ide heo e ical g ounds o expec ing heigh ened ins abili y and
inconsis ency du ing adolescence compa ed o adul hood. Poli ical a i udes end o become inc easingly
s able as indi iduals ansi ion om adolescence o ea ly adul hood (Rekke e al., 2019). Longi udinal
esea ch shows ha co e poli ical a i udes in adul hood demons a e ema kable s abili y o e ime, wi h
app oxima ely 75% o indi iduals main aining consis en belie s o e decades (Pe e son e al., 2020).
The luidi y we obse ed aligns wi h de elopmen al li e a u e sugges ing ha adolescence ep esen s a
c i ical pe iod o iden i y o ma ion (Dahl e al., 2018), cha ac e ized by explo a ion ac oss a ious domains,
including poli ical iden i y, coined by heigh ened suscep ibili y o social in luence (S einbe g & Monahan,
2007), and ongoing cogni i e de elopmen (S einbe g, 2005). Consequen ly, adolescen poli ical a i udes
migh be pa icula ly suscep ible o change and less ideologically cohe en .
Howe e , while ou indings highligh no able ideological inconsis encies and empo al ins abili ies in
adolescen s’ poli ical belie s, we acknowledge ha such pa e ns a e no unique o adolescen s. Resea ch on
Eu opean adul s iden i ied di e se belie s uc u es, which we e classi ied as “ideologue” g oups (in e nally
consis en le ‐ igh posi ions), “al e na i e” g oups (mul i‐dimensional, o en con lic ing cul u al and
economic a i udes), and “uns uc u ed” g oups (weakly co ela ed belie s; an Noo d e al., 2025). Simila ly,
Ame ican adul s equen ly hold hyb id belie s ha combine economic libe alism wi h mo al conse a ism o
ice e sa (Baldassa i & Goldbe g, 2014). Wi hou a di ec compa ison o adolescen s and adul s using
iden ical measu es, we canno conclude ha he ins abili y and ideological inconsis ency obse ed in ou
sample exceeds ha o adul s. This limi a ion highligh s he need o u u e esea ch employing
age‐compa a i e designs wi h pa allel measu es o empi ically es de elopmen al di e ences in poli ical
a i ude s abili y and consis ency.
4.3. Inclusi i y No ms and A ec i e Pola iza ion
Con a y o ou expec a ions ha pe cei ed inclusi i y no ms would p edic inc eased dialogue and social
o ien a ions o e ime (Schä e e al., 2024), c oss‐lagged analyses showed no longi udinal e ec s be ween
inclusi i y no ms and eadiness o c oss‐g oup social in e ac ions. Ins ead, bo h dialogue o ien a ion and
social dis ance we e p ima ily p edic ed by hei own p io le els. This pa e n sugges s ha na u ally occu ing
Social Inclusion •2025 •Volume 13 •A icle 10122 17
school inclusi i y no ms may be insu icien o o e come es ablished iden i y‐based a i udes. While p e ious
esea ch has demons a ed ha school in e en ions can p oduce sho ‐ e m ole ance gains (Shani e al.,
2023), ou indings indica e ha he me e pe cep ion o inclusi e no ms does no al e dialogue o ien a ion o
social dis ance owa d opposing g oups. Family socializa ion may exe a s onge in luence han school‐based
no ms (Beck & Jennings, 1991; Sapi o, 2004; Tyle & Iyenga , 2023), o adolescen s migh espond mo e
s ongly o immedia e pee ‐g oup no ms han b oade ins i u ional clima e messages. Fu u e esea ch should
examine whe he pee ‐led app oaches p o e mo e e ec i e han ins i u ional ini ia i es in shi ing in e g oup
a i udes (Mu a e al., 2020).
P esc ip i e inclusi i y no ms demons a ed s onge associa ions wi h pola iza ion measu es han
desc ip i e no ms, pa icula ly nega i e desc ip i e no ms, which showed no signi ican ela ionships ac oss
wa es. This pa e n aligns wi h Schä e e al.’s (2024) indings o s onge connec ions be ween injunc i e o
p esc ip i e no ms and ole ance ac oss Eu opean coun ies. The limi ed in luence o desc ip i e no ms may
e lec hei con ex ‐speci ici y—p ejudice o en aces a iable public condemna ion depending on he a ge
g oup (C andall e al., 2002), sugges ing ha beha io al no ms’ e ec i eness may a y be ween conc e e
ou g oups and abs ac ideological opponen s.
4.4. Theo e ical and P ac ical Implica ions
Ou indings ad ance se e al key deba es in poli ical socializa ion and in e g oup ela ions esea ch. Fi s ,
hey suppo eme ging pe spec i es ha iew poli ical iden i y de elopmen as mo e luid and complex han
adi ionally assumed. The subs an ial opinion‐based g oup ins abili y we obse ed challenges classic
models sugges ing poli ical enmi y eme ges only in la e adolescence (G eens ein, 1960; Hess &
To ney‐Pu a, 2017). Ra he , ou esul s align wi h e idence ha pa isan hos ili y can eme ge by age 11
(Tyle & Iyenga , 2023) while emaining malleable h oughou adolescen de elopmen (Hooghe &
Wilken eld, 2008; Lay e al., 2023).
