Does Living in a Protected Area Reduce Resource Use and Promote Life Satisfaction? Survey Results from and Around Three Regional Nature Parks in Switzerland

Regional nature parks in Switzerland are, for the most part, protected areas that aim to promote sustainable development and residents’ well-being. In recent years, research on regional nature parks and comparable protected areas has focused on questions regarding local populations’ acceptance of such areas, their governance, and their economic effects. However, we know surprisingly little about the impact of protected areas on environmental resource use and life satisfaction, two essential ingredients of sustainable regional development. In this study, we survey people living in and around three regional nature parks in Switzerland on their resource use and life satisfaction (gross sample n = 3358). We propose a novel measurement of resource use based on vignettes describing different lifestyles, which we validate against the carbon footprint obtained for a subsample of our respondents. With these indicators, using multiple regression analyses, we test several hypotheses derived from the literature on the relationship between resource use and life satisfaction in and around protected areas. Contrary to our expectations, we do not find differences in resource use or life satisfaction, or the relationship between resource use and life satisfaction, across park and non-park regions. We discuss potential explanations for our findings and their implications for nature park authorities and future study designs. Supplementary Information The online version contains supplementary material available at 10.1007/s11205-023-03164-z.


OLS regression models for satisfaction with infrastructure and resource use, clustered by municipalities
The simple model including the factor "satisfaction with work and financial matters" as the outcome variable likewise indicates a significant negative relation (Table S2).The coefficients in the models M14 and M15 in Table S2 are only slightly lower than those in the models using the life satisfaction variable presented in the article (Table 3, M1: b = -0.032,p < 0.001, M2: b = -0.032,p < 0.001).Both models indicate a significant negative relation between satisfaction with infrastructure and resource use, as does the model using the global satisfaction index presented in the article (see Table 3).In the "Results" section of the article, we present the results of models testing whether people living in parks have lower resource use and higher self-reported life satisfaction than people living in comparable, non-park regions (HP 2a).Here we tested the same hypothesis with the factors "satisfaction with infrastructure" and "satisfaction with work and financial matters" as outcome variables.Like the models reported in the article (Table 4; M5: b = 0.057, p > 0.5, M7: b = 0.145, p > 0.5) the models M16 and M17 in Table S3 indicate an insignificant relation between the two regions (park and non-park) and the two factors.Estimating the models with "satisfaction with infrastructure" and "satisfaction with work and financial matters" thus does not provide support for our hypothesis either.Levels of satisfaction with these specific aspects (infrastructure, work, and financial situation) are not higher for park inhabitants than for the control group.To test the validation of the life satisfaction variable in the main analyses of the article (see  In the "Results" section of the article, we present models testing whether people living in parks have lower resource use and higher self-reported satisfaction than people living in comparable, non-park regions (HP 2a).
Here we tested the same hypothesis with the global index of satisfaction.Like the models reported in the article (Table 4; M5: b = 0.057, p > 0.5, M7: b = 0.145, p > 0.5), the models M20 and M21 in Table S5 indicate an insignificant relation between the two regions (park and non-park) and the index of satisfaction.Thus, estimating the models with the index does not provide support for our hypothesis either.Table S7 reports the results of OLS regression models without multiple imputations to test Hypothesis 2a.The hypothesis postulates that people living in parks have lower resource use and higher self-reported satisfaction than people living in comparable, non-park regions.Like in the models with multiple imputations in the article (M5 and M7 in Table 4), the results of the models show insignificant relations between regions (park and nonpark) and life satisfaction (see M24 in Table S7) and between regions and resource use (see M25 in Table S7).

OLS regression models for global index of satisfaction and park and non-park regions, clustered by municipalities
Thus, the models without multiple imputations show that the resource use of individuals living in parks is not lower, and their life satisfaction is not higher, than the resource use and life satisfaction of those in the control group -just like the models reported in the "Results" section of the article (see Table 4).Accordingly, it can be assumed that the estimations with multiple imputations presented in the article are reliable.
The table lists coefficient estimates and cluster-robust standard errors (***p < 0.001, **p < 0.01, *p < 0.05, for twosided tests) of simple and multiple OLS regression models with multiple imputations of missing values.The outcome variable of models 12 and 13 is a factor of satisfaction, namely satisfaction with infrastructure.The income variable represents individuals' resource use.The number of clusters corresponds to the number of municipalities.
The table lists coefficient estimates and cluster-robust standard errors (***p < 0.001, **p < 0.01, *p < 0.05, for twosided tests) of simple and multiple OLS regression models with multiple imputations of missing values.The outcome variable in models 20 and 21 is a global index of life satisfaction.The income variable represents individuals living either in RNPs or in the control group.The number of clusters corresponds to the number of municipalities.

Table S2
OLS regression models for satisfaction with work and financial matters and resource use, clustered by municipalities Notes:The table lists coefficient estimates and standard errors (*** p < 0.001, ** p < 0.01, * p < 0.05) of a simple and multiple OLS regression model with multiple imputations of missing values.The outcome variable

Table S3
OLS regression models for satisfaction with infrastructure, satisfaction with work and financial matters and park and nonpark regions, clustered by municipalities Notes:The table lists coefficient estimates and cluster-robust standard errors (***p < 0.001, **p < 0.01, *p < 0.05, for twosided tests) of simple and multiple OLS regression models with multiple imputations of missing values.The outcome variables in models 16 and 17 are factors of satisfaction.The income variable represents individuals living either in RNPs or in the control group.The number of clusters corresponds to the number of municipalities.

Table 2
) and the factors above, we also created a global index of all of the 21 variables relating to people's life satisfaction.Here we tested in an additional analysis whether the relation between satisfaction and resource use is qualitatively different if we use the global satisfaction index (see TableS4) instead of the variable general life satisfaction, as in the model of the article.The models (18 and 19) in TableS4including the global index "satisfaction" as the outcome variable indicate a significantly negative relation between satisfaction with resource use, as does the model presented in the article that includes the single variable of general life satisfaction as the outcome variable (Table3; M1: b = -0.032,p < 0.001, M2: b = -0.032,p < 0.001).

Table S4
OLS regression models for the global index of life satisfaction and resource use, clustered by municipalities Notes:The table lists coefficient estimates and cluster-robust standard errors (*** p < 0.001, ** p < 0.01, * p < 0.05) for twosided tests of simple and multiple OLS regression models with multiple imputations of missing values.The outcome variable of models 18 and 19 is the global index of satisfaction and the income variable is the indicator of resource use.

Table S6
OLS regression models for resource use and life satisfaction, clustered by municipalitiesThe table lists coefficient estimates and cluster-robust standard errors (***p < 0.001, **p < 0.01, *p < 0.05, for twosided tests) of simple and multiple OLS regression models.The outcome variable of models 22 and 23 is the life satisfaction variable (general life satisfaction).The income variable is the resource use indicator.The number of clusters corresponds to the number of municipalities.

Table S7
OLS regression models for life satisfaction and park and non-park regions, as well as resource use and park and non-park regions, clustered by municipalitiesThe table lists coefficient estimates and cluster-robust standard errors (*** p < 0.001, ** p < 0.01, * p < 0.05, for twosided tests) of simple and multiple OLS regression models.The outcome variable of model 24 is satisfaction (life satisfaction in general).The outcome variable of model 25 is the resource use indicator.The income variable represents individuals living either in RNPs or in the control group.The number of clusters corresponds to the number of municipalities.