Research

INTRODUCTION

Beliefs and expectations regarding future climatic conditions drive private adaptation and mitigation actions as well as support for public interventions (Howden et al. 2007, Niles et al. 2016). Although expectations and actions among urban residents are especially important for many OECD countries, primary industry was responsible for 11% of New Zealand’s total GDP, 14% of its total employment, and 52.2 billion NZD in export revenue in 2022 (Ministry for the Primary Industries 2022). The primary sector was also responsible for approximately 50% of national greenhouse gas (GHG) emissions in 2020 (Ministry for the Environment 2022), cf. 17% globally (FAO 2020). Given the important role of primary production in New Zealand’s economy and identity (Stahlmann-Brown and Walsh 2022), understanding climate expectations among those working in the country’s primary sector is critical for designing policy and communicating the risks to the general population.

There exists a large literature on the heterogeneity of belief in climate change across groups (e.g., Poortinga et al. 2011, Lee et al. 2015, Hornsey et al. 2016, Milfont et al. 2017, Whitmarsh and Capstick 2018) as well as support for public policies to address climate change across groups (e.g., Rhodes et al. 2017, Sælen and Aasen 2023). A further literature has demonstrated increasing polarization in climate-related beliefs across groups (Capstick et al. 2015, Dunlap et al. 2016, Bolsen and Shapiro 2018, Merkley and Stecula 2018, Jenkins-Smith et al. 2020). Despite such extensive inquiry, hardly anything is known about the heterogeneity of climate beliefs within groups. Heterogeneity in beliefs results in less efficient information targeting, which may pose problems for policy makers or industry groups who have limited budgets with which to encourage climate mitigation or adaptation. As such, understanding heterogeneity in climate beliefs has considerable practical importance.

We use the 2021 wave of the Survey of Rural Decision Makers, a national survey of farmers, foresters, growers, and lifestyle block owners in New Zealand with over 6700 respondents, to provide novel evidence about the heterogeneity in climate beliefs at the individual level.^[1] To do so, we employ the flexible, conditional-moments approach developed by Just and Pope (1978) and Antle (1983) and more recently applied by Cissé and Barrett (2018) and Wesselbaum et al. (2023). We estimate the dispersion in climate change beliefs among individuals conditional on their individual characteristics (such as demographic factors). This is a direct measure of the heterogeneity in beliefs about climate change.

Our analysis shows that while 88% of survey respondents believe that climate change is real, there exists substantial heterogeneity in beliefs. We find that education, gender, region of residence, and industry are the key drivers of beliefs in climate change and (to varying degrees) of the heterogeneity in beliefs. Specifically, women, younger respondents, more educated respondents, and respondents who are less trusting in social media are more likely to believe in climate change. Further, beliefs are more heterogeneous among males (young and old), the less educated, and those who trust social media.

DATA

Survey data

The Survey of Rural Decision Makers is a long-running, large-scale, Internet-based survey of commercial farmers, foresters, and growers as well as lifestyle block owners from across New Zealand (Stahlmann-Brown 2021). The survey has been conducted biennially since 2013.

The 2021 wave of the Survey of Rural Decision Makers comprised 130 questions, recording up to 599 datapoints for each respondent. That said, because of branching and randomization, no respondent saw every question. The questionnaire covered land use and land-use change, land management, personal values and motivations, trust in various sources of information, demographics, and belief in climate change. With the purpose of this paper in mind, the questionnaire was strongly informed by previous findings on climate beliefs (e.g., Poortinga et al. 2011, Hornsey et al. 2016, Milfont et al. 2017, Whitmarsh and Capstick 2018, Booth et al. 2020, Diehl et al. 2021).

The survey was open from 1 June 2021 until 15 August 2021. In total, 6740 respondents representing the breadth of New Zealand primary industry across all 16 regions and all 66 districts participated. Respondent demographics are representative of the primary sector as a whole, although the dairy and sheep and beef industries have intentionally been oversampled since 2015 (Brown 2015).

