Sage Journals: Discover world-class research

Abstract

Are judgments of the fairness of the American economy based on perceptions of economic inequality, mobility, or both? In two experiments, the authors varied information on levels of U.S. inequality and mobility, measuring effects on individuals’ judgments of economic fairness and meritocracy. Although both treatments influenced perceptions of economic fairness and meritocracy, the mobility effect was generally larger. The two treatments did not interact, countering a common claim that high social mobility legitimizes high inequality. Effects on preferences for government action to reduce inequality and increase mobility were weak or nonexistent. Additional conditions that measured, rather than manipulated, inequality and mobility perceptions showed that respondents generally perceived inequality to be very high but were more optimistic about the level of mobility. Our studies suggest that Americans’ optimism about economic mobility does more to mitigate concerns about economic fairness than does underestimation of economic inequality.

Keywords

inequality social mobility fairness meritocracy survey experiment

Income inequality (hereafter “inequality”) in the United States has risen sharply in recent decades (Piketty and Saez 2014; Saez and Zucman 2016), while intergenerational income mobility (hereafter “mobility”) has fallen (Chetty et al. 2017). Although we might reasonably expect that these trends would have led to comparable increases in concerns about the extent to which the economic system is fair and meritocratic (Kluegel and Smith 1986; McCall 2013; McCall et al. 2017), these attitudes have changed little over the years and certainly have not kept up with the level of inequality (Ashok, Kuziemko, and Washington 2015; Gallup 2018; Newport 2015).

One explanation for this seeming disconnect is that Americans are unaware of the extent of these changes, especially the rise of inequality (Alesina, Stantcheva, and Teso 2018; Hauser and Norton 2017; McCall 2013; McCall et al. 2017). If so, a key implication is that informing people of actual levels of inequality would lead to heightened concerns about economic fairness (Hauser and Norton 2017; McCall et al. 2017). But other studies suggest that Americans already see existing inequality as high and prefer lower levels of inequality (Chambers, Swan, and Heesacker 2014; Eriksson and Simpson 2012; McCall 2013; Osberg and Smeeding 2006). If so, relatively limited concern about economic fairness may be driven instead by a continuing belief in mobility (Alesina et al. 2018; Kraus and Tan 2015; though see Chambers, Swan, and Heesacker 2015; Cheng and Wen 2019).

This research seeks to answer two questions: How do beliefs about economic fairness depend on perceptions of mobility levels? And does this depend on perceptions of inequality? As detailed more fully below, the existing literature points to two main possibilities. One line of reasoning suggests that people (or at least Americans) will be relatively unperturbed by high levels of inequality if they think social mobility is also high (Day and Fiske 2017; McCall 2013; Shariff, Wiwad, and Aknin 2016; Starmans, Sheskin, and Bloom 2017). Another line of thinking suggests that Americans value mobility in its own right and will be concerned about a relatively immobile America at any level of inequality (Alesina et al. 2018; Benabou and Ok 2001).

Despite extensive research on beliefs about inequality and mobility, no study to date has experimentally varied perceptions of both facets of the American economy to examine their independent and interactive effects on judgments of the economic system. This is problematic because inequality and mobility beliefs tend to be confounded (e.g. McCall 2013; Shariff et al. 2016) and because prior work typically investigates inequality and mobility separately. Thus, we currently do not know whether attitudes are more affected by high inequality, low mobility, or the interaction of the two. Although our primary focus is on perceptions of economic fairness and meritocracy, our experiments also allow us to test for potential effects on preferences for government policies to redress inequality and immobility.

In this article, we report the results of two experiments designed to manipulate participants’ perceptions of inequality and mobility. As we describe in detail, we used large, diverse samples of Americans, but they are not representative. Thus, we cannot draw population-level inferences. Instead, our focus is solely on the effects of our manipulations on our key dependent measures. We find that inequality and mobility perceptions have independent effects on beliefs about meritocracy and economic fairness, though perceived mobility tends to have larger effects than perceived inequality. Furthermore, to provide context for our finding that a mobility manipulation matters more than an inequality manipulation, we added an ancillary condition to our studies that measured, rather than manipulated, perceptions of inequality and mobility. In these conditions, participants perceived very high levels of inequality and more moderate levels of mobility. Thus, our results suggest that informing people of the high level of inequality that exists in the United States may be telling them what they already know, while informing them of low mobility rates is more likely to contradict their preexisting beliefs.

