Climate anxiety, wellbeing and pro-environmental action: correlates of negative emotional responses to climate change in 32 countries

Jou nal o En i onmen al Psychology 84 (2022) 101887 A ailable online 6 Oc obe 2022 0272-4944/© 2022 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY license (h p://c ea i ecommons.o g/licenses/by/4.0/). Clima e anxie y, wellbeing and p o-en i onmen al ac ion: co ela es o nega i e emo ional esponses o clima e change in 32 coun ies Cha les A. Ogunbode a , * , Rou en Do an b , Daniel Hanss c , Ma ia Ojala d , Ka a iina Salmela-A o e , Ka lijn L. an den B oek , g , Na jo Bhulla h , Sibele D. Aquino i , Tiago Ma o i , Julie Ai ken Sche me j , Anna Wloda czyk k , Su Lu l , Feng Jiang m , Daniela Acquad o Ma an n , Radha Yada o , Rahkman A di p , Razieh Chegeni b , Elahe Ghanba ian q , Somayeh Zand , Reza Naja i s , Joonha Pa k , Takashi Tsubaki a , Chee-Seng Tan u , JohnBosco Chika Chukwuo ji , Kehinde Ade emi Ojewumi w , Haj a Tahi b , Mai Albzou x , Ma c E ic S. Reyes y , Samuel Lins z , Viole a Enea aa , Ta iana Volkoda ab , Tomas Solla ac , Gin´ es Na a o-Ca illo ad , Jo ge To es-Ma ín ae , Win ed Mbungu a , A in H. Ayanian ag , Jihane Gho ayeb ah , Cha les Onyu ha ai , Michael J. Lomas aj , Mai Helmy ak , al , Lau a Ma ínez-Buel as am , Aydin Bayad ag , an , Mehme Ka asu ag a School o Psychology, Uni e si y o No ingham, UK b Depa men o Psychosocial Science, Facul y o Psychology, Uni e si y o Be gen, No way c Depa men o Social Sciences, Da ms ad Uni e si y o Applied Sciences, Ge many d School o Law, Psychology and Social Wo k, ¨ O eb o Uni e si y, Sweden e Depa men o Educa ional Sciences, Uni e si y o Helsinki, Finland Cope nicus Ins i u e o Sus ainable De elopmen , U ech Uni e si y, Ne he lands g Resea ch Cen e o En i onmen al Economics, Heidelbe g Uni e si y, Ge many h School o Psychology, Uni e si y o New England, New Sou h Wales, Aus alia i Psychology Depa men , Pon i ical Ca holic Uni e si y o Rio de Janei o, B azil j Depa men o Managemen and O ganisa ional S udies, Facul y o Social Science, Uni e si y o Wes e n On a io, Canada k Escuela de Psicología, Facul ad de Humanidades, Uni e sidad Ca ´ olica del No e, Chile l School o Applied Social Sciences, De Mon o Uni e si y, UK m Depa men o Human Resou ces and O ganisa ional Beha iou , Uni e si y o G eenwich, UK n Depa men o Psychology, Uni e si y o To ino, I aly o School o Business and Managemen , Ch is Uni e si y, India p Depa men o Psychology, Uni e si as Ai langga, Indonesia q Depa men o Educa ional Psychology and Counselling, Uni e si y o Teh an, I an Depa men o Psychology, Uni e si y o Milano-Bicocca, I aly s Depa men o Gene al Psychology, Uni e si y o Pado a, I aly Nagoya Uni e si y o Comme ce and Business, Japan u Depa men o Psychology and Counselling, Facul y o A s and Social Science, Uni e si i Tunku Abdul Rahman, Malaysia Depa men o Psychology, Uni e si y o Nige ia, Nige ia w Depa men o Psychology, Oba emi Awolowo Uni e si y, Nige ia x Depa men o Social and Beha iou al Science, Bi zei Uni e si y, Pales ine y Depa men o Psychology, College o Science, Uni e si y o San o Tomas, Philippines z Facul y o Psychology and Educa ion Sciences, Uni e si y o Po o, Po ugal aa Depa men o Psychology, Alexand u Ioan Cuza Uni e si y o Iasi, Romania ab Facul y o Pedagogy, Psychology and Communicology, Kuban S a e Uni e si y, Russia ac Depa men o Psychological Sciences, Facul y o Social Sciences and Heal hca e, Cons an ine he Philosophe Uni e si y in Ni a, Slo akia ad Depa men o Psychology, Uni e si y o Ja´ en, Spain ae Depa men o Resea ch Me hods in Beha iou al Sciences, Uni e si y o G anada, Spain a Depa men o Ci il and Wa e Resou ces Enginee ing, School o Enginee ing and Technology, Sokoine Uni e si y o Ag icul u e, Tanzania ag Ins i u e o In e disciplina y Resea ch on Con lic and Violence, Biele eld Uni e si y, Ge many ah Depa men o Psychology, Zayed Uni e si y, Dubai, Uni ed A ab Emi a es ai Depa men o Ci il and En i onmen al Enginee ing, Kyambogo Uni e si y, Uganda * Co esponding au ho . School o Psychology, Uni e si y o No ingham, B65 Psychology, Uni e si y Pa k, No ingham, NG7 2RD, Uni ed Kingdom. E-mail add ess: [email p o ec ed] (C.A. Ogunbode). Con en s lis s a ailable a ScienceDi ec Jou nal o En i onmen al Psychology jou nal homepage: www.else ie .com/loca e/jep h ps://doi.o g/10.1016/j.jen p.2022.101887 Recei ed 8 Feb ua y 2022; Recei ed in e ised o m 1 Augus 2022; Accep ed 1 Oc obe 2022 Jou nal o En i onmen al Psychology 84 (2022) 101887 2 aj School o Heal h and Socie y, Uni e si y o Sal o d, UK ak Depa men o Psychology, Sul an Qaboos Uni e si y, Oman al Psychology Depa men , Facul y o A s, Menuo ia Uni e si y, Egyp am Uni e sidad Tecnol´ ogica de Bolí a , Ca agena, Colombia an Van Yüzüncü Yil Uni e si y, Tu key ARTICLE INFO Handling Edi o : W. Schul z Keywo ds: Clima e change anxie y Clima e change Wellbeing P o-en i onmen al beha iou Clima e ac i ism Emo ions ABSTRACT This s udy explo ed he co ela es o clima e anxie y in a di e se ange o na ional con ex s. We analysed c oss- sec ional da a ga he ed in 32 coun ies (N =12,246). Ou esul s show ha clima e anxie y is posi i ely ela ed o a e o exposu e o in o ma ion abou clima e change impac s, he amoun o a en ion people pay o clima e change in o ma ion, and pe cei ed desc ip i e no ms abou emo ional esponding o clima e change. Clima e anxie y was also posi i ely linked o p o-en i onmen al beha iou s and nega i ely linked o men al wellbeing. No ably, clima e anxie y had a signi ican in e se associa ion wi h men al wellbeing in 31 ou o 32 coun ies. In con as , i had a signi ican associa ion wi h p o-en i onmen al beha iou in 24 coun ies, and wi h en i on- men al ac i ism in 12 coun ies. Ou indings highligh con ex ual bounda ies o engagemen in en i onmen al ac ion as an an ido e o clima e anxie y, and he b oad in e na ional signi icance o conside ing nega i e clima e- ela ed emo ions as a plausible h ea o wellbeing. 