Questions regarding accessibility for rural populations and representativeness are often levied against online survey platforms. However, New Zealand’s Internet penetration rate was 94.9% at the start of 2022 (Datareportal 2022). Moreover, the survey was optimized for mobile phones and other devices to increase accessibly for those without high-speed Internet access in their homes.

Dependent variable

We analyze belief in climate change, considering both the belief itself and heterogeneity in that belief. The 2021 Survey of Rural Decision Makers included the following question: “Which of the following statements best describes your personal thoughts about climate change?” Respondents chose from the following answer options:

1. Climate change is real and is already affecting New Zealand (positively or negatively)
2. Climate change is real. Although it is not yet affecting New Zealand, it will in the next 10 years (positively or negatively)
3. Climate change is real. Although it will not affect New Zealand in the next 10 years, it will in the future (positively or negatively)
4. Climate change is real, but it will not affect New Zealand
5. Climate change is not real

To evaluate belief, we constructed a dummy variable that is equal to one if the respondent selected answer 1, 2, or 3 and zero if the respondent selected answer 4 or 5. There are three reasons for dichotomizing our response variable. First, it follows best practice in the related literature (e.g., Hornsey et al. 2016, Lewis et al. 2019, Stahlmann-Brown and Walsh 2022). Second, ordinal regression models do not allow to compute residuals that are used in the second step of our estimation approach. Third, we aimed to distinguish those who think that climate change is real and has impacts from those who believe that it is not real and/or does not have impacts.

Independent variables

The extensive literature on belief in climate change has consistently found that older people, men, and people with less education are more skeptical of climate change and its anthropogenic causes both internationally (e.g., Poortinga et al. 2011, McCright et al. 2016, Lewis et al. 2019) and in New Zealand (Milfont et al. 2017, Stahlmann-Brown and Walsh 2022). Similarly, indigenous peoples with environmentally attuned belief systems and climate-vulnerable populations often have greater awareness of climate change (e.g., Boillat and Berkes 2013, Salick et al. 2013, Eisenstadt and West 2017, Aria-Bustamante and Innes 2020). Surveys of farmers and other primary producers have further identified that the length of family farming history, an indicator of intergenerational experience and knowledge (Fiske 2016, Booth et al. 2020, Stahlmann-Brown and Walsh 2022), also negatively influences belief in climate change. Therefore, we analyzed these demographic factors in understanding belief in climate change and heterogeneity thereof.

To account for nonlinearities, age was measured both as a level and its square. Gender was measured as a dummy for male. Indigenous identity was measured as a dummy indicating whether the respondent identifies as Māori, the indigenous population of New Zealand. We constructed three indicator variables to measure education: whether respondents had secondary education or less, professional training in the form of a certificate or diploma, or higher education in the form of a bachelor’s degree, post-graduate diploma, master’s degree, or doctoral degree. Family farm experience was derived from the question, “How many generations has your family been involved in farming, forestry, and growing food in Aotearoa New Zealand?”

Because personal values also inform belief in climate change (e.g., Poortinga et al. 2011, Hornsey et al. 2016, Whitmarsh and Capstick 2018), we also accounted for several attitudinal measures in our analysis. For example, because risk preferences have been shown to negatively correlate with climate adaptation (e.g., Sara et al. 2016), we included Dohmen et al. (2011)’s self-reported propensity for risk. The question was: “Are you generally a person who is fully prepared to take risks or do you try to avoid taking risks?” with values ranging from 0 (don’t like taking risks) to 10 (fully prepared to take risks). Adhering to social norms has also been shown to positively correlate with belief in climate change (Cialdini and Jacobsen 2021). In our study, social norms were measured via the extent to which respondents agreed with the statements “My/our values are similar to those of other commercial operators in the industry (or other lifestyle block owners)” and “My/our management practices are similar to those of other commercial operators in the industry (or other lifestyle block owners),” recorded on a scale from 0 (strongly disagree) to 10 (strongly agree). Last, higher trust in social media platforms (e.g., YouTube, Facebook, and Twitter) as a credible news source is associated with lower climate skepticism (Diehl et al. 2021). Thus, we also analyzed the extent to which trust in social media impacts belief in climate change; specifically, our measure indicated whether survey respondents considered social media to be among the most important sources of information on topics (including climate change) facing primary industry in New Zealand. Using and trusting social media can close information gaps because of the different nature of consuming social media news: users infer information from peers in their network and may thus be more exposed to news about the climate, for example, breaking news to weather-related disasters (Diehl et al. 2021).