Second, ou indings illumina e he di e en ial impac o no m ypes on adolescen in e g oup a i udes.
The s onge associa ions o a ec i e pola iza ion beha io al indica o s wi h p esc ip i e no ms compa ed
o desc ip i e no ms (see Supplemen a y File, Table S17) sugges s ha explici ins i u ional messages abou
inclusion may ca y mo e weigh han pe cei ed pee beha io , suppo ing ecen wo k on p esc ip i e
no ms as egula o y mechanisms in in e g oup ela ions (Bicchie i e al., 2023; Cialdini & Jacobson, 2021).
Howe e , hese no ma i e in luences compe e wi h o he socializa ion o ces, pa icula ly amily in luence in
he ansmission o pa isan loyal ies (Jennings & Niemi, 1968; Ped aza & Pe y, 2020; Tyle & Iyenga ,
2023). Ou measu e o inclusi i y no ms p ima ily con ex ualizes hese p inciples wi hin cul u al di e si y,
which may in luence hei ela ionship wi h opinion‐based pola iza ion ou comes. P ejudice ope a es
di e en ly ac oss ideological lines, wi h bo h libe als and conse a i es exp essing p ejudice owa d
dissimila ou g oups (C aw o d & B and , 2020). The modes associa ions we ound be ween p esc ip i e
inclusi i y no ms and opinion‐based pola iza ion may e lec his dis inc ion be ween cul u al and ideological
inclusi i y domains. Ne e heless, pa allel esea ch wi h adul s using mo e gene ic inclusi i y no m measu es
inds simila pa e ns o ela ionships be ween inclusi i y no ms and willingness o engage wi h ideological
ou g oups (Schä e e al., 2024), sugges ing some ans e abili y o inclusi i y p inciples ac oss di e en
in e g oup con ex s.
Social Inclusion •2025 •Volume 13 •A icle 10122 18
This aises he possibili y o a bidi ec ional ela ionship be ween ideological o ien a ion ( a he han hei
a ec i e consequences) and pe cep ions o inclusi i y no ms. Libe al‐iden i ying adolescen s demons a ed
he s onges inclusi i y‐o ien ed disposi ions, consis en ly pe cei ing posi i e p esc ip i e and desc ip i e
no ms while exhibi ing enhanced dialogue eadiness. I ideological g oups di e in hei app oaches o
a ious o ms o di e si y, wi h libe als o en demons a ing g ea e endo semen o di e si y‐ ela ed
a i udes (C aw o d & B and , 2020), inclusi i y no ms may no ep esen a uly independen cons uc ha
in luences pola iza ion, bu a he pa ially e lec exis ing ideological posi ions. The co ela ional na u e o
ou s udy limi s causal in e ences, and he s abili y o bo h inclusi i y no ms and poli ical posi ions o e he
one‐yea in e al sugges s a po en ial unde lying ela ionship. Fu u e esea ch employing expe imen al
manipula ions o pe cei ed no ms could be e disen angle hese e ec s and es whe he inclusi i y no ms
can unc ion as an in e en ion s a egy ha anscends ideological di e ences.
Thi d, he a ying e ec s ac oss con ex s highligh he impo ance o conside ing mul iple mode a ing
ac o s. The enhanced dialogue‐social openness link in nondi e se schools sugges s ha dialogue may be
pa icula ly c ucial in se ings whe e in e g oup con ac oppo uni ies a e limi ed. These indings b idge
classic con ac heo y (Allpo , 1954; B acegi dle e al., 2022; Tausch e al., 2024) wi h con empo a y
esea ch on how no ma i e clima es a ec pola iza ion in he absence o sus ained ou g oup in e ac ion
(Kingze e e al., 2021; Mullinix & Ly hgoe, 2023). Al hough dialogue o ien a ion was ope a ionalized
dis inc ly om pe cei ed inclusi i y no ms, his esul o e s p elimina y suppo o he heo e ical ole o
inclusi i y no ms in educing pola iza ion, gi en ha dialogue ep esen s one o he co e alues associa ed
wi h ole ance owa d poli ical ad e sa ies (Schä e e al., 2024).
These insigh s sugges se e al ac ionable in e en ions. Since adolescen s o en o e es ima e pee hos ili y,
in e en ions ha e eal mo e mode a e ac ual pee a i udes can help ecalib a e social pe cep ions (You &
Lee, 2024). Face‐ o‐ ace discussions combined wi h inclusi i y messages can e ec i ely educe s e eo ypes
and p omo e c oss‐g oup iendships (Le endusky & S ecula, 2021). Gi en he emo ional ounda ions o
pola iza ion (Bakke & Lelkes, 2024), in eg a ing inclusi i y no ms wi h emo ion egula ion s a egies
(E e s e al., 2014; Simonsson e al., 2022) may enhance in e en ion e ec i eness. P og ams should be
ailo ed o ideological o ien a ions—emphasizing communal alues o conse a i e you h (Mullinix &
Ly hgoe, 2023) while adop ing nuanced app oaches wi h hose holding hyb id belie s o sus ain dialogue
wi hou elici ing de ensi eness.