1. In oduc ion Clima e change is ecognised as a majo h ea by mos people a ound he wo ld. In a 2021 su ey o 76,328 Facebook use s ac oss 31 coun ies, mos esponden s epo ed ha hey we e “somewha ” o “ e y” wo ied abou clima e change (Leise owi z e al., 2021). As he acu e and long- e m e ec s o clima e change become a li ed expe ience o a g owing p opo ion o he wo ld’s popula ion, he e also appea s o be a co esponding ise in he numbe o people expe iencing psycho- logical dis ess in ela ion o he en i onmen and clima e c isis (Ojala e al., 2021; Pihkala, 2020). Commonly e med “eco-anxie y” o “clima e anxie y”, he expe ience o en i onmen - ela ed dis ess en- compasses nega i e emo ions like ea , wo y, guil , shame, hopeless- ness and despai (Ojala e al., 2021). I has been no ed ha anxie y has simply eme ged as a concep used by esea che s o ep esen a wide ange o nega i e emo ions people ha e abou clima e change (Clay on, 2020). To da e, mains eam discou se and esea ch ega ding nega i e emo ional esponses o clima e change ha e p edominan ly ocused on he expe iences and pe spec i es o a whi e and Wes e n demog aphic (Ray, 2021). Ye , he h ea posed by clima e change is by no means limi ed o he Global No h. In a ecen in e na ional poll o young adul s, mos esponden s in ou coun ies o he Global Sou h: B azil (86%), India (80%), Nige ia (70%), and he Philippines (92%), epo ed eeling igh ened abou he u u e because o clima e change (Hickman e al., 2021). Clima e anxie y has also been linked o poo men al heal h among a sample o Filipino young adul s (Reyes e al., 2021). The cu - en s udy ad ances eme ging e o s o di e si y clima e- ela ed emo- ions esea ch. We explo ed plausible p edic o s o clima e anxie y, and examined how clima e anxie y ela es o p o-en i onmen al ac ion and men al wellbeing in 32 coun ies. 1.1. Ope a ionalising clima e anxie y Nega i e emo ions a e key d i e s o human ac ion (Webe , 2006). The e o e, widesp ead nega i e emo ional esponses o clima e change a e a posi i e indica o o he po en ial o la ge-scale social and poli - ical ac ion on he clima e c isis. Howe e , nega i e emo ions can also ha e ad e se implica ions o human wellbeing and psychological unc ioning. Ea lie esea ch showed a link be ween clima e change conce n and psychological dis ess (Sea le & Gow, 2010). Mo e ecen schola ship on he ela ionship be ween nega i e clima e- ela ed emo- ions and wellbeing ha e coalesced a ound he concep o clima e anx- ie y – a concep closely ela ed o ea and wo y (Pihkala, 2020). Clima e anxie y can be conside ed a subse o eco-anxie y which is de ined by (Clay on e al., 2017) as a ch onic ea o ecological doom. Schola s sugges ha clima e anxie y can be igge ed by a loss o places, ac i i ies and adi ions due o clima e change o ea o he po en ial scope and impac o dange ous clima e change (Clay on, 2021). I is no conside ed o be inhe en ly pa hological (i.e., i is no a clinical diso de ). None heless, Clay on and Ka azsia (2020) de eloped a measu e o clima e anxie y ha emphasizes he po en ial o psycho- logical impai men . Thei scale is based on clinical measu es o unc- ional impai men and umina ion. I assesses he deg ee o which hinking abou clima e change makes i di icul o people o sleep, socialise, egula e hei emo ions, and concen a e on wo k o s udying. In hei s udies o US esiden s, hey ound ha a ound one- i h o people expe ience hese e ec s a leas some o he ime. The cu en s udy di e ges om Clay on and Ka aszia’s (2020) app oach by ocusing on he nega i e emo ionali y embodied by clima e anxie y. 1 We do no p esume an o e lap wi h psychological impai men . Ins ead, ou app oach aligns wi h Pihkala’s (2020) a gumen o a oid na owly es ic ing concep ualisa ions o clima e anxie y only o he s onge anxie y symp oms (pg. 12). We used a measu e o nega i e clima e- ela ed emo ions ha is based on he ‘S a e Anxie y’ dimension o he S a e-T ai Anxie y In en o y (STAI: Spielbe ge , 1983). S a e anxie y cap u es he in ensi y o an indi idual’s eeling o anxie y as an emo ional s a e a a gi en poin in ime. High le els o s a e anxie y a e in e p e ed as e lec ing nega i e emo ional esponses o si ua ional in luences such as an icipa ed o p esen dange , o pas auma ic e en s associa ed wi h he p esen si ua ion (Spielbe ge & Reheise , 2009) Co espondingly, ou ope a ionalisa ion o clima e anxie y e- lec s a s a e o heigh ened nega i e emo ionali y due o pe cei ed h ea om clima e change. 2 This cons uc di e s om concep s like clima e change wo y o conce n by cap u ing no jus he eeling o being 1 Clay on’s (2020) accoun o he e idence o clima e anxie y a ound he wo ld de ails epo s o s ong emo ional esponses o clima e, such as high le els o wo y and emo ional dis ess, among a ious popula ions. This ac- coun , as well as ha o o he s such as Pihkala (2020), does no suppo he assump ion o unc ional o emo ional impai men as de ining elemen s o clima e anxie y. 2 In addi ion o being based on he STAI, ou clima e anxie y measu e also closely mi o s he ‘clima e anxie y’ dimension (Thinking abou clima e change makes me eel – ense, anxious, wo ied, ang y, conce ned, s essed, sad, sca ed, dep essed) o Sea le and Gow’s (2010) clima e dis ess scale. Ange is no ep esen ed in ou clima e anxie y measu e because i e lec s a di e en deg ee o ac i a ion han anxie y o wo y and has been shown o ha e a di e en pa e n o associa ion wi h men al heal h and p o-en i onmen al ac- ion (e.g., S anley e al., 2021). C.A. Ogunbode e al. Jou nal o En i onmen al Psychology 84 (2022) 101887 3 wo ied o anxious, bu also he associa ed men al s a es o being ense, calm, peace ul o un elaxed. I is he e o e a mo e holis ic ep esen a- ion o he expe ience o in ense nega i e emo ionali y. 1.2. An eceden s o clima e anxie y One way nega i e emo ions a ise is om app aisals o e en s o si ua ions in ligh o hei implica ion o wellbeing (Moo s e al., 2013). He e, implica ions o wellbeing encompass he po en ial ul ilmen o obs uc ion o he app aise ’s needs, alues, goals, belie s and o he in e es s. Social cogni i e heo y (Bandu a, 1986) indica es ha in- di iduals’ emo ions and cogni ions a e shaped by con ex ual in luences ha p o ide in o ma ion h ough ins uc ion, modelling and social pe suasion. We p opose ha he knowledge ha unde lies