Not all negative: Macro justice principles predict positive attitudes towards asylum seekers in Australia

Abstract

Objective

Public opinion towards asylum seekers within Australia has become increasingly hostile over the past decade. In particular, such negative attitudes are associated with questioning the legitimacy of those who seek asylum and the fairness of granting their refugee status. The major aim of this paper is to test the role of macro and micro principles of social justice in predicting attitudes towards asylum seekers, beyond the established role of social dominance orientation (SDO) and right‐wing authoritarianism (RWA).

Method

A sample of 100 students (M_age = 22.83 years, SD_age = 8.26 years) responded to measures of macro and micro principles of social justice, SDO, RWA, and a measure of Attitudes Towards Asylum Seekers.

Results

Using multiple hierarchical regression analyses, we show that macro justice social principles (i.e., the belief in equal distribution of resources across a society) predict positive attitudes towards asylum seekers beyond the variation accounted for by SDO and RWA in predicting negative attitudes.

Conclusions

These results underscore the importance of taking into account individual orientations towards justice; we argue that these findings have important implications for the development of communication designed to reduce prejudice towards asylum seekers.

Key words:

• asylum seekers
• ATAS
• justice principles
• refugee
• social justice

Declaration of conflicting interests: We declare no potential conflicts of interest with respect to the research, authorship, and /or publication of this article. This study was unfunded.

The population of individuals seeking refuge worldwide is steadily increasing, and this trend has extended to applications for asylum seekers arriving in Australia. However, Australia only accepts a small fraction of the worldwide total of refugees, both per capita and in absolute terms (Refugee Council of Australia, 2012). For example, in 2012 over two million applications for asylum were received globally. However, Australia hosted only 16,000 individuals out of this worldwide population (United Nations High Commissioner for Refugees, 2012).

Despite Australia's relatively small intake of asylum seekers, the acceptance and treatment of refugee claimants have reached the status of a major electoral issue, particularly pertaining to claimants who travel to Australia by boat (for details, see Hall & Swan, 2013; Lusher, Balvin, Nethery, & Tropea, 2007). Australia's stance towards asylum seekers can be characterised by largely hostile community attitudes and punitive public policy, increasingly so over the last two decades (see Haslam & Holland, 2012). The arrival of asylum seekers by boat has been constructed as a threat to the integrity of Australia's borders, security, and international sovereignty (Marr & Wilkinson, 2003), leading to a range of policies based on the goal of deterrence, such as mandatory and indefinite detention, turning back boats, and offshore processing. Such policies have been criticised by the United Nations, the Australian Human Rights Commission, and other humanitarian groups who argue that these measures violate Australia's human rights obligations (e.g., Australian Human Rights Commission, 2013). Most recently, in 2012, the Australian government reintroduced offshore processing (previously dismantled in 2008), whereby all asylum seekers attempting to arrive in Australia by boat are held in detention in neighbouring countries. Significant concerns have been raised with the reported conditions in these offshore detention centres (e.g., hygiene, safety), and over whether it is assured that asylum seekers will not be returned to a country where they face human rights violations (Amnesty International Australia, 2013).

At the community level, Australia's stance towards asylum seekers can be characterised by largely hostile attitudes. Indeed, research consistently shows that a large proportion of the Australian public hold openly negative attitudes towards asylum seekers. For example, in an analysis of community attitudes in Port Augusta, Klocker (2004) found that 70% of participants saw asylum seekers as ‘illegal’, ‘unlawful’, ‘an economic burden’, ‘problematic’, ‘unwelcome’, and ‘ungrateful’. Other terms used were ‘illegitimate’, ‘queue jumpers’, and ‘terrorists’. In a similar vein, Pedersen and her colleagues have identified a range of commonly held false beliefs about asylum seekers that are endorsed by a significant proportion of the Australian community, such as the belief that asylum seekers who arrive by boat are queue jumpers, illegals, not genuine refugees, ‘cashed up’, and receive excessive government handouts. Such false beliefs are strongly and consistently associated with negative attitudes towards asylum seekers (Pedersen, Attwell, & Heveli, 2005; Pedersen, Watt, & Hanser, 2006; Suhnan, Pedersen, & Hartley, 2012). We argue that the common thread drawing these beliefs together is that they tend to be based around the notion of violations of fairness.