4.5. Limi a ions and Fu u e Di ec ions
Se e al limi a ions quali y ou conclusions. While ou pola iza ion measu e enabled he assessmen o
opinion‐based g oups in an adolescen sample, i s complexi y and na ow ocus on only wo issues wi h wo
ques ions each esul ed in a high p opo ion o wi hin‐domain inconsis en esponses ha complica ed
app op ia e p o iling and classi ica ion in o opinion‐based g oups. Fu u e esea ch migh de elop mo e
comp ehensi e p o iles using ques ions ha add ess a b oade ange o socie al issues, o ocus on pa isan
g oup a ilia ions a he han opinion‐based g oups.
Rega ding ou esea ch design, he wo‐wa e s uc u e wi h a one‐yea in e al limi ed ou abili y o de ec
po en ial sho ‐ e m o delayed shi s in no ms and pola iza ion. Fu u e s udies employing longe
ime ames wi h mo e equen measu emen s (e.g., using andom‐in e cep c oss‐lagged panel model,
Social Inclusion •2025 •Volume 13 •A icle 10122 19
Hamake e al., 2015) could be e illumina e whe he and how no m‐based changes eme ge g adually.
Finally, while we ocused on pe cei ed school no ms, adolescen s simul aneously na iga e mul iple social
ne wo ks—including amily, social media, and neighbo hoods— ha may ansmi compe ing messages abou
ou g oup engagemen .
A en ion should also be paid o he absence o a compa able adul sample. While ou indings e eal
conside able luidi y in adolescen s’ s ances on LGBTQIA+ and e ugee igh s o e a one‐yea pe iod,
wi hou adul compa ison using iden ical measu es, we canno de ini i ely de e mine whe he he ins abili y
obse ed is unique o adolescence o simply e lec s b oade pa e ns o opinion change ac oss he li espan.
Fu u e esea ch should di ec ly compa e adolescen and adul samples using longi udinal designs and pa allel
measu es o es ablish he ela i e s abili y and consis ency o poli ical a i udes ac oss de elopmen al s ages.
Mo eo e , ins abili y and inconsis ency may ha e esul ed om ou opinion‐based pola iza ion measu e,
which may ha e o e simpli ied adolescen s’ complex ideological explo a ions. Fu u e s udies should
inco po a e a i udinal ambi alence measu es (Bu ge , 2024) and neu al esponse op ions o be e
dis inguish be ween genuine poli ical unce ain y, ac i e ideological explo a ion, and measu emen ‐ ela ed
esponse inconsis ency. Addi ionally, we did no assess s eng h o iden i ica ion wi h opinion‐based g oups,
hough his ac o o en in luences how ing oup no ms a e in e nalized o esis ed (You & Lee, 2024).
The age ange o ou sample (g ades 7–11) limi s he gene alizabili y o younge child en o eme ging adul s,
as poli ical a i udes con inue o de elop ac oss adolescence (Nieuwelink e al., 2018). Fu u e s udies should
employ coho ‐sequen ial designs ha ollow mul iple age g oups (including adolescen s and adul s o
a ying ages) o e se e al yea s o disen angle age e ec s om coho and pe iod e ec s, and o iden i y
he ajec o ies o opinion s abili y ac oss he li espan.
Fu he mo e, a key limi a ion pe ains o ou measu emen o inclusi i y no ms. Following cogni i e
p e es ing, we con ex ualized inclusi i y no m i ems wi hin cul u al, eligious, and na ional di e si y o
enhance comp ehensibili y o adolescen pa icipan s. This app oach, while de elopmen ally app op ia e,
may ha e na owed he concep ual scope o inclusi i y no ms, po en ially limi ing hei applicabili y o
opinion‐based pola iza ion. Fu u e esea ch should de elop and alida e measu es o inclusi i y no ms ha
mo e explici ly encompass ideological di e si y alongside cul u al di e si y, allowing esea che s o examine
whe he hese no ms unc ion simila ly o di e en ly ac oss domains.
Finally, while his s udy p o ides aluable insigh s in o a ec i e pola iza ion among adolescen s in Ge many,
se e al scope condi ions should be conside ed when gene alizing hese indings o o he con ex s. Ou
measu emen app oach ocusing on opinion‐based a he han pa isan pola iza ion, while app op ia e o
adolescen s wi h de eloping poli ical iden i ies in a mul ipa y sys em, migh yield di e en esul s in
con ex s whe e pa isan iden i ies o m ea lie o whe e poli ical con lic is s uc u ed di e en ly.
