Cosmopolitan Civil Societies: An Interdisciplinary Journal Vol. 15, No. 2 2023 © 2023 by the author(s). This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International (CC BY 4.0) License (https:// creativecommons.org/ licenses/by/4.0/), allowing third parties to copy and redistribute the material in any medium or format and to remix, transform, and build upon the material for any purpose, even commercially, provided the original work is properly cited and states its license. Citation: Kamp, A., Dunn, K., Sharples, R., Denson, N., Diallo, T. 2023. Understanding Trust in Contemporary Australia Using Latent Class Analysis. Cosmopolitan Civil Societies: An Interdisciplinary Journal, 15:2, 84– 104. https:// doi.org/10.5130/ccs.v15.i2.8595 ISSN 1837- 5391 | Published by UTS ePRESS | https://epress. lib.uts.edu.au/journals/index. php/mcs 84 ARTICLE (REFEREED) Understanding Trust in Contemporary Australia Using Latent Class Analysis Alanna Kamp*, Kevin Dunn, Rachel Sharples, Nida Denson, Thierno Diallo Western Sydney University, Australia Corresponding Author: Alanna Kamp, Western Sydney University, Locked Bag 1797, Penrith, NSW 2751, Australia, a.kamp@westernsydney.edu.au DOI: https://doi.org/10.5130/ccs.v15.i2.8595 Article History: Received 29/03/2023; Revised 20/06/2023; Accepted 14/07/2023; Published 30/07/2023 Abstract In 2019, an online survey of 2,015 Australian residents examined the extent of trust of various groups and institutions. A Latent Class Analysis (LCA) of the results generated a typology of trust in Australia. The LCA uncovered four classes based on levels of trust as well as associated demographic profiles and attitudes. The four groups were: those that are very distrusting (15%); those that are largely unsure about how much they can trust various groups and institutions (17%); those that are somewhat trusting (42%); and those that are largely trusting (26%). The largely trusting group was differentiated by their holistic trust in institutions and trust in other Australians (no matter their background). Discomfort with cultural difference was a defining characteristic of the very distrusting class. Examination of these four groups helps understand concerns of Australians and enable the development of strategies to address institutional and interpersonal distrust. Keywords Trust; Social Harmony; Latent Class Analysis; Australia; National Survey DECLARATION OF CONFLICTING INTEREST The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. FUNDING This research was funded by the Special Broadcasting Service (SBS). https://creativecommons.org/licenses/by/4.0/ https://creativecommons.org/licenses/by/4.0/ https://creativecommons.org/licenses/by/4.0/ https://doi.org/10.5130/ccs.v15.i2.8595 https://doi.org/10.5130/ccs.v15.i2.8595 https://epress.lib.uts.edu.au/journals/index.php/mcs https://epress.lib.uts.edu.au/journals/index.php/mcs https://epress.lib.uts.edu.au/journals/index.php/mcs mailto:a.kamp@westernsydney.edu.au https://doi.org/10.5130/ccs.v15.i2.8595 Introduction Trust is essential for the successful functioning of society – politically, economically, and socially. For example, trust in institutions and social trust is a key variable in civic and social cooperation and thus economic growth and the performance of society’s institutions (Putnam 1993; Fukuyama 1995; Moesen, van Puyenbroeck, & Cherchye 2000; Zak & Knack 2001; Fazio et al. 2018). In the words of La Porta et al. (1997) trust facilitates ‘a propensity of people in a society to cooperate to produce socially efficient outcomes’ (p. 333). Trust (and corresponding cooperation) is also central to wellbeing. Personal wellbeing is higher in societies with higher levels of institutional trust and cooperation (Putnam 2000; Hudson 2006). From a sociological perspective, trust is relational, not purely psychological as it ‘characterises the state of a relationship, not just the actors involved’ (Cook & Santana 2020, p. 191). From this standpoint, trust is not only ‘a person’s attitude toward others or simply the disposition to be trusting’ but is reliant on the incentives the parties in the mutual relationship have to be trustworthy with respect to one another, and understandings of each other’s competence, integrity, and the commitment to ‘do no harm’ (Cook & Santana 2020, p. 191; see also Cook, Hardin, & Levi 2005, Hardin 2002). The media, and the quality of public information, is key to trust. The World Economic Forum (2018) called online misinformation ‘a digital wildfire’ and numerous international studies have found a link between conspiracy theory beliefs and distrust of government (Georgiou, Delfabbro, & Balzan 2020; Miller, Saunders, & Farhart 2016; van Prooijen & Douglas 2017). In recent ‘crisis’ times, the issue of trust has come to the fore regarding the relationship between distrust and conspiratorial thinking during the COVID-19 pandemic, particularly distrust in science (Agley & Xiao 2021) and in government (Georgiou, Delfabbro, & Balzan 2020). There are also indicators that the COVID-19 pandemic has seen a bifurcation in trust in public institutions (that is, some public institutions being more trusted than others), across different countries (some countries showing higher levels of trust than others) and by type of institution, for example commercial media versus public media, business and NGOs, and different branches of government (Kye & Hwang 2020; Edelman 2021; Karić & Međedović 2021; Scandurra et al. 2021). Trust in the media and public information varies across partisanship and demography. A 2019-2020 survey of more than 20,000 USA citizens found that political party affiliation remains the key predictor of attitudes towards and trust in the media, with Republican participants (led by Trump at the time of research) expressing more negative sentiments on every aspect of media performance compared to Democrats and Independents (Gallup/Knight Foundation 2020; see also van der Linden, Panagopoulos, & Roozenbeek 2020). While Republicans were the most distrusting of the media, the research found that the majority of Americans felt the media’s role to inform and hold those in power accountable had been compromised by increasing bias: 46% saw ‘a great deal’ of political bias in news coverage and 37% saw ‘a fair amount’ (p. 5). This is concerning as the majority (81%) of participants also upheld the ideal that the news media is fundamental to a healthy democracy (Gallup/Knight Foundation 2020, p. 14). This relationship between distrust in the media and Republican/Trump support in the USA is perhaps not surprising given that distrust has been linked to support for populist politics (Mudde & Kaltwasser 2017). In Europe it has been found that the decline in trust, alongside economic precarity and social media narrow casting, has provided what is referred to as a ‘political opportunity structure’ for more extreme politics, generating fear, and a reality, of social division (Ernst et al. 2019; Koopmans & Olzak 2004; Meyer 2004). Within this context, there is an urgent need for better understandings of the contributing factors to distrust and measures to counter a pivot towards a distrusting society. Most of the key studies of trust have favoured what Ruelens and Nicaise (2020) characterise as ‘variable-centred approaches’. In these approaches, the variables themselves (be they demographic variables or attitudinal variables) are the central conceptual unit and they are analysed via descriptive statistics, correlation, and regression techniques. Within the trust literature, focus is placed on the relationship Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202385 between demographic variables (e.g., age groups, educational groups, voting groups) and attitudes on trust variables. This means that any remedies or solutions that emerge are anchored to the action in regard to a variable, rather than a group or class based on trust dispositions. As Ruelens and Nicaise (2020) argue ‘variable-oriented research has focused too much on attitudes [such as trust] as variables and not enough attention has been given to citizens who hold these attitudes’ (p. 1). A person-centred approach, such as Latent Class Analysis (LCA), ‘inherently adopts a multidimensional view of attitudinal phenomena by focusing on a set of attitudes’ (Ruelens & Nicaise 2020, emphasis added; see also Hooghe, Marien, & Oser 2017). This approach is ‘effective in identifying groups in the population that share a similar response pattern on a series of indicators’ (Ruelens & Nicaise 2020, p. 2, emphasis added). Internationally, the utility of understanding trust (and its impacts on social processes) via LCA is increasingly acknowledged. For example, in the Italian context, Fazio et al. (2018) developed a measure of ‘trust in others’ and a measure of ‘trust in institutions’ via LCA which was mapped at the regional level in Italy. In the European context more broadly, Ruelens and Nicaise (2020) used LCA to investigate the relationship between institutional trust at the national and supranational (EU) level. Also examining the broader European citizenry, Hooghe, Marien, & Oser (2017) utilised LCA to investigate the impact of democratic ideals on levels of political trust and developed a typology of distinct groups of respondents characterised by similar combinations of attitudes. Suh, Chang, and Lim (2012) have examined the links between trust in political and non-political institutions in South Korea using LCA. Finally, as mentioned previously, in the US, Agley and Xiao (2021) have utilised LCA to understand the relationship between trust in science, political orientation, religiosity, education and conspiratorial thinking in the COVID-19 context. The research presented in this paper responds to the urgent need for better understanding of the contributing factors to ongoing public distrust in institutions and interpersonal contacts with a particular focus on the Australian context. More specifically, it builds upon the international literature by contributing a typology of trust, and the development of trust measures, to facilitate a more nuanced person-centred understanding of trust in Australia. While the existing research (reviewed in the following section) has provided important insights into demographic variations in Australians’ trust, the results of Latent Class Analysis and multinomial logistic regression offers a more advanced and multidimensional approach to existing descriptive, variable focused methods. To our knowledge, our study is the first of its kind to utilise LCA to understand trust in the Australian context. Identifying groups or segments of the population, sorted in part by trust, will enable remedies to be developed that address the specific trust issues for that group. The aim of this paper is to therefore develop and present a typology of trust in Australia. The second aim is to examine the demographic and attitudinal variables associated with each class1 to better understand each class. Thirdly, we will analyse what distinguishes each class, with a view to identifying the means to prevent Australians moving into less trusting classes. Trust in Australia Recent research has identified discontent among the wider Australian population regarding trust in government, media, law enforcement agencies and educational institutions. Using the Edelman Trust Index (measuring trust in institutions such as NGOs, business, government, and media2), Australia was ranked the twelfth (of twenty-six) most ‘distrusting’ nations in 2019. In that year, Australia was one of the 1 We use the term ‘class’ throughout this paper as a statistical term pertaining to the groups or categories resulting from Latent Class Analysis rather than a term referring to socio-economic class stratification. 