PREPRINT I know people who can and who cannot: A measure of the perception of economic inequality in everyday life Juan Diego García-Castro University of Costa Rica and University of Granada Guillermo B. Willis University of Granada Rosa Rodríguez-Bailón University of Granada CORRESPONDING AUTHOR: Juan Diego García-Castro juandiego.garcia@ucr.ac.cr University of Costa Rica and University of Granada ORCID of the author: 0000-0002-9662-6547 Number: +34 958240690 Centro de Investigación Mente Cerebro y Comportamiento (CIMCYC), Campus Cartuja S/N 18071, Granada, Spain Funding: The present research was supported by a fellowship from the UNIVERSITY OF COSTA RICA to the first author and a grant from the SPANISH MINISTRY OF ECONOMY AND COMPETITIVENESS (PSI2016-78839-P). Declarations of interest: none How to cite: García-Castro, JD., Willis, G., & Rodríguez-Bailón, R. (2018). I know people who can and who cannot: A measure of the perception of economic inequality in everyday life. The Social Science Journal, doi: 10.1016/j.soscij.2018.09.008 Highlights • A scale of Perceived Economic Inequality in Everyday Life (PEIEL) is presented • The scale is unidimensional and shows valid and reliable psychometric properties • PEIEL predicted tolerance of economic inequality over and above wage gap estimates • PEIEL had an impact on the tolerance towards inequality of right-wing individuals I know people who can and who cannot: A measure of the perception of economic inequality in everyday life Abstract This paper describes the development of the Perceived Economic Inequality in Everyday Life (PEIEL) scale. It is written and validated in Spanish. We first carried out an exploratory study, using a sample of 205 participants (52.2% men and 47.8% women; age: M = 24.69, SD = 8.95). We then conducted a confirmatory study with a sample size of 215 individuals (43.7% men and 56.3% women; age: M = 23.83, SD = 6.46). Results showed that the PEIEL scale is a valid and reliable unidimensional instrument. This scale negatively predicted tolerance of economic inequality over and above perceived inequality measured by wage gap estimates. In addition, perceived economic inequality in everyday life was negatively associated with tolerance of inequality, particularly in individuals with right-wing political ideology. Keywords: economic inequality, income inequality, tolerance of inequality, political ideology, social class Introduction Contemporary societies are characterized by increasing economic inequality, which impacts social relationships, health, life expectancy, academic performance, and violence, among other factors (Wilkinson & Pickett, 2009; Wilkinson & Pickett, 2017). Recent studies have highlighted that objective conditions alone are not sufficient to fully understand the consequences of inequality; subjective perceptions of economic inequality are equally relevant in the analysis of the psychosocial effects of the unequal distribution of resources (Gimpelson & Treisman, 2018; Kteily, Sheehy-Skeffington & Ho, 2017). Perceptions of economic inequality are not directly observable. It is an abstract construct that has been measured by different indicators. The methods most frequently used to measure it involve asking participants questions regarding differences in wages between better- and less-paid individuals in a certain context. Other methods use diagrammatic measures in which participants are shown a chart of the distribution of resources representing the proportions of the population belonging to different social strata. Additionally, some surveys use a generic item (see Bavetta, Li Donni & Marino, 2017; Castillo, Miranda & Carrasco, 2012). These methods produce differences in the estimation of perceived inequality levels (Bavetta et al., 2017). All of the indicators of perception of inequality used so far have been poorly conceptualized and have assessed economic inequality abstractly. They have not been based on immediate experiences or close reference groups, which have great effects on people’s attitudes (Helgason & Mérola, 2017; Stephany, 2017). Moreover, and as we will next argue, some of these measures present other problems. One way of investigating perceptions of economic inequality is through diagrammatic measures (Castillo et al., 2012), which consist of graphical evaluations of economic distribution within a society. In the specific case of the measure used by Castillo et al. (2012), participants were requested to indicate which one among five diagrams best described their country. However, it is not clear if this measure can be used as a continuous or categorical variable. Another way of investigating perceptions of economic inequality is through surveys that use a generic indicator. This indicator generally consists of an item measured on a Likert scale of agreement- disagreement regarding the existence of inequalities in a society, and it is frequently incorporated into national and comparative public opinion surveys. It is arguable that such an indicator can comprehensively encompass such a complex phenomenon, such as the perception of inequality (Castillo et al., 2012). Furthermore, it is a normative inquiry that cannot be taken as an assessment of perceived actual inequality (Knell & Stix, 2017). The most widely used measure is the wage gap estimation. This measure has been used in several studies by the General Social Survey in United States and by the International Social Survey Programme (ISPP) (Pedersen & Mutz, 2018). However, when using wage gaps as an indicator of inequality perception, many people have difficulty estimating how much a person on the highest or lowest rungs of a company earns. Thus, ignorance generates a high variance in the responses, and uncertainty promotes the use of malleable and biased heuristics as a function of arbitrary context information (Knell & Stix, 2017; Page & Goldstein, 2016). This produces an anchoring effect wherein people are unsure about how much more or less individuals earn in a certain context, and they answer by starting randomly at one value (i.e., an “anchor”) and then amend away from this number (Pedersen & Mutz, 2018). This is especially true when the references respondents estimate are outside of their daily experience or reference groups (Liebig, Sauer & Friedhoff, 2015; Markovsky & Eriksson, 2012; Pedersen & Mutz, 2018). An additional concern with wage gap estimation is that it implies a personal evaluation of the degree of perceived inequality but does not take into consideration potential biases from inappropriate self- positioning (Engelhardt & Wagener, 2014). Most people fail to accurately designate their own location within income distribution. Scientific literature points out that incorrect self-positioning in income distribution generates a biased view and an underestimation of economic inequality (Cruces, Perez-Truglia & Tetaz, 2013; Engelhardt & Wagener, 2014). Some authors argue in favor of developing different measures of economic inequality perceptions that capture the effects of inequality in the nearby reference groups (García-Sánchez et al., 2018; Grundler & Kollner, 2017; Nair, 2018; Stephany, 2017). Our goal in the present research was to concretely explore perceived economic inequality in individuals’ everyday lives based on their experience by developing a specific measure: the Perceived Economic Inequality in Everyday Life (PEIEL) scale. Individuals estimate economic inequality by assessing economic differences according to the information available to them in their everyday lives. They use the availability heuristic (Schwarz, Bless, Wanke & Winkielman, 2003), thus overestimating the presence of those who are part of their everyday experience by using them as a reference group to estimate both actual and ideal inequality (Bisgaard, Thisted & Mannemar, 2016; Schneider, 2012). These immediate social contexts may not be representative of society as a whole, and this can create differences in perceived economic inequality within the same context depending on the social groups individuals are most frequently in contact with (Dawtry, Sutton & Sibley, 2015). The PEIEL scale considers individuals’ everyday experiences, which, to our knowledge, are overlooked by all other measures. For this reason, we hypothesized that this measure would be a better predictor of many of the psychosocial effects of economic inequality than traditional abstract evaluation methods. Perceived economic inequality Perceived economic inequality is the individual perception of the way resources are distributed between the members of a society (Kim, Huh, Choi & Lee, 2017). Subjective perception of income inequality, and not just objective measures of it, have been shown to psychologically impact people’s lives (Curhan et al., 2014) and to perpetuate social inequality (Jost & Van der Torn, 2012). For example, it has recently been discovered that high-perceived inequality leads to a more individualistic self-concept (Sánchez-Rodríguez, Willis & Rodríguez-Bailón, 2017). Similarly, other studies have shown that higher perceived economic inequality inhibits the will to cooperate (Nishi, Shirado, Rand & Christakis, 2015), leads to increased perceptions of threat and negative attitudes towards minority groups (De Botton, 2005), and has negative consequences on individuals’ health (Chung & Lee, 2015; Oshio & Urakawa, 2014). Perceived economic inequality has generally been associated with social context (Hauser & Norton, 2017) and political ideology (Rodríguez-Bailón et al., 2017). Different economic contexts produce diverse images of inequality (Hadler, 2005; Oisho & Urakawa, 2014), and, despite not being the only variable at stake, experiences of economic inequality are conditioned by social class. Some scholars (Evans & Kelley, 2004; Irwin, 2018) argued that the combination of reference groups and social indicators, like income or education, form perceptions and understandings of economic inequality. For example, individuals of higher social class with higher levels of income and education tend to better discriminate economic inequality than those of lower social class. Additionally, political ideology influences perceptions of economic inequality (Anderson & Singer, 2008; Hadler, 2005; Janmaat, 2013). For example, individuals with a right-wing political ideology tend to have lower perceptions of economic inequality than left-wing individuals because, in general, they are more satisfied with the system, they prefer principles of equity over equality, they believe more strongly that success is due to individual efforts, and they have a more optimistic perception of life (Chambers, Swan & Heesacker, 2014). Previous research showed that perceptions of economic inequality can be moderated by political ideology (Willis, Rodríguez-Bailón, López-Rodríguez & García-Sánchez, 2015), and people with more liberal, or leftist, beliefs have greater perceptions of inequality (Anderson & Singer, 2008; Hadler, 2005; Franko, 2017). However, to date, such a hypothesis has been tested using abstract measures of perceived inequality. Another effect of perceived economic inequality that has been explored is tolerance of inequality, that is, the level of economic inequality that individuals are ready to accept or consider reasonable (Gonthier, 2017; Schröder, 2017). The fact that some people perceive large economic differences does not imply that they consider them unfair or unethical (Janmaat, 2013). The importance of tolerance of inequality is based on its relationship with the type of society individuals wish to live in, their attitude regarding redistribution policies, and their level of involvement in efforts to reduce inequality (Shariff, Wiwad & Aknin, 2016). However, existing literature on this topic does not show conclusive results on perceived economic inequality and tolerance of it. For example, the variance of tolerance towards inequality predicted by the wage gap estimation, one of the abstract measures of economic inequality, is limited and generally negative (Castillo et al., 2012; Knell & Stix, 2017; Shariff et al., 2016). Furthermore, tolerance towards inequality decreases with drops in socioeconomic status. For instance, people with lower socioeconomic status believe income differences in their society should be reduced (Hadler, 2005; Knell & Stix, 2017). Considering this, the aim of the present research was to provide a valid and reliable instrument to measure perceived economic inequality in a more direct and experiential way than with those developed so far in the hope of finding more robust effects of perceived economic inequality. We expected the PEIEL scale to negatively predict tolerance of inequality better than perceived inequality assessed with the abstract measure of wage gap estimation. Overview of the current research Following DeVellis (2017), we first conceptually defined the construct “perceived economic inequality in everyday life.” Next, the definition was assessed by a panel of five experts—researchers in the field of social psychology—who gave their opinion on its conceptualization and operationalization. After that, we developed individual items aimed at the Spanish general population in a unidimensional instrument. On the basis of the experts’ opinions, 17 items were constructed and, in turn, presented to another five experts to assess their representativeness, intelligibility, ambiguity, and conciseness. Following this assessment, certain changes were made to the items, which were then reassessed and measured with a Content Validity Index (CVI). The CVI is the grade to which a scale has a proper sample of items for the concept being measured. As a general