The di e gence be ween Ge many’s declining ideological pola iza ion and inc easing a ec i e pola iza ion
(Co é e al., 2025; Fe ei a Da Sil a & Ga zia, 2024) ep esen s a unique poli ical clima e ha may shape
adolescen poli ical socializa ion di e en ly han in coun ies whe e ideological and a ec i e pola iza ion
align. Addi ionally, Ge many’s his o ical legacies, including i s Nazi pas and Eas –Wes di ision and
euni ica ion, ha e c ea ed dis inc poli ical socializa ion pa e ns ha in luence how young people o m and
exp ess poli ical a i udes. The coun y’s ongoing deba es a ound LGBTQIA+ igh s and e ugee policies,
which we used as ocal issues, e lec pa icula na ional conce ns ha may mani es di e en ly in coun ies
Social Inclusion •2025 •Volume 13 •A icle 10122 20
wi h dis inc his o ical legacies and con empo a y poli ical na a i es, ye hese also a y ac oss he ede al
s a es wi hin Ge many (Angenend & B ause, 2024). Fu u e esea ch should explo e hese dynamics ac oss
di e se egional, na ional, cul u al, and poli ical se ings o de e mine which aspec s o adolescen a ec i e
pola iza ion ep esen uni e sal de elopmen al p ocesses e sus con ex ‐speci ic mani es a ions shaped by
pa icula poli ical and educa ional en i onmen s.
These limi a ions sugges se e al p omising esea ch di ec ions. The g owing in luence o algo i hm‐d i en
social media eeds and i al poli ical con en on adolescen a i udes may ampli y o coun e ac
ole ance‐p omo ing messages in educa ional se ings (Chan & Fu, 2017; Daniel, 2024), wa an ing
in es iga ion in o how hese digi al en i onmen s in e ac wi h school‐based no m in e en ions.
Resea che s should also examine he compa a i e e ec i eness o pee ‐led ini ia i es e sus ins i u ional
policies, explo e how emo ion egula ion s a egies migh bu e agains pola iza ion‐induced hos ili y, and
in es iga e he dynamic in e play be ween e ol ing ideological alignmen s and no ma i e in luences du ing
adolescen de elopmen .
Acknowledgmen s
We hank pa icipa ing schools, he eache s, and he s uden s who dedica ed hei ime and e o s o enable
his s udy. In addi ion, we hank he s uden assis an s and in e ns who suppo ed us a a ious s ages o
da a collec ion.
Funding
This esea ch was suppo ed by he Volkswagen Founda ion unde he Inclusi i y No ms o Coun e
Pola iza ion in Eu opean Socie ies (INCLUSIVITY) p ojec (9B060), and by a g an om he S a e C ime
P e en ion Council o Lowe Saxony (Lande p ä en ions a Niede sachsen).
Con lic o In e es s
The au ho s decla e no con lic o in e es s. In his a icle, edi o ial decisions we e unde aken by Ul
R. Hede o (Uni e si y o Copenhagen, Denma k).
Da a A ailabili y
Da a and ma e ials used in his s udy a e a ailable upon eques o he co esponding au ho .
Supplemen a y Ma e ial
Supplemen a y ma e ial o his a icle is a ailable online in he o ma p o ided by he au ho (unedi ed).
Re e ences
Allpo , G. W. (1954). The na u e o p ejudice. Addison‐Wesley.
Angenend , M., & B ause, S. D. (2024). The long way owa ds pola ized plu alism: Pa y and pa y sys em
change in Ge many. In T. Pogun ke & W. Ho meis e (Eds.), Poli ical pa ies and he c isis o democ acy
(1s ed., pp. 82–107). Ox o d Uni e si y P ess. h ps://doi.o g/10.1093/oso/9780198888734.003.0005
Bakke , B. N., & Lelkes, Y. (2024). Pu ing he a ec in o a ec i e pola isa ion. Cogni ion and Emo ion,
38(4), 418–436. h ps://doi.o g/10.1080/02699931.2024.2362366
Baldassa i, D., & Goldbe g, A. (2014). Nei he ideologues no agnos ics: Al e na i e o e s’ belie sys em in
an age o pa isan poli ics. Ame ican Jou nal o Sociology,120(1), 45–95. h ps://doi.o g/10.1086/676042
Social Inclusion •2025 •Volume 13 •A icle 10122 21

Ban el, I. (2023). Camps, no jus pa ies. The dynamic ounda ions o a ec i e pola iza ion in mul i‐pa y
sys ems. Elec o al S udies,83, A icle 102614. h ps://doi.o g/10.1016/j.elec s ud.2023.102614
Ba es, D., Mächle , M., Bolke , B., & Walke , S. (2015). Fi ing linea mixed‐e ec s models using lme4. Jou nal
o S a is ical So wa e,67(1). h ps://doi.o g/10.18637/jss. 067.i01
Beck, P. A., & Jennings, M. K. (1991). Family adi ions, poli ical pe iods, and he de elopmen o pa isan
o ien a ions. The Jou nal o Poli ics,53(3), 742–763. h ps://doi.o g/10.2307/2131578
Be n zen, L. E., Kelsall, H., & Ha e eld, E. (2024). Consequences o a ec i e pola iza ion: A oidance,
in ole ance and suppo o iolence in he Uni ed Kingdom and No way. Eu opean Jou nal o Poli ical
Resea ch,63(3), 927–949. h ps://doi.o g/10.1111/1475‐6765.12623
Bicchie i, C., Diman , E., Gel and, M., & Sonde egge , S. (2023). Social no ms and beha io change: The
in e disciplina y esea ch on ie . Jou nal o Economic Beha io & O ganiza ion,205, A4–A7. h ps://doi.