2 The Edelman Trust Index represents the average percentage trust in each market/nation (Edelman 2020). Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202386 most distrustful nations of the media (behind only Russia, UK, France, Ireland, and Japan), reporting a ‘neutral’ standing in its trust towards NGOs, and was also ‘neutral’ in its trust towards business (Edelman 2020). In 2019, the Edelman Trust Index also ranked Australians as distrusting of government – being five points below the global average. Research conducted by Stoker, Evans, & Halupka (2018) and the Scanlon Foundation (Markus 2019a) mirrored such low levels of trust of the Australian government. The Scanlon Foundation has been tracking Australians’ trust in government since 2007 and according to their surveys, Australians’ trust in government had been low for over a decade (Markus 2021). The COVID-19 context saw a sharp increase in Australians’ trust in government. In 2020, Australians’ trust in government became a majority view – rising to 55 percent in November 2020 (Markus 2021, p. 34). The Edelman Trust Barometer also reported increases in trust towards the Australian government, and other institutions, between 2018 and 2021. Australia moved from being a ‘distrusting’ nation that sits around the global average to a ‘trusting’ nation well above the global average (Edelman 2021, p. 44). In fact, the 2020 Edelman Trust Barometer saw Australians’ trust in all institutional sectors increase (NGOs, business, media, and government) (Edelman 2021). Similarly, in a smaller survey of 500 Australians in July 2020, Goldfinch, Taplin, & Gauld (2021) found that Australians’ trust in government rose steeply to 80%, up from 49% in 2009 (see Goldfinch, Gauld, & Herbison 2009). This was linked to overall perceptions that the government had managed the COVID-19 pandemic well. However, media reports and polling suggested that Australians’ trust in government once again degraded as the Morrison Federal Government grappled with the COVID-19 vaccination roll-out, the ending of JobKeeper payments, the poor handling of sexual violence and harassment allegations in parliament, and broader calls for workplace reforms to tackle casualisation, real wages decline and the prevalence of sexual harassment in Australian workplaces (Speers 2021; Grattan 2021; Murphy 2021). Indeed, in 2022, the Edelman survey reported that Australia’s ‘trust bubble’ had burst (p. 8), with Australia among the top three countries showing an annual decline in trust in government, business, NGOs, and media (along with Germany and The Netherlands (p. 5)). The change of federal government in 2022 has not seen a change to the negative trend, in fact, trust in the government has further declined: ‘Government [has] join[ed] the media in the realm of ‘distrust for the first time since 2020’ (Edelman 2023a). Trust in journalists and institutional leaders (CEOs and government leaders) has also declined the second year in a row and the media continues to be the most distrusted institution (distrusted by 38% of people and seen as a source of misleading or false information by 48%; Edelman 2023b). These trends indicate that the levels of trust are dynamic, and that they can be improved and worsened depending on political settings and structural circumstances. Some groups (in Australia and internationally as discussed in the previous section) are much more exposed to distrust than others, and these assessments of trust across social categories are useful in contemplating how distrust can be remedied. The Edelman Trust Barometer provides insight into demographic variations in Australians’ trust towards key institutions. The barometer divides the sample demographically into two groupings: ‘informed public’ and ‘mass public’. The ‘informed public’ are defined by Edelman (2021) as individuals that are: 1) 25-64 years of age; 2) College (or equivalent) educated; 3) in top 25% of household income per age group in each market; and 4) report significant media consumption and engagement in public policy and business news. ‘Mass public’ are the remaining participants who do not meet the ‘informed public’ criteria. While the Edelman Trust Barometer has seen Australians’ trust in all institutional sectors increase between 2019 and 2020, there is what is termed a ‘trust inequality’ between the ‘informed public’ and ‘mass public’ in Australia. In 2020, Australia’s ‘informed public’ had an average percentage trust in NGOs, business, government, and media of 68% (falling into the ‘Trust’ category of the Index). In comparison, the average percentage trust among the Australian ‘mass public’ was 45% (falling into the ‘Distrust’ category of the Index). The trust inequality between the two groups was not unique to Australia, with record trust inequalities recorded globally (Edelman 2020: 8). In the period Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202387 between 2020 and 2021, when trust in Australia increased (previously described as ‘the bubble’), the divide worsened (2022, p. 5). Edelman examined what would close that gap, and respondents indicated that quality information was key (2022, p. 31). For example, lower income respondents’ trust levels would exceed the average of not well-informed high-income respondents, if they were well-informed (p. 31). Similar to the international literature, Markus (2019a, 2019b, 2021) has found a strong link between Australians’ support for populist politics and lack of trust in key institutions of the state (criminal justice system [police excluded]; government, politicians) and media, as well as their fellow citizens. For example, in 2019, only 9% of One Nation Party3 voters agreed that the government in Canberra can be trusted to do the right thing for Australia almost always or most of the time. This was in stark contrast to the rate of agreement among Liberal/National Party (centre-right) voters (49% agreed that the government can be trusted) and Australian Labor Party (centre-left) voters (22% agreed that the government can be trusted; Markus 2019a, p. 42). While these levels of trust increased across all voting groups in 2020, the association between populist politics and distrust in the government remained: 31% One Nation Party voters agreed that the government in Canberra can be trusted to do the right thing for Australia almost always or most of the time in 2020, compared to 75% Liberal/National and 43% Labor (see Markus 2021, p. 113). Moving away from trust in government, One Nation Party voters had the lowest level of agreement that ‘most people can be trusted’ (28% in 2020 and 23% in 2018-19; see Markus 2021: 72). This distrust in fellow citizens sits alongside One Nation Party voters’ heightened concerns with immigration, their support for discriminatory immigration policies, and lack of support for multiculturalism (Markus 2019a; 2019b). With trust a key ingredient for economic prosperity, individual and collective wellbeing, social harmony in culturally diverse circumstances, and the staving off of populist politics, the extent of the distrust evident amongst One Nation Party voters (and in the Australian population more broadly), is cause for concern. Method We utilised a LCA of Australian national survey data obtained in October 2019 (n: 2,015) to provide deeper understanding of the relationship between respondents’ trust in institutions, and social or interpersonal trust. We adapted the categorisations of ‘institutional trust’ and ‘interpersonal trust’ from Suh, Chang, and Lim (2012) and Fazio et al. (2018). For the purposes of our study, ‘institutional trust’ (a term used by both Suh, Chang, and Lim 2012 and Fazio et al. 2018) refers to individuals’ trust in public and private institutions including political institutions (governments, government departments, political parties), courts and other institutions relating to law and justice systems such as police, private enterprises, civil associations and media organisations. ‘Institutional trust’ variables that were tested in our research included trust in the Australian Family Courts, trust in the Australian media, and trust in the Police. ‘Social’ or ‘interpersonal trust’ (used by Suh, Chang, & Lim 2012 but termed ‘trust in others’ by Fazio et al. 2018) refers to individuals’ trust of other groups or individuals. ‘Social’ or ‘interpersonal trust’ variables tested in our research included trust in neighbours, family members, friends, and people of different religious or cultural backgrounds. We also measured ‘trust in the science of climate change’ and investigated, via LCA, the relationship between this kind of trust and institutional and social/interpersonal trust. Trust in the science of climate change does not fall under ‘institutional’ or ‘interpersonal’ trust. A second stage in the methodology (multinomial logistic regression) also enabled us to consider the relationship between individuals’ latent class membership, demographic indicators (e.g., gender and age) and attitudes to other aspects of Australian life such as national economic prosperity and job security, quality of healthcare, cultural diversity, and quality of leadership. 