eelings o clima e anxie y may be acqui ed h ough di ec pe sonal expe iences (Rese e al., 2014), o indi ec ly h ough social in e ac ion and media consump ion (Rosen hal, 2022). Below, we ou line he empi ical con- side a ions ha amed ou hypo heses ega ding he plausible an e- ceden s o clima e anxie y. 1.2.1. Pe sonal expe ience Exposu e o os ensible clima e change impac s, as well as people’s pe cei ed subjec i e expe iences o clima e change, a e an in eg al aspec o isk pe cep ions and emo ional engagemen wi h he issue (Rese e al., 2014). People o en d aw on hei pe sonal expe iences wi h ex eme wea he o make in e ences abou he eali y o clima e change (McDonald e al., 2015). Exposu e o ex eme e en s like looding and hu icanes can heigh en nega i e eelings abou clima e change by making clima e change eel mo e conc e e and immedia e (Be gquis e al., 2019; Mye s e al., 2013; Spence e al., 2011). Th ough a p ocess o expe ien ial lea ning, nega i e emo ions igge ed by ex eme wea he expe iences o en become in ui i ely associa ed wi h clima e change and can subsequen ly eed in o he way people pe cei e clima e isks (Ma x e al., 2007). Some s udies sugges ha ex eme wea he expe iences may ha e a ansien e ec on clima e change conce n (Konisky e al., 2016), o no e ec a all (Whi ma sh, 2008). O he s indica e ha indi iduals’ alues, p e-exis ing belie s and subjec i e a ibu ion o hei expe iences o en mode a e how ex eme wea he expe iences shape esponses o clima e change (B uine de B uin e al., 2014; Ogunbode e al., 2017, 2019, 2020b). Like clima e- ela ed emo ions, much o he esea ch on ex eme wea he expe iences and clima e change engagemen has ocused on Global No h con ex s. In he cu en s udy, we explo ed how pe sonal expe iences o ex eme wea he , pa icula ly looding expe iences, ela e o clima e anxie y. We chose o ocus on looding expe iences because looding is he mos common wea he - ela ed disas e occu ing a ound he wo ld (CRED, 2015). Speci ically, we hypo hesized ha : H1. P io looding expe ience is posi i ely associa ed wi h clima e anxie y 1.2.2. Social no ms D awing om he ocus heo y o no ma i e conduc (Cialdini e al., 1991), he e a e wo ca ego ies o social no ms. The i s is desc ip i e no ms, which e e s o a pe son’s pe cep ion o wha mos people do in a gi en si ua ion. The second is injunc i e no ms, which e e s o pe cep- ions o wha beha iou s impo an social e e en s (e.g., amily, iends, colleagues o neighbou s) app o e o disapp o e o in speci ic si ua- ions. People a e mo e likely o enac adap i e beha iou al esponses o clima e change when hey pe cei e social no ms o be suppo i e o such beha iou ( an Valkengoed & S eg, 2019). Social no ms also seem o ma e o clima e change pe cep ions and emo ions mo e b oadly. Fo example, people epo a g ea e sense o h ea om clima e change when hey belie e ha impo an social e e en s a e engaging in clima e ac ion o when hey eel social p essu e o pe sonally ac on clima e change ( an de Linden, 2014). Simila ly, people a e mo e likely o exp ess wo y abou clima e change when close o he s, such as iends and amily membe s, a e pe cei ed o also ca e abou clima e change (Goldbe g, an de Linden, Leise owi z, & Maibach, 2020). Unce ain y abou he speci ic iming, loca ion and e ec s o clima e change impac s is said o be an elemen o clima e anxie y (Clay on, 2021). We also know ha people a e mo e likely o use he beha iou o o he s as guidance on how o beha e in si ua ions ha a e no el, unce ain o ambiguous (Cialdini, 2001). The e o e, pe cei ing s ong nega i e emo ional esponses o clima e change among key social e e en s like iends and amily membe s (desc ip i e no ms) may os e eelings o clima e anxie y by emphasizing he h ea posed by clima e change, o in o he wo ds p o iding social p oo ha s ong nega i e emo ional esponses a e an app op ia e esponse o he h ea . We hypo hesized ha : H2. Pe cei ed desc ip i e no ms o nega i e emo ional esponses o clima e change among key social e e en s a e posi i ely associa ed wi h clima e anxie y 1.2.3. Media exposu e Schola s ha e specula ed ha ising a es o exposu e o en i on- men al in o ma ion ia he media may be leading o inc eased nega i e emo ions and poo wellbeing (S okols e al., 2009), bu his specula ion has ye o be suppo ed wi h empi ical e idence in he con ex o clima e anxie y. Mo e b oadly, e idence o he e ec o media exposu e on clima e change a i udes is mixed. A p io s udy obse ed no signi ican e ec o in o ma ion exposu e on clima e change pe cep ions among an indigenous communi y in Boli ia (Fe n´ andez-Llamaza es e al., 2015). Ano he s udy showed ha exposu e o clima e change- ela ed media h ough he ele ision p edic ed g ea e isk pe cep ions in India, bu in e ne use showed a nega i e e ec and newspape use showed no e ec (Thake e al., 2017). Exposu e o clima e change- ela ed media has also been posi i ely linked wi h clima e change conce n in Japan (Sampei & Aoyagi-Usui, 2009). In No way, exposu e o he 2018 IPCC epo on 1.5 ◦C global wa ming was ound o be associa ed wi h heigh ened clima e change conce n, and mos esponden s in he s udy encoun e ed he epo h ough he media (Ogunbode e al., 2020a). The mixed e idence o he e ec s o media exposu e on a i udes owa d clima e change may be a esul o he ac ha mos p io s udies ail o explici ly accoun o di e ences in he con en and alence o in o ma ion con eyed o audiences by he media. Whe e he ocus o esea ch has explici ly been on media co e age o e en s ha migh be desc ibed as plausible impac s o clima e change, pa icula ly na u al disas e s, signi ican associa ions ha e consis en ly been ound wi h in ensi ied nega i e emo ions (Hous on, Fi s , & Dan o h, 2018; P e - e baum e al., 2014). This associa ion is especially p ominen o pos - auma ic s ess esponses, bu o he eac ions like dis ess and anxie y ha e also been epo ed (P e e baum e al., 2014). Exposu e o media co e age o clima e change impac s like na u al disas e s may os e s ong nega i e emo ional eac ions by eeding ex emely nega- i e app aisals o he isks posed o onesel o alued o he s by such impac s (Ehle s & Cla k, 2000; Hous on, Spialek, & Fi s , 