Inequality and Mobility Perceptions

As noted earlier, a common explanation for the relatively tepid reaction to extreme inequality is that Americans believe that inequality is far lower than it actually is (Hauser and Norton 2017; Norton and Ariely 2011). In this view, if Americans simply knew the actual extent of inequality, they would become more concerned and more supportive of redistributive policies. Yet in a number of studies with American samples, participants report that inequality is higher than they want it to be (Eriksson and Simpson 2012; McCall 2013; Osberg and Smeeding 2006; Pedersen and Mutz 2018). Furthermore, though informing or reminding Americans of the high levels of inequality that exist in the United States may make them more concerned about the issue, research is more mixed on whether perceiving higher inequality leads to changes in policy preferences (Kuziemko et al. 2015; McCall et al. 2017). For example, McCall et al. (2017) gave participants a treatment about the level of inequality. Across several experiments, they found that participants’ beliefs in the importance of structural factors for success increased, while their beliefs in the power of individual factors decreased (d ≈ .50, authors’ calculation). In contrast, they found a smaller effect on the belief that government should reduce inequality (d ≈ .25, authors’ calculation).

Another possible explanation for the disconnect between high inequality and increased concern is that Americans are intolerant of inequality if they believe that citizens lack opportunities to move up in society but are more willing to accept high inequality if they think people have the ability to change their station through hard work (Alesina and La Ferrara 2005; Alesina et al. 2018; Benabou and Ok 2001; McCall 2013; Shariff et al. 2016). Importantly, as inequality rises, attitudes about merit and mobility may change to accommodate the new, more unequal, status quo (Mijs forthcoming; Schröder 2017; Solt et al. 2016). Belief in mobility through merit allows people to head off concerns about fairness by attributing inequality to personal characteristics such as drive and competence, rather than structural factors such as extreme inequality (Della Fave 1980; Heiserman and Simpson 2017; Hunt and Bullock 2016; Kluegel and Smith 1986; Lepianka, van Oorschot, and Gelissen 2009; though see Davidai 2018; McCall et al. 2017).

American society, in particular, is also awash in cultural narratives about upward mobility and the “American dream” (Kluegel and Smith 1986; McCall 2013). Because Americans tend to derive their beliefs about economic conditions from their social surroundings (Dawtry, Sutton, and Sibley 2015), such pervasive “rags to riches” stories help bolster the belief that anyone can rise up the economic ladder and that inequality is thus a result of individual achievement in a highly mobile system (McCoy and Major 2007).

Indeed, although social mobility has decreased in recent decades (Chetty et al. 2017), Americans continue to believe that mobility is relatively high. Most studies find that Americans overestimate the chances that people on the bottom rungs of the socioeconomic ladder are able to climb it (Alesina et al. 2018; Davidai and Gilovich 2015; Kraus and Tan 2015; though see Chambers et al. 2015; Cheng and Wen 2019). Similarly, Americans tend to think that the United States ranks much higher internationally in mobility than the objective evidence shows (Davidai and Gilovich 2018).

It follows from the above that perceptions of inequality and mobility are linked in such a way that changes in one will likely be confounded with changes in the other. If higher perceived inequality can increase beliefs in its own justifications (e.g., “inequality is high because the meritorious are getting what they deserve”), it is less surprising that higher perceived inequality does not lead to greater concerns about economic fairness.

To clearly identify how perceived inequality, perceived mobility, and their interaction affect outcomes such as belief in meritocracy and policy attitudes, we draw on recent studies that experimentally manipulate beliefs about inequality or mobility (Côté, House, and Willer 2015; Davidai 2018; Day and Fiske 2017; McCall et al. 2017; Shariff et al. 2016). Unlike this prior work, however, our experiments orthogonally manipulate inequality and mobility. This is particularly important given that beliefs about inequality and mobility perceptions are intertwined. Thus we cannot manipulate a single factor (e.g., perceived inequality) and draw inferences only from that factor, because its manipulation will also likely affect the other. To allow clear causal attributions, the experiments we outline below manipulate both perceptions of inequality and mobility to measure their independent and interactive effects.