Descriptive statistics

Figure 1 presents a map of New Zealand showing heterogeneity in climate change beliefs at the postcode level. The map suggests substantial heterogeneity within and across postcodes.

Table 1 presents descriptive statistics.^[2] In our sample, 88% of respondents believe that climate change is real and that it will affect New Zealand. The average respondent was 61 years old, male (68.9%), and was a third-generation farmer. Most respondents had a university degree (40.3%) or professional training (33.2%). The average score for risk preferences was 5.13 on a 0–10 scale, with a large standard deviation. Most respondents stated that they have common values (5.94) and management practices (5.66) relative to their peers. Approximately 10% of respondents considered social media to be among their most trusted sources of information. Finally, 5.4% of respondents identified as Māori.

Methodology

Given the rich survey data, we estimated the heterogeneity in climate change beliefs directly from individual-level data by applying the two-step method proposed by Just and Pope (1978) and Antle (1983). The reason for using this method is that we want to estimate the inequality in beliefs directly from individual-level data allowing us to include as much individual-level information about the survey respondents. An alternative to this approach would be to use multiple survey waves and compute the inequality in beliefs across individuals for each survey wave. However, this approach would rely on aggregating variables, which we consider to be less informative and tenable.

In this method, we first estimated the conditional mean of climate change beliefs on our set of controls and then calculated the squared residuals. Because squared residuals are mean zero, we constructed the conditional variance directly as the conditional expectation of the squared residuals. Second, we ran the same regression used for the conditional mean using the squared residuals as the dependent variable. Intuitively, the conditional variance estimated in the second step reflects the interpersonal dispersion of climate change beliefs conditional on individual-level observables and can therefore be interpreted as a direct measure of heterogeneity.

Step 1 uses the climate change beliefs (Y) of individual i in region j in industry k as the dependent variable, Y_i,j,k. X_i,j,t is a vector of variables that may influence belief. Step 2 then uses the squared residuals from the first step, ε_i,j,k, as dependent variable.

In short, the two-steps involve estimating the following models using OLS:

(1)

(2)

We were interested in the estimated coefficients, β, in both steps capturing the effects of our control variables on mean and heterogeneity in beliefs. Both models included region, μ_j, and industry, μ_k, fixed effects. Standard errors were robust to heteroscedasticity.

We then wanted to quantify the impact each variable (or group of variables) has on the variation in the mean and heterogeneity of beliefs. To do so, we applied the Shorrocks-Shapley decomposition (Shapley 1953, Shorrocks 1982, 2013). Intuitively, the method works as follows: Assume that factors X_i potentially drive outcome Y, where i = 1, ... , K. Then, assume that there exists some aggregator function, f, such that Y = f(X₁, ... , X_K), linking the K factors to the outcome variable in a potentially nonlinear way. The role of the decomposition method is to assign contributions, C_i, to each of the K factors. The solution to this problem turns out to be equivalent to the solution found by Shapley (1953) for the problem of allocating output to a set of contributors and relies on calculating the so-called Shapley values.

RESULTS

Table 2 presents the results from estimating the two-step approach (eq. 1 and 2). We present a core model with only key demographic variables (columns 1 and 2) and an extended model that includes additional, potentially endogenous variables of interest (columns 3 and 4). Finally, columns 5 and 6 present the Shapley decomposition results for the extended model (values shown are the percentage contributions to the variation in R²).