2018). The e o e, we hypo hesized ha : H3a. Ra e o exposu e o in o ma ion abou clima e change impac s in he media is posi i ely associa ed wi h clima e anxie y In addi ion, we also conside ed he possibili y ha exposu e o in- o ma ion ha ocuses on clima e change solu ions could educe nega- i e clima e- ela ed emo ions. Fo example, p e ious esea ch has ound ha people epo educed le els o dis ess when p esen ed wi h op i- mis ic in o ma ion abou p og ess in educing global ca bon emissions (Ho nsey & Fielding, 2016). Thus, we hypo hesized ha : H3b. Ra e o exposu e o in o ma ion abou clima e change solu ions in he media is in e sely ela ed o clima e anxie y O e and abo e he me e a e o exposu e o clima e- ela ed media, C.A. Ogunbode e al. Jou nal o En i onmen al Psychology 84 (2022) 101887 4 le els o a en ion o such in o ma ion may also signi ican ly p edic he in ensi y o nega i e emo ional esponses. Media exposu e and media a en ion a e concep ually dis inc and ha e unique ela ionships wi h speci ic media e ec s (D ew & Wea e , 2016). While media exposu e e e s o he quan i y o media iewed o hea d ega ding a speci ic opic, media a en ion e e s o he inclina ion o channel cogni i e e- sou ces in o p ocessing he messages con eyed ia hese media (Lee & Cho, 2019). Some s udies sugges ha a en ion o media may be a be e p edic o o media e ec s han media exposu e alone. Fo example, a en ion o news abou social isks, like c ime, has been shown o be mo e s ongly p edic i e o isk judgemen s han sel - epo ed media use (Sla e & Rasinski, 2005). Simila ly, a en ion o sus ainabili y media has been shown o p edic p o-en i onmen al beha iou and g een consume ism mo e s ongly han exposu e o sus ainabili y media (Lee & Cho, 2019). Hence, we hypo hesized ha : H3c. A en ion o clima e change- ela ed media is posi i ely associ- a ed wi h clima e anxie y The li e a u e is unclea abou how media exposu e and media a en ion in e ac . None heless, i is no possible o a end o in o ma ion o which one is no exposed, and media exposu e is unlikely o ha e any signi ican e ec s i he e is li le mo i a ion o a end o o p ocess he in o ma ion con eyed. Consequen ly, we es ed wo explo a o y hypo heses: H3d. A en ion o clima e change- ela ed media in e ac s wi h expo- su e o in o ma ion abou clima e change impac s in p edic ing clima e anxie y, whe eby exposu e o in o ma ion abou clima e change impac s has a s onge posi i e ela ionship clima e anxie y when a en ion is high. H3e. A en ion o clima e change- ela ed media in e ac s wi h expo- su e o in o ma ion abou clima e change solu ions in p edic ing clima e anxie y, whe eby exposu e o in o ma ion abou clima e change solu- ions has a s onge in e se ela ionship wi h clima e anxie y when a en ion is high. 1.3. Ou comes o clima e anxie y: p o-en i onmen al ac ion and wellbeing Some au ho s ha e sugges ed ha clima e anxie y o en mani es s i sel as a o m o ‘p ac ical anxie y’ by engende ing in o ma ion-seeking and causing people o e hink hei li es yles and beha iou (Pihkala, 2020). Indeed, wo y and anxie y abou clima e change ha e been linked wi h clima e change policy suppo and p o-en i onmen al beha iou (Bouman e al., 2020; S anley e al., 2021; Wullenko d e al., 2021). Howe e , i is impo an o dis inguish be ween di e en ca ego ies o p o-en i onmen al beha iou s as di e en combina ions o causal ac o s may be associa ed wi h each ca ego y. Acco ding o S e n (2000), en i onmen ally-signi ican beha iou s can be di ided in o ou main ca ego ies: en i onmen al ac i ism (e.g., in ol emen in en i- onmen al p o es s o demons a ions), non-ac i is beha iou s in he public sphe e (e.g., pe i ioning on en i onmen al issues), p i a e-sphe e beha iou s (e.g., ecycling and g een consume ism) and beha iou s in o ganisa ions (e.g., a endance o en i onmen al c i e ia in decision-making and o ganisa ional p ac ices). In he cu en s udy, we assessed how clima e anxie y ela es o p i a e-sphe e p o-- en i onmen al beha iou s (hence o h e med p o-en i onmen al beha iou ) and engagemen in en i onmen al ac i ism. Impo an ly, clima e anxie y may also ha e signi ican nega i e implica ions o men al wellbeing (Ojala e al., 2021). A g owing num- be o s udies show ha clima e anxie y is ela ed o dep essi e symp- oms, poo men al heal h, and lowe ed psychological wellbeing (Ogunbode e al., 2021; Reyes e al., 2021; S anley e al., 2021; Wul- lenko d e al., 2021). The na u e o he ela ionship be ween clima e anxie y and wellbeing is complica ed, and he la gely co ela ional na- u e o he a ailable e idence makes i di icul o asce ain he di ec ion o causal in luence be ween he wo ac o s. Clima e anxie y can be expec ed o ha e a deg ee o co ela ion wi h poo men al wellbeing due o he sha ed elemen o nega i e a ec (Ojala e al., 2021). By and la ge, a signi ican ela ionship has mo e commonly been ound be ween clima e anxie y and men al heal h o men al wellbeing when he mea- su e o clima e anxie y also includes indices o pa hology o dys unc ion (e.g., Clay on & Ka azsia, 2020; Reyes e al., 2021). S udies ha ocus pu ely on nega i e clima e emo ions like wo y ha e ended o ind no meaning ul ela ionships wi h men al heal h (Be y & Peel, 2015; Ve - planken & Roy, 2013). Conside ing ha clima e change has become mo e acu e in he ime since hese la e s udies we e published, we we e in e es ed in explo ing how ou emo ionali y- ocused measu e o clima e anxie y ela es o men al wellbeing. The e o e, we hypo hesized ha : H4a. Clima e anxie y is posi i ely associa ed wi h p o-en i onmen al beha iou H4b. Clima e anxie y is posi i ely associa ed wi h engagemen in en i onmen al ac i ism H5. Clima e anxie y is in e sely ela ed o men al wellbeing. 