Hypotheses

Above, we outlined two possible ways in which perceptions of inequality and mobility will affect judgments about economic fairness and meritocracy. One line of reasoning suggests that concerns about economic fairness and preferences for redistributive policies should increase when people believe inequality is high or when they believe that mobility is low (Alesina and La Ferrara 2005; Benabou and Ok 2001; Kluegel and Smith 1986; McCall 2013).

Hypothesis 1a: Net of perceived mobility, higher perceived inequality will lead to greater concerns about economic fairness and meritocracy.

Hypothesis 1b: Net of perceived inequality, lower perceived mobility will lead to greater concerns about economic fairness and meritocracy.

As noted earlier, in keeping with these “independent effects” predictions, most existing research on perceptions of inequality and mobility (and their consequences for attitudes) examined either inequality (Chambers et al. 2014; Eriksson and Simpson 2012; Kiatpongsan and Norton 2014; Kraus and Tan 2015; McCall et al. 2017; Norton and Ariely 2011; Osberg and Smeeding 2006) or mobility (Alesina et al. 2018; Chambers et al. 2015; Cheng and Wen 2019; Davidai and Gilovich 2015, 2018), but not both.

The alternative possibility is that perceptions of inequality and mobility interact. Specifically, we noted that high levels of social mobility may moderate or “soften” the effects of inequality on perceived fairness. In this view, social mobility justifies high inequality by recasting it as an innocuous by-product of an ideal meritocratic system of just deserts, wherein people are afforded similar opportunities in life, and individual outcomes result from ability and effort (Alesina and La Ferrara 2005; Bobo 1991; Davidai 2018; Day and Fiske 2017; Evans, Kelley, and Peoples 2010; Hunt and Bullock 2016; Reynolds and Xian 2014; Shariff et al. 2016; Starmans et al. 2017). From this perspective, higher perceived inequality will cause greater concerns when perceived mobility is low but will matter less when perceived mobility is higher.

Hypothesis 2: Higher perceived inequality will lead to greater concerns about economic fairness and meritocracy, and this effect will be larger when mobility is also seen as low.

In addition to measuring concerns about economic fairness, our experiments allow us to measure the impact of perceived inequality and immobility on policy preferences. Prior work suggests that changes in perceptions of the level of inequality or mobility may have inconsistent or limited effects on policy preferences (Alesina et al. 2018; Ashok et al. 2015; Kuziemko et al. 2015), in part because policy preferences are so strongly tied to political ideology (Alesina et al. 2018; Ashok et al. 2015; Chambers et al. 2014, 2015; Cech 2017; Eriksson and Simpson 2012; Pedersen and Mutz 2018). Thus, although we do not outline any hypotheses regarding policy preferences, our experiments also measured support for various government actions linked to inequality and immobility.

Overview of Studies

We conducted two Web-based experiments (experiment 1, n = 744; experiment 2, n = 981) that manipulated perceptions of inequality and mobility in the United States and then measured concerns about inequality, meritocracy, and the fairness of the American economy. We also measured preferences for government actions to address inequality and mobility. To put our findings in context, we added an ancillary condition to each study (an additional n = 179 in experiment 1 and n = 252 in experiment 2) that measured, rather than manipulated, beliefs about inequality and mobility. Much of the literature on perceptions of inequality and mobility has focused specifically on misperceptions, reasoning that if individuals underestimate inequality or overestimate mobility, they will have overly optimistic assessments of society, and lower preferences for interventionist policies than they would if they had accurate perceptions (Alesina et al. 2018; Davidai and Gilovich 2015, 2018; Hauser and Norton 2017; Kraus and Tan 2015). It is therefore important that we assess whether, at the baseline, participants in our sample tend to see inequality as high and mobility as low. Findings from the ancillary conditions provide a benchmark for assessing how participants in our samples would view inequality and mobility in the absence of manipulated feedback about them.