Applications of psychology to understanding community attitudes towards asylum seekers have primarily focused on known individual difference predictors of prejudice (Hodson & Costello, 2007; Louis, Duck, Terry, Schuller, & Lalonde, 2007; Pedersen et al., 2005). However, we argue that an individual's definition of what constitutes a fair allocation of resources may also be relevant to understanding negative attitudes towards asylum seekers. Indeed, social justice literature suggests that there are psychologically distinct types of justice beliefs. In this article, we focus on equity versus equality‐based justice principles, sometimes called macro versus micro orientations of justice (Brickman, Folger, Goode, & Schul, 1981; Zdaniuk & Bobocel, 2011), that is, the belief that people should be allocated resources based on meritocracy versus the belief that societal resources should be allocated in an egalitarian fashion, with consideration for disadvantaged social groups. Therefore, in this study, we examine how prejudice towards asylum seekers is related to different social justice orientations. Specifically, we explore the relationship between attitudes towards asylum seekers and support for equity‐based justice principles (since application for asylum is based on need, rather than merit; i.e., micro justice principles), and whether positive attitudes towards asylum seekers are underpinned by a belief in equality‐based principles (i.e., macro justice principles).

Psychological theories underlying negative perceptions of asylum seekers

One prominent approach to understanding negative attitudes towards asylum seekers is Duckitt and Sibley's dual process model of prejudice (Duckitt & Sibley, 2010). The model suggests that prejudice is derived from threat‐ and competition‐based dual processes, which relate to authoritarianism and traditionalism (i.e., right‐wing authoritarianism, RWA; Altemeyer, 1981), and hierarchy and inequality (i.e., social dominance orientation, SDO; Sidanius & Pratto, 1999), respectively. Threat‐based cognitive motivational processes manifest as RWA, a construct currently understood as an individual difference factor that measures the attitudinal components of authoritarian submission, authoritarian aggression, and conventionalism. People scoring high on RWA tend to be intolerant of societal deviance and champion social control and conformity, which in turn leads to more negative attitudes of ‘deviant’ out‐groups, perceived to be a threat to the social order and the security of their in‐group. Therefore, RWA is thought to lead to intergroup prejudice due to a view of the world as dangerous and threatening.

Competition‐based cognitive motivational processes manifest as SDO, which stems from notions that societies function optimally with a few ‘dominant and hegemonic groups’ in a superordinate position, and with a number of subordinate groups below them (Sidanius & Pratto, 1999, p. 31). SDO is the individual difference manifestation of social dominance theory, which is characterised by a competitive worldview and preferences for social power, hierarchy, and group‐based dominance. People scoring high on SDO show negative attitudes towards out‐groups due to the belief that some groups in society rightfully deserve more power than others. SDO has routinely been shown to predict negative attitudes towards group‐based minorities over and above both political identities and political orientations (e.g., Sanders & Mahalingam, 2012). Duckitt and Sibley (2010) demonstrate that while RWA and SDO tend to be moderately correlated, they are conceptually distinct, and are meaningful and statistically powerful predictors of prejudices. Each construct has distinct motivational antecedents and contributes unique variance in models that predict prejudice (e.g., Altemeyer, 1988; Duckitt, 2006; Duckitt & Sibley, 2010; Heaven & Connors, 2001).

Drawing on Duckitt and Sibley's dual process model of prejudice (e.g., 2010), recent Australian research has found support for the link among RWA, SDO, and negative attitudes towards asylum seekers (e.g., Louis et al., 2007; Nickerson & Louis, 2008). Nickerson and Louis (2008) found that RWA and SDO independently predict negative responses to asylum seekers in Australia. Similarly, Louis et al. (2007) demonstrated that participants who preferred a hierarchical rather than egalitarian social structure (SDO) were more likely to endorse the restriction of asylum seekers' access to Australia.

Although this model is well established and useful in understanding prejudice, much of the hostility towards asylum seekers in Australia seems to be driven by perceptions of violations of fairness and justice (e.g., the idea that asylum seekers are ‘queue jumpers’ Pedersen et al., 2005, 2006; Suhnan et al., 2012). SDO and RWA revolve around perceptions of hierarchical preferences, or preferences for conformity and authority, respectively, so the application of these ideological domains may provide an incomplete picture. We suggest that prejudice towards asylum seekers may also arise from different perceptions of distributive justice, that is, an individual's belief in the ideal way to allocate resources within a given society. We discuss this area of research below.