2. Me hod 2.1. Pa icipan s and p ocedu e We ga he ed da a in 32 coun ies (N =12, 246; M age =23.1 yea s, SD age =7.0 yea s, Women =63.9%, Men =32.7%) using a s uc u ed ques ionnai e. Da a collec ion was conduc ed in 25 o he coun ies in au umn 2019 bu con inued in h ee coun ies (India, Pales ine, and he Uni ed A ab Emi a es) un il la e summe 2020, and a u he ou coun ies un il sp ing 2022 (Colombia, Egyp , Oman and Tu key) due o in e up ion by he onse o he COVID-19 pandemic. The ques ionnai e was adminis e ed in pape -and-pencil o ma o online, depending on wha was mos p ac ical in each coun y. In all coun ies, we ec ui ed esponden s om uni e si y esea ch pa icipan pools o h ough op- po uni y sampling o uni e si y s uden s (See Supplemen a y Da a Table 1 o a coun y-wise demog aphic b eakdown o he sample). Pa icipa ion in he s udy was s ic ly by olun a y consen . Rewa ds (e. g., cou se c edi , en y in o a a le) we e o e ed o pa icipan s in a ew coun ies in acco dance wi h local con en ions. A he ou se , we se a minimum a ge o N =200 pa icipan s o each coun y, bu he inal sample size was de e mined by he numbe o accessible olun a y pa icipan s. The ques ionnai e was o iginally p epa ed in English and subse- quen ly ansla ed o o he ele an languages using a sys em o ansla ion-back- ansla ion in ol ing a leas one bilingual speake . Whe e a language was spoken in mo e han one coun y (e.g., A abic, Spanish, Po uguese), he ansla o s wo ked in c oss-coun y eams o de elop and check he ansla ions be o e ielding he su ey. E hical app o al o he s udy was g an ed by he De Mon o Uni e si y Heal h and Li e Sciences Facul y Resea ch E hics Commi ee (Re : 3434), wi h supplemen a y app o al also ob ained in coun ies whe e his was equi ed. 2.2. Measu es 2.2.1. Clima e anxie y We measu ed clima e anxie y wi h a 7-i em scale based on he s a e anxie y componen o he S a e-T ai Anxie y In en o y (Spielbe ge , 1983). Pa icipan s we e ins uc ed o collec hei hough s and ocus on hei eelings ega ding clima e change. Subsequen ly, hey we e di ec ed o indica e he deg ee o which hey el calm, ense, elaxed, anxious, peace ul, wo ied o e i ied abou clima e change a ha momen in ime. Ra ings o each emo ion we e eco ded wi h a 5-poin esponse scale anging om 1 (no a all) o 5 (ex emely) and sco es o C.A. Ogunbode e al. Jou nal o En i onmen al Psychology 84 (2022) 101887 5 ‘calm’, ‘ elaxed’, and ‘peace ul’ we e e e sed p io o analysis. The scale showed a good le el o eliabili y ac oss coun ies wi h C onbach’s alpha sco es anging om 0.71 o 0.92 (Table 1). 2.2.2. Flooding expe ience Pas expe ience wi h looding was measu ed wi h an i em adop ed om an de Linden (2014). Pa icipan s we e asked o indica e how o en hey had expe ienced (a) looding in hei local a ea o e he las i e yea s. Responses we e eco ded on a 5-poin scale anging om Table 1 Desc ip i e s a is ics and eliabili y indices o clima e anxie y ac oss coun ies. Coun y α M(SD) % e y o ex emely “ ense” % e y o ex emely “anxious” % e y o ex emely “wo ied” % e y o ex emely “ e i ied” N 1. Aus alia .90 3.22 (.90) 25.2 30.6 44.9 14.0 314 2. B azil .91 3.76 (.91) 51.4 43.6 73.6 33.3 280 3. Canada .88 3.12 (.78) 18.4 21.7 40.1 14.9 309 4. Chile .83 3.31 (.74) 22.5 30.7 47.8 21.5 391 5. China .89 2.86 (.81) 19.1 22.1 33.7 15.0 267 6. Colombia .79 3.13 (.68) 16.4 19.5 47.1 17.6 159 7. Egyp .69 2.96 (.66) 28.5 26.0 31.8 25.0 428 8. Finland .92 3.26 (.96) 29.1 38.7 65.7 12.3 633 9. Ge many .87 3.39 (.84) 32.1 27.6 63.0 36.2 246 10. I aly .89 3.41 (.73) 15.6 23.8 45.6 24.1 294 11. India .81 3.12 (.76) 25.3 22.3 37.6 21.4 228 12. Indonesia .84 3.32 (.68) 14.8 28.7 42.9 29.6 345 13. I an .86 3.32 (.84) 34.7 28.5 42.1 26.6 322 14. Japan .73 3.05 (.65) 8.6 25.8 29.8 11.6 302 15. Malaysia .86 3.06 (.71) 14.8 15.2 33.3 18.9 264 16. Ne he lands .88 3.20 (.78) 21.7 24.5 57.7 11.8 415 17. Nige ia .71 2.84 (.73) 22.0 26.8 27.2 25.1 584 18. No way .87 3.09 (.81) 15.6 27.5 48.5 9.5 262 19. Oman .65 2.78 (.63) 19.8 20.9 23.7 18.4 278 20. Pakis an .81 3.03 (.80) 26.5 21.6 37.6 26.9 245 21. Pales ine .80 2.98 (.78) 21.6 28.2 26.0 19.7 360 22. Philippines .85 3.55 (.73) 28.4 37.1 63.0 44.5 1418 23. Po ugal .87 3.43 (.77) 31.8 30.2 68.2 16.3 258 24. Romania .86 2.90 (.84) 13.9 14.4 30.6 17.7 395 25. Russia .75 2.71 (.62) 7.5 5.0 9.6 7.8 477 26. Slo akia .83 3.00 (.77) 11.2 27.1 32.6 14.7 258 27. Spain .86 3.66 (.72) 35.8 24.6 77.6 34.9 590 28. Tanzania .75 2.91 (.72) 19.0 19.0 28.6 21.4 248 29. Tu key .83 3.56 (.68) 38.7 50.0 49.0 23.0 418 30. Uganda .84 3.18 (.87) 27.0 31.4 45.6 35.1 653 31. UAE .87 3.33 (.90) 29.2 39.4 48.7 32.7 219 32. UK .88 3.26 (.80) 20.3 29.2 44.1 15.4 370 To al .85 3.21 (.82) 24.2 28.2 45.9 24.0 12,246 C.A. Ogunbode e al. Jou nal o En i onmen al Psychology 84 (2022) 101887 6 “ne e ” o “mo e han 3 imes”. Sco es we e subsequen ly unca ed o c ea e 3 ca ego ies (0 =no lood expe ience, 1 =expe ienced looding once o wice, 2 =expe ienced looding 3 imes o mo e). Responden s who indica ed ha hey could no emembe i hey had expe ienced any such e en we e coded as missing. 2.2.3. Desc ip i e no ms We measu ed desc ip i e no ms ega ding nega i e emo ional e- sponses o clima e change wi h ou i ems (e.g., “mos people close o me a e wo ied abou he u u e e ec s o clima e change”, “mos people close o me exp ess eelings o dis ess when alking abou clima e change”). These i ems we e combined o ob ain an agg ega e indica o o pe cei ed desc ip i e no ms ( α ange =0.63–0.84) ha su passed he con en ional eliabili y h eshold ( α =0.70; Nunnally, 1970) in all bu a e y small mino i y o coun ies. A coun y-wise b eakdown o he eliabili y indices o he pe cei ed desc ip i e no ms measu e is p o- ided as supplemen a y da a (Supplemen a y Table 2). 2.2.4. Media exposu e and a en ion We asked pa icipan s o indica e he equency a which hey ead o hea abou clima e change impac s and clima e change solu ions. Re- sponses o hese ques ions we e eco ded using he 9-poin esponse scale anging om “Ne e ” o “Mo e han 10 imes pe day”. We also asked pa icipan s o indica e he amoun o a en ion hey pay o clima e change- ela ed in o ma ion on a 4-poin scale anging om ‘none’ o ‘a lo ’ (Sla e & Rasinski, 2005). Desc ip i e s a is ics o hese measu es a e p esen ed in Supplemen a y Table 3. 