ETHNIC VAKIIWION IN ENVIRONMENTAL 13ELI13E' AND I3EHAVIOR: AII Examirlatioli of tlie New l~cologic;~l Paradigm in a Social Psychological Context

We use nntion:tl-level data to test a n~otlified version of Stern, Dietz, & Guagnano's calrsal rt~odcl ofc11viroii1i1~11t;I) lwlief;~r~d I)ctt;tvior: We focils nt;tittly on ethnic varizttion in enviro11nlcnt:ti belief, as measitred by the New Ecological Paradigm (NEP), and cthnic vasi:ttion for four eiiviro~~nlerttal behaviors: environntent;~l realling, housetiold recycling, envit.orlment:tl group joining, and participation i n nature-based outdoor recreation. Blacks and foreigtl-horn Latinos were less likely than Whites to score liiglteion tlie NEI? Any behnvior;tl clitferences between Whites arid the respective ~riinosity groups were expectetl to dintinisli witti the inclusion of the NIlP as an intervening v;~ri:tble in the model between etlt~ticity aiid beltavior. However, ethnic tlifferences senrained stable and strong even when environ~nentai belief ENVIIIONMI~N'I ' A N [ ) l l l l l lAVIOI( , \'ol 36 Nu. 2 , h1;itcli 2004 157-IS6 1)OI: 10.1177/(H)I 3~J105lI325 !-I78 Q) 2004 S;tgc I ' I I ~ I ~ ~ C : I ~ I < I I I \ wits ntlcletl. Over;tll, Asian An~crican mtl U.S.-born I.nfirlo environrrtenfalisr~i was iiiost si~uilitr to Whites. Al'rican Alllcsican concern and behavior was least similar to White e~lvironnielit;tlis~n. Gender, age, and liberal political orientation were also consistent explicators litr botti envir-ortrner~tal concern ant1 betrz~vior. 'I'l~is article exar~iirles the convergence of etliriic diversity and environmentalism in the United States. We address environtnental belief anti behavior for five ethnic groups: African Americans, US.-born Latinos, foreign-born Latinos, Asian Americans, and E~~ropeati ~mer icans . ' We respond to Stern, Dietz, & Guagnano (1 995) and Dietz, Stern, & Guagnano (1998) concerning relationships between social structural factors, environmental beliefs, arid environmental behavior. Stern et al. (1 995) and Dietzet al. (1 998) argued that the environmental values literature, although containing important inforrnation on environnientnl trencls ancl public opinions, lacks a social, psychologically derived thcory base relating environmental valt~es to social structure, environmental bcliels and attitutles, and behavior. The authors proposecl an environmental concern model in wlticlt generalized environmental beliefs or worldview is positioned within a causal relationship where social strtictural variables are a pl-ectirsolof s ~ ~ c h beliefs. In tiirn, beliefs ant1 attitucfes about the environmen~ are predictors of environmental behavior. We expand 011 Stcrn et al. (1995) and Dietz et al. (1998) by focusing on ethnic variation in tlle nlodel. Stern et al. (1995) stressetl the importance of structural factors such as ethnicity in influencing environmental perception ancl behavior, but tllcy did not inclucle an ethnic indicator in their analysis; Dietz et al. prescrttetl only a limited discussion and analysis of ethnicity by examining only t3lacks and Whites. Along with Dietz et al.'s study, ours is one of a limited number o f studies that examines caiisal relationships between structural, perceptnal, ancl bcliavioral variables using national-level environmental clata. Following Dietz et al.'s (1998) recommendation, we employ the New Ecological Par-acfigrn (NEP), which Dietz et al. (1998) believed to be a better indicator of environmental belief than the environmental cjuestions i~scd in their analyses. CAUSAI, hIOI)EI, 017 ENVIRONhllSN'I'AI, UIS1,IEF AND BEEIAVIOR ITigtire I shows the Stern et al. (1905) aiid Dietz et al. (1998) model. 'Tt~e c:~t~sal inoclel contains I'our basic parts: (a) \oci:il strt~ctttral vari:~bles, (h) ~~rrspecifiecl life viiltres, (c) t~oth ge~ier;~li/ecl i r ~ l r l specific cr~vir.ortriier~l;~I beliefs, aricl ( ( I ) erivi~.on~~ler~taI t)ellavior. 'l'l~e I'irst level corltaiils tliosc sociitl str.uctt~ral ilctors Illat ;we Ii~rgely i~rflcxible, li)r e x a ~ ~ ~ p l c , etl~llicity, age, arttl gentler. Next, are general life values that undergird one's life slicl~ as ideas aboi~t God, cl~ild rearing, or (lie goverrlrirellt. 'The third level incltlcles envir-o~trt~er~tal worlclview or ge~iel.;~l hcliefs 'snnrl values conce~.nirig the e~tvirot~tneot. This level precedes more specific environnlental perceptio~l sirch as atlitrides about energy conservation or protectiort of rvildlife habitats. The fourth level is behavioral corn~nitrnent and intention proceeded by actual behavior. Central to the rnodel is environmental worldview operationitli~etl by the NEP (ilunlap & Van Liere, 1978; Ihnlap, Van Liere, Mertig, Catton, & tiowell, 1992; Dtrnlap, Van Liere, Mertig, & Jones, 2000). Stern et al. (1995) argued that the NEP measures a,fi)lk ecology or lay person's view of relationships in the natural worlcl. I t Iins bee11 i~rgtrecl that the niotlerr~ environmental movement, which gaineel rnorner~turn in the late 1960s (Inglehnrt, 1990), ~~slicretl in an era ol' f~~nt la~l~cr~t i r l allittr le sliil'ts witlr r.cspect to rli~ltrl-e (herice the tlescriptor paratligr-n), anti (lie NEI' is tlrought to rneasure generalizeti beliefs about the hiophysicnl erlvironment nricl tlic 11111uan relatio~lship to it. 'I'he NEP challenges the Donlinant Social Paradigm, which 1)unlap & Van Liere (1978) rnaintainetl j~ervntled Western societies l i r l t i l the 1960s with its ; r~l thro~)o~t~or~) l~ic e i~pllitsis on 11attll.e clorl~i~~atior~ anel t,csotrrce extractiot~ (Taylor, 2000; see also Pirages & El1rlich. 1974). We operationalize a sirl~pler version of the Stern et :I/. ( i 995) moclel-one which tloes not ir~cl~lcle g neral life values, specific environmental beliefs1 attitudes, or behavioral intentions (Figure 2 ) . 2 This motlel is tested for ethnic variation in beliefs and behavior. We propose a causal rnodel in which ethnicity operates on environmental behavior primarily tllrough environolental worldview. h4odel ii~terpretntion flows from top to bottorn where one's position in the social stnrcture is expected to affect his or her general beliefs about nature. Generalized beliefs are then pretlicted to have a direct causal effect on environmer~tal behavior. If the NEP cloes ~ilei~surc a l'olk ecology, the cl~tcstior~ for the present sti~cly is whether this ecology is coilsistei~t arirong etll~lic grot~ps. I t would be prestrrnptuous to assert I11;rt one fi)lk ecology Iiolcls 1 1 L I ~ for different e t l~~ i i c groupings even in the United States where subcultural contrasts among ethnic groups Ilnve not beer1 exl~~erne (Allxr, 1995; Wolf, 190%). 'l'i~is is an irlslit~itional constrni~ils, worlclview, folk ecological theory