The contribution of social justice principles

Attitudes towards asylum seekers may differ based on the justice orientation adopted by a given individual. Within the social justice literature, a distinction between a micro and macro justice orientations has been proposed (Brickman et al., 1981; Lillie & Janoff‐Bulman, 2007; Zdaniuk & Bobocel, 2011). A macro justice view of distributive justice is based on the principle of restoring equality (i.e., the belief in maximising the outcomes for the currently most disadvantaged social groups in a society). This orientation is thought to be derived from a focus on society's overall well‐being. Conversely, a micro justice orientation operates from the principle of equity, which is the belief that a fair society is one where individuals are allocated wealth, goods, and services in proportion to their output. Seen in this light, there should be different outcomes for members of a society based on their individual merit (i.e., wealth should be allocated according to how much a person contributes to the society, the effort they put into their job, and this skills they possess). This orientation tends to be driven by a concern for societal productivity.

In a longitudinal study, Smith and Matějů (2012) concluded that meritocratic (micro justice) orientations to justice have become more dominant over time. Duru‐Bellat and Tenret (2012) also found generally high levels of pro‐meritocratic attitudes in a recent survey encompassing 26 countries and suggested this may be a product of modern individualistic societies. In practice, this global and temporal increase in support for micro justice principles might explain Australia's current asylum seeker policies, based on the assumption that resources should be allocated to those who possess the highest merit, rather than need. To our knowledge, no research has explored the role of social justice principles in predicting negative attitudes towards those seeking asylum.

Aims and hypotheses

The current study aims to extend the literature predicting individual‐level differences in attitudes towards asylum seekers in Australia. We have argued that debates surrounding the treatment of those seeking asylum might be informed by notions of social justice. Therefore, the current study explores the predictive potential of two novel factors, specifically macro and micro justice principles in the prediction of attitudes towards asylum seekers. Individual differences that posit a preference for hierarchy (i.e., SDO) and tradition (i.e., RWA) have previously been shown to predict negative attitudes towards asylum seekers (e.g., Nickerson & Louis, 2008). As such, in line with Duckitt and Sibley's (2010) dual process model of prejudice, our first prediction is that SDO and RWA will predict negative attitudes towards asylum seekers.

Previous research has found that negative perceptions of asylum seekers tend to be associated with uncertainty about whether or not their claims for asylum are legitimate and if they are in genuine need of refuge, and the perception that they take resources that belong to Australians (e.g., Louis et al., 2007). Thus, we expect that negative attitudes towards asylum seekers may be underpinned by micro justice principles because asylum seekers are seen as violating meritocratic processes by trying to ‘cheat’ the system or expect handouts. Conversely, principles of macro justice would be linked to support for asylum seekers due to ideals of equality, that is, seeing asylum seekers as being in a position of severe disadvantage, and therefore in need of resources. As such, our second prediction is that those who endorse micro justice principles will be more likely to hold negative attitudes towards asylum seekers, whereas those who endorse macro justice principles will be more likely to hold positive attitudes towards asylum seekers. If micro justice and macro justice principles are indeed conceptually separate from SDO and RWA, then they should predict variation in prejudice beyond that accounted for by SDO and RWA.

Method

Participants

The sample comprised 100 undergraduate psychology students from an Australian university (M_age = 22.83 years, SD_age = 8.26 years, 79 females) who participated in exchange for partial course credit. All participants were born in Australia, and the majority identified as Caucasian (n = 71). The remaining participants identified as Asian, Middle Eastern, Indian, or as Indigenous Australian (n's = 11, 5, 2, and 1, respectively).

Materials and procedure

Participants had the option to complete the battery online or in a supervised testing lab session. When tested in a supervised session, contact with the tester was limited until the debriefing. The battery comprised demographic questions followed by the measures described below, which were presented in a randomised order to limit presentation effects. Upon completion, all participants were debriefed.

The Attitudes Toward Asylum Seekers scale (ATAS: Pedersen et al., 2005) is designed to measure affective evaluations of refugees and asylum seekers in an Australian context. Comprising 18 items (e.g., ‘Asylum seekers are ungrateful by protesting in the manner that they do’), the ATAS measures participants' levels of endorsement to statements on a scale ranging from 1 (very strongly disagree) to 7 (very strongly agree). Appropriate items were reverse‐scored, and an average score was computed for each participant; higher scores indicated more negative attitudes towards asylum seekers.