We recruited American participants for both studies from Amazon Mechanical Turk (MTurk) between December 2017 and March 2018 (experiment 1) and November and December 2018 (experiment 2). Although the demographics and characteristics of MTurk workers tend to differ from those of the general population (Hargittai and Shaw 2020), prior work shows that MTurk samples yield reliable, high-quality data (Buhrmester, Kwang, and Gosling 2011; Clifford, Jewell, and Waggoner 2015; Paolacci and Chandler 2014), provided appropriate precautions are taken. Most important for present purposes, effects of experimental manipulations in MTurk samples tend to parallel effects obtained from more representative samples (Coppock, Leeper, and Mullinix 2018; Weinberg, Freese, and McElhattan 2014).

Given that attitudes about inequality and opportunity vary with political orientation (Cech 2017; Eriksson and Simpson 2012; Pedersen and Mutz 2018), we wanted to be sure that our key findings are not skewed by political orientation. Thus, in both experiments we recruited similar numbers of liberals and conservatives. In experiment 2 we used an MTurk panel to quota sample equal numbers of liberals, independents, and conservatives. Experiment 2 also imposed a quota of 60 percent to 70 percent non-college-educated participants to obtain greater representation of lower educated groups. Full demographics for both studies are given in Table 1.

Table 1.

Sample Demographics for All Studies.

	Experiment 1	Experiment 2	Survey A1	Survey A2
Median age (years)	34.0	47.0	35.0	47.0
Percentage female	50.2	57.3	48.0	57.8
Percentage white	73.2	77.5	72.6	71.7
Percentage bachelor’s degree or above	50.4	36.5	51.4	38.7
Median income	$40,000–$60,000	$40,000–$50,000	$40,000–$60,000	$40,000–$50,000
Mean (SD) subjective SES	5.1 (1.7)	6.4 (1.7)	5.2 (1.8)	6.3 (1.6)
Mean (SD) political orientation	47.6 (28.9)	3.9 (1.8)	48.3 (27.7)	4.1 (1.9)
2016 presidential vote (%)
Democrat	36.6	33.8	39.1	31.9
Republican	34.4	33.1	30.7	32.3
n	744	981	179	252

Note: Subjective socioeconomic status (SES) ranges from 1 to 10; political orientation ranged from 1 to 100 in experiment 1 and survey A1 and from 1 to 7 in experiment 2 and survey A2.

Experiment 1

Participants (n = 744 for our experimental conditions and n = 179 for our ancillary condition, discussed later) selected into the experiment through a link describing it as a survey of social attitudes. They first completed the consent and demographics forms. Participants were then randomly assigned to one of our experimental conditions, which provided them with information about the extent of American income inequality and mobility.

Following previous studies (McCall et al. 2017; Shariff et al. 2016), we manipulated perceptions of inequality and mobility using a realistic (but fabricated) excerpt from a news article, which described the levels of inequality and social mobility in the United States as either high or low (the full text of the treatments is provided in Appendix B). We fully crossed level of inequality (high vs. low) with level of mobility (high vs. low) to allow clearer causal inferences about their independent or interactive effects. Inequality was expressed as the share of income earned by each of 10 income deciles, and social mobility was expressed as the percentage of today’s 30-year-olds who earn more, less, or about the same as their parents did when they were 30. Our treatments are thus intended to affect participants’ beliefs about broader socioeconomic conditions, not necessarily their own income rank or personal mobility.

The income shares used in the high-inequality conditions correspond to the actual level of inequality in the United States, while the income shares in the low-inequality conditions are more similar to those in Denmark and Norway (WID.WORLD 2018). The level of mobility described in the high-mobility conditions is qualitatively similar to those found in Australia or Sweden, while the level of mobility described in the low-mobility condition is closer to that found in the United States (Alesina et al. 2018; Chetty et al. 2017; Corak 2016). Experiment 1 used information on absolute mobility, or the likelihood that a person obtains a better standard of living than his or her parents (Chetty et al. 2017).

Participants completed three comprehension checks that required them to correctly report information from the article on inequality and mobility: the income share of the bottom half and the top decile of the income distribution and the proportion of 30-year-olds better off than their parents. We excluded from analyses 67 participants (9 percent) who answered at least two items incorrectly and 28 (3.8 percent) who indicated suspicions about the veracity of the article in open-ended suspicion-check questions at the end of the study. This left an analytic n of 649. Power analysis shows that this sample size gives us greater than 80 percent power to detect a main effect of .20, which is generally considered small (Cohen 1988).