2.2.5. P o-en i onmen al beha iou and en i onmen al ac i ism P o-en i onmen al beha iou was measu ed wi h a mul i-i em index o sus ainable consump ion beha iou s adap ed om Ojala (2012). Pa icipan s we e asked o indica e how o en hey engage in each o eigh p o-en i onmen al beha iou s: (1) cycle o walk ins ead o d i ing, (2) es ain onesel om buying unneeded new clo hes, (3) choose no o ly, (4) y o in luence amily and iends o ac p o-en i onmen ally, (5) sa e ene gy in he household, (6) ake public anspo a ion ins ead o he ca , (7) a oid ood was e, and (8) make clima e- iendly ood choices. Responses o hese i ems we e a ed be- ween “1 =almos ne e ” and “5 =almos always”. The agg ega ed scale su passed he con en ional eliabili y h eshold ( α =0.70) in 24 ou o 32 coun ies ( α ange =0.56–0.85). We also measu ed en i onmen al ac i ism by asking pa icipan s i hey had a ended a clima e p o es a any poin in he pas yea up ill he ime o da a collec ion. Responses o his ques ion we e coded as “No” =0 and “Yes” =1. A small mino i y (13.4%) o he o al sample indica ed ha hey had a ended a clima e p o es (Table 2). 2.2.6. Men al wellbeing We measu ed men al wellbeing using he WHO-5 wellbeing index (WHO, 1998). The scale includes 5 i ems in which pa icipan s a e asked o a e how o en o e he p eceding wo weeks: (1) hey ha e el chee ul and in good spi i s, (2) hey ha e el calm and elaxed, (3) hey ha e el ac i e and igo ous, (4) hey woke up eeling esh and es ed, and (5) hei li e has been illed wi h hings ha in e es hem. Re- sponses o hese ques ions a e eco ded on a 6-poin scale anging om 0 (A no ime) o 5 (All o he ime). The scale i ems we e agg ega ed o o m a eliable measu e ac oss all he coun ies ( α ange =0.71–0.89). Coun y-wise desc ip i e s a is ics o men al wellbeing a e p esen ed in Table 2. 2.3. Analysis Due o he nes ed na u e o he da a (indi iduals wi hin coun ies), we es ed ou hypo heses wi h hie a chical linea models using he R Table 2 Desc ip i e s a is ics and eliabili y indices o p o-en i onmen al beha iou , en i onmen al ac i ism, and men al wellbeing ac oss coun ies. Coun y P o-en i onmen al beha iou A ended clima e p o es WHO-5 wellbeing index N α M(SD) %Yes α M(SD) 1. Aus alia .85 3.01 (0.87) 9.2 .89 2.72 (1.00) 314 2. B azil .78 3.39 (0.81) 15.0 .86 2.37 (1.01) 280 3. Canada .81 2.99 (0.80) 5.8 .86 2.59 (0.99) 309 4. Chile .69 3.35 (0.74) 21.0 .85 2.63 (0.96) 391 5. China .79 3.54 (0.72) 0.4 .87 2.59 (0.86) 267 6. Colombia .70 3.45 (0.73) 10.7 .86 2.83 (0.97) 159 7. Egyp .68 3.31 (0.68) 11.4 .75 2.57 (1.05) 428 8. Finland .84 3.85 (0.76) 31.4 .82 2.73 (0.82) 633 9. Ge many .79 3.48 (0.77) 28.5 .71 2.67 (0.79) 246 10. I aly .72 3.39 (0.70) 23.5 .80 2.44 (0.80) 294 11. India .72 3.57 (0.68) 19.7 .81 3.17 (0.98) 228 12. Indonesia .56 3.33 (0.53) 7.8 .82 2.65 (0.82) 345 13. I an .69 3.37 (0.68) 5.3 .89 2.59 (1.17) 322 14. Japan .60 2.53 (0.66) 2.0 .81 2.86 (1.00) 302 15. Malaysia .75 3.22 (0.66) 5.3 .87 2.90 (0.90) 264 16. Ne he lands .82 3.73 (0.75) 35.3 .85 2.80 (0.88) 415 17. Nige ia .72 3.07 (0.72) 10.3 .77 3.19 (0.93) 584 18. No way .81 3.61 (0.75) 23.3 .79 2.90 (0.85) 262 19. Oman .65 2.87 (0.59) 16.5 .81 2.89 (0.98) 278 20. Pakis an .61 3.14 (0.64) 18.4 .75 2.59 (1.02) 245 21. Pales ine .74 3.06 (0.71) 6.8 .75 2.94 (0.99) 360 22. Philippines .70 3.67 (0.63) 5.1 .85 2.65 (0.93) 1418 23. Po ugal .72 3.52 (0.71) 13.6 .88 2.72 (0.97) 258 24. Romania .76 2.94 (0.78) 3.0 .80 2.85 (0.86) 395 25. Russia .69 3.43 (0.75) 7.8 .80 3.04 (0.89) 477 26. Slo akia .71 3.23 (0.74) 2.7 .78 2.47 (0.79) 258 27. Spain .70 3.79 (0.66) 32.0 .80 2.57 (0.80) 590 28. Tanzania .69 3.55 (0.69) 25.4 .76 3.42 (0.95) 248 29. Tu key .73 3.53 (0.70) 8.4 .82 2.27 (0.93) 418 30. Uganda .73 3.55 (0.73) 11.6 .74 2.99 (0.95) 653 31. UAE .70 2.96 (0.67) 5.8 .83 2.47 (0.89) 219 32. UK .76 3.14 (0.75) 7.6 .84 2.48 (0.94) 370 To al .75 3.39 (0.77) 13.4 .82 2.71 (0.96) 12,246 C.A. Ogunbode e al. Jou nal o En i onmen al Psychology 84 (2022) 101887 7 package lme4 o mixed e ec s models (Ba es e al., 2014). Le el 1 ixed e ec s o he model p edic ing clima e anxie y we e indi idual-le el a iables: looding expe ience, desc ip i e no ms, media exposu e and a en ion, wi h coun y-le el andom in e cep s added. Fo models p edic ing p o-en i onmen al beha iou , ac i ism and men al well- being, clima e anxie y was included as a Le el 1 p edic o . In he i s s ep o model-building, we es ima ed uncons ained models o each dependen a iable wi h no p edic o s. These models showed ha he g ouping ac o (coun y), explained app oxima ely 8% a iance in men al wellbeing, 10% a iance in nega i e clima e- ela ed emo ions, 15% a iance in p o-en i onmen al beha iou , and 21% a iance in en i onmen al ac i ism. This demons a es a signi ican clus e ing e ec in he da a. Addi ionally, he analysis accoun ed o po en ial c oss-na ional a ia ion in how clima e anxie y ela es o he beha iou al and well- being ou comes. The e o e, we es ima ed a model con aining Le el 1 p edic o s and andom coun y-le el in e cep s only o each ou come a iable (p o-en i onmen al beha iou , en i onmen al ac i ism, and men al wellbeing) and compa ed ha wi h ano he model con aining he Le el 1 p edic o s, andom coun y-le el in e cep s and andom slopes o clima e anxie y using a likelihood a io es . The esul s o he model wi h andom slopes o clima e anxie y is epo ed whe e he likelihood a io es shows ha his model had a be e i o he da a. P edic o a iables we e g and mean-cen ed p io o analysis and missing alues we e add essed wi h lis wise dele ion. The analyses con olled o pa icipan s’ age and gende . 3. Resul s 3.1. P elimina y obse a ions A conside able p opo ion o pa icipan s epo ed s ong nega i e emo ional esponses o clima e change, bu he e was no able a ia ion be ween coun ies (Table 1). The highes p opo ion o pa icipan s indica ing ha hey we e ‘ e y’ o ‘ex emely’ wo ied abou clima e change was eco ded in Spain (77.6%) and he lowes p opo ion was eco ded in Russia (9.6%). The a e age p opo ion o pa icipan s indica ing ha hey we e ‘ e y’ o ‘ex emely’ wo ied abou clima e change ac oss he o al sample was 46.8%. As he samples a e no na ionally- ep esen a i e, hese obse a ions do no necessa ily e lec b oade pa e ns o eeling abou clima e change in each coun y. Ra he , hey a e p esen ed he e o illus a e he pa icula p o ile o indi iduals ha olun ee ed o pa icipa e in he s udy. 3.2. P edic o s o clima e anxie y 3.2.1. Flooding expe ience Ou hypo hesis (H1) was con adic ed by he da a (Table 3). We obse ed ha p io expe ience o looding in i e yea s p io o he ime o da a collec ion did no signi ican ly p edic clima e anxie y; i e- spec i e o whe he looding had been expe ienced only one o wo imes (B =0.003, SE =0.019, p =.867, CI: −0.034, 0.040), o h ee imes and mo e (B = − 0.035, SE =0.018, p =.051, CI: −0.070, 0.000). 