Public support for restoration: Does including ecosystem services as a goal engage a different set of values and attitudes than biodiversity protection alone?

Methods A large sample of adult Australians (n = 1,902) was recruited by a professional survey research firm and demographically matched to the Australian population on the basis of sex, age, state of residence, income, and education. The sample frame was the firm’s panel of 350,000+ Aus- tralian consumers with Internet access and respondents were recruited with small incentives (valued at �AUD$4), with quotas used to ensure demographic representation. Prior to the launch of the online survey, the questionnaire was pre-tested for length, intelligibility, word- ing, format, and question order among more than 25 respondents from six Australian states. Pre-test respondents were chosen to have no special knowledge of restoration, or concern for the environment, and were interviewed using verbal probing to understand their cognitive processes in reacting to the survey questions. Interviewees were also asked to suggest possible drawbacks of restoring woodlands in Australia, and we used the most commonly cited con- cerns to construct the question about ecosystem disservices. To measure respondents’ values and attitudes, we used existing scales whose validity has been independently established, rather than inventing a new metric. For core values, we used the 12-item value scale of DeGroot and Steg [16], which scores responses on two self-transcen- dent value orientations (“Biospheric” and “Altruistic”) and on a self-enhancement value orien- tation (“Egoistic”) (S1 Appendix). For environmental attitudes, we used a short version of the

EAI (Milfont & Duckitt 2010), with four statements on the pro-“preservation” axis and four statements on the pro-“utilization” axis (S2 Appendix). We presented two restoration scenarios, a “biodiversity-only” scenario and a “biodiversity- plus-ecosystem-services” scenario (S3 Appendix). The biodiversity-only scenario (hereafter, BO) involved eradication of invasive species on an offshore island, and was written to preclude the possibility of humans benefiting directly from ecosystem services. The biodiversity-plus- ecosystem-services scenario (hereafter, BES) involved revegetation of a local woodland and the promise of a diverse set of potential ecosystem services and biodiversity improvements, includ- ing carbon sequestration, habitat, and recreation. Respondents viewed regionally specific ver- sions of the scenario, using photographs and descriptions of local plant and animal species, to make each scenario more plausible as a potentially local restoration effort; otherwise, scenarios did not vary in wording or in the services promised. After viewing the BES scenario, respon- dents were asked to choose their preferred benefit from a list of those that could result from the restoration. Half of the respondents saw the BO scenario first and the other half saw the BES scenario first. In both cases, respondents were told that private donations were needed to accomplish each restoration project and asked if they would be willing to donate. If so, respondents chose amounts from a payment card with exponentially increasing amounts, ranging from $1 to

“more than $500” [47]. Those who were unwilling to pay (i.e., answered “no”) were asked to explain why, by choosing a set response or giving an open-ended explanation. Answers to this question were used to distinguish “true zeroes,” or people truly unwilling or unable to pay, from “protest noes,” those who objected to aspects of the payment scenario. When willingness to pay was treated as a binary category, only true zeroes were treated as “zero” and any specific dollar amounts were treated as “nonzero.” For analyses of WTP as a continuous variable, true zeroes were coded as $0 WTP, responses of “more than $500” were coded as $750, and protest noes were eliminated. The scenarios devised to test whether the presence of an ecosystem services rationale increased WTP necessarily differed in other aspects, most importantly the location and type of restoration activity. We tolerated these differences because there are few plausible ecological restoration efforts that totally lack ES benefits (and none at all that would produce exclusively PLOS ONE Values and attitudes towards biodiversity and ecosystem service goals of restoration PLOS ONE | https://doi.org/10.1371/journal.pone.0245074 January 19, 2021 4 / 15