Following the comprehension checks, participants completed the dependent measures. (Participants also completed exploratory items about how trustworthy and effective they thought government and businesses are, the merit of the rich and poor, satisfaction with their own status, and desire for more business-oriented policy actions. Analysis of these measures is available upon request.) All measures in experiment 1 used 100-point sliders. Full descriptions of scales and items for both experiments can be found in Appendix B.

Most work on attitudes related to perceived inequality and mobility traces them to beliefs about how fair (whether people generally get what they deserve) and meritocratic (whether success is generally due to hard work and ability or to other factors) people see society as being. We therefore measured these attitudes in our experiments. We measured perceptions of meritocracy by asking participants how important each of five factors (hard work, ambition, a good economy, coming from a wealthy family, and luck) are to success in the United States. We adapted these items from previous studies of attitudes about inequality and meritocracy (McCall 2013; Reynolds and Xian 2014). We used these items to create a variable summarizing net belief in merit versus other factors (Reynolds and Xian 2014) by subtracting the mean of a good economy, coming from a wealthy family, and luck from the mean of hard work and ambition. The larger this summary variable, the more the participant indicated that hard work and ambition are more important than other factors in success.

We measured participants’ concerns about inequality and mobility using two separate scales, each with two items. One item asked them to agree or disagree that inequality or mobility is an issue, and the other that the issue should be a priority for leaders in society. We also assessed preferences for government actions aimed at reducing inequality and increasing opportunity. Although this link is often assumed, few studies have tested whether heightened concern also translates into changes in support for policies designed to address economic unfairness. Those that do so generally find smaller effects on policy preferences than on more general attitudes about fairness and meritocracy (Alesina et al. 2018; Kuziemko et al. 2015; McCall et al. 2017). We assessed these preferences using a four-item scale that asked participants to indicate whether they would approve or disapprove if the government made it a high priority to enact policies aimed at inequality and mobility, such as increasing taxes on the wealthy or education funding.

Analysis

In all analyses, we treat the low-inequality, high-mobility condition as the baseline, as it represents the “best-case scenario” in which America has both relatively low inequality and high social mobility. These conditions prevailed in the United States in the decades between the end of World War II and the end of the Vietnam War (Chetty et al. 2017; Piketty and Saez 2014). As a result, this period has been used as a baseline for qualitative assessments of whether inequality is too high and mobility too low (Putnam 2016).

We analyzed the direct effects of inequality (low inequality coded 0, high inequality coded 1) and immobility (high mobility coded 0, low mobility coded 1) and their interaction in simple regressions for each dependent variable with df = 648. To facilitate comparisons of effects, we present unstandardized (b) as well as standardized (d; the raw coefficient divided by the dependent variable’s standard deviation) effects. For brevity, we provide standardized main effects in the text; full statistics for all dependent variables are reported in Table 2.

Table 2.

Main Effects of Condition on Dependent Variables, Experiment 1.

	Inequality Concern		Mobility Concern		Perception of Meritocracy		Government Action
	d	b	d	b	d	b	d	b
Inequality	.42	10.65*** (1.97)	.18	4.32* (1.82)	−.06	−1.44 (2.00)	.15	3.50^† (1.85)
Immobility	.11	2.95 (1.97)	.52	12.66*** (1.82)	−.35	−8.90*** (2.00)	.10	2.32 (1.85)
Constant	−.68	53.85	−.53	50.84	.26	15.04	−.27	65.79

Note: Standard errors shown in parentheses. †p < .10. *p < .01. ***p < .001.

Independent versus Interaction Effects

For each of the dependent variables, we tested for interaction effects of the inequality and mobility treatments. As shown in Appendix C, we did not find any significant interaction effects, and interaction terms were descriptively small (|d| < .20 for all), consistent with the claim that each perception is important in its own right (hypotheses 1a and 1b). Table 2 thus summarizes main effects of condition on our dependent variables.

