Search

Griffith University’s Climate Action Beacon conducted the second of five planned Climate Action Surveys in September-December, 2022. The survey aimed to discover what Australians think, feel, and do about climate change and related environmental and climatic events, conditions, and issues. This report gives details of the background to the survey, as well as its methods, major findings, and potential implications. Comparisons are made with findings from the corresponding 2021 survey and with other recent survey research. In 2022, the survey was conducted in two stages. First, to obtain longitudinal data and monitor within-person changes in responses, all 3,915 people who satisfactorily completed the survey questionnaire in 2021 were invited to participate again in 2022. Usable responses were obtained from 1,263 members (32.3%) of this cohort (51.9% female, Mage = 54.2 years). Second, to boost the total sample size to the target of N = 4,000, and to permit estimations of nationwide trends over time, a quota sample of Australian resident adults, stratified by gender, age, and state of Australia (in proportion to the representation of these categories in the national population), was recruited. Data collection closed when usable questionnaires had been received from 2,767 people (50.2% female, Mage = 47.3 years). Therefore, the 2022 sample comprised 4,030 Australian adults (50.7% female, Mage = 49.4 years). Two versions of the online questionnaire were used in 2022 - one for the repeat respondents, and one for the new respondents. The latter questionnaire closely resembled that used in 2021. For the repeat respondents, questions that did not warrant asking a second time in two years were replaced by questions tapping new topics. Both questionnaires comprised almost 200 single items/questions, approximately 30 multi-item composite scales, and several openended questions. Each could be completed in approximately 30 minutes. The survey content pertained to the extent and distribution of different views about climate change; feelings/concerns about the threat and reality of climate change; knowledge of climate change and information sources used to obtain this knowledge; experiences of extreme weather events, natural disasters (including the 2022 Australian floods), and climate change impacts; pro-environmental behaviours and lifestyles; barriers to engaging in these behaviours and lifestyles; and self-views, worldviews and socio-political opinions. Demographic data enabled the identification of group differences in the climate change variables. As was the case in 2021, the survey demonstrates the high prevalence of beliefs in, and concerns about, climate change, and the overwhelming support for government policies that facilitate mitigation of the rate and extent of climate change. Findings have implications for climate change interventions, government policy, future research, and theory development. At the time of writing, more detailed analyses of the quantitative and qualitative data, and deeper consideration of the implications of the findings, are ongoing. Planning has commenced to conduct a third iteration of the survey in the second half of 2023.

The Gaussian chain graph model simultaneously parametrizes (i) the direct effects of $p$ predictors on $q$ correlated outcomes and (ii) the residual partial covariance between pair of outcomes. We introduce a new method for fitting sparse Gaussian chain graph models with spike-and-slab LASSO (SSL) priors. We develop an Expectation-Conditional Maximization algorithm to obtain sparse estimates of the $p \times q$ matrix of direct effects and the $q \times q$ residual precision matrix. Our algorithm iteratively solves a sequence of penalized maximum likelihood problems with self-adaptive penalties that gradually filter out negligible regression coefficients and partial covariances. Because it adaptively penalizes model parameters, our method is seen to outperform fixed-penalty competitors on simulated data. We establish the posterior concentration rate for our model, buttressing our method's excellent empirical performance with strong theoretical guarantees. We use our method to reanalyze a dataset from a study of the effects of diet and residence type on the composition of the gut microbiome of elderly adults.

We study the asymptotic properties of the multivariate spike-and-slab LASSO (mSSL) proposed by Deshpande et al.(2019) for simultaneous variable and covariance selection. Specifically, we consider the sparse multivariate linear regression problem where $q$ correlated responses are regressed onto $p$ covariates. In this problem, the goal is to estimate a sparse matrix $B$ of marginal covariate effects and a sparse precision matrix $\Omega$, which captures the residual conditional dependence structure of the outcomes. The mSSL works by placing continuous spike and slab priors on all the entries of $B$ and on all the off-diagonal elements in the lower-triangle of $\Omega$. Under mild assumptions, we establish the posterior contraction rate for the slightly modified mSSL posterior in the asymptotic regime where both $p$ and $q$ diverge with $n.$ Our results imply that a slightly modified version of Deshpande et al.~(2019)'s mSSL procedure is asymptotically consistent., Comment: arXiv admin note: text overlap with arXiv:2207.07020