3 One Nation is an Australian political party associated with far-right populist politics (Grant, Moore, & Lynch 2019). Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202388 PARTICIPANTS To assess Australians’ (dis)trust in various groups and institutions, we drew upon an online national survey conducted in October 2019. The Australia-wide online survey comprised a sample of 2,015 respondents (Australian residents, 18+ years of age). The sampling frame was provided to the online panel provider Dynata, who have a pool of approximately 300,000 Australian panellists. Dynata employs quality controls, recruits participants in a variety of ways to ensure a diverse sample, and randomly selects panel participants (within the criteria) to participate in the surveys. One agreed and obvious limitation of this survey design is that participants must have access to the internet and be registered panellists of Dynata. The demographic characteristics of the sample are provided in Table 1. MEASURES The survey included a variety of closed-response questions used to obtain information regarding Australians’ attitudes and experiences, identity and culture, community and belonging, trust and anxiety, and contemporary social, economic, and political changes. The majority of questions used a Likert-type scale response option, allowing respondents to indicate whether they agreed or disagreed with a statement and to what extent. The 18 questions on trust were largely drawn/adapted from the World Values Survey (2014) and Scanlon Mapping Social Cohesion Survey (Markus 2017). LCA INDICATORS There were seven items selected as indicators for the Latent Class Analysis based on prior research (e.g., Markus 2017) and sufficient cell sizes, to represent Australians’ trust in social groups, issues, or institutions (see Table 3). Two items asked about the extent to which Australians trust their neighbours and people of another religion (all ranging from 1 = do not trust at all to 4 = trust completely). Four items asked about trust in the Australian Family, Australian media, police, and the science of climate change (all ranging from 1 = strongly disagree to 5 = strongly agree). The final item asked whether “in some circumstances a non- democratic government can be preferred” (ranging from 1 = strongly disagree to 5 = strongly agree). COVARIATES There were 15 items used as covariates, and included demographic and background characteristics, attitudes towards Australia and diversity, and participants’ future life. These were selected based on a series of chi- square tests to determine whether the participant characteristics and their attitudes were related to the latent classes. These 15 items were used as predictors in the multinomial logistic regression analyses. The demographic and background characteristics consisted of seven variables: age, gender, education, income, religion, location, and country of birth. Age was grouped into three age ranges (i.e., 18-34 years; 35-64 years; 65+ years), and gender was grouped into two (females and males). Both education (i.e., high school or less; other tertiary/trade/TAFE; and university/postgraduate) and income (low, medium, and high) had three categories. Religion (no religion and religion), location (metropolitan and non-metropolitan), and country of birth (Australia and overseas) were all dichotomous. Attitudes towards Australia and diversity consisted of eight items. Three items asked Australians the extent to which they agree that Australia has strong leadership, is economically prosperous, and that the Australian government can be trusted to do the right thing for Australians (ranging from 1 = strongly disagree/disagree to 3 = agree/strongly agree). Two items asked about concern regarding the “quality of healthcare” and “employment and job security” (ranging from 1 = not concerned to 3 = very/extremely concerned). Attitudes towards diversity included two items: “I would not object to a religious place of worship being built in my own community” and “Australia is weakened by people of different ethnic origins Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202389 Table 1. Demographic Characteristics of Survey Sample, 2019 Demographic Variable % State of Residence New South Wales 33 Victoria 22 Queensland 21 Western Australia 11 Tasmania 2 Australian Capital Territory 2 Northern Territory 1 Metro/non-Metro Metropolitan 53 Non-Metropolitan 47 Country of Birth Australia 72 United Kingdom 8 India 2 New Zealand 2 Indigeneity Aboriginal and Torres Strait Islander 3 Non-Indigenous 97 Gender Female 51 Male 49 Non-binary/gender fluid <0.5 Age 18-34 years of age 41 35-54 years of age 30 55-75 22 75+ years of age 8 Education (highest level) University degree or postgraduate qualification 36 Other tertiary qualifications including a trade or TAFE qualification 34 Year 12 or equivalent 23 No formal qualifications. 6 Employment Employed 48 Unemployed 8 Retired 23 Income $80,000+ 21 $50,000-$79,999 20 $30,000-$49,999 16 <$30,000 31 Religion Judaism 0.5 Buddhism 2 Hinduism 2 Islam 2 Christianity 40 No religion, agnostics and atheists 45 Political affiliation Labor 31 Liberal 32 Greens 13 One Nation 6 Independent 8 Other 11 Notes: The sample is representative across all states and territories. Aboriginal and Torres Strait Islander Australians were represented in line with national representation. The high proportion of retirees reflects the older age cohort of the sample. The unemployed were also over-represented [Australian Bureau of Stastistics (ABS) 2019a]. Buddhism, Hinduism, Islam and Judaism were at representative national levels. No religion, agnostics and atheists were overrepresented. Christianity was underrepresented (ABS 2017). Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202390 sticking to their old ways” (ranging from 1 = strongly disagree/disagree to 3 = agree/strongly agree). The final item asked participants about their future life: “In five years my life in Australia will be much worse” (ranging from 1 = strongly disagree/disagree to 3 = agree/strongly agree). STATISTICAL ANALYSES Latent class analysis (LCA) and multinomial logistic regressions were employed using a structural equation modeling framework (SEM) using Mplus Version 8.3 (Muthén & Muthén 2017). LCA is a person-centred analysis in which relationships among individuals, rather than relationships among variables, are of primary interest. Therefore, LCA was chosen as the most appropriate statistical approach to identifying a typology of trust in Australia. Mplus was used to estimate LCA (a) estimating model parameters and (b) allowing for the inclusion of all available data through the full information maximum likelihood (FIML) estimation to account for missing data (Enders 2010). FIML uses all available data for each case during the estimation process, has been shown in recent simulation studies to be unbiased when data are missing at random and to be robust to non-normal distributions (Enders 2010; Schafer & Graham 2002). LCA is a model-based approach in which the population is considered to consist of k latent groups or classes where the number of classes is not known a priori. Individuals in the same class share a common joint probability distribution among the observed variables (e.g., levels of trust). In the current study, individuals were assigned to the latent class for which their posterior probability was the highest. To identify the optimum number of classes, a series of LCA models of between two and six classes were estimated. Given that there is no single indicator reflecting an optimal model fit, the model selection was based on a balance of parsimony, substantive consideration, and used ‘fit and test’ statistics. First, we examined the following fit indices: The Akaike’s Information Criterion (AIC; Akaike 1987); Bayesian Information Criterion (BIC, Schwartz 1978); sample size adjusted BIC (SBIC; Sclove 1987); and the Lo-Mendell- Rubin likelihood ratio test (LMR; Lo, Mendell, & Rubin 2001). Recent simulation studies have indicated that the BIC, SBIC, and LRT are the most effective for identifying the correct number of classes (Diallo, Morin & Lu 2016a, 2016b, 2017; Peugh & Fan 2013). Second, we examined classification uncertainty using the entropy indicator. Entropy measures the probability that person i is a member of a given class. Values near zero suggest low certainty in classification and values near one suggest high certainty in classification. Third, we examined class size to ensure that each class had at least 1% of the entire sample included (Nylund, Asparouhov, & Muthén 2007; Shanahan et al. 2013). Fourth, we examined whether the latent class solution made sense with respect to parsimonious coverage and the relevant literature. Furthermore, when fitting mixture models like LCA, issues such as convergence are important to consider (Hipp & Bauer 2006). In this study, all LCA were conducted using 800 sets of random starting values for 80 iterations each, and then we selected the 40 best sets of random starting values associated with the highest likelihood values for final optimization to minimize the risk of convergence to local maxima, in other words, to make sure the model made sense mathematically (Hipp & Bauer 2006). Using this method, we fitted models with class one through seven to determine the optimal number of classes for the data. Table 2 shows the fit statistic for each model specification. All LCA models converged. The log-likelihood increased while no minimum was found for the iCs as their values decreased across the range of models considered. The LMR pointed to the five-class solution since the test of the four-class model against the five-class model has a p-value less than 0.000, suggesting rejection, whereas the test of the five-class against the six-profile has a p-value of 0.760. Further examination of the LCA models indicated that the five- and six-class models each included small classes that seemed to have splintered off from larger classes in the four-class model. The four-class model resulted in a log-likelihood value of -18023.494 with 107 parameters, an AIC of 36260.988, a BIC of 36865.25, a SBIC of 36525.3, and a high entropy value of 0.725. Moreover, the four-class solution satisfied the minimum class size required to be useful Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202391 (each comprised at least 1% of individual within the overall sample) and meaningful. The four-class model also made