The relationship between perceived health and physical activity indoors, outdoors in built environments, and outdoors in nature

The Rela ionship be ween Pe cei ed Heal h and Physical Ac i i y Indoo s, Ou doo s in Buil En i onmen s, and Ou doo s in Na u e Ty i P. Pasanen* Uni e si y o Tampe e, Finland Liisa Ty äinen Finnish Fo es Resea ch Ins i u e Me la, Finland Kale i M. Ko pela Uni e si y o Tampe e, Finland Backg ound: A body o e idence shows ha bo h physical ac i i y and exposu e o na u e a e connec ed o imp o ed gene al and men al heal h. Expe imen al s udies ha e consis en ly ound sho e m posi i e e ec s o physical ac i i y in na u e compa ed wi h buil en i onmen s. This s udy explo es whe he hese bene i s a e also e iden in e e yday li e, pe cei ed o e epea ed con ac wi h na u e. The opic is impo an om he pe spec i es o ci y planning, indi idual well-being, and public heal h. Me hods: Na ional su ey da a (n=2,070) om Finland was analysed using s uc u al eg ession analyses. Pe cei ed gene al heal h, emo ional well-being, and sleep quali y we e eg essed on he weekly equency o physical ac i i y indoo s, ou doo s in buil en i onmen s, and in na u e. Socioeconomic ac o s and o he plausible con ounde s we e con olled o . Resul s: Emo ional well-being showed he mos consis en posi i e connec- ion o physical ac i i y in na u e, whe eas gene al heal h was posi i ely asso- cia ed wi h physical ac i i y in bo h buil and na u al ou doo se ings. Be e sleep quali y was weakly connec ed o equen physical ac i i y in na u e, bu he connec ion was ou weighed by o he ac o s. Conclusion: The esul s indi- ca e ha na u e p o ides an added alue o he known bene i s o physical ac i i y. Repea ed exe cise in na u e is, in pa icula , connec ed o be e emo- ional well-being. Keywo ds: Finland, men al heal h, na u al en i onmen , physical ac i i y, sleep, well-being * Add ess o co espondence: Ty i Pasanen, School o Social Sciences and Humani ies, Kale an ie 5, 33014 Uni e si y o Tampe e, Tampe e, Finland. Email: [email p o ec ed] bs_bs_banne APPLIED PSYCHOLOGY: HEALTH AND WELL-BEING, 2014, 6(3), 324–346 doi:10.1111/aphw.12031 © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology. This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion- NonComme cial-NoDe i s License, which pe mi s use and dis ibu ion in any medium, p o ided he o iginal wo k is p ope ly ci ed, he use is non-comme cial and no modi ica ions o adap a ions a e made. INTRODUCTION Being exposed o na u e, whe he i be iewing pic u es o na u al landscapes o physical ac i i y (PA) in he wild, has been ound o ha e nume ous heal h bene i s. P e ious epidemiological s udies ha e ound posi i e co ela ions be ween neighbou hood g eene y and long e m heal h indica o s such as mo bidi y and longe i y (Maas e al., 2009; Takano, Nakamu a, & Wa anabe, 2002). Expe imen al s udies ha e ocused on he sho e m posi- i e e ec s o being exposed o g een en i onmen s, such as changes in mood and physiological s ess measu es (see o example, Ha ig, E ans, Jamne , Da is, & Gä ling, 2003; Lee, Pa k, Tsune sugu, Kagawa, & Miyazaki, 2009; P e y, Peacock, Sellens, & G i in, 2005). These bene i s ha e been gene ally explained by in olun a y a en ion eco e y (A en ion Res o a ion Theo y, ART; Kaplan, 1995) and s ess educ ion on psychological and physiological le els (Ul ich, 1983; Ul ich e al., 1991; see Beu e & de Ko , 2014, o a ecen e iew). Epidemiological s udies ha e explo ed he co ela ion be ween g een space indica o s and imp o ed well-being. The ela ionship has been es ablished o a a ie y o measu es o neighbou hood g eene y, such as p oximi y, pe - cei ed quali y, and he ex en o he g eene y (de Jong, Albin, Skä bäck, G ahn, & Bjö k, 2012; S igsdo e e al., 2010; an Dillen, de V ies, G oenewegen, & Sp eeuwenbe g, 2012; Van He zele & de V ies, 2012; Wa d Thompson & Aspinall, 2011). Plausible mechanisms behind his ela ion include inc eased le els o PA (My on, Townsend, Ru e , & Fos e , 2012), educed s ess, and social cohesion (de V ies, an Dillen, G oenewegen, & Sp eeuwenbe g, 2013). In e es ingly, a ecen epidemiological s udy ound ha g een exe cise, bu no he o al amoun o PA, media ed he connec ion be ween g een space p oximi y and well-being (de V ies e al., 2013). By encou aging he abo e-men ioned heal hy beha iou pa icula ly in dep i ed popula ions, exposu e o g een space has been ound o balance socioeco- nomic heal h inequali ies (Mi chell & Popham, 2008; Wa d Thompson & Aspinall, 2011). The s ong posi i e connec ion be ween PA and gene al and men al heal h has been widely accep ed in p e ious esea ch (Fox, 1999; Penedo & Dahn, 2005; S ephens, 1988). Whe he exe cising in na u e, as opposed o o he en i onmen s, p o ides added alue o hese bene icial e ec s has been he p ima y in e es in many ecen expe imen al s udies. A me a-analysis by Bowle , Buyung-Ali, Knigh , and Pullin (2010) ou lined ha exe cising in a na u al en i onmen , a he han in buil en i onmen s indoo s o ou doo s, educes nega i e emo ions and imp o es a en ion, while physical measu es (blood p essu e and co isol concen a ion) had no shown sys ema ic di e - ences in he s udies hey e iewed. Mo e speci ically, he imp o emen s in mood and sel -es eem a e expe ienced wi hin i e minu es o exposu e PERCEIVED HEALTH AND ACTIVITY IN NATURE 325 © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology (Ba on & P e y, 2010). A sys ema ic e iew by Thompson Coon e al. (2011) summa ised ha he bene i s ollowing exe cise ou doo s, compa ed wi h simila ac i i ies indoo s, include inc eased ene gy and eelings o e i- alisa