principle, CVI values should be at least >.70 (Delgado-Rico, Carretero-Dios & Ruch, 2012). The second round of item assessments revealed a global CVI >.70, which indicates an acceptable agreement among experts for these criteria. These items were used in Study 1; the purpose of Study 2 was to confirm the findings of Study 1. Study 1 In this study we conducted descriptive analyses of the items proposed to measure perceived economic inequality in everyday life, and we also explored their factor structure. To this end, the items were subjected to an exploratory factor analysis and an internal consistency analysis. Participants and procedure. A total of 205 individuals took part in the study: 107 (52.2%) men and 98 (47.8%) women. Their ages ranged between 18 and 60 years (M = 24.69, SD = 8.95), and 80.9% were college students or already held a university degree. Their participation was voluntary, no compensation was involved, and the anonymity of their responses was guaranteed in an informed consent form. Data collection was carried out at the bus terminal in Granada, Spain, using a community convenience sample. The lead researcher asked for the voluntary participation of the people present in the waiting room. Participants required an average time of 15 minutes to complete the questionnaire. Instruments. The questionnaire, which was self-administered, included sociodemographic questions such as participant’s gender, age, and level of education, as well as the PEIEL 17-item scale. The scale had a 7- point Likert response format ranging from 1 (completely disagree) to 7 (completely agree). Examples of the items presented included “Conozco a personas con niveles de ingresos muy diferentes” (“I know people with very different levels of income”) and “Entre las personas que conozco algunas cuentan con viviendas más grandes y lujosas que otras” (“Among the people I know, some have bigger and more luxurious homes than others”). Participants were also asked about their net monthly family income on a 10-option scale ranging from 1 (under 650 euros) to 10 (over 5,800 euros) (M = 4.1, SD = 2.0). They also indicated their mother’s level of education on a 5-option scale ranging from 1 (no education) to 5 (university degree) (M = 3.2, SD = 1.3) and their father’s level of education (M = 3.3, SD = 1.2). Data analyses. We conducted descriptive analyses of each item and assessed their discrimination index with the corrected item-total correlation method. The item-total correlation method determines if an item’s performance relates to the other items within the scale. We also performed an Exploratory Factor Analysis (EFA) using the principal axis method to reveal the internal structure of the scale. In the EFA, we used the Predictive Mean Matching (PMM) to deal with missing data. A regression model was used to identify possible donor cases based on variables with full data that have similar covariate values, in this case the other items of the scale. After potential donor cases are identified, a potential donor case is randomly selected, and PMM directly assigns the donor case’s value. Predictive mean matching is the method recommended in EFA analysis because it estimates factor loadings without bias (McNeish, 2017). We also used Bartlett’s test of sphericity to determine if the items were related to each other and to find a suitable structure detection. The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy was used to evaluate the proportion of variance in the items that could be influenced by causal factors (DeVellis, 2017). Finally, Horn’s parallel analysis (1965) was used to create a set of data simulating a study like ours to determine the number of factors to retain and compare this result with the output of our EFA. Results and discussion Item analysis. The discrimination index (corrected item-total correlation) was greater than .40 in most of the items, with the exception of three. Standard deviations were higher than 1 in all cases. Average item scores ranged between 3.14 (SD = 1.44) and 5.74 (SD = 1.50). See the supplemental materials for a full analysis (osf.io/c68bg). Exploratory factor analysis and internal consistency. Bartlett’s test of sphericity, χ2 (136) = 1179.744; p < .001, and a KMO index value of .86 supported the validity of the correlation matrix to conduct the analysis. Horn’s parallel analysis (1965) created 100 sets of random data simulating (n = 205, 17 variables) with the Watkins program (2000). A comparison of these results with those of a principal component analysis of the 17 items of the PEIEL scale with oblimin rotation revealed a two-factor solution. The first factor accounted for 34.3% of the variance, and the second accounted for 8.6%. Of the 17 items, 12 (> .50) were loaded on the first factor and two items were loaded both on the first and second factors. Three items did not show high loadings on either of these factors. A content analysis of the two items that loaded on the second factor showed that they had no theoretical relationship with one another. Examples of the deleted items are “En mi trabajo hay personas que ganan mucho más que otras” (“Some of my work colleagues earn much more than others”) and “Conozco a personas que se pueden permitir económicamente aficiones que otros/as de mis conocidos/as no pueden permitirse” (“Some of my acquaintances can afford hobbies that others can’t”). We decided to remove this second factor and propose a unifactorial scale to measure PEIEL comprising the 12 items that loaded exclusively on the first factor. In this new version of the 12-item scale, all items had factor loadings > .50, and the variance explained was 42.6%. Finally, a reliability analysis of the 12 selected items showed adequate values, α = .87. The reliability of the scale did not increase by eliminating any of the items. The result of using the split-half Spearman- Brown formula was .86. The total mean of the 12 sample items was 5.08 (SD = 1.06). The results produced a set of 12 items assessing perceived economic inequality in everyday life that conformed to a unifactorial structure. The purpose of Study 2 was to replicate the results of Study 1 by conducting a confirmatory factor analysis on an independent sample. Study 2 The main goals of Study 2 were: a) to confirm the findings of Study 1 by conducting a confirmatory factor analysis using the 12 items suggested by the first study; and b) to explore the convergent validity evidence of the PEIEL scale. Specifically, we intended to explore the effect of PEIEL on tolerance of inequality compared to the abstract perception of inequality measured through wage gap estimation. We hypothesized that PEIEL would negatively predict tolerance of inequality better than perceived inequality measured by wage gap estimates. In this study, we also conducted exploratory analyses to determine