theoretical sense, it had interpretability, and the four classes looked to have utility for making recommendations on improving the trust landscape (Muthén 2004). The four-class solution is described below. Table 2. Fit Indices for LCA Models for Trust Number of Latent Classes Log likelihood # Free Parameters AIC BIC SBIC LMR LRT Entropy 1 -19443.007 26 38938.015 39084.845 39002.24 NA NA 2 -18710.865 53 37527.729 37827.037 37658.65 p <.000 0.731 3 -18226.248 80 36612.496 37064.281 36810.113 p <.000 0.732 4 -18023.494 107 36260.988 36865.25 36525.3 p <.000 0.725 5 -17849.572 134 35967.144 36723.883 36298.152 p <.000 0.745 6 -17754.223 161 35830.446 36739.663 36228.15 0.760 0.712 7 -17694.095 188 35764.19 36825.884 36218.589 0.828 0.721 Note. # = number; AIC = Akaike Information Criterion; BIC = Bayesian Information Criterion; SBIC = Sample Size Adjusted BIC; p LMR = p-values for the Lo-Mendell-Rubin Likelihood ratio test for k versus k+1classes. Lastly, multinomial logistic regression analyses were conducted to examine the relationship between the 15 covariates and the latent classes. Class membership was the dependent variable, and the 15 covariates were the independent variables. Multinomial logistic regression analysis was used to investigate the relationship between the covariates and class membership. Results and Discussion DESCRIPTIVE STATISTICS Descriptive statistics (i.e., frequencies, bivariate cross-tabulations) generated a snapshot of current views and attitudes of Australians in relation to social trust (trust in people and social groups) and trust in institutions. More particularly, the descriptive statistics were used to examine the extent to which social and institutional trust vary across selected demographic and background characteristics (e.g., age, gender, country of birth). Due to the large number of cross-tabulations, only the significant results are presented. Social Trust Overall, Australians are trusting of their neighbours (74%) and people of another religion (67%). Trust in neighbours is high across all age groups, but there was a positive and significant (although weak) association between age and trust, such that as age increased, so too did the extent of these forms of trust. For example, the proportion of those trusting neighbours completely was 10% for those aged 18-34, while it was 34% among those aged 75 and older. Complete trust in people of another religion was lower, but there is a majority with some level of trust. For example, when asked to indicate how much they trust people of another religion, only 8% of participants indicated ‘trust completely’ while 59% indicated ‘trust somewhat’. Age was consistently and positively associated with trust in people of another religion. For example, the proportion of those trusting people of another religion completely was 8% for those aged 18-34, while Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202392 it was 17% among those aged 75 and older. This confounds the established research that has found that younger people’s experience of super diversity generates more comfortable views about difference (Halafoff et al. 2020). Rather, these findings support contact theory which suggests that interpersonal contact between groups (which will expand over a persons’ lifetime) can reduce prejudices (Allport 1954), and lead to greater levels of trust between groups. This may indicate an interesting disjuncture between philosophy and practice around difference, in which younger Australians are pro-diversity but cautious about actual encounters, whereas older Australians a more likely to have an ideological antipathy to diversity but less of a problem with actual cross-cultural encounters. Institutional Trust Three-quarters (74%) of the sample have some level of distrust in key institutions (government, media, political parties). However, trust varies across different types of institutions. Trust in the Australian media (20%), political parties (20%) and government (29%) is weaker. There were similar levels of trust in institutions like the family court and criminal courts, while 62% of Australians have trust in the Police. Trust in the science of climate change was held by about half of the sample, and only a quarter said they did not trust it. Such distrust in key institutions aligns with global trends (Edelman 2021). Significant patterns in institutional trust are evident across demographic variables. Men were more trusting of the police, as were heterosexuals (compared to non-heterosexuals), older people, non-Indigenous people, and those who were Christians. Unlike trust in police, older Australians were more likely to be distrusting of the Australian Family Court and Media compared to younger Australians. Similarly, men were more distrusting of the Australian Family Court and political parties compared to women. Australian- born participants were more distrusting of the Australian Family Court, political parties, the government, and media compared to participants born overseas. Interestingly, there were no observable patterns across age groups for trust in government. These variations in institutional trust have interesting correlations with positions of (more) privilege and (more) power. That is, the results suggest that those with more societal power and privilege (males, heterosexuals, older Australians, Australian-born, non-Indigenous and Christians) have greater trust in institutions which regulate the less powerful (e.g., Police). In contrast, the less powerful and less privileged (women, those born overseas, younger people), trust more in democratic institutions that protect the more vulnerable (i.e., Family Court). Australians are divided on whether in some circumstances a non-democratic government can be preferred, with 28% agreeing, 28% disagreeing, and 44% who neither agree nor disagree. However, younger Australians, men, and those born overseas were more likely to agree with this statement as compared to older Australians, women, and those born in Australia. This is an interesting finding that somewhat contradicts the aforementioned finding that those born overseas, and younger people trust more in democratic institutions. It is perhaps the case, as found in existing Australian and international research, that younger people (and males) are more susceptible to ideological and political extremism (Nilan et al. 2023; Adam-Troian, Tecmen, & Kaya 2021; Forscher & Kteily 2020), and that those born overseas may have lived-experience of non-democratic governments in their countries of birth. Overall, almost half (48%) of the survey sample agreed that they have trust in the science of climate change (23% disagreed). Bivariate analysis revealed that younger people were more trusting in this science than were older people. For example, distrusters were about 16% of the young respondents, but were 29% of those aged 55 to 74, and 33% of those aged 75 and over. There were expected variations according to political party affiliations, with Green, ALP and Independent voters more trusting of the science of climate change, while One Nation Party voters were distrusting. The latter were the most distrusting of all institutions and groups (including neighbours for example). Our finding that younger respondents are generally more trusting of key Australian institutions aligns with previous research that found that younger Australians feel especially reliant on public institutions in Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202393 the age of precarity, dynamism and diversity (Wyn & White 2012; Harris 2010). As Australians get older, their faith in institutions, like the courts and the media degrades. The same is true of trust in the science of climate change. This may be because older people’s greater exposure to or experience of institutions (compared to younger people) has degraded their trust. This may also explain the propensity for Australian- born participants to have less trust in key institutions such as the Australian Family Court, political parties, the government, and media compared to overseas-born Australians. There is one exception to this trend, with three quarters of older Australians agreeing they have faith in policing (74%), with only about half of younger Australians agreeing (54%). As noted by Sindall, McCarthy, & Brunton-Smith (2016), ‘It is widely recognized that trust and confidence in the police are more fragile amongst young people than amongst adults’ (see also Dwyer 2014; Kennelly 2011; Flexon, Lurigio, & Greenleaf 2009; Hurst & Frank 2000; Hinds 2007). Explanations for this include young people’s greater use of public space, higher visibility in public space, assumed positionality as being ‘at risk’ or ‘a risk’, and therefore ‘high levels of state surveillance, intervention and regulation’ (Fileborn 2019, p. 435) which may amount to ‘heightened contact and conflict with the police’ and/or particularly negative experiences of policing (Sindall, McCarthy, & Brunton-Smith 2016, see also Dwyer 2014; Kennelly 2011; Loader 1996; McAra & McVie 2005, 2010; Carr, Napolitano, & Keating 2007; Fagan & Tyler 2005; Hinds 2007; Piquero et al. 2005). As previous research has shown, trust is key to a well-functioning society. High levels of distrust have negative impacts on economic prosperity and social and civic cooperation, population wellbeing, a robust and critical media, and contributes to political opportunities for extremist politics and conspiracy theories (Koopmans & Olzak 2004; Meyer 2004; Mudde & Kaltwasser 2017; Ernst et al. 2019). Trust, particularly in institutions, is a necessary requirement for a strong democracy. Given such high overall levels of distrust illustrated in our survey findings, there is urgent need for a better understanding of the contributing factors to distrust and to explore remedies to counter a pivot towards a distrusting society. The Latent Class Analysis results we present in the following section may assist in developing such remedies. FOUR SEGMENTS OF AUSTRALIAN TRUST Table 3 presents the LCA results for the four-class model. The last two rows represent the proportion based on posterior probabilities and the class count based on the most likely latent class. The proportions refer to the estimated proportions of individuals that belong to a particular latent class: 15% in Class 1, 17% in Class 2, 42% in Class 3, and 26% in Class 4 (described next in more detail). The class counts refer to the number of individuals (rather than proportions) that belong to a particular latent class. Class 1: ‘Very Distrusting’ In the four-class solution of the latent class analysis (see Table 2), Class 1 represents 15% of the sample and was labelled ‘very distrusting’. Individuals in this class tend to strongly disagree that they trust the Australian Family Court (72%), the Australian media (79%), and the science of climate change (54%). A bare majority in this class still had some level of trust in neighbours (56%), but most did not trust people of another religion very much if at all (55%). This group were also the least trusting of the police (44% distrust), although their trust of the police (36%) was well ahead of other institutions (Family Court 1%; media 3%). Class 2: ‘Unsure’ (worried) Class 2 represents 17% of the sample and has been labelled ‘unsure’. Respondents in this class neither agreed nor disagreed that they have trust in the Australian Family Court (78%), the Australian media (79%), the science of climate change (79%), and in the police (54%). They were also unsure as to whether a non- Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202394 democratic government can be preferred in some circumstances (83%). Just under two-thirds (65%) tended to trust neighbours, and just over half trusted people of other faiths (58%), but mostly only somewhat. A third did not have very much trust at all in neighbours (32%) and in individuals of other faiths (37%). Class 3: ‘Somewhat Trusting’ Class 3 represents 42% of the sample (the most populous class). This class has been labelled ‘somewhat trusting’. The individuals in this class ‘somewhat’ trust their neighbours (63% plus 13% who trust neighbours completely), and they disagreed that they trust the Australian Family Court (47%) and the Australian media (56%), but unlike the group in the ‘Very distrusting’ class, they did not hold to this as ‘strongly’. The class has the same variations in trust across neighbours and institutions, with the main difference being the strength of that trust. Those in the ‘somewhat trusting’ class do not have the same level of distrust towards people of other religions as the very distrusting, with 65% trusting somewhat people from other religions (and only 4% trusting completely). If we combine Class 2 and Class 3, we have a significant middle group who comprise the majority of the sample (59%) where there is some level of trust, but where that trust is uneven. Class 4: ‘Trusting’ Class 4 represents 26% of the sample and has been labelled ‘Trusting’. These respondents agreed that they have trust in the Australian Family Court (75%), in the Australian media (62%), the science of climate change (81%) and the police (91%). They trusted neighbours somewhat (57%) or completely (27%) and people of other faiths somewhat (61%) or completely (19%). THE FOUR CLASSES AND COVARIATES We used multinomial logistic regression to predict the probabilities of each person belonging to the four latent classes. Specifically, we used multinomial logistic regression to test the association between 15 variables (covariates) with the four latent classes (Table 4). For example, there is one statistically significant finding for age in the first row (i.e., 65+ years), which is an odds ratio of 2.530. This means that the odds of being in the ‘somewhat trusting’ class as compared to the ‘trusting’ class is 2.530 times higher for those aged 65 years and older as compared to those aged 18-34 years, controlling for all the other variables. Due to space constraints, we highlight the significant findings in our interpretation below. One of the most distinguishing results from our analysis is that compared to the Trusting class (26% of sample) the Very Distrusting class of respondents (15% of sample) are especially distrusting of the Australian government. The latter variable had a particularly strong effect, as shown in the odds ratio, and was significant at p <.001. Another telling variation between the Trusting and Very Distrusting classes were that the latter were more likely to object to a religious place of worship being built in their own community. Therefore, individuals in this group are also more likely to be associated with negative dispositions towards minority cultural groups and towards diversity. Other significant variations were economic, with less agreement that Australia was economically prosperous, yet slightly less concerned about their own job and employment security. The largest class are the Somewhat Trusting (42%). Like the Very Distrusting class, this class had a higher level of distrust of the Australian government than the control group (‘Trusting’). There was also a significant effect of age (in particular with those aged 65 and over), education (associated with those with tertiary/Trade/TAFE education as compared to those with no post-school education) and being Australian born (versus those born overseas). Being less likely to agree that Australia is economically prosperous was also an effect on this class, that also marked them as different from the Trusting class. They were also less likely to agree that they would not object to a religious place of worship being built in their own community. Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202395 Table 3. LCA (trust) four-class model results   C1 Very distrusting C2 Unsure C3 Somewhat trusting C4 Trusting Whole Sample Trust your neighbours Trust Completely 0.13 0.07 0.13 0.27 15.8 Trust Somewhat 0.43 0.58 0.63 0.57 57.5 Do not trust very much 0.22 0.32 0.2 0.13 20.2 Do not trust at all 0.23 0.03 0.04 0.04 6.5 Trust people of another religion   Trust Completely 0.05 0.03 0.04 0.19 7.9 Trust Somewhat 0.4 0.54 0.65 0.61 58.2 Do not trust very much 0.32 0.37 0.27 0.16 26.5 Do not trust at all 0.23 0.06 0.04 0.04 7.4 I have trust in the Australian Family Courts Strongly Agree 0 0.01 0 0.16 4.5 Agree 0.01 0.12 0.12 0.59 22.7 Neither Agree nor Disagree 0.09 0.78 0.33 0.2 33.9 Disagree 0.18 0.06 0.47 0.03 24 Strongly Disagree 0.72 0.03 0.08 0.02 14.9 I have trust in the Australian media Strongly Agree 0.02 0 0 0.1 3 Agree 0.01 0.09 0.05 0.52 17.5 Neither Agree nor Disagree 0.07 0.79 0.23 0.28 31.7 Disagree 0.1 0.11 0.56 0.07 28.2 Strongly Disagree 0.79 0.01 0.16 0.03 19.5 I have trust in the science of climate change Strongly Agree 0.15 0 0.17 0.28 16.6 Agree 0.09 0.19 0.32 0.53 31.8 Neither Agree nor Disagree 0.13 0.79 0.23 0.14 28.5 Disagree 0.1 0.02 0.21 0.04 11.4 Strongly Disagree 0.54 0 0.08 0.01 11.6 I have trust in the Police Strongly Agree 0.09 0.04 0.07 0.32 13.3 Agree 0.27 0.37 0.54 0.59 48.4 Neither Agree nor Disagree 0.2 0.54 0.26 0.07 25 Disagree 0.16 0.04 0.11 0.02 8.1 Strongly Disagree 0.28 0.01 0.02 0 5.1 In some circumstances a non-democratic government can be preferred Strongly Agree 0.09 0 0.02 0.11 5.1 Agree 0.12 0.14 0.22 0.37 23 Neither Agree nor Disagree 0.35 0.83 0.41 0.27 43.9 Disagree 0.07 0.03 0.26 0.14 16 Strongly Disagree 0.37 0 0.09 0.11 12 Proportion based on posterior probabilities 0.15 0.17 0.42 0.26   Class count based on most likely latent Class 306 355 885 549   n= 1713 Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202396 Table 4. Demographic and attitude profiles of ‘Trust’ classes using multinomial logistic regression estimates and odds ratios, against ‘Trusting’ Very distrusting Unsure Somewhat trusting B (SE) OR B (SE) OR B (SE) OR Age 18 -34 years 0a 0a 0a 35-64 years -0.064 (0.230) 0.938 -0.016 (0.204) 0.984 0.301 (0.166) 1.351 65+ years 0.248 (0.303) 1.281 -0.353 (0.296) 0.703 0.928 (0.218)*** 2.530 Gender Male 0a 0a 0a Female -0.304 (0.199) 0.738 0.545 (0.192)** 1.725 0.090 (0.146) 1.095 Education High School/no-formal 0a 0a 0a Other tertiary/Trade/ TAFE 0.431 (0.243) 1.539 0.264 (0.244) 1.302 0.386 (0.189)* 1.471 Postgrad/University -0.309 (0.265) 0.734 -0.109 (0.253) 0.896 -0.179 (0.187) 0.836 Income Low 0a 0a 0a Medium -0.325 (0.232) 0.722 -0.162 (0.219) 0.851 -0.097 (0.173) 0.908 High -0.380 (0.265) 0.688 -0.454 (0.256) 0.635 -0.292 (0.191) 0.747 Religion No religion 0a 0a 0a Religious -0.091 (0.202) 0.913 0.163 (0.191) 1.117 0.002 (0.147) 1.002 Location Metropolitan 0a 0a 0a Non-metropolitan 0.026 (0.198) 1.026 -0.296 (0.192) 0.744 0.032 (0.146) 1.032 Country of birth Australia 0a 0a 0a Overseas -0.169 (0.220) 0.845 0.292 (0.196) 1.342 -0.308 (0.156)* 0.735 Australia has strong leadership Neither agree/disagree 0a 0a 0a Disagree -0.197 (0.262) 0.821 -0.961 (0.259)*** 0.382 0.072 (0.205) 1.075 Agree 0.170 (0.257) 1.186 -0.557 (0.229)* 0.573 -0.255 (0.178) 0.775 Australia is economically prosperous Neither agree/disagree 0a 0a 0a Disagree -0.205 (0.315) 0.814 -0.927 (0.363)* 0.396 -0.286 (0.278) 0.751 Agree -0.611 (0.236)* 0.543 -0.984 (0.212)*** 0.374 -0.383 (0.178)* 0.682 Quality of healthcare Not concerned 0a 0a 0a Slightly concerned/ concerned 0.043 (0.404) 1.044 -0.044 (0.351) 0.957 0.331 (0.271) 1.392 Very/extremely concerned 0.644 (0.418) 1.905 -0.221 (0.371) 0.802 0.291 (0.292) 1.338 Employment and job security Not concerned 0a 0a 0a Slightly concerned/ concerned -0.854 (0.396)* 0.426 0.100 (0.380) 1.105 0.042 (0.292) 1.043 Very/extremely concerned -0.647 (0.417) 0.524 0.032 (0.399) 1.033 0.045 (0.318) 1.046 The Australian government can be trusted to do the right thing for Australians Neither agree/disagree 0a 0a 0a Disagree 1.996 (0.277)*** 7.358 -0.434 (0.265) 0.648 0.987 (0.207)*** 2.683 Agree -1.683 (0.302)*** 0.186 -1.893 (0.232)*** 0.151 -1.443 (0.174)*** 0.236 In five years my life in Australia will be much worse Neither agree/disagree 0a 0a 0a Disagree -0.345 (0.250) 0.709 -0.642 (0.222)** 0.526 0.035 (0.169) 1.036 Agree 0.031 (0.241) 1.031 -0.101 (0.241) 0.904 -0.433 (0.193)* 0.649 I would not object to a religious place of worship being built in my own community Neither agree/disagree 0a 0a 0a Disagree 1.102 (0.275)*** 3.011 -0.060 (0.285) 0.942 0.282 (0.221) 1.325 Agree -0.522 (0.236)* 0.593 -0.779 (0.210)*** 0.459 -0.658 (0.167)*** 0.518 Australia is weakened by people of different ethnic origins sticking to their old ways Neither agree/disagree 0a 0a 0a Disagree 0.089 (0.261) 1.093 -0.614 (0.244)* 0.541 0.170 (0.190) 1.185 Agree -0.057 (-.241) 0.945 -0.748 (0.224)** 0.473 -0.096 (0.179) 0.909 Note. Parameter estimates (B) for each latent class are relative to the reference latent class (Trusting), adjusted for all other variables in the model. SE: standard Error of the estimates. OR: Odds Ratio. Odds-ratios are exponentiated parameter estimates. 