3.2.2. Desc ip i e no ms Ou hypo hesis (H2) was suppo ed as we ound a signi ican ela- ionship be ween clima e anxie y and pe cei ed desc ip i e no ms (B = .230, SE =0.009, p <.001, CI: 0.213, 0.247). This means ha he deg ee o clima e anxie y exp essed by pa icipan s in he s udy was signi i- can ly associa ed wi h he ex en o which hey obse ed o he people a ound hem exp essing nega i e eelings abou clima e change. 3.2.3. Media exposu e In line wi h ou hypo hesis (H3a), clima e anxie y was signi ican ly ela ed o he a e o exposu e o media in o ma ion abou clima e change impac s (B =0.032, SE =0.005, p <.001, CI =0.022, 0.041). Howe e , we ound no signi ican associa ion be ween clima e anxie y and a e o exposu e o in o ma ion abou clima e change solu ions in he media (B =0.001, SE =0.005, p =.895, CI: −0.009, 0.010); meaning ha ou hypo hesis (H3b) was no suppo ed. In suppo o ou hypo hesis (H3c), we obse ed a signi ican asso- cia ion be ween clima e anxie y and a en ion o clima e change in o - ma ion (B =0.120, SE =0.010, p <.001, CI: 0.102, 0.139). Ye , a en ion did no signi ican ly in e ac wi h a e o exposu e o in o - ma ion abou clima e impac s (B =0.009, SE =0.005, p =.108, CI: −0.002, 0.019) o a e o exposu e o in o ma ion abou clima e change solu ions (B = − 0.006, SE =0.005, p =.260, CI: −0.016, 0.004). This means ha ou hypo heses (H3d) and (H3e) we e no suppo ed by he da a. These esul s p o ide some suppo o he idea ha he e ec o media exposu e on clima e change anxie y is a leas pa ly subjec o he con en o he in o ma ion. In his case, in o ma ion abou clima e change impac s appea s o be a mo e impo an p edic o o clima e anxie y han in o ma ion abou clima e change solu ions. 3.3. Clima e anxie y as a p edic o o p o-en i onmen al beha iou , en i onmen al ac i ism and men al wellbeing O e all, clima e anxie y showed a signi ican posi i e ela ionship wi h p o-en i onmen al beha iou (H4a: B =0.174, SE =0.021, p < .001, CI: 0.132, 0.216) and wi h en i onmen al ac i ism (H4b: B = 0.250, SE =0.080, p =.002, CI: 0.083, 0.407) – which was ep esen ed by pa icipa ion in clima e p o es s (Table 4). A likelihood a io es showed ha adding andom slopes o nega i e clima e- ela ed emo- ions o he model p edic ing p o-en i onmen al beha iou p oduced a signi ican ly be e i o he da a han a model wi h only andom in- e cep s ( χ 2 =100.82, p <.001). This means ha he ela ionship be- ween clima e anxie y and p o-en i onmen al beha iou a ies signi ican ly ac oss coun ies. Inspec ing he plo o andom slopes ac oss coun ies (Fig. 1) shows ha clima e anxie y had a signi ican posi i e ela ionship wi h p o-en i onmen al beha iou in all coun ies excep Egyp , I an, Japan, Malaysia, Oman, Pakis an, and Tanzania – meaning ha he hypo hesis (H4a) was suppo ed in 25 ou o 32 Table 3 P edic o s o clima e anxie y. P edic o s Es ima es 95%CI (UL, CL) p (In e cep ) 3.22 3.14, 3.30 <.001 Flooding Exp. ( e =no exp) Once o wice .00 −.03, .04 .867 3 imes o mo e −.03 −.07, −.00 .051 Media exposu e CC Impac .04 .03, .05 <.001 CC Solu ion .00 −.01, .01 .389 A en ion .12 .10, .14 <.001 CC Impac x A en ion .01 −.00, .02 .108 CC Solu ion x A en ion −.01 −.02, .00 .260 Desc ip i e no ms .23 .21, .25 <.001 Demog aphic co a ia es Age −.00 −.01, −.00 <.001 Gende (Female) .22 .19, .25 <.001 Random e ec s σ 2 .52 τ 00 Coun y .05 ICC .09 N Coun y 32 Obse a ions 11,461 Ma ginal R 2 /Condi ional R 2 .13/.21 C.A. Ogunbode e al. Jou nal o En i onmen al Psychology 84 (2022) 101887 8 coun ies. The weakes associa ion be ween clima e anxie y and p o- en i onmen al beha iou was obse ed in Pakis an and he s onges associa ion was obse ed in Finland. Like p o-en i onmen al beha iou , we ound ha adding andom slopes o clima e anxie y o he model p edic ing en i onmen al ac i ism showed a be e i o he da a han a model only comp ising andom in e cep s ( χ 2 =58.36, p <.001), which again sugges s ha he ela ionship be ween clima e anxie y and engagemen in en i onmen al ac i ism a ies signi ican ly ac oss coun ies. Inspec ing he plo o andom slopes ac oss coun ies (Fig. 2) e ealed ha clima e anxie y only signi ican ly p edic ed engagemen in en i onmen al ac i ism in 12 coun ies (Aus alia, B azil, Finland, Ge many, I aly, Ne he lands, No - way, Oman, Po ugal, Russia, Spain, and he UK). The weakes associ- a ion be ween clima e anxie y and engagemen in en i onmen al ac i ism was obse ed in China, while he s onges associa ion was obse ed in Finland. Finally, in suppo o ou hypo hesis (H5), we ound ha clima e anxie y had a signi ican in e se ela ionship wi h men al wellbeing (B = − .240, SE =0.020, p <.001, CI: −0.281, −0.199). Howe e , adding andom slopes o clima e anxie y also i he da a be e han only ha ing andom in e cep s in he model p edic ing men al wellbeing ( χ 2 =33.02, p <.001). This sugges s ha he ela ionship be ween clima e anxie y and men al wellbeing a ies ac oss coun ies. Inspec ing he andom slopes plo showed ha he ela ionship be ween clima e anx- ie y and men al wellbeing was signi ican in e e y coun y ep esen ed in he da ase excep Japan. The magni ude o he ela ionship also a ied signi ican ly ac oss coun ies – wi h he s onges associa ion be ween clima e anxie y and men al wellbeing obse ed in Pales ine and he weakes obse ed in Japan (Fig. 3). 