ion, as well as dec eases in ension, con usion, ange , and dep ession. The ollowing sec ions p o ide empi ical e idence showing he po en ial e ec s o exposu e o na u e on s ess and a en ion es o a ion, and he consequen implica ions on well-being and heal h. We conclude he in oduc- ion by p esen ing he ou line and hypo hesis o he p esen s udy. Exposu e o Na u e and Men al Well-Being Maas e al. (2009) ound an in e se ela ionship be ween li ing less han one kilome e away om g een space, and dep ession and anxie y diso de s. The e is also e idence ha he p opo ion o g eene y in he neighbou hood is connec ed o inc eased happiness (Van He zele & de V ies, 2012), educed men al dis ess (Whi e, Alcock, Wheele , & Depledge, 2013), and educed s ess on subjec i e and objec i e le els (Wa d Thompson e al., 2012). Simi- la ly, he quan i y o g een a eas, as well as he quan i y and quali y o s ee scape g eene y, has been ela ed o be e men al heal h ( an Dillen e al., 2012). Being ac i e ou doo s has been associa ed wi h enhanced men al well-being: he elde ly showed ewe dep essi e symp oms he mo e ime hey spen ou doo s and being physically ac i e (Ke e al., 2012), while a s udy in Sco land es ima ed ha each addi ional weekly use o na u al en i onmen lowe s he isk o poo men al heal h by 6 pe cen (Mi chell, 2013). Simila ly, a me a-analysis by Ba on and P e y (2010) ound ha hose wi h men al illnesses bene i ed sys ema ically om exposu e o na u e mo e han o he s. On he o he hand, in a popula ion-based su ey s udy, a posi i e measu e o men al well-being exhibi ed a s onge posi i e connec ion o he use o non-na u al en i onmen s han o na u e (Mi chell, 2013). Exposu e o Na u e and Gene al Heal h Good heal h has been associa ed wi h p oximi y o he nea es g een space (S igsdo e e al., 2010) and he p opo ion o g eene y in he su ounding en i onmen (Maas, Ve heij, G oenewegen, de V ies, & Sp eeuwenbe g, 2006). Howe e , he la e associa ion has ecen ly been ques ioned (de Jong e al., 2012; Van He zele & de V ies, 2012). In e es ingly, al hough de Jong e al. (2012) did no ind a signi ican ela ionship be ween objec i e measu es o neighbou hood g eene y ( ia GIS) and pe cei ed heal h, heal h co e- la ed posi i ely wi h pe cei ed g eene y. The la e esul was also weakly suppo ed by a s udy by Wa d Thompson and Aspinall (2011). Besides quan i y, be e obse ed quali y o he g een a eas and s ee s o he neigh- bou hood ha e also shown a ela ion o highe pe cei ed gene al heal h (Van 326 PASANEN ET AL. © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology Dillen e al., 2012). On an epidemiological le el, li ing close o g een elemen s has been connec ed o longe i y (Takano e al., 2002) and dec eased mo al- i y a es (Mi chell & Popham, 2008). Exposu e o Na u e and Sleep Quali y An epidemiological s udy in Aus alia ound ha people who li e close o less g een a eas a e mo e likely o sleep ewe hou s pe nigh (As ell-Bu , Feng, & Kol , 2013). Apa om his s udy, he opic has ecei ed li le esea ch a en ion, al hough i has high ele ance o public heal h. In Finland, o example, equen insomnia- ela ed symp oms a e p e alen in 10 pe cen o men and 14 pe cen o women (Lallukka e al., 2012). Noc u nal awakening has been ela ed o dec eased quali y o li e (Vää äinen e al., 2013). Being in na u e is likely o in ol e exposu e o na u al ligh and physical exe cise, bo h o which ha e been ound o ha e independen associa ions wi h sleep quali y (D i e & Taylo , 2000). Simila ly, a s udy on he elde ly indica ed ha noc u nal sleep quali y is associa ed wi h being ac i e du ing he day as well as exposu e o na u alis ic ligh (Hood, B uck, & Kennedy, 2004). The e i- dence on he heal h e ec s o ligh exposu e on sleep is cu en ly mo e ex ensi e han he e idence ega ding exposu e o na u e and sleep (Beu e & de Ko , 2014). As exposu e o na u e and o dayligh o en coincides, hei associa ed heal h bene i s a e likely o o e lap (Beu e & de Ko , 2014). The P esen S udy In line wi h he abo e s udies, we hypo hesise ha he well-being bene i s o PA di e be ween ac i i ies ha occu in na u e, indoo s, and in buil ou doo en i onmen s. We assume ha a en ion and s ess es o a ion, shown in p e ious expe imen al and epidemiological s udies, p oduces long- e m heal h and well-being bene i s in an e e yday con ex o e epea ed con ac s wi h na u e. Two o ou ou comes, emo ional well-being and pe - cei ed gene al heal h, ha e al eady been widely s udied, whe eas he hi d, sleep quali y, has no been p e iously explo ed in his con ex . MATERIALS AND METHODS Da a Collec ion Ou da a we e collec ed in wo ounds (wi h a andom sample o 4,000 people pe ound) in he win e and sp ing o 2009. Thus, he na u al en i onmen s on which he esponses a e based include g een, aqua ic (“blue”), and snow- co e ed (“whi e”) na u e a eas. The su ey was pa o he na ionwide Ou doo Rec ea ion Demand In en o y (LVVI2), conduc ed by he Finnish PERCEIVED HEALTH AND ACTIVITY IN NATURE 327 © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology Fo es Resea ch Ins i u e, and i consis ed o six su ey ounds al oge he wi h pa ly a ying hemes. In he i s wo su ey ounds, analysed in his s udy, 3,060 Finnish esponden s ( esponse a e 38%), aged 15–74 yea s, comple ed and e u ned he ques ionnai e. When hey we e ini ially con- ac ed, a le e was sen o each esponden wi h a link o an online ques ion- nai e, pe sonal use name and passwo d, along wi h a b ochu e abou he s udy in gene al. The second eminde (o h ee), in addi ion, included a pape copy o he ques ionnai e. The esponden s ep esen ed he o iginal sample ai ly well al