0a: reference category. The coefficients and OR reflects the effects of the predictors on the likelihood of membership in the listed latent class relative to the reference class (Trusting). * p<.05, ** p<.01, *** p<.001 Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202397 This Somewhat Trusting class were also less likely than the Trusting class to agree that their life in Australia would be much worse in five years. The Unsure class were 17% of the sample. They varied from the Trusting class by gender, being more likely to be female. They were also more likely to be unable to agree or disagree (being ‘Unsure’) that: 1) the government can be trusted; 2) Australia is economically prosperous; 3) Australia has strong leadership; 4) Australia is weakened by people of different ethnic origins sticking to their own ways; 5) they would not object to a religious place of worship being built in their own community; and 6) that their life in Australia will be much worse in five years. This Unsure class could be more appropriately characterised as worried, as shown in their uncertainties. Finally, the covariate assessments across all of the forms reveal that the Trusting class (26%) are marked by their enhanced agreement that the Australian government can be trusted to do the right thing for Australians. While other classes shared some level of trust across family, neighbours, people of other faiths and various institutions, it was the Trusting Class alone that had the highest trust in the Australian government. This indicates that it is trust in government specifically that is required to move Australians from the other classes into this class of trust. In other words, trust in government is a key stone for fortifying general trust and social cohesion. They were also distinct from the other three groups in their embrace of religious diversity (not disposed to oppose local places of worship). This class also looks to institutions for leadership (e.g., on climate change). This is a class of Australians whose trust and confidence could be leveraged to influence other Australians, for whom contact and confidence could be key. Conclusion In the age of global populism and disruption, a new ‘red-button’ issue of public trust has emerged. Trust towards institutions is of global concern, but even trust between neighbours is increasingly a matter of concern. Our research has identified that distrust of institutions is a key feature in contemporary Australian society, however, just like ‘social trust’, ‘institutional trust’ is differentiated across demographic categories. For example, trust in family, close friends and neighbours was higher for older Australians. This was also the case for trust in people of other faiths and ethnicities. Trust in the science of climate change was a feature of younger Australians rather than older Australians. Older Australians, and those born in Australia, had lesser levels of trust in the courts, media, political parties and government, but higher trust in their neighbours and people from different cultural backgrounds. Moving beyond demographic analyses, a Latent Class Analysis identified four classes of Australians regarding their trust levels. The Trusting class were about quarter of Australians (26%), and were remarkable for trusting institutions, family, neighbours, as well as people of other faiths and ethnicities. This class has relatively high levels of trust, is associated with younger Australians and was the only class to show high levels of trust in the government. At the other end of spectrum was the Very Distrusting class (15%) who were highly distrustful of institutions, family and neighbours and other faiths and ethnicities. The remainder of the sample, a large middle group, comprised two classes whose trust was uneven. The Somewhat Trusting class (42%) distrust the key institutions, but had trust in their neighbours, and some trust in other faiths and ethnicities. This class was more likely to comprise older Australians, and those with higher levels of education. The Unsure class (17%) shows concerns or worry across all trust indicators, but was most sceptical around institutions. They are not sure whether they should or can trust institutions, they are unsure about the economic fortunes of Australia, they are less committed to democratic forms of government, they were unsure on matters of cultural retention among ethnic minority groups (assimilationist), they were more likely to be unsure about a place of worship being built in their own community, and they were more likely to be women. This would suggest that the Unsure class is worried and uncertain, making them a prime target for anti-diversity political forces and tactics. Confidence building in Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202398 the public institutions, democracy and their fellow citizenry will help inoculate this large and worried group from cynical dispositions promulgated by populist politics. We argue that by taking a multidimensional view of the attitudinal data of individual respondents, we have been able to identify four classes in the population. The Australians within each of those classes share a similar response pattern when it comes to the issues of trust. By segmenting the population in this way, we have developed a typology of trust that facilitates a more nuanced person-centred understanding of Australians’ trust. Whilst it is a typology of trust in Australia at a single, cross-sectional point in time, it can be used as a baseline to address distrust in Australia; a key and urgent consideration must be to arrest the high levels of distrust in institutions, and in particular government institutions. Our findings indicate that a negative view of religious diversity and assimilationist views is associated with the most distrusting Australians (15%), and there is a sense that such a disposition is present among the 17% who are unsure in their trust. Building confidence and positivity toward diversity will work to contain the proportion of Australians in these two classes. A pressing societal need is to ensure that more Australians in the unsure class do not move into the very distrusting class, and trust in the media, courts and science is key here. More generally, our interpretations suggest that confidence is generally associated with trust in institutions, as demonstrated by the demographic variations in which institutions are trusted: police more trusted by male and older Australians, and the Family Court more trusted by women. The background literature suggests that quality information is key in building confidence and therefore trust. The low trust in mainstream media, and the perceptions of its bias and divisiveness (Edelman 2022), make quite difficult the project of enhancing the extent to which people feel well-informed. The temporal variations in trust point to its dynamism, reflecting the theorising around political opportunity structures and political tactics (Ernst et al., 2019; Koopmans & Olzak 2004; Meyer 2004). This also reminds us of the possibility to enhance and build trust, and this paper provides some data and analysis on the most propitious levers to pull. References Adam-Troian, J., Tecmen, A., & Kaya, A. 2021, ‘Youth extremism as a response to global threats? A threat-regulation perspective on violent extremism among the youth’, European Psychologist, vol. 26, no. 1, pp. 15-28. https://doi. org/10.1027/1016-9040/a000415 Agley, J. & Xiao, Y. 2021, ‘Misinformation about COVID-19: evidence for differential latent profiles and a strong association with trust in science’, BMC Public Health, vol. 21, article 89. https://doi.org/10.1186/s12889-020-10103-x Akaike, H. 1987, ‘Factor analysis and AIC’, Psychometrika, vol. 52, no. 3, pp. 317–332. https://doi.org/10.1007/ BF02294359 Allport, G. W. 1954, The Nature of Prejudice, Addison-Wesley, Reading, MA. Australian Bureau of Statistics 2017, 2016 Census data reveals ‘no religion’ is rising fast, Media Release 074/2017, Australian Bureau of Statistics, Canberra. https://www.abs.gov.au/AUSSTATS/abs@.nsf/ mediareleasesbyReleaseDate/7E65A144540551D7CA258148000E2B85 Australian Bureau of Statistics 2019a, 6202.0 – Labour Force, Australia, Dec 2019. https://www.abs.gov.au/statistics/ labour/employment-and-unemployment/labour-force-australia/dec-2019 Australian Bureau of Statistics 2019b, 6302.0 – Average Weekly Earnings, Australia, Nov 2019. https://www.abs.gov.au/ statistics/labour/earnings-and-working-conditions/average-weekly-earnings-australia/nov-2019 Carr, P. J., Napolitano L., & Keating J. 2007, ‘We never call the cops and here is why: A qualitative examination of legal cynicism in three Philadelphia neighborhoods’, Criminology, vol. 45, no. 2, pp. 445–480. https://doi.org/10.1111/j.1745- 9125.2007.00084.x Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 202399 https://doi.org/10.1027/1016-9040/a000415 https://doi.org/10.1027/1016-9040/a000415 https://doi.org/10.1186/s12889-020-10103-x https://doi.org/10.1007/BF02294359 https://doi.org/10.1007/BF02294359 https://www.abs.gov.au/AUSSTATS/abs@.nsf/mediareleasesbyReleaseDate/7E65A144540551D7CA258148000E2B85 https://www.abs.gov.au/AUSSTATS/abs@.nsf/mediareleasesbyReleaseDate/7E65A144540551D7CA258148000E2B85 https://www.abs.gov.au/statistics/labour/employment-and-unemployment/labour-force-australia/dec-2019 https://www.abs.gov.au/statistics/labour/employment-and-unemployment/labour-force-australia/dec-2019 https://www.abs.gov.au/statistics/labour/earnings-and-working-conditions/average-weekly-earnings-australia/nov-2019 https://www.abs.gov.au/statistics/labour/earnings-and-working-conditions/average-weekly-earnings-australia/nov-2019 https://doi.org/10.1111/j.1745-9125.2007.00084.x https://doi.org/10.1111/j.1745-9125.2007.00084.x Cook, K. S. & Santana, J. J. 2020, ‘Trust: perspectives in sociology’, In: Simon, J. (ed), The Routledge Handbook of Trust and Philosophy. Routledge, New York, NY, pp. 189-204. https://doi.org/10.4324/9781315542294-15 Cook, K. S., Hardin, R., & Levi, M. 2005, Cooperation without Trust?, Russell Sage Foundation, New York, NY. Diallo, T. M. O., Morin, A. J. S., & Lu, H. 2016a, ‘Impact of misspecifications of the latent variance–covariance and residual matrices on the class enumeration accuracy of growth mixture models’, Structural Equation Modeling: A Multidisciplinary Journal, vol. 23, no. 4, pp. 507–531. https://doi.org/10.1080/10705511.2016.1169188 Diallo, T. M. O., Morin, A. J. S., & Lu, H. 