The short‐form of the Right‐Wing Authoritarianism scale (RWA; Altemeyer, 1981) was used to assess support for authority and tradition. This version comprises 14 items (e.g., ‘Obedience and respect for authority are the most important values children should learn’.) to which participants indicate their agreement to statements on a scale from 1 (very strongly disagree) to 7 (very strongly agree). Appropriate items were reverse‐scored, and an average score was computed for each participant; higher scores indicate stronger endorsement of RWA beliefs.

The Social Dominance scale (SDO; Pratto, Sidanius, Stallworth, & Malle, 1994) measures preference for inequality among social groups. It comprises 16 items (e.g., ‘Some groups of people are simply inferior to other groups’) on which participants indicate their agreement to statements on a scale from 1 (very strongly disagree) to 7 (very strongly agree). Appropriate items were reverse‐scored, and an average score was computed for each participant; higher scores indicate a stronger SDO.

The Micro Justice and Macro Justice Principles Scale comprises two dimensions that measure preferences of social fairness (Zdaniuk & Bobocel, 2011). Items were adjusted to suit the Australian context (i.e., changing the social groups in the items to Australians and asylum seekers). Micro justice principles emphasise what is fair for the individual (eight items; e.g., ‘Each person's income should be based on how hard he or she works relative to others’.). Macro justice principles emphasise what is fair for the society (eight items; e.g., ‘The income differences between asylum seekers and Australian‐born citizens should not be too large’.). Participants indicate their agreement to statements on a scale from 1 (very strongly disagree) to 7 (very strongly agree). Appropriate items were reverse‐scored, and an average score was computed for each subscale; a higher micro justice orientation score indicates social justice preferences based on equity, and a higher macro justice orientation score indicates a social justice preference based on equality.

Results

In order to estimate the proportion of variance in attitudes towards asylum seekers that can be accounted for by attitude‐relevant factors, we conducted a multiple hierarchical regression analysis.¹ The sample reported relatively low levels of SDO, and an average level and range of the remaining variables. Scores on the ATAS scale had a strong positive relationship with SDO, a moderate, positive relationship with RWA, and a moderate but negative relationship with principles of macro justice (see Table 1 for descriptive statistics and correlation analysis for all variables).

Table 1. Summary of descriptive statistics, correlations, and reliabilities for attitudes towards asylum seekers scale and attitude‐relevant measures

Scale	1	2	3	4	5
1. ATAS	(.91)
2. RWA	.22*	(.72)
3. SDO	.55***	−.04	(.85)
4. Micro justice	−.16	.20*	−.08	(.75)
5. Macro justice	−.35***	.15	−.19	.57**	(.98)
Mean	3.44	4.37	2.95	5.31	4.69
SD	.95	.70	.89	.86	.88

Notes. ATAS = Attitudes Towards Asylum Seekers; SDO = social dominance orientation; RW = right‐wing authoritarianism; SD, standard deviation.

*p < .05, **p < .01, ***p < .001; all correlations reaching statistical significance are highlighted in boldface; alpha Cronbach coefficients are presented in parentheses.

Step 1 of the regression included SDO and RWA (i.e., variables that are known predictors of attitudes towards asylum seekers); these variables accounted for a significant 36.00% of the variance in attitudes towards asylum seekers, F(2, 95) = 26.69, p < .001. Step 2 included the principles of micro and macro justice, accounting for an additional 6.10% of the variance ΔF(2, 93) = 4.88, p < .001. In combination, predictor variables accounted for 42.00% variance in reported attitudes towards asylum seekers, which can be considered a large effect (Cohen's f2 = .852; Cohen, 1988).²

As expected, SDO and RWA were strong predictors of negative attitudes towards asylum seekers in the first step of the model. Both predictors remained significant in the second step (uniquely accounting for 25.50% and 7.84% of the variance in attitudes, respectively), and principles of macro justice were revealed as a predictor of positive attitudes towards asylum seekers (contributing a unique 5.57% of variance in attitudes towards asylum seekers). Principles of micro justice was a not a significant predictor of attitudes towards asylum seekers. Regression coefficients and squared semi‐partial correlations for each predictor on each step of the hierarchical multiple regression analysis predicting attitudes towards asylum seekers are presented in Table 2.