Effects of Political Orientation

Appendix D details the direct effects of political orientation and interactions between treatment and political orientation for both studies. Unsurprisingly, our dependent variables are strongly related to political orientation. Across all conditions, conservatives were more likely than liberals to believe that success comes more from hard work and drive than other factors, were less likely to be concerned about inequality and mobility, and expressed weaker preferences for government intervention on these issues (mean |d| = .42, p < .001 for all).

There was a slight tendency for the treatments to increase concerns about inequality (inequality treatment × conservatism: d = .13, b = .12, SE = .06, p < .05) and mobility (immobility treatment × conservatism: d = .14, b = .12, SE = .06, p < .05) more for conservatives than for liberals. In the case of mobility concerns, this interaction was enough to negate the main effect of high inequality on liberals (i.e. people with −1 SD conservatism). However, these interactions can likely be attributed to ceiling effects among liberals, who tended to be very concerned about these issues across all conditions, leaving less room for the treatments to further increase concern compared with conservatives. Political orientation did not interact with treatment effects for any other outcomes.

Concerns about Inequality and Mobility

The inequality treatment significantly increased concerns about inequality (d = .42, p < .001) and had a smaller effect on concerns about mobility (d = .18, p < .05). Similarly, the low-mobility treatment increased concerns about mobility (d = .52, p < .001) but did not significantly affect concerns about inequality (d = .11, p = ns). The treatments both increased the sense that inequality and immobility are problems and that someone should do something about them.

Perception of Meritocracy

Participants’ beliefs about what leads to success in America depended on the mobility treatment. Those in the low-mobility conditions had lower overall belief in merit versus nonmerit factors (d = −.35, b = −8.90, SE = 2.00, p < .001). In contrast, higher inequality did not significantly affect perceptions of meritocracy (d = −.06, p = ns).

Overall, these findings show that mobility and inequality have independent effects on economic fairness, rather than mobility mattering more (or only) when inequality is high. Furthermore, we find that mobility perceptions are more strongly related to attitudes about meritocracy than are perceptions of inequality. However, as others have noted (McCall 2013; McCall and Kenworthy 2009), Americans may become more concerned about economic fairness without necessarily supporting specific policy solutions. Thus, we examined whether the treatments increased preferences for government action on inequality and mobility.

Government Action

Compared with the effects described thus far, the treatments had smaller effects on preferences for government action. Although those in the higher inequality condition expressed marginally higher preferences for government policies to reduce inequality and promote mobility (d = .15, p < .10), this effect did not reach a conventional level of significance. The low-mobility treatment did not have any significant effects (d = .10, p = ns).

Ancillary Condition

As context for the aforementioned findings, our experiment also included an ancillary condition (run concurrently with the other conditions) that asked participants to estimate the average income in each income decile. We then converted those mean incomes into a percentage of the total of the incomes participants reported (Eriksson and Simpson 2012). We also summarize these income distributions using Gini coefficients. The Gini coefficient ranges from 0 to 1, with 0 representing complete equality and 1 complete inequality. Twenty participants (11.2 percent) were excluded for providing non-rank-ordered estimates or failing an attention check, leaving an analytic n of 159. Appendix E gives detailed findings from this ancillary condition.

Overall, participants estimated a Gini coefficient of .56. This is lower than the World Inequality Database’s (WID.WORLD 2018) Gini coefficient estimate for U.S. pretax income of .60 but higher than the U.S. Census Bureau’s estimate of .49 (Semega et al. 2019) and much higher than the Organisation for Economic Co-operation and Development’s (2019) estimate of .39, Thus, participants clearly believe that U.S. inequality is very high, consistent with actual economic estimates.

Perceptions varied along partisan lines, with liberals estimating higher levels of overall inequality than conservatives (.60 vs .53, respectively; two-tailed t test: difference = .07, t = 2.75, df = 157, p < .01). This difference stemmed mainly from liberals’ tendency to believe that lower deciles had smaller shares of income and that the top decile had a larger share of income, compared with their conservative counterparts.

To measure mobility perceptions, we asked participants to estimate the percentage of today’s 30-year-olds who are better off, worse off, or in about the same socioeconomic position as their parents. To create a measure of overall mobility, we classify participants into those who perceive more upward than downward mobility, those who perceive more downward than upward mobility, and those who perceive the same amount of upward and downward mobility.