2016b, ‘Performance of growth mixture models in the presence of time- varying covariates’, Behavior Research Methods, vol. 49, no. 5, pp. 1951–1965. https://doi.org/10.3758/s13428-016-0823- 0 Diallo, T. M. O., Morin, A. J. S., & Lu, H. 2017, ‘The impact of total and partial inclusion or exclusion of active and inactive time invariant covariates in growth mixture models’, Psychological Methods, vol. 22, no. 1, pp. 166–190. https:// doi.org/10.1037/met0000084 Dwyer, A. 2014, ‘“We’re not like these weird feather boa-covered AIDS-spreading monsters”: How LGBT young people and service providers think riskiness informs LGBT youth-police interactions’, Critical Criminology, vol. 22, no. 1, pp. 65–79. https://doi.org/10.1007/s10612-013-9226-z Edelman 2020, 2020 Edelman Trust Barometer. https://www.edelman.com/trust/2020-trust-barometer Edelman 2021, 2021 Edelman Trust Barometer. https://www.edelman.com/trust/2021-trust-barometer Edelman 2022, Edelman Trust Barometer 2022: Australia Country Report. https://www.edelman.com.au/sites/g/files/ aatuss381/files/2022-02/Edelman%20Trust%20Barometer%202022%20-%20Australia%20Country%20Report.pdf Edelman 2023a, Australia on a path to polarisation: Edelman Trust Barometer 2023, February 23. https://www.edelman. com.au/australia-path-polarisation-edelman-trust-barometer-2023 Edelman 2023b, Edelman Trust Barometer 2023: Australia Report. https://www.edelman.com.au/sites/g/files/aatuss381/ files/2023-02/2023%20Edelman%20Trust%20Barometer%20Report%20-%20AUS%2002-2023.pdf Enders, C. K. 2010, Applied Missing Data Analysis, Guilford Press, New York. Ernst, N., Esser, F., Blassnig, S., & Engesser, S. 2019, ‘Favorable opportunity structures for populist communication: Comparing different types of politicians and issues in social media, television and the press’, The International Journal of Press/Politics, vol. 24, no. 2, pp. 165-188. https://doi.org/10.1177/1940161218819430 Fagan, J. & Tyler, T. R. 2005, ‘Legal socialization of children and adolescents’, Social Justice Research, vol. 18, no. 3, pp. 217–241. https://doi.org/10.1007/s11211-005-6823-3 Fazio, G., Giambona, F., Vassallo, E., & Vassiliadis, E. 2018, ‘A measure of trust: The Italian regional divide in a Latent Class Approach’, Social Indicators Research, vol. 140, pp. 209–242. https://doi.org/10.1007/s11205-017-1756-4 Fileborn, B. 2019, ‘Policing youth and queerness: the experiences and perceptions of young LGBTQ+ people from regional Victoria’, Current Issues in Criminal Justice, vol. 31, no. 3, pp. 433-451. https://doi.org/10.1080/10345329.2019. 1633727 Flexon, J. L., Lurigio, A. J., & Greenleaf, R. G. 2009, ‘Exploring the dimensions of trust in the police among Chicago juveniles’, Journal of Criminal Justice, vol. 37, no. 2, pp. 180–189. https://doi.org/10.1016/j.jcrimjus.2009.02.006 Forscher, P. S. & Kteily, N. S. 2020, ‘A Psychological Profile of the Alt-Right’, Perspectives on Psychological Science, vol. 15, no. 1, pp. 90–116. https://doi.org/10.1177/1745691619868208 Fukuyama, F. 1995, Trust: The Social Values and the Creation of Prosperity, Free Press, New York, NY. Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 2023100 https://doi.org/10.4324/9781315542294-15 https://doi.org/10.1080/10705511.2016.1169188 https://doi.org/10.3758/s13428-016-0823-0 https://doi.org/10.3758/s13428-016-0823-0 https://doi.org/10.1037/met0000084 https://doi.org/10.1037/met0000084 https://doi.org/10.1007/s10612-013-9226-z https://www.edelman.com/trust/2020-trust-barometer https://www.edelman.com/trust/2021-trust-barometer https://www.edelman.com.au/sites/g/files/aatuss381/files/2022-02/Edelman%20Trust%20Barometer%202022%20-%20Australia%20Country%20Report.pdf https://www.edelman.com.au/sites/g/files/aatuss381/files/2022-02/Edelman%20Trust%20Barometer%202022%20-%20Australia%20Country%20Report.pdf https://www.edelman.com.au/australia-path-polarisation-edelman-trust-barometer-2023 https://www.edelman.com.au/australia-path-polarisation-edelman-trust-barometer-2023 https://www.edelman.com.au/sites/g/files/aatuss381/files/2023-02/2023%20Edelman%20Trust%20Barometer%20Report%20-%20AUS%2002-2023.pdf https://www.edelman.com.au/sites/g/files/aatuss381/files/2023-02/2023%20Edelman%20Trust%20Barometer%20Report%20-%20AUS%2002-2023.pdf https://doi.org/10.1177/1940161218819430 https://doi.org/10.1007/s11211-005-6823-3 https://doi.org/10.1007/s11205-017-1756-4 https://doi.org/10.1080/10345329.2019.1633727 https://doi.org/10.1080/10345329.2019.1633727 https://doi.org/10.1016/j.jcrimjus.2009.02.006 https://doi.org/10.1177/1745691619868208 Gallup/Knight Foundation 2020, American Views 2020: Trust, Media and Democracy: A Deepening Divide, Washington D. C. https://knightfoundation.org/wp-content/uploads/2020/08/American-Views-2020-Trust-Media-and- Democracy.pdf Georgiou, N., Delfabbro, P., & Balzan, R. 2020, ‘COVID-19-related conspiracy beliefs and their relationship with perceived stress and pre-existing conspiracy beliefs’, Personality and Individual Differences, vol. 166, article 110201. https://doi.org/10.1016/j.paid.2020.110201 Goldfinch, S., Gauld, R., & Herbison, P. 2009, ‘The participation divide? Political participation, trust in government, and e-government in Australia and New Zealand’, Australian Journal of Public Administration, vol. 68, no. 3, pp. 333–350. https://doi.org/10.1111/j.1467-8500.2009.00643.x Goldfinch, S., Taplin, R., & Gauld, R. 2021, ‘Trust in government increased during the Covid‐19 pandemic in Australia and New Zealand’, Australian Journal of Public Administration, vol. 80, no. 1, pp. 3-11. https://doi.org/10.1111/1467- 8500.12459 Grant, B., Moore, T., & Lynch, T. (eds) 2019, The Rise of Right-Populism: Pauline Hanson’s One Nation and Australian Politics, Springer, Singapore. https://doi.org/10.1007/978-981-13-2670-7 Grattan, M. 2021, ‘Grattan on Friday: Vaccine rollout has enough problems without ministers politicking’, The Conversation, 1 April. https://theconversation.com/grattan-on-friday-vaccine-rollout-has-enough-problems-without- ministers-politicking-158313 Halafoff, A., Shipley, H., Young, P. D., Singleton, A., Rasmussen, M. L., & Bouma, G. 2020, ‘Complex, critical and caring: Young people’s diverse religious, spiritual and non-religious worldviews in Australia and Canada’, Religions, vol. 11, no. 4, pp. 166-178. https://doi.org/10.3390/rel11040166 Hardin, R. 2002, Trust and Trustworthiness, Russell Sage Foundation, New York. Harris A. 2010, ‘Young people, everyday civic life and the limits of social cohesion’, Journal of Intercultural Studies, vol. 31, no. 5, pp. 573-589. https://doi.org/10.1080/07256868.2010.513424 Hinds, L. 2007, ‘Building police–youth relationships: The importance of procedural justice’, Youth Justice, vol. 7, no. 3, pp. 195–209. https://doi.org/10.1177/1473225407082510 Hipp, J. R. & Bauer, D. J. 2006, ‘Local Solutions in the Estimation of Growth Mixture Models: Correction to Hipp and Bauer (2006)’, Psychological Methods, vol. 11, no. 3, p. 305. https://doi.org/10.1037/1082-989X.11.3.305 Hooghe, M., Marien, S., & Oser, J. 2017, ‘Great expectations: The effect of democratic ideals on political trust in European democracies’, Contemporary Politics, vol. 23, no. 2, pp. 214-230. https://doi.org/10.1080/13569775.2016.1210 875 Hudson, J. 2006, ‘Institutional trust and subjective well-being across the EU’, Kyklos: International Review for Social Sciences, vol. 59, no. 1, pp. 43–62. https://doi.org/10.1111/j.1467-6435.2006.00319.x Hurst, Y. G. & Frank, J. 2000, ‘How kids view cops: The nature of juvenile attitudes toward the police’, Journal of Criminal Justice, vol. 28, no. 3, pp. 189–202. https://doi.org/10.1016/S0047-2352(00)00035-0 Karić, T. & Međedović, J. 2021, ‘Covid-19 conspiracy beliefs and containment-related behaviour: The role of political trust’, Personality and Individual Differences, vol. 175, article 110697. https://doi.org/10.1016/j.paid.2021.110697 Kennelly, J. 2011, ‘Policing young people as citizens-in-waiting: Legitimacy, spatiality and governance’, The British Journal of Criminology, vol. 51, no. 2, pp. 336–354. https://doi.org/10.1093/bjc/azr017 Koopmans, R. & Olzak, S. 2004, ‘Discursive opportunities and the evolution of right-wing violence in Germany’, American Journal of Sociology, vol. 110, no. 1, pp. 198-230. https://doi.org/10.1086/386271 Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 2023101 https://knightfoundation.org/wp-content/uploads/2020/08/American-Views-2020-Trust-Media-and-Democracy.pdf https://knightfoundation.org/wp-content/uploads/2020/08/American-Views-2020-Trust-Media-and-Democracy.pdf https://doi.org/10.1016/j.paid.2020.110201 https://doi.org/10.1111/j.1467-8500.2009.00643.x https://doi.org/10.1111/1467-8500.12459 https://doi.org/10.1111/1467-8500.12459 https://doi.org/10.1007/978-981-13-2670-7 https://theconversation.com/grattan-on-friday-vaccine-rollout-has-enough-problems-without-ministers-politicking-158313 https://theconversation.com/grattan-on-friday-vaccine-rollout-has-enough-problems-without-ministers-politicking-158313 https://doi.org/10.3390/rel11040166 https://doi.org/10.1080/07256868.2010.513424 https://doi.org/10.1177/1473225407082510 https://doi.org/10.1037/1082-989X.11.3.305 https://doi.org/10.1080/13569775.2016.1210875 https://doi.org/10.1080/13569775.2016.1210875 https://doi.org/10.1111/j.1467-6435.2006.00319.x https://doi.org/10.1016/S0047-2352(00)00035-0 https://doi.org/10.1016/j.paid.2021.110697 https://doi.org/10.1093/bjc/azr017 https://doi.org/10.1086/386271 Kye, B. & Hwang, S-J. 2020, ‘Social trust in the midst of pandemic crisis: Implication from COVID-19 of South Korea’, Research in Social Stratification and Mobility, vol. 68, article 100523. https://doi.org/10.1016/j.rssm.2020.100523 La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. W. 1997, ‘Trust in large organizations’, The American Economic Review, vol 87, no. 2, pp. 333-338. https://www.jstor.org/stable/2950941 Lo, Y., Mendell, N., & Rubin, D. 2001, ‘Testing the number of components in a normal mixture’, Biometrika, vol. 88, no. 3, pp. 767–778. https://doi.org/10.1093/biomet/88.3.767 Loader, I. 1996, Youth, Policing, and Democracy, Macmillan, Basingstoke. https://doi.org/10.1057/9780230373839 Markus, A. 2017, Mapping Social Cohesion: The Scanlon Foundation Surveys 2017, ACJC, Faculty of Arts, Monash University, Caulfield East. https://scanloninstitute.org.au/sites/default/files/2019-11/ScanlonFoundation_ MappingSocialCohesion_2017-1.pdf Markus, A. 2019a, Mapping Social Cohesion: The