Table 2. Unstandardised (B) and standardised (β) regression coefficients, and semi‐partial correlations for predictors in a hierarchical regression model predicting attitudes towards asylum seekers (N = 100)

	B [95% CI]	SE B	β	p	Sr^2
Step 1
Constant	−.43	.78
RWA	3.38 [1.12, 5.63]	1.13	.25	.004	.24
SDO	.59 [.42, .76]	.087	.56	<.001	.55
Step 2
Constant	.711	.85
RWA	3.74 [1.39, 6.08]	1.18	.28	.001	.27
SDO	.54 [.37, .71]	.09	.51	<.001	.50
Micro justice	−.01 [−.2, .21]	.11	−.02	.874	−.01
Macro justice	−.27 [−.48, −.07]	.10	−.29	.004	−.23

Notes. CI, confidence interval; RWA = right‐wing authoritarianism; SDO = social dominance orientation.

Final model = F(4, 93) = 16.87, p < .001; findings reaching statistical significance are highlighted in boldface.

Discussion

Previous research into individual difference variables underpinning negative attitudes towards asylum seekers in Australia has primarily employed the dual process model of prejudice. In the current study, we sought to investigate the role of social justice principles in the prediction of attitudes towards asylum seekers, over and above that accounted for by the dual process model. Consistent with previous research, our results supported the hypothesis that SDO and RWA are unique predictors of negative attitudes towards asylum seekers (Duckitt & Sibley, 2010; Heaven & Connors, 2001; Jost, Glaser, Kruglanski, & Sulloway, 2003; Louis et al., 2007; Sanders & Mahalingam, 2012). SDO was a strong predictor and RWA explained relatively less, but still a significant proportion of variance in the prediction of negative attitudes towards asylum seekers. There are clear reasons for links between RWA, SDO, and negative attitudes towards asylum seekers. The association between these variables conceptually in line with common perceptions of asylum seekers as a threat to the social order, and as a source of competition for jobs and social welfare benefits. Specifically, a person high in RWA may see asylum seekers as a threat to their safety or way of life, whereas those high on SDO may perceive asylum seekers to be a direct challenge to resources, such as jobs and property.

The novel result of the current study pertains to macro justice principles, which predicted positive attitudes towards asylum seekers, and significantly contributed to the variance in attitudes beyond that predicted by SDO and RWA. We reason that a person who believes that his or her society should specifically allocate resources to disadvantaged social groups is logically more inclined to exhibit tolerant and positive attitudes towards asylum seekers, which by definition is a disadvantaged group in need of resources. Smith and Matějů (2012) posed a similar notion that macro justice is driven by principles of equality with particular consideration given to those in need, which is a category that asylum seekers fall into with relative ease.

The finding that a micro justice orientation does not predict attitudes towards asylum seekers was unexpected. Zdaniuk and Bobocel's (2011) research linked micro justice principles with more meritocratic beliefs and the reduced likelihood of accepting affirmative action. Thus, it is inconsistent with the theory that our data failed to find micro justice principles to be a significant predictor of negative attitudes towards asylum seekers. However, it is plausible that in the specific case of asylum seekers, meritocratic beliefs may not be applicable to a social group who are not perceived as part of a society in which they can prove their merit. Thus, it may be that micro justice principles are not related to these attitudes because of the lack of relevance to the target social group.

While micro justice principles, and meritocracy, are found to be on the rise globally (Smith & Matějů, 2012), there is a promising line of evidence to suggest that people desire more societal equality than currently exists (e.g., Norton, Neal, Govan, Ariely, & Holland, 2014). Studies, such as these demonstrate that the disparities existing between social groups can be made more apparent by experimentally manipulating a macro justice perspective, and this can in turn increase support for the restoration of equality.

Together, our findings have important implications for the strategies that are employed to reduce community hostility towards asylum seekers. Louis and her colleagues (2007) have suggested that Australian attitudes towards asylum seekers are polarised and are unlikely to reconcile in the near future. However, Lillie and Janoff‐Bulman (2007) propose that justice orientations are changeable and often subject to situational factors, which offers promising future directions for potential prejudice reduction interventions (see also Hartley, Pedersen, & Dandy, 2012). Likewise, we suggest that encouraging people to adopt a macro justice perspective may be a useful addition to community interventions, along with interventions established in prior research (e.g., tackling false beliefs about asylum seekers receiving government handouts, or being ‘cashed up’; Pedersen et al., 2005).