Mean beliefs

We begin by discussing the drivers of belief in climate change to verify that the commonly observed drivers of the beliefs in climate change are also found in our study. We find that age had a significant effect on believing in climate change. The relationship is inverted-U shaped with a maximum at 59.54 years. This is somewhat unexpected as the literature generally finds that younger people are more likely to believe in climate change (Hornsey et al. 2016). We find that males are less likely (p < 0.01) to believe in climate change than women: Being male reduces the likelihood of believing in climate change relative to females by 4.3%, in line with the literature (Hornsey et al. 2016). Family farming history also affects belief in climate change (p < 0.01), with respondents who have longer family histories of farming being less likely to believe in climate change. However, the effect is quantitatively small: each subsequent generation more reduces the likelihood of belief by 0.9%. We find that more educated respondents are more likely to believe in climate change compared to those with secondary education or less (p < 0.01), also in line with the results from the meta-analysis by Hornsey et al. (2016). For example, a university degree increases the likelihood of belief in climate change by 9.9% over having a secondary education or less (p < 0.01). Finally, we find that trusting in social media reduces the likelihood of belief in climate change by 3.6% (p < 0.05). Having similar management practices marginally significantly (p < 0.1) increases the likelihood of belief in climate change.

Our identified drivers of climate change beliefs are in line with the commonly found drivers in the literature. These include personal characteristics, social identity, and personal experience (e.g., Borick and Rabe 2010, Hornsey et al. 2016). In contrast to the literature (e.g., Diehl et al. 2021), however, we find that trust in social media reduces the belief in climate change and increases the variance in beliefs. The reason for why we find differences could be that the study by Diehl et al. (2021) uses cross-country survey data from the general population rather than focusing on farmers as we do.

Beyond estimating the effect of these variables, we also want to quantity their relative importance for explaining the variation in mean beliefs. Column 5 shows the Shapley decomposition results for the mean beliefs. We find that education is the key driver and accounts for 32.88% of the R² of mean climate change beliefs. The other key drivers are industry (22.30%), generations (14.08%), region (11.54%), and gender (10.22%). That is, education and industry jointly explain more than half of the variation in mean climate change beliefs.

Heterogeneity in beliefs

We now turn to our main outcome variable of interest: the heterogeneity in climate change beliefs (columns 2 and 4). Although we can relate our results for (mean) beliefs to the literature, we are not aware of research on the drivers of heterogeneity in beliefs in climate change. There are also no theoretical arguments that the effect of a variable on the heterogeneity needs to be of a given sign or the same (or opposite) to the effect of this variable on mean beliefs.

Our results indicate that men exhibit larger heterogeneity in belief in climate change than women (p < 0.01). Put differently, climate change beliefs are more equally distributed around the mean for women than men. Higher education also reduces the heterogeneity in beliefs (p < 0.01): the higher the education, the more consistent beliefs are around the mean. The length of family farming history is associated with higher heterogeneity in beliefs (p < 0.01). Higher trust in social media is also associated with greater heterogeneity in beliefs (p < 0.05). Finally, we find that age has a marginally significant (p < 0.1) effect on heterogeneity in beliefs. The relationship is U-shaped, with higher heterogeneity amongst younger and older respondents.

Having discussed which factors affect the heterogeneity in beliefs, it is important to indicate their relative importance. Column (6) uses the Shapley values to quantify the relative importance of these variables for explaining the heterogeneity in beliefs. We find that the key drivers of the variation in the heterogeneity in beliefs are education (36.8%), industry (24.2%), region (10.4%), and gender (11.9%).

CONCLUSION

Beliefs matter for actions. In the context of climate policy, understanding who believes in climate change and who does not is paramount for designing effective and efficient communication designed to foster climate action uptake and buy-in for climate policies (e.g., Capstick et al. 2015, Bolsen and Shapiro 2018).