Participants believed that there was more upward than downward mobility. Most were optimistic (55 percent), assuming that the net direction of mobility was upward, with more people having attained better socioeconomic status than their parents. About a third (32 percent) were pessimistic, thinking that more people were worse off than their parents. Only 13 percent of participants thought there was as much downward as upward mobility. Differences between liberals and conservatives were not statistically significant.

These findings show that in the absence of manipulated information about inequality and mobility levels, participants in our sample tended to perceive the United States as highly unequal but that they tended to perceive relatively high levels of upward mobility.

Experiment 1 Discussion

Our inequality and mobility treatments tended to increase concerns about meritocracy and economic fairness, as well as the belief that something should be done to address inequality and mobility. We found no evidence for interaction effects, and interaction terms were descriptively small. Furthermore, perceptions of meritocracy were affected only by perceived mobility, not by perceived inequality. The results of the ancillary condition, which measured existing perceptions of inequality and mobility, help explain why. Participants in this condition believed that inequality was high and that mobility in the United States is generally upward rather than downward. This suggests that telling participants in the “high inequality” conditions that inequality in high may have tended to confirm their preexisting beliefs, whereas “low mobility” information likely ran counter to their preexisting beliefs.

Although the results of experiment 1 are suggestive, there are several limitations we sought to address in experiment 2. First, experiment 1 did not incorporate manipulation checks to measure whether the treatments affected participants’ views of the levels of inequality and mobility in the United States. Consequently, experiment 1 cannot tell us how much the manipulations altered perceptions of inequality or mobility.

Second, experiment 1 manipulated absolute mobility, but much prior work has focused on relative mobility (Alesina et al. 2018; Davidai and Gilovich 2015, 2018; Kraus and Tan 2015), the rate at which individuals born in a given quantile (most commonly the bottom 20 percent) transition to a different quantile as an adult (e.g., the top 20 percent). By manipulating relative mobility in experiment 2, we can assess whether the conclusions of experiment 1 are robust to different conceptions of mobility.

Third, although the inequality and mobility treatments clearly affected beliefs about the importance of merit and overall concerns about inequality and mobility, the effect on preferences for government action was much weaker. This is consistent with some prior work that has found relatively small or null effects of inequality or mobility treatments on policy attitudes compared with other social attitudes (Alesina et al. 2018; Kuziemko et al. 2015). But we wanted to ensure that the relative lack of effects was not an artifact of the specific items used in experiment 1. Thus, experiment 2 introduces several new items to assess preferences for inequality- and opportunity-oriented policies, as well as a general preference for government to make economic outcomes fairer.

In addition to these changes, for experiment 2 we also revised our response scales to more precisely measure our attitudes of interest, and we imposed sample quotas to include a larger number of lower socioeconomic status participants in our sample.

Experiment 2

For experiment 2 we used the same overall method as experiment 1, with several improvements. First, to better ensure that participants understood the concepts of income inequality and social mobility, we added a brief information screen before the treatments that explained these terms. Two comprehension check items then assessed participants’ understanding of these terms.

We also replaced the absolute mobility treatment used in experiment 1 with a relative mobility treatment, depicting the probability of a person from the bottom quintile ending up in each of the five income quintiles later in life. The diagrams and statistics were adapted from prior work (Swan et al. 2017). In the low-mobility version, 53 percent of children from the bottom quintile stayed at the bottom, and only 1 percent rose to the top quintile. In the high-mobility version, 33 percent stayed at the bottom, while 9 percent rose to the top. Actual mobility rates in the United States are between the rates in our high- and low-mobility conditions (The Pew Charitable Trusts 2012; Swan et al. 2017). Most accounts of relative mobility focus on the mobility of people in the bottom quintile (Alesina et al. 2018; Davidai and Gilovich 2015; Kraus and Tan 2015) and generally show that Americans overestimate mobility rates (though see Chambers et al. 2015; Cheng and Wen 2019). Using a relative mobility treatment will show whether our findings are robust to the type of mobility involved.