Our findings also have broader theoretical implications. Despite the long‐established tradition of micro and macro justice principles (in political philosophy especially; Rawls, 1971), relatively few psychological studies examine the role of these principles in understanding inequality or prejudice, particularly in comparison to more recently predominant constructs of SDO and RWA. Yet understanding prejudice requires a multifaceted approach, as there are often different ideologies that lead to the positive or negative evaluation of particular social groups, and these can have their separate underpinnings. Our finding that macro justice principles predict positive attitudes towards asylum seekers suggests that there is a need for further theoretical development of the justice principles. As far as we are aware, most prior work manipulating macro justice orientation experimentally has focused on inequality in wealth distribution (e.g., Norton & Ariely, 2011; Norton et al., 2014), rather than on prejudice towards social groups per se (cf., Lillie & Janoff‐Bulman, 2007, who found support for a truth and reconciliation commission when priming macro‐level processing). Thus, future research may consider individual difference factors underlying social justice orientations, and environmental factors that may moderate the ability of people to adopt a given justice orientation, and how both of these relate to prejudice. Furthermore, it is likely that some of the statements within the ATAS may be more or less related to social justice principles, and so further research that breaks down the components of attitudes could determine whether different justice principles lead to different types of prejudice (or tolerance) towards asylum seekers.

We acknowledge some limitations of the current study. Previous research suggests that age and education are related to broad constructs of prejudice and towards asylum seekers attitudes in particular (Morris & Heaven, 1986; Pedersen et al., 2005). The current sample comprised self‐selecting university students, and the age range was both relatively restricted and young in nature. As such, this may have limited the range of responses in the data. Further, there is robust evidence that education is negatively correlated with prejudice (Nekuee, 1999; Wagner & Zick, 1995). Indeed, our sample revealed relatively moderate attitudes towards asylum seekers, and therefore we may be underestimating hostility towards asylum seekers, typical of the general population (e.g., Pedersen et al., 2005 reported findings from a community sample of M = 4.66, SD = 1.51, relative to our student sample findings of M = 3.44, SD = .95). However, we would expect to find a similar pattern of results (arguably with larger effect sizes) from a sample more representative of the general population. Nevertheless, it would be beneficial to replicate our findings in the wider community. We also acknowledge that our reasoned explanations for the observed relationships are correlational only, and causal possibilities will need to be explored in further research, such as how a general macro justice orientation might lead to support for specific disadvantaged social groups.

Despite these limitations, our research is an important contribution to understanding the psychological underpinnings of attitudes towards asylum seekers. Most prejudice research looks for (and finds) predictors of negative attitudes. However, this study contributes a novel predictor of positive attitudes towards asylum seekers, grounded in support for macro justice. Importantly, our results suggest that framing an issue in terms of macro justice may influence people's positive attitudes towards asylum seekers, which may in turn make them more supportive of policies aiming to restore fairness in the treatment of disadvantaged groups, such as asylum seekers in the Australian context.

Acknowledgements

This research was conceived during the Society of Australasian Social Psychologists PhD summer school 2012. The authors would like to thank their mentors from the summer school, Kelly Fielding and Heather Smith, for their guidance, enthusiasm, and contributions during the preliminary stages of this research.

Notes

Declaration of conflicting interests: We declare no potential conflicts of interest with respect to the research, authorship, and /or publication of this article. This study was unfunded.

1. A natural log transformation was applied to the RWA scale. A single case exceeded 3 standard deviations (z = 3.59), and was replaced with $\bar{\mathrm{x}}+2\times \text{standard}\text{deviations}$(i.e., new value = 5.84; Tabachnick & Fidell, 2007). This replacement did not affect the results in any substantial manner. It is also worth reporting the restricted range in responses to the principles of micro justice scale (i.e., minimum $\bar{\mathrm{x}}=3.43$). Statistical analyses were performed on transformed data, but for the sake of interpretability the non‐transformed means and standard deviations are reported.

2. Assumptions were checked and one multivariate outlier was treated (replaced with $\bar{\mathrm{x}}+2\text{SD}$; Tabachnick & Fidell, 2007). There were some concerns regarding heteroscedastic variance in the residuals of the ATAS scale. Robust regression techniques were used to explore this violation (Wilcox, 2005, chapter 10), and further transformations of the raw data were applied (Field, 2009), but these did not affect the outcomes.