Although there exists a large literature on the drivers of climate change beliefs (e.g., Hornsey et al. 2016), less is known about what drives heterogeneity in beliefs. Although perceiving one’s management practices to be similar to those of one’s peers is associated with higher belief in climate change but not with greater or lesser heterogeneity in those beliefs, several drivers of mean beliefs also underpin heterogeneity in belief, including age, gender, education, family farming history, and trust in social media. Indeed, young age, male gender, lower education, lengthier family histories of farming, and higher trust in social media are all associated with both lower belief in climate change and greater heterogeneity in those beliefs.

Many governments have implemented campaigns to encourage farmers to adopt practices to reduce greenhouse gas emissions and/or to adapt to climate change (Detenber et al. 2016). Given heterogeneity in belief in climate change across groups, different messages may fruitfully be targeted toward different audiences, as shown by Bostrom et al. (2013), Hine et al. (2014), and others. Unfortunately, greater heterogeneity in beliefs renders information targeting less efficient, and thus policy makers with constrained budgets may either be forced to ignore some segments altogether or to increase the efficiency of messaging by reducing heterogeneity within segments. Informed by the Shapley decomposition results, targeting men, less educated people, and farmers with longer family histories of farming for information dissemination appears most promising to reduce heterogeneity in beliefs.

__________

^[1] “Lifestyle blocks” is the New Zealand term for “rural residential living” (Australia) or “hobby farming” (UK and US). These are rural properties that are maintained as a secondary source of income and/or for lifestyle choice rather than as a primary source of income.
^[2] Table A.1 and A.2 in Appendix 1 present the break-down of respondent by industry and region, respectively.

LITERATURE CITED

Antle, J. M. 1983. Testing the stochastic structure of production: a flexible moment-based approach. Journal of Business & Economic Statistics 1(3):192-201. https://doi.org/10.1080/07350015.1983.10509339

Arias-Bustamante, J. R., and J. L. Innes. 2020. Mapuche spirituality and its contribution to climate change mitigation. Pages 1-32 in W. Leal Filho, J. Luetz, D. Ayal, editors. Handbook of climate change management: research, leadership, transformation. Springer, Cham, Switzerland. https://doi.org/10.1007/978-3-030-22759-3_119-1

Boillat, S., and F. Berkes. 2013. Perception and interpretation of climate change among Quechua farmers of Bolivia: indigenous knowledge as a resource for adaptive capacity. Ecology and Society 18(4):21. https://doi.org/10.5751/ES-05894-180421

Bolsen, T., and M. A. Shapiro. 2018. The US news media, polarization on climate change, and pathways to effective communication. Environmental Communication 12(2):149-163. https://doi.org/10.1080/17524032.2017.1397039

Booth, P., P. J. Walsh, and P. Stahlmann-Brown. 2020. Drought intensity, future expectations, and the resilience of climate beliefs. Ecological Economics 176:106735. https://doi.org/10.1016/j.ecolecon.2020.106735

Borick, C. P., and B. G. Rabe. 2010. A reason to believe: examining the factors that determine individual views on global warming. Social Science Quarterly 91(3):777-800. https://doi.org/10.1111/j.1540-6237.2010.00719.x

Bostrom, A., G. Böhm, and R. E. O'Connor. 2013. Targeting and tailoring climate change communications. Wiley Interdisciplinary Reviews: Climate Change 4(5):447-455. https://doi.org/10.1002/wcc.234

Brown, P. 2015. Survey of rural decision makers. Manaaki Whenua Landcare Research, Lincoln, New Zealand.

Capstick, S., L. Whitmarsh, W. Poortinga, N. Pidgeon, and P. Upham. 2015. International trends in public perceptions of climate change over the past quarter century. Wiley Interdisciplinary Reviews: Climate Change 6(1):35-61. https://doi.org/10.1002/wcc.321

Cialdini, R. B., and R. P. Jacobson. 2021. Influences of social norms on climate change-related behaviors. Current Opinion in Behavioral Sciences 42:1-8. https://doi.org/10.1016/j.cobeha.2021.01.005

Cissé, J. D., and C. B. Barrett. 2018. Estimating development resilience: a conditional moments-based approach. Journal of Development Economics 135:272-284. https://doi.org/10.1016/j.jdeveco.2018.04.002

Datareportal. 2022. Digital 2022: Global overview report. https://datareportal.com/reports/digital-2022-global-overview-report

Detenber, B., S. Rosenthal, Y. Liao, and S. S. Ho. 2016. Climate and sustainability| audience segmentation for campaign design: addressing climate change in Singapore. International Journal of Communication 10:23.