hough women we e o e ep esen ed ela i e o men by 3.4 pe cen age poin s, and he younge age g oups we e unde ep esen ed (15- o 24-yea -olds by 4.9 pe cen age poin s and 25- o 44-yea -olds by 3.1 pe cen age poin s) ela i e o he olde age g oups (45- o 64-yea -olds o e ep esen ed by 5.1 pe cen age poin s and 65- o 75-yea -olds by 2.9 pe cen age poin s; o iginal igu es om Vi anen, Nybe g, Salonen, Neu onen, & Sie änen, 2011). To inspec he alidi y o he LVVI2 su ey se ies, a phone su ey o a andom sample o he non- esponden s in he inal su ey ound (n=301) was conduc ed (Vi anen e al., 2011). Pa icipa ion in he su ey was olun a y and based on in o med consen ; i did no expose he esponden s o any ha m, and he esponden s we e gi en all he necessa y in o ma ion ega ding he s udy. The e o e, acco ding o he Finnish Ad iso y Boa d on Resea ch In eg i y, an e hical e iew o he s udy was no equi ed (h p://www. enk. i/en/). Measu es Ou comes. Emo ional well-being was measu ed by i e s a emen s ha comp ise he Emo ional Well-Being subscale in he RAND 36-i em heal h su ey 1.0 (Hays, She bou ne, & Mazel, 1993; alida ed in Finland by Aal o, A o, & Tepe i, 1999; see Appendix 1, in he Supplemen a y Da a, o he comple e i ems o he scale). The i ems we e measu ed on a 6-poin Like scale anging om “All o he ime” (1) o “No a all” (6). In he analyses, i necessa y, he ou comes we e in e ed so ha highe alues indica ed be e well-being. Pe cei ed gene al heal h was in es iga ed wi h a single ques ion ph ased, “In gene al, would you say you heal h is”: wi h he op ions “good” (1), “ ai ly good” (2), “a e age” (3), “ ai ly poo ” (4) and “poo ” (5). Assess- ing pe cei ed gene al heal h by a single i em is a widely acknowledged p ac- ice in heal h esea ch (Idle & Benyamini, 1997; Miilunpalo, Vuo i, Oja, Pasanen, & U ponen, 1997). Likewise, sleep quali y was in es iga ed wi h a single ques ion asking how o en in he pas ou weeks he esponden s had expe ienced sleep p oblems o poo sleep quali y, wi h he same op ions as he i ems o he emo ional well-being scale. A comp ehensi e and widely used sleep assessmen scale, he Pi sbu gh Sleep Quali y Index (PSQI; Buysse, 328 PASANEN ET AL. © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology Reynolds, Monk, Be man, & Kup e , 1989), also assesses sleep quali y in he pas ou weeks by a single i em, and he i em has shown he highes co e- la ion wi h he o e all sco e o he same index (Ca pen e & And ykowski, 1998). Physical Ac i i y. The PA measu es we e quan i ied by me ging wo a iables, one es ima ing he weekly equency o PA and he o he eco ding he loca ions whe e he PA akes place. Rega ding he loca ions o PA, he esponden s we e ins uc ed o es ima e he p opo ions o hei spa e ime PA ha ake place in se en ypes o en i onmen (wi h examples in b acke s). These en i onmen s we e g ouped in o h ee ca ego ies o he pu poses o his s udy. Indoo spo s se ings (such as indoo spo s hall, gym, o swim- ming pool) and indoo s a home (such as hea y housewo k o home gym- nas ics) we e classi ied in he g oup indoo s.Ou doo s in a buil se ing (such as s ee s, cycle lanes, o spo s ields) o med i s own ca ego y. Ou doo s a ound home (such as ga dening, clea ing he snow, o ball games) was excluded om he analysis because i could e e o bo h na u al and buil se ings. The inal ca ego y, na u e, included he na u al en i onmen nea home (such as nea by o es s o u ban pa ks), he na u al en i onmen a ound one’s second home, and he na u al en i onmen elsewhe e. Rega d- ing he weekly equency o PA, he esponden s we e ins uc ed o only conside ac i e bou s ha las ed a leas 20 minu es (a common measu e in p e ious Finnish popula ion s udies; o example, Pel onen e al., 2008), du ing which hey b ea hed mo e hea ily and b oke in o a swea (excluding walking, cycling, unning, e c. as pa o hei no mal daily ac i i ies). Finally, he measu es in he analyses we e ob ained by mul iplying he weekly equency o PA by he p opo ions aking place in each ype o se ing. Fo example, i he esponden claimed o exe cise wice a week, 50 pe cen o he ime indoo s and 50 pe cen in na u e, he new a iables PA indoo s and PA in na u e would bo h be equal o 2 ×0.5 =1. An es ima e o how ac i e he esponden had been in he pas ou weeks, indica ing he in ensi y o PA, was de i ed om a a iable gene al ac i i y. The ins umen , adap ed om he widely used scale “Sel - epo ed Physical Ac i i y”, de eloped by Sal in and G imby (1968), has been ound o ha e s onge connec ions o men al heal h han an objec i e measu e o physical i ness (Lindwall, Ljung, Hadžibaj amo ic´, & Jonsdo i , 2012). The measu e consis s o ou esponse ca ego ies ha depic he esponden ’s ac i i y and exe ion le el (see Appendix 1, Supplemen a y Da a). Co a ia es. The i s demog aphic co a ia e was gende . Wi hin he Finnish popula ion, gende has no shown a ela ionship o pe cei ed heal h (Kallio, 2006) o emo ional well-being (Aal o e al., 1999). Con e sely, sleep p oblems a e mo e common in emales (Lallukka e al., 2012; Ohayon & PERCEIVED HEALTH AND ACTIVITY IN NATURE 329 © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology Pa inen, 2002), wi h he excep ion o sleep apnoea, which is mo e p e alen in men (K onholm e al., 2009). The second demog aphic co a ia e, age, has been ound o co ela e nega i ely wi h pe cei ed heal h (in Finland, Aal o e al., 1999) and sleep quali y (D i e & Taylo , 2000) bu no wi h men al heal h (Aal o e al., 1999). The e is addi ional e idence ha di e en age g oups espond di e en ly o exposu e o na u e (Ba on & P e y, 2010; Maas e al., 2009). Low le els o he hi d co a ia e, mon hly household