o g/10.1016/j.jebo.2022.11.007
Binde , J., Zage ka, H., B own, R., Funke, F., Kessle , T., Mummendey, A., Maquil, A., Demoulin, S., & Leyens, J.‐P.
(2009). Does con ac educe p ejudice o does p ejudice educe con ac ? A longi udinal es o he con ac
hypo hesis among majo i y and mino i y g oups in h ee Eu opean coun ies. Jou nal o Pe sonali y and
Social Psychology,96(4), 843–856. h ps://doi.o g/10.1037/a0013470
Bliuc, A., McGa y, C., Reynolds, K., & Mun ele, D. (2007). Opinion‐based g oup membe ship as a p edic o
o commi men o poli ical ac ion. Eu opean Jou nal o Social Psychology,37(1), 19–32. h ps://doi.o g/
10.1002/ejsp.334
Boga dus, E. S. (1947). Measu emen o pe sonal‐g oup ela ions. Sociome y,10(4), A icle 306. h ps://doi.
o g/10.2307/2785570
Boonen, J. (2015). The de elopmen o s able pa y p e e ences: Explaining indi idual‐le el s abili y among
adolescen s in Belgium. YOUNG,23(4), 313–335. h ps://doi.o g/10.1177/1103308815596887
Boonen, J. (2019). Lea ning who no o o e o : The ole o pa en al socializa ion in he de elopmen o
nega i e pa isanship. Elec o al S udies,59, 109–119. h ps://doi.o g/10.1016/j.elec s ud.2019.04.001
B acegi dle, C., Reime , N. K., Van Zalk, M., Hews one, M., & Wöl e , R. (2022). Disen angling con ac and
socializa ion e ec s on ou g oup a i udes in di e se iendship ne wo ks. Jou nal o Pe sonali y and Social
Psychology,122(1), 1–15. h ps://doi.o g/10.1037/pspa0000240
B and , M. J., & C aw o d, J. T. (2020). Wo ld iew con lic and p ejudice. In B. Gaw onski (Ed.), Ad ances in
expe imen al social psychology (Vol. 61, pp. 1–66). Else ie . h ps://doi.o g/10.1016/bs.aesp.2019.09.002
B oockman, D. E., Kalla, J. L., & Wes wood, S. J. (2023). Does a ec i e pola iza ion unde mine democ a ic
no ms o accoun abili y? Maybe no . Ame ican Jou nal o Poli ical Science,67(3), 808–828. h ps://doi.o g/
10.1111/ajps.12719
B own, G. D. A., Lewandowsky, S., & Huang, Z. (2022). Social sampling and exp essed a i udes: Au hen ici y
p e e ence and social ex emeness a e sion lead o social no m e ec s and pola iza ion. Psychological
Re iew,129(1), 18–48. h ps://doi.o g/10.1037/ e 0000342
B own, J. K., & Hohman, Z. P. (2022). Ex eme pa y animals: E ec s o poli ical iden i ica ion and ideological
ex emi y. Jou nal o Applied Social Psychology,52(5), 351–362. h ps://doi.o g/10.1111/jasp.12863
Bu ge , A. M. (2024). Poli ical ideology and a i udinal ambi alence: In es iga ing he ole o ideological
ex emi y. Social Psychological and Pe sonali y Science,16(4), A icle 19485506231222958. h ps://doi.o g/
10.1177/19485506231222958
Chan, C., & Fu, K. (2017). The ela ionship be ween Cybe balkaniza ion and opinion pola iza ion: Time‐se ies
analysis on Facebook pages and opinion polls du ing he Hong Kong occupy mo emen and he associa ed
deba e on poli ical e o m: Cybe balkaniza ion and opinion pola iza ion. Jou nal o Compu e ‐Media ed
Communica ion,22(5), 266–283. h ps://doi.o g/10.1111/jcc4.12192
Social Inclusion •2025 •Volume 13 •A icle 10122 22
Cheung, G. W., & Rens old, R. B. (2002). E alua ing goodness‐o ‐ i indexes o es ing measu emen
in a iance. S uc u al Equa ion Modeling: A Mul idisciplina y Jou nal,9(2), 233–255. h ps://doi.o g/
10.1207/S15328007SEM0902_5
Cialdini, R. B., & Jacobson, R. P. (2021). In luences o social no ms on clima e change‐ ela ed beha io s. Cu en
Opinion in Beha io al Sciences,42, 1–8. h ps://doi.o g/10.1016/j.cobeha.2021.01.005
Cialdini, R. B., Reno, R. R., & Kallg en, C. A. (1990). A ocus heo y o no ma i e conduc : Recycling he concep
o no ms o educe li e ing in public places. Jou nal o Pe sonali y and Social Psychology,58(6), 1015–1026.