3.4. Supplemen a y analyses We conduc ed some u he analyses o explo e po en ial explana- ions o he coun y-le el di e ences in associa ions o clima e anxie y wi h p o-en i onmen al ac ions and men al wellbeing. P e ious esea ch has shown ha cul u al o ien a ions ha e a signi ican ole in he ela ionship be ween en i onmen al conce n and p o-en i onmen al beha iou s such ha conce n is mo e s ongly linked o p o- en i onmen al beha iou in socie ies wi h highe le els o indi idu- alism (Tam & Chan, 2017). The e o e, we expec ed ha c oss-na ional di e ences in indi idualism-collec i ism may be signi ican ly linked o he ela ionship be ween clima e anxie y and p o-en i onmen al ac ion. We did no ha e any expec a ions o how indi idualism-collec i ism Table 4 P o-en i onmen al ac ions and men al wellbeing eg essed on clima e anxie y. P edic o s DV: P o-en i onmen al beha iou DV: En i onmen al ac i ism DV: Men al wellbeing (WHO-5) Es ima e 95%CI (UL, CL) p Odds a io 95%CI (UL, CL) p Es ima e 95%CI (UL, CL) p (In e cep ) 3.36 3.23, 3.42 <.001 0.11 .08, .16 <.001 2.70 2.62, 2.78 <.001 Clima e anxie y .17 .13, .21 <.001 1.28 1.10, 1.50 .002 −.24 −.28, −.20 <.001 Demog aphic co a ia es Age .00 −.00, .00 .069 1.01 1.00, 1.02 .039 .01 .00, .01 <.001 Gende (Female) .05 .02, .08 .001 .95 .83, 1.08 .414 −.09 −.13, −.06 <.001 Random e ec s σ 2 .48 3.29 .81 τ 00 Coun y .08 .77 .06 τ 11 Coun y.CA .01 .13 .01 ICC .15 .20 .07 ρ 01 Coun y .24 .62 .40 N Coun y 32 32 32 Obse a ions 11,503 11,485 11,503 Ma ginal R 2 /Condi ional R 2 .04/.18 .01/.21 .05/.12 Fig. 1. Random slopes o clima e anxie y (CA) as a p edic o o p o- en i onmen al beha iou (PEB). Coun ies a e iden i ied by hei h ee-le e alpha-3 coun y codes. Con idence in e als c ossing ze o indica es a non- signi ican ela ionship be ween CA and PEB in he speci ied coun y. Fig. 2. Random slopes o clima e anxie y (CA) as a p edic o o en i onmen al ac i ism (pa icipa ion in clima e p o es s). Coun ies a e iden i ied by hei h ee-le e alpha-3 coun y codes. Con idence in e als c ossing ze o indica es a non-signi ican ela ionship be ween CA and en i onmen al ac i ism in he speci ied coun y. C.A. Ogunbode e al. Jou nal o En i onmen al Psychology 84 (2022) 101887 9 a ec he ela ionship be ween clima e anxie y and men al heal h. To explo e his, we compu ed pa ial co ela ions o clima e anxie y wi h p o-en i onmen al beha iou , ac i ism and men al wellbeing o each coun y while con olling o pa icipan age and gende . Nex , we plo ed hese co ela ions es ima es agains he indi idualism-collec i ism sco es o each coun y in he da ase (Fig. 4). We ound ha indi idualism was posi i ely co ela ed wi h he s eng h o he ela ionship be ween clima e anxie y and p o-en i onmen al beha iou ac oss coun ies ( =.54, p <.001, N = 29), as well as he ela ionship be ween clima e anxie y and clima e ac i ism ( =.59, p <.001, N =29). Howe e , indi idualism was no signi ican ly co ela ed wi h he ela ionship o clima e anxie y o men al heal h a he coun y le el ( =.05, p =.775, N =29). These esul s sugges ha clima e anxie y ends o be mo e s ongly associa ed wi h p o-en i onmen al beha iou and clima e ac i ism in coun ies wi h highe le els o indi idualism. Nex , we explo ed a po en ial ole o na ional le el a luence (Fig. 5). Recen esea ch by Pohjolainen e al. (2021) shows ha conce n abou clima e change is mo e s ongly associa ed wi h clima e policy suppo and pe sonal clima e ac ions in Eu opean coun ies wi h highe g oss domes ic p oduc (GDP). Consequen ly, we an icipa ed ha GDP may also show signi ican co ela ions wi h he associa ion o clima e anxie y wi h p o-en i onmen al beha iou and engagemen in clima e ac i ism. Ou da a showed ha GDP is posi i ely co ela ed wi h he ela ionship o clima e anxie y o p o-en i onmen al beha iou ( =.59, p <.001, N =31), as well as he ela ionship o clima e anxie y o clima e ac i ism ( =0.53, p <.001, N =31). The e was howe e no signi ican co e- la ion o GDP wi h he ela ionship be ween clima e anxie y and men al heal h a he coun y le el ( =0.10, p =.600, N =31). Finally, a plausible ole o physical condi ions, speci ically ulne a- bili y o clima e isk as e lec ed in he clima e isk index (CRI) de el- oped by Ge manWa ch (Ecks ein e al., 2019), was explo ed (Fig. 6). The CRI indica es a coun y’s le el o exposu e and ulne abili y o ex eme e en s. I is based on ou indica o s: numbe o dea hs due o ex eme e en s, numbe o dea hs pe 100,000 inhabi an s, sum o losses in $US pu chasing powe pa i y (PPP), and losses pe uni o GDP. The CRI alues used in his s udy e lec an a e age o ulne abili y es ima es o wo decades (2000–2019) leading up o he s a o da a collec ion. We ound ha he CRI did no signi ican ly co ela e wi h he associa ion o clima e anxie y and p o-en i onmen al beha iou ( =.05, p =.768, N =31), clima e anxie y and engagemen in clima e ac i ism ( =0.05, p =.768, N =31), o clima e anxie y and men al wellbeing ( =0.05, p = .782, N =31). These esul s sugges ha exposu e o objec i e clima e change impac s may no necessa ily ha e a s ong ole in shaping how emo ional esponses o clima e change ela e o beha iou al and well- being ou comes. 4. Discussion Emo ions shape he way we p ocess in o ma ion and de e mine app op ia e cou ses o ac ion. They a e consequen ly an in eg al elemen o how we engage wi h clima e change (Da idson & Kecinski, 2021). The numbe o people epo ing psychological dis ess abou he clima e c isis has been ising in andem wi h he g owing isibili y o ad e se clima e change impac s a ound he wo ld. In his s udy, we explo ed plausible an eceden s and ou comes o such emo ional e- sponses ac oss a di e se ange o na ional con ex s. Speci ically, we examined i and how clima e anxie y ela es o pe sonal expe ience o looding, pe cei ed desc ip i e no ms and exposu e o clima e change- ela ed media. We also explo ed how clima e anxie y ela es o p o-en i onmen al beha iou , en i onmen al ac i ism and men al wellbeing. Fig. 3. Random slopes o clima e anxie y (CA) as a p edic o o men al well- being (MW). Coun ies a e iden i ied by hei h ee-le e alpha-3 coun y codes. Con idence in e als c ossing ze o indica es a non-signi ican ela ion- ship be ween CA and en i onmen al ac i ism in he speci ied coun y. Fig. 4. Associa ion be ween indi idualism and magni ude o pa ial co ela ion be ween clima e anxie y, p o-en i onmen al ac ions and men al wellbeing. Indi- idualism sco es we e ob ained om www.ho s ede-insigh s.com. Oman, Pales ine and Uganda we e no included in his se o analyses due o a lack o indi id- ualism da a. C.A. Ogunbode e al.