Empathizing With Nature: The Effects of Perspective Taking on Concern for Environmental Issues - Statistical Data Included

Journal of Social Issues, Fall, 2000 by P. Wesley Schultz

Results are based on maximum likelihood estimates produced from covariance matrices. Analyses indicated that the independence model could be rejected (df = 66, [[chi].sup.2] = 2200.28, [[chi].sup.2]/df 33.37, root mean-square error of approximation [RMSEA] = .29, goodness-of-fit index [GFI] = .38, adjusted GFI [AGFI] = .26, Tucker-Lewis index [TLI] = .00). The one-factor model showed an improved, but still unacceptable fit (df = 54, [[chi].sup.2] = 821.69, [[chi].sup.2]/df 15.22, RMSEA = .19, GFI = .68, AGFI = .54, TLI = .56). The two-factor model was significantly better, [[chi].sup.2](1) = 406.47, p [less than] .001, than the one-factor model (df = 53, [[chi].sup.2] = 415.22, [[chi].sup.2]/df 7.83, RMSEA = .13, GFI = .83, AGFI = .75, TLI = .80), but did not provide an acceptable fit--all of the fit indices were beyond my established limits. The three-factor model showed a significant, [[chi].sup.2](2) = 196.74, p [less than].001, improvement over the two-factor model and provided an overall acceptable fit (df = 51, [[chi].sup.2] = 218.48, [[chi].sup.2]/df = 4.28, RMSEA = .08, GFI = .92, AGFI = .90, TLI = .90). The unstandardized factor weights, standardized factor weights (shown in parentheses), covariances between the three factors, and correlation coefficients between the three factors (shown in parentheses) are presented in Figure 1.

Study 2

The results from Study 1 showed support for the distinction between egoistic, altruistic, and biospheric attitudes of environmental concern. The second study was an experimental attempt to activate different environmental concerns using a perspective-taking manipulation. I have argued that the types of concerns an individual has for environmental problems are associated with the degree to which the individual includes nature within his or her cognitive representations of self. Based on this perspective, I predicted that taking the perspective of another person or an animal would lead to a greater inclusiveness and subsequently, greater levels of biospheric environmental concern.

Methodology

Sample. Participants in the study were 180 undergraduates recruited from the psychology department's human participant pool.

Materials. Participants were randomly assigned to view one of three sets of pictures: people engaging in recreational activities in a natural environment (a woman meditating on the beach, a hiker in a forest, a painter near a lake, a rock climber, a person canoeing), animals in a natural environment (a caribou on a hill, gorillas in a forest, a bear, a rhinoceros on a savanna, a breaching whale), or animals being harmed by nature (a seal caught in a fishing net, an eagle on a smoky factory smokestack, an otter in an oil spill, a bear in a trash pile, a bird with a plastic bag around its neck). Color images were shown on a 15" SVGA color monitor in 8-bit color using Microsoft PowerPoint 4.0. Each image appeared on the screen for 30 s; participants viewed a total of five slides.

After viewing the slides, participants completed a questionnaire that contained several measures of environmental attitudes, including the 12 environmental concern items identified above. Separate scale scores were produced for egoistic, altruistic, and biospheric concerns by averaging the four items in each domain. Alpha reliabilities for the three subscales were all high: egoistic (alpha = .91), altruistic (alpha = .92), and biospheric (alpha = .94).