income, ha e been associa ed wi h lowe emo ional well-being (Lahelma, Laaksonen, Ma ikainen, Rahkonen, & Sa lio-Läh eenko a, 2006) and sleeping ewe hou s pe nigh (Lallukka e al., 2012) in Finland. In gene al, subjec i e socioeconomic s a us and pe cei ed heal h co ela e posi i ely (Kallio, 2006). Epidemiological s udies ha e ound ewe income- ela ed heal h inequali ies in he g eene neighbou hoods (Maas e al., 2009; Mi chell & Popham, 2008). The ou h demog aphic co a ia e, household size, was added o con ol o he bias i could cause in income. In addi ion, we es ed o al e na i e explana ions o he well-being meas- u es, as sugges ed by Spec o and B annick (2011). Recen unusual e en s migh a ec well-being; his was conside ed by asking he esponden s whe he hey had been going h ough an excep ional si ua ion in li e in he p e ious ou weeks, and i he phase had been easie o mo e di icul han usual. Pe manen o long e m illnesses o disabili ies may a ec physical and men al unc ioning, so we enqui ed whe he he esponden s had long e m disabili ies ha impede hei physical ac i i ies ou doo s. To conside any o he less pe manen disad an ages in li e he esponden s may expe ience, such as lack o ime o acili ies, we asked whe he hey had been able o exe cise ou doo s as much as hey had wished (cons ain s) in he p e ious 12 mon hs. Finally, season (win e o sp ing) a ec s na u al ligh exposu e which may cause luc ua ions in sleep quali y (Hood e al., 2004), impac emo ional well-being (G aw, Recke , Sand, K äuchi, & Wi z-Jus ice, 1999), and al e exe cise ou ines (Tucke & Gilliland, 2007). S a is ical Analysis The e we e in o al 28 (0.9%) esponden s ha had a long e m illness o disabili y ha p e en ed hem om going ou doo s, and hey we e excluded om he analyses. In addi ion, mul i a ia e ou lie s, e alua ed by Mahalanobis dis ances, we e de ec ed and excluded om he da a i e a i ely ollowing he guidelines o Tabachnick and Fidell (2007). Only he mos dis inguishable ex eme cases we e excluded each ime, and a e each exclu- sion, he Mahalanobis dis ances we e e-calcula ed and e-inspec ed. The exclusion o ou lie s was comple e when he la ges Mahalanobis alue was close o he selec ed c i ical alue (based on he χ2dis ibu ion wi h d = “numbe o a iables” −1) and he no mal p obabili y plo ceased o ha e a 330 PASANEN ET AL. © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology dis inguishable ail a he la ge end. A e he exclusion, he co ela ions we e e-examined and compa ed o he p e- il e ed da a. Fo he main analyses, all a iables we e inco po a ed in o s uc u al eg ession models using Mplus e sion 7. The main eason o choosing a s uc u al model was i s abili y o con ol o he co a iance be ween inde- penden a iables as well as he heal h measu es. Emo ional well-being was speci ied as a la en ac o wi h i e o dinal indica o s, and gene al heal h and sleep quali y we e ea ed as la en single indica o ac o s. We used he ecommended app oach in he model building whe e he models o in e es a e es ed immedia ely wi hou a p eceding measu emen model o he la en cons uc s (Hayduk & Glase , 2000). As all obse ed ou come a iables we e o dinal, a no mal ans o ma ion was pe o med and a diagonally weigh ed leas squa es (WLSMV) es ima o was used (Mu hén & Mu hén, 1998–2012). Independen a iables we e added in he cons uc ed models in ou main s eps ha we e speci ied ollowing he sugges ions o Spec o and B annick (2011). The au ho s ecommend i s examining he ela ionships ha a e heo e ically expec ed, a e which o he easible al e na i es should be es ed. The i s s ep consis ed o PA indoo s, in buil ou doo en i onmen s, and in na u e eg essing on emo ional well-being, gene al heal h, and sleep quali y. In he second s ep, a measu e o he in ensi y o ac i i y was added. In he hi d s ep, we added he co a ia es. I a co a ia e was no signi ican in mos analyses, ei he he insigni ican pa hs o he co a ia e i sel was dele ed om he ollowing models. The ou h and he inal s ep included al e na i e explana ions o he heal h ou comes. The SEM models we e e alua ed wi h he χ2 es and he ela ed i indices a ailable in he WLSMV es ima ion. The Ben le compa a i e i index (CFI) measu es how a he speci ied model is om he baseline model wi h no connec ions (Rayko & Ma coulides, 2011), and he gene al guidelines a e 0.95 o close i , whe eas less han 0.90 is conside ed unaccep able (Hu & Ben le , 1999; Yu, 2002). As o he oo mean-squa e e o o app oxima ion (RMSEA), he maximum accep able alue a ies be ween 0.05 and 0.08 (B owne & Cudeck, 1992; Hu & Ben le , 1999). In addi ion, we compa ed he obse ed ou come alues wi h he p obabili ies ha hei model-es ima ed alue is co ec . RESULTS Uni a ia e and Bi a ia e Dis ibu ions All ou comes had a skewed dis ibu ion, wi h he a e age esponses app oaching highe a he han lowe alues o well-being (see Table 1 o all obse ed dis ibu ions). The emo ional well-being scale, o which compa a- i e popula ion igu es we e eadily a ailable, was dis ibu ed wi h he same PERCEIVED HEALTH AND ACTIVITY IN NATURE 331 © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology median (80) as he na ional no ms, al hough in ou sample, mean alue was sligh ly g ea e (78 e sus 74) and s anda d de ia ion smalle (15 e sus 20) han he a ailable popula ion alues (Aal o e al., 1999). Rega ding bi a ia e dis ibu ions, all independen a iables had a signi i- can co ela ion coe icien (p<.05) wi h a leas one o he h ee well-being measu es (Table 2; see Appendix 2a o co ela ions wi hin he independen a iables and Appendix 2b, in he Supplemen a y Da a, o co ela ions wi hin he ou