whether social class or political ideology could moderate the predicted relationship between PEIEL and tolerance of inequality. As stated before, perceptions of economy inequality are influenced by social context (Hauser & Norton, 2017) and political ideology (Rodríguez-Bailón et al., 2017). Other studies have found that ideology moderates perceptions of economic inequality, suggesting a motivational basis of this relation (Willis et al., 2015). Participants and procedure. A total of 215 individuals took part in the study: 121 (56.3%) women and 94 (43.7%) men. Their ages ranged between 18 and 54 years (M = 23.83, SD = 6.46), and 76.8% of them were college students or already held a university degree. All participants provided informed consent. As in Study 1, the data were collected at the bus terminal in Granada using a community convenience sample. People were asked for their voluntary participation; those who accepted were given assurances of confidentiality and anonymity. The average time required to complete the questionnaire was 15 minutes. Instruments. We used the 12-item PEIEL scale (annex 1, M = 5.4, SD = 1.0, α = .88). Sociodemographic data: We collected data on participants’ age (M = 23.8, SD = 6.4) and gender (female = 56.3%, male = 43.7%), as well as their mother’s level of education on a 5-option scale ranging from 1 (no education) to 5 (university degree) (M = 3.3, SD = 1.2) and their father’s level of education (M = 3.4, SD = 1.3). Tolerance of inequality: The following statement was used and taken from the surveys carried out by the ISPP and Kelly and Evans (2008) (see Castillo 2011; Schröder 2017): “Las diferencias de ingreso en España son demasiado grandes” (“Income differences in Spain are too large”). Response options ranged from 1 (totally disagree) to 7 (totally agree). Score (M = 5.8, SD = 1.2). Due to the way the data were presented, higher scores showed less tolerance of inequality. Social class: In line with previous studies (Navarro-Carillo, Valor-Segura & Moya, 2018), social class measurement was computed using an index including the level of formal education achieved and the income. Net monthly income of all the family members was measured using a 10-point scale ranging from 1 (under 650 euros) to 10 (over 5,800 euros) (M = 3.9, SD = 2.1). Next, we divided net income by the number of family members who depend on it (M = 3.2, SD = 1.3). The formal educational level achieved by participants was measured using an 8-point scale ranging from 1 (primary education) to 8 (doctorate) (M = 5.1, SD = 1.2). We finally standardized both measures and computed their sum. Political ideology: Political beliefs were measured on a scale ranging from 1 (left-wing) to 7 (right-wing) (M = 3.3, SD = 1.4) (see Piurko, Schwartz & Davidov, 2011). Actual wage gap: Participants’ perception of the average monthly salary of the highest-ranking person in a typical Spanish company, as opposed to that of one of the least-skilled employees in the same company, was estimated as the logarithmic ratio between these two magnitudes: ln (perceived high-status salary/perceived low status salary) (see Castillo 2011; Willis et al., 2015) (M = 2.2, SD = 1.7). Ideal wage gap: The ideal wage gap is the perception of the wage gap that participants believe should exist in their ideal society. The specific question evaluates the average monthly salary that the highest-ranking person in a typical Spanish company should ideally receive as opposed to what an unskilled employee in the same company should ideally receive. It is estimated as the logarithmic ratio between these two magnitudes: ln (perceived ideal high-status salary/perceived ideal low status salary) (see Castillo 2011; Willis et al., 2015) (M = 1.1, SD = .9). Data analyses. We conducted descriptive analyses of each item and assessed its discrimination index with the corrected item-total correlation method. We also performed a confirmatory factor analysis to reveal the internal structure of the scale. We used the Robust Maximum Likelihood (MLR) estimation, as both the non-independence of the observations and the non-normality of the data were taken into consideration (Kaplan, 2000). The moderation analyses were performed using Model 1 of the PROCESS Macro for SPSS (Hayes, 2013). Values for the moderator were the mean and plus/minus one standard deviation from the mean. Level of confidence for the confidence intervals was 95%. Resampled bootstrapping consisted of 5000 iterations. A listwise deletion method was used in most of the analyses (except in correlations where we used pairwise), but results not change using other methods (e.g., PPM, replace for mean, pairwise) of handling missing data. Results and discussion Item analysis. All items showed adequate results in the discrimination index (corrected item-total correlation) (≥ .49). All standard deviations were higher than 1. Participants’ mean score on the scale was 5.40 (SD = 1.06). Results are summarized in Table 1. [PLEASE INSERT TABLE 1] Confirmatory factor analysis. A unifactorial model including 12 items was specified. All items had factor loadings > .50. The variance explained by all the items was 44.3%. The results revealed an acceptable fit of the unidimensional model (χ2 = 114.935, df = 54; p < .001, RMSEA = .07, SRMR = .05, CFI = .90). Results of this analysis are shown in Figure 1. [PLEASE INSERT FIGURE 1] Reliability. Internal consistency, as measured with Cronbach’s alpha index, was adequate (α = .88). The reliability of the scale did not increase by eliminating any of the items. The result of applying the split-half Spearman-Brown formula was .86. Relationship between the PEIEL scale and other variables. A correlation analysis was carried out between the PEIEL scale and all other variables included in this study. The PEIEL scale was found to correlate with social class, tolerance of inequality, and political ideology. However, no correlations were found with actual wage gap, ideal wage gap, gender, or age. [PLEASE INSERT TABLE 2] PEIEL and tolerance of inequality Subsequently, a hierarchical regression analysis was performed to predict tolerance of inequality. Gender, age, social class, and political ideology were included in Step 1 of the analysis; actual and ideal wage gap were included in Step 2; and PEIEL was included in Step 3. [PLEASE INSERT TABLE 3] Results showed that PEIEL was a better predictor of tolerance of inequality than actual and ideal wage gap, social class, and political ideology. In fact, in Step 3 the only significant predictors of tolerance of inequality were age, ideal wage gap, political ideology, and PEIEL. Exploratory moderation analyses. We also performed two moderation analyses to explore whether the relationship between PEIEL and tolerance of inequality differed depending on social class and political ideology. In both analyses, PEIEL was always the predictor variable and tolerance of inequality was the criterion variable; the moderator variables were modified between analyses. Results did not show any significant interaction between social class and PEIEL (b = .00 (.00); t = 1.7, p >. 