h ps://doi.o g/10.1037/0022‐3514.58.6.1015
Cionea, I. A., Hample, D., & Fink, E. L. (2014). Dialogue ypes: A scale de elopmen s udy. In D. Mohammed &
M. Lewiński (Eds.),Vi ues o a gumen a ion: P oceedings o he 10 h In e na ional Con e ence o he On a io
Socie y o he s udy o a gumen a ion (OSSA) (pp. 1–11). OSSA.
Co é, H., C awley, S., & Gi ens, J. (2025). G owing pola isa ion: Ideology and a i udes owa ds clima e change.
Wes Eu opean Poli ics. Ad ance online publica ion. h ps://doi.o g/10.1080/01402382.2024.2435727
C andall, C. S., Eshleman, A., & O’B ien, L. (2002). Social no ms and he exp ession and supp ession o
p ejudice: The s uggle o in e naliza ion. Jou nal o Pe sonali y and Social Psychology,82(3), 359–378.
C aw o d, J. T., & B and , M. J. (2020). Ideological (a)symme ies in p ejudice and in e g oup bias. Cu en
Opinion in Beha io al Sciences,34, 40–45. h ps://doi.o g/10.1016/j.cobeha.2019.11.007
Dahl, R. E., Allen, N. B., Wilb ech , L., & Suleiman, A. B. (2018). Impo ance o in es ing in adolescence om a
de elopmen al science pe spec i e. Na u e,554(7693), 441–450. h ps://doi.o g/10.1038/na u e25770
Daniel, J. (2024). Impac o social media usage on poli ical pola iza ion in he Uni ed S a es. Ame ican Jou nal
o Communica ion,6(1), 56–66. h ps://doi.o g/10.47672/ajc.1932
Dias, N., & Lelkes, Y. (2022). The na u e o a ec i e pola iza ion: Disen angling policy disag eemen om
pa isan iden i y. Ame ican Jou nal o Poli ical Science,66(3), 775–790.
D uckman, J. N., & Le endusky, M. S. (2019). Wha do we measu e when we measu e a ec i e pola iza ion?
Public Opinion Qua e ly,83(1), 114–122. h ps://doi.o g/10.1093/poq/n z003
Ecks ein, K., & Noack, P. (2015). Class oom clima e e ec s on adolescen s’ o ien a ions owa d poli ical
beha io s. In P. Thijssen, J. Sionge s, J. Van Lae , J. Hae s, & S. Mels (Eds.), Poli ical engagemen o he
young in Eu ope: You h in he c ucible (pp. 161–177). Rou ledge.
Ende s, C. K. (2001). The impac o nonno mali y on ull in o ma ion maximum‐likelihood es ima ion o
s uc u al equa ion models wi h missing da a. Psychological Me hods,6(4), 352–370. h ps://doi.o g/
10.1037/1082‐989X.6.4.352
E e s, C., Hopp, H., G oss, J. J., Fische , A. H., Mans ead, A. S. R., & Mauss, I. B. (2014). Emo ion esponse
cohe ence: A dual‐p ocess pe spec i e. Biological Psychology,98, 43–49. h ps://doi.o g/10.1016/
j.biopsycho.2013.11.003
Feh , E., Fischbache , U., & Gäch e , S. (2002). S ong ecip oci y, human coope a ion, and he en o cemen o
social no ms. Human Na u e,13(1), 1–25. h ps://doi.o g/10.1007/s12110‐002‐1012‐7
Feldman, S., & Johns on, C. (2014). Unde s anding he de e minan s o poli ical ideology: Implica ions o
s uc u al complexi y. Poli ical Psychology,35(3), 337–358. h ps://doi.o g/10.1111/pops.12055
Fe ei a Da Sil a, F., & Ga zia, D. (2024). A ec i e pola iza ion owa ds pa ies and leade s, and elec o al
pa icipa ion in 13 pa liamen a y democ acies, 1980–2019. Public Opinion Qua e ly,88(4), 1234–1248.
h ps://doi.o g/10.1093/poq/n ae053
Gab io, A., Plump on, C., Bane jee, S., & Leu en , B. (2022). Linea mixed models o handle missing a andom
da a in ial‐based economic e alua ions. Heal h Economics,31(6), 1276–1287. h ps://doi.o g/10.1002/
hec.4510
Social Inclusion •2025 •Volume 13 •A icle 10122 23
Ganzach, Y., & Schul, Y. (2021). Pa isan ideological a i udes: Libe als a e ole an ; he in elligen a e in ole an .