comes). Some o he signi ican co ela ions we e small in magni ude which is common in many esea ch ields, including applied TABLE 1 Dis ibu ions o he Obse ed Va iables (n= 2,070) Va iable Scale o ange Mean (SD)o % Emo ional well-being (C onbach’s α=0.846) 0–100 (summa y sco e) 77.6 (15.4) Pe cei ed gene al heal h Good (1)–Poo (5) 2.0 (0.93) Poo sleep quali y o sleep p oblems All o he ime (1)–No a all (6) 4.8 (1.16) Gende (%) Male 44.4 Female 55.6 Age 15–74 yea s 45.2 (14.8) Household size 1–8 people 2.5 (1.2) Mon hly household incomea(%) €1,000 o less 5.9 €1,001–3,000 32.6 €3,001–5,000 31.7 €5,001–7,000 19.2 €7,001–9,000 6.1 Mo e han €9,000 4.4 Long e m disabili y (%) None 82.9 Yes, bu can s ill exe cise ou doo s 17.1 Excep ional si ua ion in li e in he pas 4 weeks (%) Mo e s ess ul han usual 27.3 No excep ional 64.6 Easie han usual 8.1 Cons ain s on ou doo exe cise in he pas 12 mon hs (%) None 55.6 Some 44.4 Gene al ac i i y in he pas ou weeks (%) Inac i e 14.2 Mode a e ac i i y 53.0 Vigo ous ac i i y 30.9 Compe i i e spo s 1.9 Weekly equency o PA by loca ion (max 5 imes) Indoo s 0.7 (0.8) Ou doo s (buil ) 0.5 (0.7) In na u e 0.9 (0.9) Season Win e 51.9 Sp ing 48.1 asho ened scale, in he analysis we used 11 ca ego ies. 332 PASANEN ET AL. © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology (Sie änen & Neu onen, 2011), he esul s a e no likely o ha e been a ec ed by inequali ies in access o na u al acili ies. Finally, he e a e some al e na i e explana ions o he esul s ha we e no conside ed in he analyses. Fi s , no in o ma ion on he nea by g eene y o he esponden s’ li ing en i onmen s was a ailable. Nea by g eene y may, o ins ance, encou age engagemen in physical ac i i ies (My on e al., 2012) and impac sleep quali y h ough ligh pollu ion and noise. Second, we do no know whe he he a e age du a ion o exe cise indoo s, ou doo s in buil se ings, and in na u e a ies. I isi s o he na u al en i onmen s end o las longe , he inc eased amoun o physical ac i i y and exposu e o na u al ligh could, a leas pa ly, explain he s ong ela ionship be ween exe cise in na u e and well-being. Thi d, he b oad ca ego isa ion in o indoo , buil ou doo , and na u al en i onmen s igno ed he quali a i e di - e ences wi hin he en i onmen s in he same ca ego y. Di e en ypes o na u al en i onmen s may induce di e en es o a i e e ec s (Ko pela, Ylén, Ty äinen, & Sil ennoinen, 2010; Ty äinen e al., 2014) and he e o e, he de ec ed associa ions may no apply o all na u al en i onmen s. Validi y o he Me hods and Resul s Mos o ou measu es a e s anda d in esea ch p ac ice in heal h sciences and ga e us no eason o suspec hei alidi y o eliabili y. Measu es ha necessi a e some discussion a e he es ima ion o he amoun and in ensi y o PA. Compa ed wi h con olled expe imen s, a sel - epo ques ionnai e ine i ably p o ides less accu a e measu es o PA. None heless, es ima ing one’s ac i i ies in weekly in e als was conside ed su icien ly easy o he esponden s o es ima e. Ano he p oblem wi h he measu e was ha i assumed ha PA in di e en ypes o se ings is some hing one does in a gene ally consis en way om week o week, which migh no be he case and, consequen ly, may bias he es ima e. Ou measu e o he gene al in ensi y o PA, howe e , has p e iously been shown o be a s onge connec ion o men al heal h han an objec i e measu e o ae obic i ness (Lindwall e al., 2012). Ne e heless, we conside ou es ima es o equency and in ensi y o PA capable o a leas clea ly dis inguishing he mos ac i e om hose who exe cise only occasionally. Based on he RMSEA and he CFI, ou models 2–4 i he da a well. The ac ha he CFIs somewha wo sened and he RMSEAs imp o ed as new inde- penden a iables we e added is a common phenomenon (Kenny & McCoach, 2003). The χ2 alues, on he con a y, we e a om he ideal. We deduced h ee easons o he ailu e o his es . Fi s , he χ2 es is known o be oo sensi i e when sample sizes a e la ge (N>300; Kline, 2011). The RMSEA, a simila measu e ha is less sensi i e o sample size, suppo ed mos o he es ed models. Second, he models be e explained posi i e a he han PERCEIVED HEALTH AND ACTIVITY IN NATURE 339 © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology nega i e emo ional well-being and sleep quali y, esul ing in some la ge esid- uals ha e iden ly in luenced he χ2 alue. Thi d, he la ges esiduals be ween he obse ed and es ima ed co ela ions appea ed be ween he i ems o he RAND-36. Had we cons ained some o hei esidual a iances, he i could ha e been a i icially imp o ed, bu we ound i heo e ically unjus i iable. Conclusions This s udy p o ides a new ype o su ey e idence suppo ing he heo ies on heal h-enhancing e ec s o con ac wi h na u e by sugges ing ha epea ed exe cise in na u e is ela ed o imp o ed long e m well-being mo e explici ly han epea ed exe cise in buil en i onmen s indoo s o ou doo s. Good emo ional well-being, in pa icula , seems o be e iden ly associa ed wi h mo e equen , ac i e isi s o na u al en i onmen s. Mo eo e , pe cei ed gene al heal h appea s o be connec ed o PA bo h in na u e and in buil ou doo se ings. The e idence on sleep quali y, on he o he hand, was weake in his s udy. Mo e esea ch is needed o be e unde s and he connec ion be ween exposu e o na u e and sleep. Whe he his ela ionship is, o example, media ed by imp o ed psychological well-being o neigh- bou hood quali ies is wo hy o u he examina ion. ACKNOWLEDGEMENTS Many hanks go o Tuija Sie änen, Ma jo Neu onen, Miisa Pie ilä, and Ann Ojala o he Finnish Fo es Resea ch Ins i u e (Me la) o hei con ibu ions o he p ojec . This wo k was suppo ed by he Academy o Finland (p