05, CI 95% = -.0012, .0166). By contrast, an interaction effect was observed between PEIEL and political ideology (b = .01 (.00); t = 2.72, p < .01, CI 95% = .0044, .0277). [PLEASE INSERT TABLE 4] Specifically, as shown in Figure 2, no relationship was found between PEIEL and tolerance of inequality in individuals of a left-wing ideology (-1 SD) (b = .00 (.01), t = .56, p > .05, CI 95% = -.0170, .0307). However, in individuals with a center (b = .03 (.00); t = 4.1, p < .001, CI 95% =.0159, .0446) or center- right (+ 1 SD1) (b = .05 (.01); t = 5.16, p < .001, CI 95% = .0332, .0741) political ideology, a positive relationship was observed between PEIEL and tolerance of economic inequality. [PLEASE INSERT FIGURE 2] Results suggest that the 12 items of the PEIEL scale obtained in Study 1 fit well into a unidimensional structure. The scale showed an acceptable convergent validity, correlating with social class, tolerance of inequality, and political ideology. In addition, it showed adequate discriminant validity compared to actual wage gap and ideal wage gap. Moreover, PEIEL was found to predict tolerance of inequality over and above the actual wage gap. Only PEIEL, the ideal wage gap, political ideology, and age predicted tolerance of inequality. From an exploratory perspective, a higher everyday perception of inequality was found to lead to a lower tolerance of inequality, especially in individuals from center and center-right wing. General discussion Results of Studies 1 and 2 revealed that the PEIEL scale is a valid and reliable measure of perceived economic inequality in the everyday lives of the general population. This measure was also found to better predict tolerance of inequality than perceived inequality assessed through the abstract estimation of wage gaps, one of the measures most frequently used in research on this topic (Castillo et al., 2012; Page & Goldstein, 2016). Moreover, daily exposure to perceived economic inequality predicted a lower tolerance of inequality in individuals with a self-reported center or center-right political ideology. The development of a new instrument to measure perceived economic inequality can help to gain deeper insight into the psychosocial consequences of inequality. The main drawback of some of the abstract measures of economic inequality currently in use may be that individuals do not have access to enough information on the distribution of resources across society, and this lack of knowledge leads to inadequate responses. It is difficult for these respondents to estimate economic differences, so they do it randomly (Grundler & Kollner, 2017; Stephany, 2017). However, people do not need to have accurate information about economic inequality to have attitudes about it. Most people do not actively seek information about economic conditions in their society, but they 1 Individuals with a score +1 SD above the mean are positioned at 4.7 points on the ideological spectrum; this justifies their classification as being of center-right ideology. are exposed daily to all kinds of signals, especially through the people they interact with (Franko, 2017). The most important reference groups for people are those with which they interact the most (Clark & Senik, 2010). Individuals get information about economic distribution from their reference group and generalize it to the whole society (Cruces et al., 2013; Irwin, 2018), and they give greater weight to more immediate experiences than distant ones (Evans & Kelley, 2004, Irwin, 2018). Passively and routinely, this information influences the judgment of reality through unconscious and automatic processes. Information that is more accessible in memory is more likely to influence subsequent inferences (Bisgaard et al., 2016). Therefore, developing a measure of the perception of economic inequality in everyday life, such as PEIEL, allows researchers to explore perceived consequences of economic differences between acquaintances in a familiar context. Individuals estimate inequality through their everyday acquaintances; therefore, this measure involves a social sampling process that reflects their cognitive functioning within a given social structure and can generate systematic differences in perceived reality (Bisgaard et al., 2016; Dawtry et al., 2005). Social sampling has illustrated that social cognition is situated and causally interdependent with the environment (Dawtry et al., 2005). The concept of economic inequality in their country is distant and difficult for people to concretize (García-Sánchez et al., 2018; Minkoff & Lyons, 2017). Individuals have problems managing representative samples of the general population, and this is why they count on available samples from their social circles (Galesic, Olsson & Rieskamp, 2012). Abstract measures of perceptions of economic inequality do not take these problems into account. The PEIEL provides a cognitive anchor by inquiring about everyday reality, therefore it is likely to better predict certain psychological effects of economic inequality than abstract measuring instruments. Indeed, the results of the studies presented here show that the PEIEL scale is a better predictor of tolerance of economic inequality than the estimation of wage gaps. It has been suggested that individuals’ tolerance of inequality is associated with not experiencing such inequality in everyday life. Individuals make social comparisons with the most salient people that they know and tend to tolerate inequality less or more depending on whether they perceive it more or less in their everyday life (Kelly & Moore-Clingenpeel, 2012; Xu & Garand, 2010). A higher exposure to perceived economic inequality and its negative consequences on society and immediate social circles can trigger an intention to reduce it (Kelly & Moore-Clingenpeel, 2012). In addition, when the wealth of people with greater resources is evident, people tend to tolerate less inequality and show more positive attitudes towards redistribution (Hauser, Kraft-Todd, Rand, & Nowak, 2016). Furthermore, people who make more economic comparisons support more redistributive policies (Clark & Senik, 2010). This may explain why the effect of PEIEL on tolerance of inequality is higher than that of abstract measures of perceived economic inequality. This relationship is even higher in individuals who define themselves as center-right wing on the political spectrum. Generally, these individuals have lower perceptions of inequality than those who define themselves as left-wing in their political ideology (Castillo et al., 2012; Kim et al., 