Jou nal o Pe sonali y and Social Psychology,120(6), 1551–1566. h ps://doi.o g/10.1037/pspi0000324
Gibson, J. L. (1992). Al e na i e measu es o poli ical ole ance: Mus ole ance be “leas ‐liked”? Ame ican
Jou nal o Poli ical Science,36(2), 560. h ps://doi.o g/10.2307/2111491
Gi , K., & Gi , T. (2015). E a um o: Does poli ics in luence hi ing? E idence om a andomized expe imen .
Poli ical Beha io ,37(3), 677–677. h ps://doi.o g/10.1007/s11109‐015‐9314‐8
Glas o d, D. E., & Do idio, J. F. (2011). E plu ibus unum: Dual iden i y and mino i y g oup membe s’ mo i a ion
o engage in con ac , as well as social change. Jou nal o Expe imen al Social Psychology,47(5), 1021–1024.
h ps://doi.o g/10.1016/j.jesp.2011.03.021
González, R. (2024). The ole o no ms as p omo e s o social change. Poli ical Psychology,46(2), 455–481.
h ps://doi.o g/10.1111/pops.13025
G aham, M. H., & S olik, M. W. (2020). Democ acy in Ame ica? Pa isanship, pola iza ion, and he obus ness
o suppo o democ acy in he Uni ed S a es. Ame ican Poli ical Science Re iew,114(2), 392–409. h ps://
doi.o g/10.1017/S0003055420000052
G eene, S. (2004). Social iden i y heo y and pa y iden i ica ion*.Social Science Qua e ly,85(1), 136–153.
h ps://doi.o g/10.1111/j.0038‐4941.2004.08501010.x
G eens ein, F. I. (1960). The bene olen leade : Child en’s images o poli ical au ho i y. Ame ican Poli ical
Science Re iew,54(4), 934–943. h ps://doi.o g/10.2307/1952644
Gu in, P., Nagda, B. A., & Zúñiga, X. (2013). Dialogue ac oss di e ence: P ac ice, heo y, and esea ch on in e g oup
dialogue. Russell Sage Founda ion.
Hamake , E. L., Kuipe , R. M., & G asman, R. P. P. P. (2015). A c i ique o he c oss‐lagged panel model.
Psychological Me hods,20(1), 102–116. h ps://doi.o g/10.1037/a0038889
Ha e eld, E. (2021). F agmen ed oes: A ec i e pola iza ion in he mul ipa y con ex o he Ne he lands.
Elec o al S udies,71, A icle 102332. h ps://doi.o g/10.1016/j.elec s ud.2021.102332
Ha e eld, E., & Wagne , M. (2023). Does a ec i e pola isa ion inc ease u nou ? E idence om Ge many,
The Ne he lands and Spain. Wes Eu opean Poli ics,46(4), 732–759. h ps://doi.o g/10.1080/01402382.
2022.2087395
Helbling, M., & Jungkunz, S. (2020). Social di ides in he age o globaliza ion. Wes Eu opean Poli ics,43(6),
1187–1210.
Hess, R. D., & To ney‐Pu a, J. (2017). The de elopmen o poli ical a i udes in child en. Rou ledge.
Hobol , S. B., Leepe , T. J., & Tilley, J. (2021). Di ided by he o e: A ec i e pola iza ion in he wake o
he B exi e e endum. B i ish Jou nal o Poli ical Science,51(4), 1476–1493. h ps://doi.o g/10.1017/
S0007123420000125
Hooghe, M., & Wilken eld, B. (2008). The s abili y o poli ical a i udes and beha io s ac oss adolescence and
ea ly adul hood: A compa ison o su ey da a on adolescen s and young adul s in eigh coun ies. Jou nal
o You h and Adolescence,37(2), 155–167. h ps://doi.o g/10.1007/s10964‐007‐9199‐x
Hu, L.‐ ., & Ben le , P. M. (1999). Cu o c i e ia o i indexes in co a iance s uc u e analysis: Con en ional
c i e ia e sus new al e na i es. S uc u al Equa ion Modeling,6(1), 1–55. h ps://doi.o g/10.1080/
10705519909540118
Hube , G. A., & Malho a, N. (2017). Poli ical homophily in social ela ionships: E idence om online da ing
beha io . The Jou nal o Poli ics,79(1), 269–283. h ps://doi.o g/10.1086/687533
Hu chins, R. (2024). The poli ical pola iza ion o p e‐adolescen s: A ec i e pola iza ion in Ame ica’s you h.