ojec numbe 2501255431) and he Finnish Minis y o Educa ion and Cul u e, and he Minis y o he En i onmen (g an numbe YTA022). CONFLICT OF INTEREST None. REFERENCES Aal o, A., A o, A.R., & Tepe i, J. (1999). RAND-36 e ey een lii y än elämänlaadun mi a ina: mi a in luo e a uus ja suomalaise äes öa o [RAND-36 as a measu e o heal h- ela ed quali y o li e: Reliabili y, cons uc alidi y and e e ence alues in he Finnish gene al popula ion]. Helsinki: S akes. As ell-Bu , T., Feng, X., & Kol , G.S. (2013). Does access o neighbou hood g een space p omo e a heal hy du a ion o sleep? No el indings om a c oss- sec ional s udy o 259 319 Aus alians. BMJ Open,3(8). doi:10.1136/bmjopen- 2013-003094 340 PASANEN ET AL. © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology Ba on, J., & P e y, J. (2010). Wha is he bes dose o na u e and g een exe cise o imp o ing men al heal h? A mul i-s udy analysis. En i onmen al Science & Tech- nology,44(10), 3947–3955. doi:10.1021/es903183 Bell, S., Ty äinen, L., Sie änen, T., P öbs l, U., & Simpson, M. (2007). Ou doo ec ea ion and na u e ou ism: A Eu opean pe spec i e. Li ing Re iews in Land- scape Resea ch,1(2), 1–46. Beu e, F., & de Ko , Y.A.W. (2014). Salu ogenic e ec s o he en i onmen : Re iew o heal h p o ec i e e ec s o na u e and dayligh . Applied Psychology: Heal h and Well-Being,6(1), 67–95. doi:10.1111/aphw.12019 Bowle , D.E., Buyung-Ali, L., Knigh , T.M., & Pullin, A.S. (2010). A sys ema ic e iew o e idence o he added bene i s o heal h o exposu e o na u al en i on- men s. BMC Public Heal h,10, 456–465. B owne, M.W., & Cudeck, R. (1992). Al e na i e ways o assessing model i . Socio- logical Me hods & Resea ch,21(2), 230–258. Buysse, D.J., Reynolds III, C.F., Monk, T.H., Be man, S.R., & Kup e , D.J. (1989). The Pi sbu gh Sleep Quali y Index: A new ins umen o psychia ic p ac ice and esea ch. Psychia y Resea ch,28(2), 193–213. doi:10.1016/0165- 1781(89)90047-4 Cappelle i, J., Bushmakin, A., McDe mo , A., Sadosky, A., Pe ie, C., & Ma in, S. (2009). Psychome ic p ope ies o a single-i em scale o assess sleep quali y among indi iduals wi h ib omyalgia. Heal h and Quali y o Li e Ou comes,7(1), 54. Ca pen e , J.S., & And ykowski, M.A. (1998). Psychome ic e alua ion o he Pi s- bu gh Sleep Quali y Index. Jou nal o Psychosoma ic Resea ch,45(1), 5–13. doi:10.1016/S0022-3999(97)00298-5 Coope , C. (1998). Indi idual di e ences. London: A nold. De Jong, K., Albin, M., Skä bäck, E., G ahn, P., & Bjö k, J. (2012). Pe cei ed g een quali ies we e associa ed wi h neighbo hood sa is ac ion, physical ac i i y, and gene al heal h: Resul s om a c oss-sec ional s udy in subu ban and u al Scania, sou he n Sweden. Heal h & Place,18(6), 1374–1380. doi:10.1016/j.heal hplace .2012.07.001 De V ies, S., an Dillen, S.M.E., G oenewegen, P.P., & Sp eeuwenbe g, P. (2013). S ee scape g eene y and heal h: S ess, social cohesion and physical ac i i y as media o s. Social Science & Medicine,94(0), 26–33. doi:10.1016/j.socscimed. 2013.06.030 D i e , H.S., & Taylo , S.R. (2000). Exe cise and sleep. Sleep Medicine Re iews,4(4), 387–402. doi:10.1053/sm .2000.0110 Fox, K.R. (1999). The in luence o physical ac i i y on men al well-being. Public Heal h Nu i ion,2(3a), 411–418. G aw, P., Recke , S., Sand, L., K äuchi, K., & Wi z-Jus ice, A. (1999). Win e and summe ou doo ligh exposu e in women wi h and wi hou seasonal a ec i e diso de . Jou nal o A ec i e Diso de s,56(2–3), 163–169. doi:10.1016/S0165- 0327(99)00037-3 Ha ig, T., E ans, G.W., Jamne , L.D., Da is, D.S., & Gä ling, T. (2003). T acking es o a ion in na u al and u ban ield se ings. Jou nal o En i onmen al Psychol- ogy,23(2), 109–123. doi:10.1016/S0272-4944(02)00109-3 PERCEIVED HEALTH AND ACTIVITY IN NATURE 341 © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology Ha ig, T., Mi chell, R., de V ies, S., & F umkin, H. (2014). Na u e and heal h. Annual Re iew o Public Heal h,35(1), 207–228. doi:10.1146/annu e -publheal h- 032013-182443 Hayduk, L.A., & Glase , D.N. (2000). Ji ing he ou -s ep, wal zing a ound ac o analysis, and o he se ious un. S uc u al Equa ion Modeling: A Mul idisciplina y Jou nal,7(1), 1–35. doi:10.1207/S15328007SEM0701_01 Hays, R.D., She bou ne, C.D., & Mazel, R.M. (1993). The RAND 36-i em heal h su ey 1.0. Heal h Economics,2(3), 217–227. doi:10.1002/hec.4730020305 Hood, B., B uck, D., & Kennedy, G. (2004). De e minan s o sleep quali y in he heal hy aged: The ole o physical, psychological, ci cadian and na u alis ic ligh a iables. Age and Ageing,33(2), 159–165. 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 Mod- eling: A Mul idisciplina y Jou nal,6(1), 1–55. Idle , E.L., & Benyamini, Y. (1997). Sel - a ed heal h and mo ali y: A e iew o wen y-se en communi y s udies. Jou nal o Heal h and Social Beha io ,38(1), 21–37. Kallio, J. (2006). Koe u e eys ja e eydellinen asa-a o kuudessa hy in oin i al iossa [Sel -pe cei ed heal h and heal h di e ences in six wel a e s a es]. Sosiaalilääke ie eellinen Aikakausleh i—Jou nal o Social Medicine,43(3), 218–230. Kaplan, S. (1995). The es o a i e bene i s o na u e: Towa d an in eg a i e ame- wo k. Jou nal o En i onmen al Psychology,15(3), 169–182. doi:10.1016/0272- 4944(95)90001-2 Kenny, D.A., & McCoach, D.B. (2003). E ec o he numbe o a iables on measu es o i in s uc u al equa ion modeling. S uc u al Equa ion Modeling,10(3), 333– 351. Ke , J., Ma shall, S., Godbole, S., Neukam, S., C is , K., Wasilenko, K. e al. (2012). The ela ionship be ween ou doo ac i i y and heal h in olde adul s using GPS. In e na ional Jou nal o En i onmen al Resea ch and Public Heal h,9(12), 4615– 