Diehl, T., B. Huber, H. Gil de Zúñiga, and J. Liu. 2021. Social media and beliefs about climate change: a cross-national analysis of news use, political ideology, and trust in science. International Journal of Public Opinion Research 33(2):197-213. https://doi.org/10.1093/ijpor/edz040

Dohmen, T., A. Falk, D. Huffman, U. Sunde, J. Schupp, and G. G. Wagner. 2011. Individual risk attitudes: measurement, determinants, and behavioural consequences. Journal of the European Economic Association 9:522-550. https://doi.org/10.1111/j.1542-4774.2011.01015.x

Dunlap, R. E., A. M. McCright, and J. H. Yarosh. 2016. The political divide on climate change: partisan polarization widens in the U.S. Environment: Science and Policy for Sustainable Development 58(5):4-23. https://doi.org/10.1080/00139157.2016.1208995

Eisenstadt, T. A., and K. J. West. 2017. Indigenous belief systems, science, and resource extraction: climate change attitudes in Ecuador. Global Environmental Politics 17(1):40-58. https://doi.org/10.1162/GLEP_a_00389

Fiske, S. J. 2016. Global climate change from the bottom up. Pages 143-172 in C. Wasson, M. Odell Butler, and J. Copeland-Carson, editors. Applying anthropology in the global village. Routledge, New York, New York, USA. https://doi.org/10.4324/9781315434650-6

Food and Agriculture Organization (FAO). 2020. Emissions due to agriculture. Global, regional and country trends 2000-2018. FAOSTAT Analytical Brief Series No 18. FAO, Rome, Italy.

Hine, D. W., J. P. Reser, M. Morrison, W. J. Phillips, P. Nunn, and R. Cooksey. 2014. Audience segmentation and climate change communication: conceptual and methodological considerations. Wiley Interdisciplinary Reviews: Climate Change 5(4):441-459. https://doi.org/10.1002/wcc.279

Hornsey, M. J., E. A. Harris, P. G. Bain, and K. S. Fielding. 2016. Meta-analyses of the determinants and outcomes of belief in climate change. Nature Climate Change 6(6):622-626. https://doi.org/10.1038/nclimate2943

Howden, S. M., J. F. Soussana, F. N. Tubiello, N. Chhetri, M. Dunlop, and H. Meinke. 2007. Adapting agriculture to climate change. Proceedings of the National Academy of Sciences 104(50):19691-19696. https://doi.org/10.1073/pnas.0701890104

Jenkins-Smith, H. C., J. T. Ripberger, and C. L. Silva, D. E. Carlson, K. Gupta, N. Carlson, A. Ter-Mkrtchyan, and R. E. Dunlap. 2020. Partisan asymmetry in temporal stability of climate change beliefs. Nature Climate Change 10:322-328. https://doi.org/10.1038/s41558-020-0719-y

Just, R. E., and R. D. Pope. 1978. Stochastic specification of production functions and economic implications. Journal of Econometrics 7:67-86. https://doi.org/10.1016/0304-4076(78)90006-4

Lee, T. M., E. M. Markowitz, P. D. Howe, C.-Y. Ko, A. A. Leiserowitz. 2015. Predictors of public climate change awareness and risk perception around the world. Nature Climate Change 5:1014-1020. https://doi.org/10.1038/nclimate2728

Lewis, G. B., R. Palm, and B. Feng. 2019. Cross-national variation in determinants of climate change concern. Environmental Politics 28(5):793-821. https://doi.org/10.1080/09644016.2018.1512261