Social Educa ion,88(1), 20–23. h ps://doi.o g/10.3102/00346543241288245
Iyenga , S., Lelkes, Y., Le endusky, M., Malho a, N., & Wes wood, S. J. (2019). The o igins and consequences
Social Inclusion •2025 •Volume 13 •A icle 10122 24
o a ec i e pola iza ion in he Uni ed S a es. Annual Re iew o Poli ical Science,22(1), 129–146. h ps://
doi.o g/10.1146/annu e ‐polisci‐051117‐073034
Iyenga , S., Sood, G., & Lelkes, Y. (2012). A ec , no ideology. Public Opinion Qua e ly,76(3), 405–431. h ps://
doi.o g/10.1093/poq/n s038
Iyenga , S., & Wes wood, S. J. (2015). Fea and loa hing ac oss pa y lines: New e idence on g oup pola iza ion.
Ame ican Jou nal o Poli ical Science,59(3), 690–707. h ps://doi.o g/10.1111/ajps.12152
Janssen, L., & Tu kenbu g, E. (2024). B eaking ee om linea assump ions: Un a elling he ela ionship
be ween a ec i e pola iza ion and democ a ic suppo . Eu opean Jou nal o Poli ical Resea ch. Ad ance
online publica ion. h ps://doi.o g/10.1111/1475‐6765.12725
Jennings, M. K., & Niemi, R. G. (1968). The ansmission o poli ical alues om pa en o child. Ame ican
Poli ical Science Re iew,62(1), 169–184.
Jos , J. T. (2017). Ideological asymme ies and he essence o poli ical psychology. Poli ical Psychology,38(2),
167–208. h ps://doi.o g/10.1111/pops.12407
Jos , J. T., Fede ico, C. M., & Napie , J. L. (2009). Poli ical ideology: I s s uc u e, unc ions, and elec i e a ini ies.
Annual Re iew o Psychology,60(1), 307–337. h ps://doi.o g/10.1146/annu e .psych.60.110707.163600
Kaise , J., Vacca i, C., & Chadwick, A. (2022). Pa isan blocking: Biased esponses o sha ed misin o ma ion
con ibu e o ne wo k pola iza ion on social media. Jou nal o Communica ion,72(2), 214–240. h ps://
doi.o g/10.1093/joc/jqac002
Kekkonen, A., Suu onen, A., Kawecki, D., & S andbe g, K. (2022). Puzzles in a ec i e pola iza ion esea ch:
Pa y a i udes, pa isan social dis ance, and mul iple pa y iden i ica ion. F on ie s in Poli ical Science,4,
A icle 920567. h ps://doi.o g/10.3389/ pos.2022.920567
Kinde , D. R., & Sea s, D. O. (1985). Public opinion and poli ical ac ion. In G. Lindzey & E. A onson (Eds.),
The handbook o social psychology (3 d ed., pp. 659–741). Random House.
Kingze e, J., D uckman, J. N., Kla , S., K upniko , Y., Le endusky, M., & Ryan, J. B. (2021). How a ec i e
pola iza ion unde mines suppo o democ a ic no ms. Public Opinion Qua e ly,85(2), 663–677. h ps://
doi.o g/10.1093/poq/n ab029
Kline, R. B., & Li le, T. D. (2023). P inciples and p ac ice o s uc u al equa ion modeling (5 h ed.). The Guil o d
P ess.
Kluge, A., Adle , E., Ni , L., Halpe in, E., Sams, M., & Le y, J. (2024). Asymme y in poli ical pola iza ion a
mul iple le els o bias. Poli ical Psychology,45(6), 1105–1121. h ps://doi.o g/10.1111/pops.12967
La ineu , C., Kaspe , J., B inke, L. T., Schumache , G., & Bakke , B. N. (2024). Poli ical ac i ism, in e es ,
and a ec i e pola iza ion: Du ch adolescen s and adul s a e no ha di e en . PsyA Xi . h ps://doi.o g/
10.31234/os .io/aux5h
Lay, J. C., Holman, M. R., G eenlee, J. S., Oxley, Z. M., & Bos, A. L. (2023). Pa isanship on he playg ound:
Exp essi e pa y poli ics among child en. Poli ical Resea ch Qua e ly,76(3), 1249–1264. h ps://doi.o g/
10.1177/10659129221132223
Lelkes, Y., & Wes wood, S. J. (2017). The limi s o pa isan p ejudice. The Jou nal o Poli ics,79(2), 485–501.
h ps://doi.o g/10.1086/688223
Len h, R. V. (2021). emmeans: Es ima ed ma ginal means, aka leas ‐squa es means (R package e sion 1.7.0).
h ps://CRAN.R‐p ojec .o g/package=emmeans
Le endusky, M. S. (2013). Why do pa isan media pola ize iewe s? Ame ican Jou nal o Poli ical Science,57(3),
611–623. h ps://doi.o g/10.1111/ajps.12008
Le endusky, M. S., & S ecula, D. A. (2021). We need o alk: How c oss‐pa y dialogue educes a ec i e pola iza ion
(1s ed.). Camb idge Uni e si y P ess. h ps://doi.o g/10.1017/9781009042192
Social Inclusion •2025 •Volume 13 •A icle 10122 25