2017) and consider these social differences to be inevitable (Jost, Glaser, Kruglanski & Sulloway, 2003). According to the results of Study 2, the effect of PEIEL on tolerance of inequality in these individuals is even greater. Left-wing individuals tend to consider economic inequality as unfair because they do not believe that economic success is always due to personal effort or, in general, to individual factors that can be controlled (Anderson & Singer, 2008; Bavetta et al., 2017; Hadler, 2005). Accordingly, for left-wing individuals, the PEIEL scale does not predict tolerance of inequality in the same way, probably because these individuals are already predisposed to reject it. By contrast, previous research has revealed that conservative individuals are more easily influenced by situational and salient elements within their environment due to their higher need of cognitive closure (Critcher, Huber, Ho & Koleva, 2009). Whenever a situation in their everyday life generates ambiguity or uncertainty, as might happen with the negative effects of economic inequality on people close to them, they are more likely to change their opinion on inequality to satisfy this need for cognitive balance. Moreover, the conclusion that social class does not have a moderating effect on the relationship between the PEIEL and tolerance of inequality, as opposed to political ideology, further reinforces the importance of considering these ideological variables when assessing the psychosocial impacts of inequality. The practical implications of having a measuring instrument, such as the PEIEL scale, open up the possibility of exploring the effects of perceived economic inequality on the psychosocial consequences of such inequality. For example, it could be used to reduce tolerance of economic inequality by presenting its negative effects on people’s everyday lives. It can also provide a basis for the development of more accurate theoretical models, not only of tolerance of inequality but also of other psychosocial effects of perceived inequality. The main limitation of this research is the specific context in which it was carried out. This can be solved by replicating it and adapting the measure to other circumstances and cultural contexts. Future studies should further explore the possible moderating role of perceived economic inequality in everyday life regarding the psychosocial processes associated with inequality (e.g., the support of redistribution policies or status-related anxiety). It would also be useful to conduct experimental manipulations to test causal relationships between PEIEL and its possible psychosocial effects. The aim of this research was to improve and expand the existing knowledge on the effects of economic inequality in everyday life, given that economic inequality is one of the main defining characteristics of current societies. Our hope is that the results can be useful in the long term to develop social programs and policies aimed at reducing inequality and its associated effects. References Anderson, C., & Singer, M. (2008). The Sensitive Left and the Impervious Right. Multilevel Models and the Politics of Inequality, Ideology, and Legitimacy in Europe. Comparative Political Studies, 41(4/5), 564-599. Bavetta, S., Li Donni, P., & Marino, M. (2017). An empirical analysis of the determinants of perceived inequality. Review of Income and Wealth, https://doi.org/10.1111/roiw.12351 Bisgaard, M., Thisted, P., & Mannemar, K. (2016). Reconsidering the Neighborhood effect: does exposure to residential unemployment influence voters’ perceptions of the national economy? The Journal of Politics, 78(3), 719-732. Castillo, J. C. (2011). Legitimacy of inequality in a highly unequal context: Evidence from the Chilean Case. Social Justice Research, 24, 314-340. Castillo, J.C., Miranda, D., & Carrasco, D. (2012). Percepción de Desigualdad Económica en Chile: Medición, Diferencias y Determinantes. Psyke, 21(1), 99-114. Chambers, J., Swan, L., & Heesacker, M. (2014). Better off than we know: distorted perceptions of incomes and income inequality in America. Psychological Science, 25(2), 613-618. Chung, S., & Lee, S. (2015). Poverty and perceived income inequality and changes in growth trajectory of problem drinking. Korean Journal of Health Education and Promotion, 32(5), 43-51. Clark, A., & Senik, C. (2010). Who compares to whom? The anatomy of income comparison in Europe. The Economic Journal, 120, 573-594. Critcher, C., Huber, M., Ho, A., & Koleva, S. (2009). Political Orientation and Ideological Inconsistencies: (Dis)comfort with values Tradeoffs. Social Justice Research, 22, 181-205. Cruces, G., Perez-Truglia, R., & Tetaz, M. (2013). Biased perceptions of income distribution and preferences for redistribution: evidence from a survey experiment. Journal of Public Economics, 98, 100- 112. Curhan, K., Levine, C., Rose, H., Kitayama, S, Park, J., Karasawa, M., Kawakami, N., Love, G., Coe, C., Miyamoto, Y., & Ryff, D. (2014). Subjective and Objective Hierarchies and Their Relations to Psychological Well-Being: A U.S./Japan Comparison. Social Psychological and Personality Science, 5 (8), 855-864. Dawtry, R., Sutton, R., & Sibley, C. (2015). Why wealthier people think people are wealthier, and why it matters: from social sampling to attitudes to redistribution. Psychological Science, 26(9), 1389-1400. De Botton, A. (2005). Status anxiety. Nueva York: Vintage. Delgado-Rico, E., Carretero-Dios, H., & Ruch, W. (2012). Content validity evidences in test development: an applied perspective. International Journal of Clinical and Health Psychology, 12(3), 449-460. DeVellis, R. F. (2017). Scale development: Theory and application (Fourth Edition). Los Angeles: Sage. Engelhardt, C., & Wagener, A. (2014). Biased Perceptions of Income Inequality and Redistribution. CESifo Working Paper Series No. 4838. https://www.wipol.unihannover.de/fileadmin/sopo/pdf/engelhardt/cesifo1_wp4838.pdf Evans, M., & Kelley, J. (2004). Subjective social location: Data from 21 nations. International Journal of Public Opinion Research, 16(1), 3–38. Franko, W. (2017). Understanding public perceptions of growing economic inequality. State Politics & Policy Quarterly, 17(3), 319-348. Galesic, M., Olsson, H., & Rieskamp, J. (2012). Social Sampling Explains Apparent Biases in Judgments of Social Environments. Psychological Science, 23(12), 1515-1523. García-Sánchez, E., Willis, G., Rodríguez-Bailón, R., García-Castro, J.D., Palacio-Sañudo, J., Polo, J., & Rentería, E. (2018). Perceptions of economic inequality in Colombian daily life: More than unequal distribution of economic resources. Frontiers in Psychology, 9, https://doi.org/10.3389/fpsyg.2018.01660 Gimpelson, V., & Treisman, D. (2018). Misperceiving Inequality. Economics & Politics, 30(1), 27-54. Gonthier, F. (2017). Baby Boomers Still in the Driver’s Seat? How Generational Renewal Shapes the Dynamics of Tolerance for Income Inequality. International Journal of Sociology, 47(1), 26-42. Grundler, K., & Kollner, S. (2017). Determinants of governmental redistribution: income distribution, development levels, and the role of perceptions. Journal of Comparative Economics, 45, 930-962. Hadler, M. (2005). Why do people accept different income ratios? A multi-level comparison of thirty countries. Acta Sociologica, 48(2), 131-154. Hauser, O., Kraft-Todd, G., Rand, D., & Nowak, M. (2016). Invisible Inequality Leads to Punishing the Poor and Rewarding the Rich. Academy of Management Proceedings, 1, 13841. Hauser, O., & Norton, M. (2017). (Mis)perceptions of inequality. Current Opinion in Psychology, 18, 21- 25. Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York: Guilford Press. Helgason, A., & Mérola, V. (2017). Employment insecurity, incumbent partisanship, and voting behavior in comparative perspective. Comparative Political Studies, 50(11), 1489-1523. Horn, J.L. (1965). A rationale and technique for estimating the number of factors in factor analysis. Psychometrika, 30, 179-185. Irwin, S. (2018). Lay perceptions of inequality and social structure. Sociology, 52(2), 211-227. Janmaat, J. (2013). Subjective inequality: a review of international comparative studies on people’s views about inequality. European Journal of Sociology, 54(3), 357-389. Jost, J. T., Glaser, J., Kruglanski, A., & Sulloway, F. (2003). Political conservatism as motivated social cognition. Psychological Bulletin, 129, 339-375. Jost, J. T., & Van der Toorn, J. (2012). System justification theory. In P. A. M. van Lange, A. W. Kruglanski., & E. T. Higgins (Eds.), Handbook of theories of social psychology (Vol. 2, pp. 313-343). London: Sage. Kaplan, D. (2000). Structural Equation Modeling: Foundations and Extensions. Thousand Oakes, CA: SAGE. Kelly, J., & Evans, M. (2008). Economic Development Reduces Tolerance for Inequality: A Comparative Analysis of 30 Nations. In Max Haller et al. (Eds.), Charting the Globe: The International Social Survey Programme 1984-2009 (pp.72-90). Routledge: New York. Kelly, S., & Moore-Clingenpeel, M. (2012). Contextual Effects on Attitudes toward Inequality. Population Review, 51(2), 127-150. Kim, H., Huh, S., Choi, S., & Lee, Y. (2017). Perceptions of inequality and attitudes towards redistribution in four East Asian welfare states. International Journal of Social Welfare, 00, 1-12. Knell, M., & Stix, H. (2017). Perceptions of Inequality. Oesterreichische Nationalbank Working paper No 216. https://www.oenb.at/en/Publications/Economics/Working-Papers.html Kteily, N., Sheehy-Skeffington., & Ho, A. (2017). Hierarchy in the Eye of the Beholder: (Anti-) Egalitarianism Shapes Perceived Levels of Social Inequality. Journal of Personality and Social Psychology, 112(1), 136-159. Liebig, S., Sauer, C., & Friendhoff, S. (2015). Using Factorial Surveys to Study Justice Perceptions: Five Methodological Problems of Attitudinal Justice Research. Social Justice Research, 28, 415-434. Markovsky, B., & Eriksson, K. (2012). Comparing direct and indirect measures of just rewards. Sociological methods & research, 41(1), 199-216. McNeish, D. (2017). Exploratory factor analysis with small samples and missing data. Journal of Personality Assessment, 99(6), 637-652. Minkoff, S., & Lyons, J. (2017). Living With Inequality: Neighborhood Income Diversity and Perceptions of the Income Gap. American Politics Research, 00(0), 1-33. Nair, G. (2018). Misperceptions of relative affluence and support for international redistribution. The Journal of Politics, 80(3), 815-830. Navarro-Carillo, G., Valor-Segura, I., & Moya, M. (2018). Do you trust strangers, close acquaintance, and members of your ingroup? Differences in trust based on social in Spain. Social Indicators Research, 135(2), 585-597. Nishi, A., Shirado, H., Rand, D., & Christakis, N. (2015). Inequality and visibility of wealth in experimental social networks. Nature, 526, 426-429. Oshio, T., & Urakawa, K. (2014). The Association Between Perceived Income Inequality and Subjective Well-being: Evidence from a Social Survey in Japan. Social Indicators Research, 116(3), 755-770. Page, L., & Goldstein, D. (2016). Subjective beliefs about the income distribution and preferences for redistribution. Social Choice and Welfare, 47(1), 25-61. Pedersen, R., & Mutz, D. (2018). Attitudes toward economic inequality: the illusory agreement. Political Science Research and Methods, 1-17. Piurko, Y., Schwartz, S., & Davidov, E. (2011). Basic Personal Values and the Meaning of Left-Right Political Orientations in 20 Countries. Political Psychology, 32(4), 537-561. Rodríguez-Bailón, R., Bratanova, B., Willis, G., López-Rodríguez, L., Sturrock, A., & Loughnan, S. (2017). Social Class and Ideologies of Inequality: how they uphold unequal societies. Journal of Social Issues, 73(1), 99-116. Sánchez-Rodríguez, A., Willis, G., & Rodríguez-Bailón, R. (2017). Economic and social distance: Perceived income inequality negatively predicts an interdependent self-construal. International Journal of Psychology, http://dx.doi.org/10.1002/ijop.12437 Schneider, S. (2012). Income inequality and its consequences for life satisfaction: what role do social cognitions play? Social Indicators Research, 106, 419-438. Schröder, M. (2017). Is Income Inequality Related to Tolerance for Inequality? Social Justice Research, 30, 23-47. Schwarz, N., Bless, H., Wanke, M., & Winkielman, P. (2003). Accessibility Revisited. In G. V. Bodenhausen., & A. J. Lambert (Eds.), Foundations of social cognition: Afestschrift in honor of Robert S. Wyer, Jr (pp. 51-77). Hillsdale, NJ: Lawrence Erlbaum Associates. Shariff, A., Wiwad, D., & Aknin, L. (2016). Income Mobility Breeds Tolerance for Income Inequality: Cross-National and Experimental Evidence. Perspectives on Psychological Science, 11(13), 373-380. Stephany, F. (2017). Who are your Joneses? Socio-specific income inequality and trust. Social Indicators Research, 134(3), 877-898. Watkins, M.W. (2000). Monte Carlo PCA for parallel analysis [computer software]. State College, PA: Ed & Psych Associates. Wilkinson, R., & Pickett, K. (2009). The Spirit Level: Why More Equal Societies Almost Always Do Better. London: Allen Lane. Wilkinson, R., & Pickett, K. (2017). The enemy between us: The psychological and social costs of inequality. European Journal of Social Psychology, 47, 11-24. Willis, G., Rodríguez-Bailón, R., López-Rodríguez., & García-Sánchez, E. (2015). Legitimacy moderates the relation between perceived and ideal economic inequalities. Social Justice Research, 28, 493-508. Xu, P., & Garand, J. (2010). Economic Context and Americans’ Perceptions of Income Inequality. Social Science Quarterly, 91(5), 1220-1241. Annex 1. PEIEL Scale* 1. Conozco a personas con niveles de ingresos muy diferentes (I know people with very different levels of income) 2. Entre la gente de la que me rodeo, hay algunas personas que pueden permitirse económicamente acceder a mejores servicios sanitarios que otras (Among the people I surround myself with, there are some people who can afford access to a better health service than others) 3. Entre las personas de las que me rodeo, hay quiénes pueden irse de vacaciones al menos una semana al año y quiénes no van a ningún sitio porque no tienen dinero suficiente (Among the people I surround myself with, there are those who can go on vacation at least one week per year and those who cannot afford it) 4. Entre las personas que conozco algunas cuentan con viviendas más grandes y lujosas que otras (Among the people I know, some have bigger and more luxurious homes than others) 5. Entre la gente de la que me rodeo, hay algunas personas que pueden permitirse económicamente acceder a mejor educación que otras (Among the people I surround myself with, there are some people who can afford access to a better education than others) 6. Conozco tanto a personas que tienen muchos problemas para pagar su vivienda (alquileres, hipotecas) como a otras que no (I know both: people who undergo many problems to pay for their home expenses (rents, mortgages) and others who do not) 7. Conozco tanto a personas que pueden permitirse ahorrar como a otras que no llegan a final de mes (I know people who can afford to save money and others who struggle to reach the end of the month) 8. Entre las personas que conozco hay quiénes no pueden afrontar gastos imprevistos y hay quiénes los solventan sin ninguna dificultad (Among the people I know some cannot afford unforeseen expenses and others cope with them without any difficulty) 9. Entre la gente de la que me rodeo, algunas personas pueden permitirse comprar muchas más y mejores cosas que otras (Among the people I surround myself with, some people can afford to buy a lot more and better things than others) 10. En mi vida cotidiana percibo situaciones de desigualdad económica (In my everyday life I perceive situations of economic inequality) 11. Conozco tanto a personas que pueden permitirse económicamente tener su vivienda a una temperatura adecuada como a otras que no se lo pueden permitir (I know both people who can afford adequate heating at home and others who cannot afford it) 12. Conozco a personas que tienen que trabajar más que otras para poder ganar lo mismo (I know people who have to work more than others to earn the same amount of money) *The English version has not been validated. It is presented for illustration purposes. Tables Table 1. Analysis of the items in the Perception of Economic Inequality in Everyday Life (PEIEL) Scale Items N r corrected item-total M DT Asymmetry Kurtosis ITEM 1 215 .53 5.62 1.52 -1.03 .535 ITEM 2 215 .49 5.25 1.71 -.855 -.242 ITEM 3 215 .47 5.86 1.54 -1.61 1.968 ITEM 4 215 .57 5.74 1.45 -1.23 .949 ITEM 5 214 .63 5.07 1.74 -.681 -.444 ITEM 6 215 .63 5.24 1.71 -.774 -.281 ITEM 7 215 .62 5.37 1.61 -.927 .149 ITEM 8 215 .67 5.38 1.43 -.806 .482 ITEM 9 215 .71 5.75 1.34 -1.29 1.72 ITEM 10 215 .62 5.43 1.60 -.732 -.641 ITEM 11 215 .51 4.49 1.93 -.243 -1.18 ITEM 12 214 .49 5.53 1.69 -1.15 .479 Table 2. Relations between the Perception of Economic Inequality in Everyday Life (PEIEL) and other variables included in Study 2 N 1 2 3 4 5 6 7 8 1. PEIEL 213 1 .33** -.15* -.17* -.01 -.08 .02 .00 2. Inequality tolerance 215 1 -.07 -.29** -.08 -.25** .07 -.07 3. Social class 197 1 .06 .17* .20** -.08 .07 4. Political Ideology 175 1 .01 .15* -.16* -.10 5. Actual wage gap 196 1 .54* -.14* -.01 6. Ideal wage gap 191 1 -.18* -.02 7. Sex 215 1 -.21** 8. Age 209 1 **p <.01, *p<.05 Table 3. Hierarchical regression analysis to predict inequality tolerance Variable B ET Β Step 1 Sex .162 .216 .062 Age -.031 .019 -.135 Social class -.054 .071 -.063 Political Ideology -.224 .074 -.248** Step 2 Sex .018 .213 .007 Age -.041 .019 -.178* Social class -.011 .070 -.012 Political Ideology -.186 .072 -.206* Actual wage gap .095 .067 .128 Ideal wage gap -.457 .124 -.344** Step 3 Sex .027 .20 .010 Age -.044 .018 -.195* Social class .056 .068 .065 Political Ideology -.137 .069 -.151* Actual wage gap .069 .063 .093 Ideal wage gap -.424 .117 -.320*** PEIEL .033 .008 .326*** R2=.097 for step 1 (p<.01). R2=142, ∆R2=.080 for step 2 (p< .001). R2=273, ∆ R2=.096 for step 3 (p< .001). *p<.05, **p<.01, *** p<.001. Sex: 0=men, 1=women. N=151. Table 4. Moderated regression analysis to predict inequality tolerance Variable CI 95% B SE t Lower Upper Political Ideology -.2276*** .0615 -3.697 -.3491 -.1061 PEIEL .0302*** .0073 4.158 .0159 .0446 Political Ideology X PEIEL .0161** .0059 2.722 .0044 .0277 R2=.21 (p<.001). ∆R2 due to interaction=.03 (p< .01). **p<.01, *** p<.001. N=174. CI=Confidence Interval Figures Figure 1. Factor structure of the PEIEL Scale. 7 6 5 4 3 2 1 Low PEIEL Medium PEIEL High PEIEL Left Center Right center Figure 2. Moderation of political ideology in the relation between the perception of economic inequality in everyday life and inequality tolerance INEQUALITY INTOLERANCE Supplemental Materials Study 1 Table 1. Analysis of the items in the Perception of Economic Inequality in Everyday Life (PEIEL) scale, version of the study 1 with 17 items (raw data) Items N r corrected item-total M DT Asymmetry Kurtosis ITEM 1 205 .56 5.74 1.50 -1.24 .959 ITEM 2 205 .43 4.64 1.88 -.467 -.881 ITEM 3 205 .60 5.34 1.71 -.886 -.331 ITEM 4 205 .71 5.38 1.52 -.841 .087 ITEM 5 203 .53 4.74 1.72 -.414 -.647 ITEM 6 205 .50 4.57 1.86 -.260 -1.094 ITEM 7 205 .60 5.04 1.69 -.532 -.702 ITEM 8 205 .54 5.15 1.35 -.443 -.574 ITEM 9 205 .68 5.42 1.32 -.805 .436 ITEM 10 205 .61 5.23 1.59 -.702 -.310 ITEM 11 204 .55 4.22 1.91 -.005 -1.18 ITEM 12 205 .51 5.43 1.59 -.933 .067 …deleted items ITEM 13 205 .33 3.14 1.44 .491 -.246 ITEM 14 205 .43 4.40 1.76 -.108 -1.056 ITEM 15 205 .51 5.71 1.56 -1.311 1.166 ITEM 16 188 .24 4.32 1.81 -.199 -.831 ITEM 17 205 -.10 3.96 1.60 .028 -.852 Table 2. Analysis of the items in the Perception of Economic Inequality in Everyday Life (PEIEL) scale, version of the study 1 with 17 items (after PMM) Items N r corrected item-total M DT Asymmetry Kurtosis ITEM 1 205 .54 5.74 1.50 -1.24 .959 ITEM 2 205 .45 4.64 1.88 -.467 -.881 ITEM 3 205 .57 5.34 1.71 -.886 -.331 ITEM 4 205 .69 5.38 1.52 -.841 .087 ITEM 5 205 .51 4.74 1.71 -.417 -.637 ITEM 6 205 .49 4.57 1.86 -.260 -1.094 ITEM 7 205 .59 5.04 1.69 -.532 -.702 ITEM 8 205 .52 5.15 1.35 -.443 -.574 ITEM 9 205 .67 5.42 1.32 -.805 .436 ITEM 10 205 .60 5.23 1.59 -.702 -.310 ITEM 11 205 .52 4.22 1.91 -.005 -1.18 ITEM 12 205 .51 5.43 1.59 -.933 .067 …deleted items ITEM 13 205 .34 3.14 1.44 .491 -.246 ITEM 14 205 .43 4.40 1.76 -.108 -1.056 ITEM 15 205 .51 5.71 1.56 -1.311 1.166 ITEM 16 205 .23 4.35 1.80 -.299 -.791 ITEM 17 205 -.12 3.96 1.60 .028 -.852