4625. Kline, R.B. (2011). P inciples and p ac ice o s uc u al equa ion modeling (3 d edn.). New Yo k: Guil o d P ess. Ko pela, K., Bo odulin, K., Neu onen, M., Pa onen, O., & Ty äinen, L. (2014). Analyzing he media o s be ween na u e-based ou doo ec ea ion and emo ional well-being. Jou nal o En i onmen al Psychology,37, 1–7. doi:10.1016/ j.jen p.2013.11.003 Ko pela, K.M., Ylén, M., Ty äinen, L., & Sil ennoinen, H. (2010). Fa o i e g een, wa e side and u ban en i onmen s, es o a i e expe iences and pe cei ed heal h in Finland. Heal h P omo ion In e na ional,25(2), 200–209. doi:10.1093/heap o/ daq007 K onholm, E., Sallinen, M., Suu ama, T., Sulka a, R., E a, P., & Pa onen, T. (2009). Sel - epo ed sleep du a ion and cogni i e unc ioning in he gene al popula ion. Jou nal o Sleep Resea ch,18(4), 436–446. doi:10.1111/j.1365-2869.2009.00765.x Lahelma, E., Laaksonen, M., Ma ikainen, P., Rahkonen, O., & Sa lio- Läh eenko a, S. (2006). Mul iple measu es o socioeconomic ci cums ances 342 PASANEN ET AL. © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology and common men al diso de s. Social Science & Medicine,63(5), 1383–1399. doi:10.1016/j.socscimed.2006.03.027 Lallukka, T., Sa es-Jaske, L., K onholm, E., Saaksja i, K., Lundq is , A., Pa onen, T. e al. (2012). Sociodemog aphic and socioeconomic di e ences in sleep du a ion and insomnia- ela ed symp oms in Finnish adul s. BMC Public Heal h,12(1), 565–586. Lee, J., Pa k, B., Tsune sugu, Y., Kagawa, T., & Miyazaki, Y. (2009). Res o a i e e ec s o iewing eal o es landscapes, based on a compa ison wi h u ban land- scapes. Scandina ian Jou nal o Fo es Resea ch,24(3), 227–234. doi:10.1080/ 02827580902903341 Lindwall, M., Ljung, T., Hadžibaj amo ic´, E., & Jonsdo i , I.H. (2012). Sel - epo ed physical ac i i y and ae obic i ness a e di e en ly ela ed o men al heal h. Men al Heal h and Physical Ac i i y,5(1), 28–34. Lomeli, H.A., Pe ez-Olmos, I., Tale o-Gu ie ez, C., Mo eno, C.B., Gonzalez-Reyes, R., Palacios, L. e al. (2008). Sleep e alua ion scales and ques ionnai es: A e iew. Ac as Espanolas De Psiquia ia,36(1), 50–59. Maas, J., Ve heij, R.A., de V ies, S., Sp eeuwenbe g, P., Schelle is, F.G., & G oenewegen, P.P. (2009). Mo bidi y is ela ed o a g een li ing en i onmen . Jou nal o Epidemiology and Communi y Heal h,63(12), 967–973. doi:10.1136/ jech.2008.079038 Maas, J., Ve heij, R.A., G oenewegen, P.P., de V ies, S., & Sp eeuwenbe g, P. (2006). G een space, u bani y, and heal h: How s ong is he ela ion? Jou nal o Epidemiology and Communi y Heal h,60(7), 587–592. doi:10.1136/jech .2005.043125 Miilunpalo, S., Vuo i, I., Oja, P., Pasanen, M., & U ponen, H. (1997). Sel - a ed heal h s a us as a heal h measu e: The p edic i e alue o sel - epo ed heal h s a us on he use o physician se ices and on mo ali y in he wo king-age popu- la ion. Jou nal o Clinical Epidemiology,50(5), 517–528. doi:10.1016/S0895- 4356(97)00045-0 Mi chell, R. (2013). Is physical ac i i y in na u al en i onmen s be e o men al heal h han physical ac i i y in o he en i onmen s? Social Science & Medicine,91, 130–134. doi:10.1016/j.socscimed.2012.04.012 Mi chell, R., & Popham, F. (2008). E ec o exposu e o na u al en i onmen on heal h inequali ies: An obse a ional popula ion s udy. The Lance ,372(9650), 1655–1660. doi:10.1016/S0140-6736(08)61689-X Mølle , A., & Jennions, M.D. (2002). How much a iance can be explained by ecologis s and e olu iona y biologis s? Oecologia,132(4), 492–500. Mu hén, L.K., & Mu hén, B.O. (1998–2012). Mplus use ’s guide (7 h edn.). Los Angeles, CA: Mu hén & Mu hén. My on, O.T., Townsend, N., Ru e , H., & Fos e , C. (2012). G een space and physical ac i i y: An obse a ional s udy using Heal h Su ey o England da a. Heal h & Place,18(5), 1034–1041. doi:10.1016/j.heal hplace.2012.06.003 Ohayon, M.M., & Pa inen, M. (2002). Insomnia and global sleep dissa is ac ion in Finland. Jou nal o Sleep Resea ch,11(4), 339–346. Pel onen, M., Ha ald, K., Männis ö, S., Saa ikoski, L., Lund, L., Sund all, J. e al. (2008). Kansallinen FINRISKI 2007- e eys u kimus: Tu kimuksen o eu us ja PERCEIVED HEALTH AND ACTIVITY IN NATURE 343 © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology ulokse . aulukkolii e [The Na ional FINRISKI 2007 Heal h S udy: P ocedu e and esul s. Table appendix]. Kansan e eyslai os. Penedo, F.J., & Dahn, J.R. (2005). Exe cise and well-being: A e iew o men al and physical heal h bene i s associa ed wi h physical ac i i y. Cu en Opinion in Psy- chia y,18(2), 189–193. P e y, J., Peacock, J., Sellens, M., & G i in, M. (2005). The men al and physical heal h ou comes o g een exe cise. In e na ional Jou nal o En i onmen al Heal h Resea ch,15(5), 319–337. Rayko , T., & Ma coulides, G.A. (2011). In oduc ion o psychome ic heo y. New Yo k: Rou ledge. Sal in, B., & G imby, G. (1968). Physiological analysis o middle-aged and old o me a hle es: Compa ison wi h s ill ac i e a hle es o he same ages. Ci cula ion,38(6), 1104–1115. doi:10.1161/01.CIR.38.6.1104 Sie änen, T., & Neu onen, M. (2011). Luonnon i kis yskäy ön kysyn ä 2010 ja kysynnän muu os. [Ou doo ec ea ion demand 2010 and changes in he demand]. In T. Sie änen & M. Neu onen (Eds.), Luonnon i kis yskäy ö 2010 [Ou doo ec ea ion 2010]. Wo king Pape s o he Finnish Fo es Resea ch Ins i u e No. 212 (pp. 37–73). Van aa: Me sän u kimuslai os. Spec o , P.E., & B annick, M.T. (2011). Me hodological u ban legends: The misuse o s a is ical con ol a iables. O ganiza ional Resea ch Me hods,14(2), 287–305. doi:10.1177/1094428110369842 S ephens, T. (1988). Physical ac i i y and men al heal h in he Uni ed S a es and Canada: E idence om ou