Scanlon Foundation Surveys 2019, ACJC, Faculty of Arts, Monash University, Caulfield East. https://scanloninstitute.org.au/sites/default/files/2019-11/Mapping%20Social%20 Cohesion%202019%20FINAL%29.pdf Markus, A. 2019b, ‘The Values of One Nation Voters’, In: Peucker, M. & Smith, D. (eds), The Far-Right in Contemporary Australia, Palgrave Macmillan, Singapore, pp. 53-72. https://doi.org/10.1007/978-981-13-8351-9_3 Markus, A. 2021, Mapping Social Cohesion: The Scanlon Foundation Surveys 2020, ACJC, Faculty of Arts, Monash University, Caulfield East. https://scanloninstitute.org.au/sites/default/files/2021-02/SC2020%20Report%20Final.pdf McAra, L. & McVie, S. 2005, ‘The usual suspects? Street-life, young people and the police’, Criminology & Criminal Justice, vol. 5, no. 1, pp. 5–36. https://doi.org/10.1177/1466802505050977 McAra, L. & McVie, S. 2007, ‘Youth justice? The impact of system contact on patterns of desistance from offending’, European Journal of Criminology, vol. 4, no. 3, pp. 315–345. https://doi.org/10.1177/1477370807077186 Meyer, D. S. 2004, ‘Protest and political opportunities’, Annual Review of Sociology, vol. 30, pp. 125-145. https://doi. org/10.1146/annurev.soc.30.012703.110545 Miller, J. M., Saunders, K. L., & Farhart, C. E. 2016, ‘Conspiracy endorsement as motivated reasoning: The moderating roles of political knowledge and trust’, American Journal of Political Science, vol. 60, no. 4, pp. 824-844. https://doi. org/10.1111/ajps.12234 Moesen, W., van Puyenbroeck, T., & Cherchye, L. 2000, Trust as societal capital: Economic growth in European regions, Discussions paper series 1, Center for Economic Studies, Catholic University of Leuven. Mudde, C. & Kaltwasser, C. R. 2017, Populism: A Very Short Introduction, Oxford University Press, Oxford. https://doi. org/10.1093/actrade/9780190234874.001.0001 Murphy, K. 2021, ‘More than half of Australians think Covid vaccine rollout is too slow, poll suggests’, The Guardian, 13 April. https://www.theguardian.com/australia-news/2021/apr/13/more-than-half-of-australians-think-covid- vaccine-rollout-is-too-slow-poll-suggests Muthén, B. 2004, ‘Latent variable analysis: Growth mixture modelling and related techniques for longitudinal data’, In Kaplan, D. (ed), Handbook of Quantitative Methodology for the Social Sciences, Sage Publishing, Thousand Oaks, CA, pp. 345–368. https://doi.org/10.4135/9781412986311.n19 Muthén L. K. & Muthén, B. O. 2017, Mplus User’s Guide, 8th edn, Muthén & Muthén, Los Angeles, CA. Nilan, P., Roose, J., Peucker, M., & Turner, B. S. 2023, ‘Young masculinities and right-wing populism in Australia’, Youth, vol. 3, no. 1, pp. 285-299. https://doi.org/10.3390/youth3010019 Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 2023102 https://doi.org/10.1016/j.rssm.2020.100523 https://www.jstor.org/stable/2950941 https://doi.org/10.1093/biomet/88.3.767 https://doi.org/10.1057/9780230373839 https://scanloninstitute.org.au/sites/default/files/2019-11/ScanlonFoundation_MappingSocialCohesion_2017-1.pdf https://scanloninstitute.org.au/sites/default/files/2019-11/ScanlonFoundation_MappingSocialCohesion_2017-1.pdf https://scanloninstitute.org.au/sites/default/files/2019-11/Mapping%20Social%20Cohesion%202019%20FINAL%29.pdf https://scanloninstitute.org.au/sites/default/files/2019-11/Mapping%20Social%20Cohesion%202019%20FINAL%29.pdf https://doi.org/10.1007/978-981-13-8351-9_3 https://scanloninstitute.org.au/sites/default/files/2021-02/SC2020%20Report%20Final.pdf https://doi.org/10.1177/1466802505050977 https://doi.org/10.1177/1477370807077186 https://doi.org/10.1146/annurev.soc.30.012703.110545 https://doi.org/10.1146/annurev.soc.30.012703.110545 https://doi.org/10.1111/ajps.12234 https://doi.org/10.1111/ajps.12234 https://doi.org/10.1093/actrade/9780190234874.001.0001 https://doi.org/10.1093/actrade/9780190234874.001.0001 https://www.theguardian.com/australia-news/2021/apr/13/more-than-half-of-australians-think-covid-vaccine-rollout-is-too-slow-poll-suggests https://www.theguardian.com/australia-news/2021/apr/13/more-than-half-of-australians-think-covid-vaccine-rollout-is-too-slow-poll-suggests https://doi.org/10.4135/9781412986311.n19 https://doi.org/10.3390/youth3010019 Nylund, K. L., Asparouhov, T., & Muthén, B. O. 2007, ‘Deciding on the number of classes in latent class analysis and growth mixture modelling: A Monte Carlo simulation study’, Structural Equation Modeling: A Multidisciplinary Journal, vol. 14, no. 4, pp. 535-569. https://doi.org/10.1080/10705510701575396 Peugh, J. & Fan, X. 2013, ‘Modeling unobserved heterogeneity using latent profile analysis: A Monte Carlo simulation’, Structural Equation Modeling: A Multidisciplinary Journal, vol. 20, no. 4, pp. 616–639. https://doi.org/10.1080/10705511. 2013.824780 Piquero, A. R., Fagan, J., Mulvey, E. P., Steinberg, L., & Odgers, C. 2005, ‘Developmental trajectories of legal socialization among serious adolescent offenders’, Journal of Criminal Law and Criminology, vol. 96, no. 1, pp. 267–298. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2776646/ Putnam, R. D. 1993, ‘What makes democracy work?’, National Civic Review, vol. 82, no. 2, pp. 101–107. https://doi. org/10.1002/ncr.4100820204 Putnam, R. D. 2000, Bowling Alone: The Collapse and Revival of American Community, Simon and Schuster, New York, NY. https://doi.org/10.1145/358916.361990 Ruelens, A. & Nicaise, I. 2020 ‘Investigating a typology of trust orientations towards national and European institutions: A person-centered approach’, Social Science Research, vol. 87, article 102414. https://doi.org/10.1016/j. ssresearch.2020.102414 Schafer, J. L. & Graham, J. W. 2002, ‘Missing data: Our view of the state of the art’, Psychological Methods, vol. 7, no. 2, pp. 147–177. https://doi.org/10.1037/1082-989X.7.2.147 Schwarz, G. E. 1978, ‘Estimating the dimension of a model’, Annals of Statistics, vol. 6, no. 2, pp. 461–464. https://doi. org/10.1214/aos/1176344136 Sclove S. L. 1987, ‘Application of model-selection criteria to some problems in multivariate analysis’, Psychometrika, vol. 52, no. 3, pp. 333-343. https://doi.org/10.1007/BF02294360 Scandurra, C., Bochicchio, V., Dolce, P., Valerio, P., Muzii, B., & Maldonato, N. M. 2021, ‘Why people were less compliant with public health regulations during the second wave of the Covid-19 outbreak: The role of trust in governmental organizations, future anxiety, fatigue, and Covid-19 risk perception’, Current Psychology, vol. 42, no. 9, pp. 7403-7413. https://doi.org/10.1007/s12144-021-02059-x Shanahan, L., Copeland, W. E., Worthman, C. M., Erkanli, A., Angold, A., & Costello, E. J. 2013, ‘Sex-differentiated changes in C-reactive protein from ages 9 to 21: The contributions of BMI and physical/sexual maturation’, Psychoneuroendocrinology, vol. 38, no. 10, pp. 2209-2217. https://doi.org/10.1016/j.psyneuen.2013.04.010 Sindall, K., McCarthy, D. J., & Brunton-Smith, I. 2017, ‘Young people and the formation of attitudes towards the police’, European Journal of Criminology, vol. 14, no. 3, pp. 344–364. https://doi.org/10.1177/1477370816661739 Speers, D. 2021, ‘Swing voters in marginal seats like Bass will be vital for Scott Morrison’, ABC News, 8 April. https:// www.abc.net.au/news/2021-04-08/marginal-seats-react-coalition-scott-morrison/100053376 Stoker, G., Evans, M., & Halupka, M. 2018, Trust and Democracy in Australia: Democratic decline and renewal, Democracy 2025 Report no. 1, Museum of Australian Democracy, Canberra. https://apo.org.au/sites/default/files/ resource-files/2018-12/apo-nid208536.pdf Suh, C. S., Chang, P. Y., & Lim, Y. 2012, ‘Spill‐up and spill‐over of trust: An extended test of cultural and institutional theories of trust in South Korea’, Sociological Forum, vol. 27, no. 2, pp. 504-526. https://doi.org/10.1111/j.1573- 7861.2012.01328.x van der Linden, S., Panagopoulos, C., & Roozenbeek, J. 2020, ‘You are fake news: Political bias in perceptions of fake news’, Media, Culture & Society, vol. 42, no. 3, pp. 460-470. https://doi.org/10.1177/0163443720906992 Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 2023103 https://doi.org/10.1080/10705510701575396 https://doi.org/10.1080/10705511.2013.824780 https://doi.org/10.1080/10705511.2013.824780 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2776646/ https://doi.org/10.1002/ncr.4100820204 https://doi.org/10.1002/ncr.4100820204 https://doi.org/10.1145/358916.361990 https://doi.org/10.1016/j.ssresearch.2020.102414 https://doi.org/10.1016/j.ssresearch.2020.102414 https://doi.org/10.1037/1082-989X.7.2.147 https://doi.org/10.1214/aos/1176344136 https://doi.org/10.1214/aos/1176344136 https://doi.org/10.1007/BF02294360 https://doi.org/10.1007/s12144-021-02059-x https://doi.org/10.1016/j.psyneuen.2013.04.010 https://doi.org/10.1177/1477370816661739 https://www.abc.net.au/news/2021-04-08/marginal-seats-react-coalition-scott-morrison/100053376 https://www.abc.net.au/news/2021-04-08/marginal-seats-react-coalition-scott-morrison/100053376 https://apo.org.au/sites/default/files/resource-files/2018-12/apo-nid208536.pdf https://apo.org.au/sites/default/files/resource-files/2018-12/apo-nid208536.pdf https://doi.org/10.1111/j.1573-7861.2012.01328.x https://doi.org/10.1111/j.1573-7861.2012.01328.x https://doi.org/10.1177/0163443720906992 van Prooijen, J. W. & Douglas, K. M. 2017, ‘Belief in conspiracy theories: Basic principles of an emerging research domain’, European Journal of Social Psychology, vol. 48, no. 7, pp. 897-908. https://doi.org/10.1002/ejsp.2530 World Economic Forum 2018, The Global Risks Report 2018, 13th Edition, Insight Report, Geneva. https://www3. weforum.org/docs/WEF_GRR18_Report.pdf World Values Survey 2014, World Values Survey Wave 6 (2010-2014) Official Questionnaire, JD Systems, Madrid. https:// www.worldvaluessurvey.org/WVSDocumentationWV6.jsp Wyn, J. & White, R. 2012, Rethinking Youth, Sage Publications, London. https://doi.org/10.4135/9781446250297 Zak, P. J., & Knack, S. 2001, ‘Trust and growth’, The Economic Journal, vol. 111, no. 470, pp. 295–321. https://doi. org/10.1111/1468-0297.00609 Kamp, et al. Cosmopolitan Civil Societies: An Interdisciplinary Journal, Vol. 15, No. 2 2023104 https://doi.org/10.1002/ejsp.2530 https://www3.weforum.org/docs/WEF_GRR18_Report.pdf https://www3.weforum.org/docs/WEF_GRR18_Report.pdf https://www.worldvaluessurvey.org/WVSDocumentationWV6.jsp https://www.worldvaluessurvey.org/WVSDocumentationWV6.jsp https://doi.org/10.4135/9781446250297 https://doi.org/10.1111/1468-0297.00609 https://doi.org/10.1111/1468-0297.00609