McCright, A. M., S. T. Marquart-Pyatt, R. L. Shwom, S. R. Brechin, and S. Allen. 2016. Ideology, capitalism, and climate: explaining public views about climate change in the United States. Energy Research & Social Science 21:180-189. https://doi.org/10.1016/j.erss.2016.08.003

Merkley, E., and D. A. Stecula. 2018. Party elites or manufactured doubt? The informational context of climate change polarization. Science Communication 40(2):258-274. https://doi.org/10.1177/1075547018760334

Milfont, T. L., M. S. Wilson, and C. G. Sibley. 2017. The public’s belief in climate change and its human causes are increasing over time. PLoS ONE 12(3):e0174246. https://doi.org/10.1371/journal.pone.0174246

Ministry for the Environment. 2022. New Zealand’s Greenhouse Gas Inventory 1990-2020. Ministry for the Environment, Wellington, New Zealand.

Ministry for Primary Industries. 2022. Situation and Outlook for Primary Industries, June 2022. Ministry for Primary Industries, Wellington, New Zealand.

Niles, M. T., M. Brown, and R. Dynes. 2016. Farmer’s intended and actual adoption of climate change mitigation and adaptation strategies. Climatic Change 135(2):277-295. https://doi.org/10.1007/s10584-015-1558-0

Poortinga, W., A. Spence, L. Whitmarsh, S. Capstick, and N. F. Pidgeon. 2011. Uncertain climate: an investigation into public scepticism about anthropogenic climate change. Global Environmental Change 21(3):1015-1024. https://doi.org/10.1016/j.gloenvcha.2011.03.001

Rhodes, E., J. Axsen, and M. Jaccard, M. 2017. Exploring citizen support for different types of climate policy. Ecological Economics 137:56-69. https://doi.org/10.1016/j.ecolecon.2017.02.027

Sælen, H. G., and M. Aasen. 2023. Exploring public opposition and support across different climate policies: poles apart? Ecological Economics 209:107838. https://doi.org/10.1016/j.ecolecon.2023.107838

Salick, J., A. Byg, and K. Bauer. 2013. Contemporary Tibetan cosmology of climate change. Journal for the Study of Religion, Nature & Culture 6(4). https://doi.org/10.1558/jsrnc.v6i4.447

Sara, L. M., S. Jameson, K. Pfeffer, and I. Baud. 2016. Risk perception: the social construction of spatial knowledge around climate change-related scenarios in Lima. Habitat International 54:136-149. https://doi.org/10.1016/j.habitatint.2015.12.025

Shapley, L. 1953. A value for n-person games. Pages 307-317 in H. W. Kuhn and A. W. Tucker, editors. Contributions to the theory of games, Vol. 2. Princeton University Press, Princeton, New Jersey, USA.

Shorrocks, A. F. 1982. Inequality decomposition by factor components. Econometrica 50(1):193-211. https://doi.org/10.2307/1912537

Shorrocks, A. F. 2013. Decomposition procedures for distributional analysis: a unified framework based on the Shapley value. Journal of Economic Inequality 11(1):99-126. https://doi.org/10.1007/s10888-011-9214-z

Stahlmann-Brown, P. 2021. Survey of rural decision makers 2021. Manaaki Whenua Landcare Research, Lincoln, New Zealand https://doi.org/10.7931/3tcs-wb24

Stahlmann-Brown, P., and P. Walsh. 2022. Soil moisture and expectations regarding future climate: evidence from panel data. Climatic Change 171:1. https://doi.org/10.1007/s10584-022-03317-y

Wesselbaum, D., M. D. Smith, C. B. Barrett, and A. Aiyar. 2023. A food insecurity Kuznets Curve? World Development 165:106189. https://doi.org/10.1016/j.worlddev.2023.106189

Whitmarsh, L., and S. Capstick. 2018. Perceptions of climate change. Pages 13-33 in S. Clayton and C. Manning, editors. Psychology and climate change. Academic Press, London, UK. https://doi.org/10.1016/B978-0-12-813130-5.00002-3