popula ion su eys. P e en i e Medicine,17(1), 35–47. doi:10.1016/0091-7435(88)90070-9 S igsdo e , U.K., Ekholm, O., Schippe ijn, J., To age , M., Kampe -Jø gensen, F., & Rand up, T.B. (2010). Heal h p omo ing ou doo en i onmen s—associa ions be ween g een space, and heal h, heal h- ela ed quali y o li e and s ess based on a Danish na ional ep esen a i e su ey. Scandina ian Jou nal o Public Heal h, 38(4), 411–417. doi:10.1177/1403494810367468 Tabachnick, B.G., & Fidell, L.S. (2007). Using mul i a ia e s a is ics (5 h edn.). Bos on, MA: Pea son Educa ion. Takano, T., Nakamu a, K., & Wa anabe, M. (2002). U ban esiden ial en i onmen s and senio ci izens’ longe i y in megaci y a eas: The impo ance o walkable g een spaces. Jou nal o Epidemiology and Communi y Heal h,56(12), 913–918. doi:10.1136/jech.56.12.913 Thompson Coon, J., Boddy, K., S ein, K., Whea , R., Ba on, J., & Depledge, M.H. (2011). Does pa icipa ing in physical ac i i y in ou doo na u al en i onmen s ha e a g ea e e ec on physical and men al wellbeing han physical ac i i y indoo s? A sys ema ic e iew. En i onmen al Science & Technology,45(5), 1761– 1772. doi:10.1021/es102947 Tucke , P., & Gilliland, J. (2007). The e ec o season and wea he on physical ac i- i y: A sys ema ic e iew. Public Heal h,121(12), 909–922. doi:10.1016/j.puhe .2007.04.009 Ty äinen, L., Ojala, A., Ko pela, K., Lanki, T., Tsune sugu, Y., & Kagawa, T. (2014). The in luence o u ban g een en i onmen s on s ess elie measu es: A 344 PASANEN ET AL. © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology ield expe imen . Jou nal o En i onmen al Psychology,38, 1–9. doi:10.1016/ j.jen p.2013.12.005 Ul ich, R.S. (1983). Aes he ic and a ec i e esponse o na u al en i onmen . Human Beha io & En i onmen : Ad ances in Theo y & Resea ch,6, 85–125. Ul ich, R.S., Simons, R.F., Losi o, B.D., Fio i o, E., Miles, M.A., & Zelson, M. (1991). S ess eco e y du ing exposu e o na u al and u ban en i onmen s. Jou nal o En i onmen al Psychology,11(3), 201–230. doi:10.1016/S0272- 4944(05)80184-7 Vää äinen, S., Tuomileh o, H., Sa amies, J., Tuomileh o, J., Uusi alo, H., Hussi, E. e al. (2013). The heal h- ela ed quali y-o -li e impac o noc u nal awakenings in he middle-aged and olde Finnish popula ion. Quali y o Li e Resea ch,22(10), 2737–2748. doi:10.1007/s11136-013-0404-y Van Dillen, S.M.E., de V ies, S., G oenewegen, P.P., & Sp eeuwenbe g, P. (2012). G eenspace in u ban neighbou hoods and esiden s’ heal h: Adding quali y o quan i y. Jou nal o Epidemiology and Communi y Heal h,66(6), e8. doi:10.1136/ jech.2009.104695 Van He zele, A., & de V ies, S. (2012). Linking g een space o heal h: A compa a i e s udy o wo u ban neighbou hoods in Ghen , Belgium. Popula ion and En i on- men ,34(2), 171–193. doi:10.1007/s11111-011-0153-1 Vi anen, V., Nybe g, R., Salonen, R., Neu onen, M., & Sie änen, T. (2011). LVVI 2—seu an a u kimuksen aineis o ja mene elmä [The da a and me hods o he LVVI2 ollow-up su ey]. In T. Sie änen, & M. Neu onen (Eds.), Luonnon i kis yskäy ö 2010 [Ou doo ec ea ion 2010] (pp. 19–36). Van aa: Me sän u kimuslai os. Wa d Thompson, C., & Aspinall, P.A. (2011). Na u al en i onmen s and hei impac on ac i i y, heal h, and quali y o li e. Applied Psychology: Heal h and Well-Being, 3(3), 230–260. doi:10.1111/j.1758-0854.2011.01053.x Wa d Thompson, C., Roe, J., Aspinall, P., Mi chell, R., Clow, A., & Mille , D. (2012). Mo e g een space is linked o less s ess in dep i ed communi ies: E idence om sali a y co isol pa e ns. Landscape and U ban Planning,105(3), 221–229. doi:10.1016/j.landu bplan.2011.12.015 Whi e, M.P., Alcock, I., Wheele , B.W., & Depledge, M.H. (2013). Would you be happie li ing in a g eene u ban a ea? A ixed-e ec s analysis o panel da a. Psychological Science,24(6), 920–928. doi:10.1177/0956797612464659 Yu, C. (2002). E alua ing cu o c i e ia o model i indices o la en a iable models wi h bina y and con inuous ou comes. Doc o al disse a ion, Uni e si y o Cali- o nia, Los Angeles. SUPPORTING INFORMATION Addi ional Suppo ing In o ma ion may be ound in he online e sion o his a icle a he publishe ’s web-si e: Appendix 1. Scale o emo ional well-being (RAND-36 subscale; Hays e al., 1992), Scale o gene al ac i i y (adap ed om Sal in & G imby, 1968). PERCEIVED HEALTH AND ACTIVITY IN NATURE 345 © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology Appendix 2a. Co ela ions be ween he obse ed independen a iables (Spea man i one o bo h o he a iable pai a e o dinal, Pea son in i alics), n=2070. Appendix 2b. Spea man co ela ions be ween he ou come a iables (emo- ional well-being as he summa y sco e o he i e RAND-36 i ems), n=2070. Appendix 3a. Diag am o model 1. χ2=405, d =25, p<0.0001, CFI =0.97, RMSEA =0.09. Only signi ican (p<0.05) connec ions a e shown (a ows). The eg ession coe icien s be ween independen and dependen a iables a e p o ided in Table 3. Appendix 3b. Diag am o model 2. χ2=388, d =29, p<0.0001, CFI =0.97, RMSEA =0.08. Only signi ican (p<0.05) connec ions a e shown (a ows). The dashed a ow ep esen s a nega i e connec ion. The eg ession coe i- cien s be ween independen and dependen a iables a e p o ided in Table 3. Appendix 3c. Diag am o model 3. χ2=463, d =43, p<0.0001, CFI =0.97, RMSEA =0.07. Only signi ican (p<0.05) connec ions a e shown (a ows). The dashed a ow ep esen s a nega i e connec ion. The eg ession coe i- cien s be ween independen and dependen a iables a e p o ided in Table 3. 346 PASANEN ET AL. © 2014 The Au ho s. Applied Psychology: Heal h and Well-Being published by John Wiley & Sons L d on behal o The In e na ional Associa ion o Applied Psychology