We also added manipulation checks to experiment 2. In pretests, these checks revealed that perceptions of the level of inequality did not differ significantly between the inequality treatments. We therefore strengthened the low-inequality treatment to emphasize the low level of inequality. We did not revise the high-inequality treatment, as participants in that condition already perceived levels of inequality near the top of our scale. This revised treatment succeeded in reducing perceived inequality in the low-inequality condition, relative to the high-inequality treatment.

After reading the article, participants completed four more comprehension checks of the information presented in the article. We combined these checks with the two checks asking participants to correctly define “income inequality” and “social mobility.” Comprehension was very high, with 899 participants (91.6 percent) answering all six questions correctly.

After the comprehension and manipulation checks, participants completed (1) two scales measuring the judgement that inequality and mobility are too low or too high, regardless of the actual level; (2) two scales measuring participants’ belief that the U.S. economy is fair and meritocratic, modified from Day and Fiske (2017); and (3) three scales measuring participants’ overall desire for government to make economic outcomes more fair, reduce inequality, and increase opportunity. All scales used seven-point Likert-type scales. Full descriptions of the scales are given in Appendix B.

We administered experiment 2 using a panel of MTurk participants, from which we quota-sampled one third liberals, one third independents, and one third conservatives, as well as 30 percent to 40 percent college-educated participants. These quotas mean that our sample is more politically and educationally balanced than typical MTurk samples.

Finally, we recruited a larger sample size to improve statistical power. As in experiment 1, experiment 2 participants (n = 981 for our experimental conditions and n = 252 for the ancillary condition, described below) selected into the experiment through a survey link and completed consent and demographics forms. We excluded 15 participants (1.5 percent) who answered at least half of the comprehension checks incorrectly. Another 10 (1.0 percent) were excluded for failing several checks designed to identify bots and 38 (3.9 percent) for indicating in open-response questions at the end of the study that they suspected that the article was manipulated. In total, we excluded 53 participants (5.4 percent), leaving an analytic sample of 928. This sample size gives 80 percent power to detect main effects of about d = .15 at the .05 level and 75 percent power to detect interactions of about d = .15 at the .10 level. Overall, experiment 2 is methodologically more refined and has a larger, better powered, and more economically diverse sample compared with experiment 1.

As in experiment 1, we also included an ancillary condition that measured, rather than manipulated, perceived inequality and mobility. Unlike the ancillary condition in experiment 1, however, we measured these perceptions using more subjective Likert-type scales. We did so because some prior work has shown that quantitative estimates of inequality and mobility can be vulnerable to estimation biases and other method effects (Chambers et al. 2014, 2015; Eriksson and Simpson 2012; Swan et al. 2017). Measuring inequality and mobility perceptions with both quantitative and subjective measures helps ensure that our conclusions are not attributable to estimation biases.

Analysis

As in experiment 1, we use the low-inequality, high-mobility condition as the baseline and examine effects of inequality and immobility. Effects refer to the direct effects of the inequality and immobility treatments with df = 927. Main effects of condition for all dependent variables are reported in Tables 3 and 4.

Table 3.

Main Effects of Condition on Dependent Variables, Experiment 2.

	Perceived Inequality		Perceived Mobility		Inequality Is Too High		Mobility Is Too Low
	d	b	d	b	d	b	d	b
Inequality	.57	.68*** (.07)	.07	.10 (.08)	.28	.44*** (.10)	.17	.25** (.10)
Immobility	.13	.16* (.07)	−.98	−1.45*** (.08)	.08	.12 (.10)	.45	.68*** (.10)
Constant	−.36	5.23	.46	3.78	−.19	5.04	−.31	4.60

Note: Standard errors shown in parentheses. *p < .01. **p < .01. ***p < .001.

Table 4.

Main Effects of Condition on Dependent Variables, Experiment 2.

	Economic Fairness		Meritocracy		General Government Action		Reducing Inequality		Increasing Opportunity
	d	b	d	b	d	b	d	b	d	b
Inequality	−.17	−.26** (.10)	−.13	−.19* (.09)	.05	.08 (.10)	.09	.16 (.11)	.13	.16^† (.08)
Immobility	−.24	−.37*** (.10)	−.39	−.55*** (.09)	.02	.03 (.10)	.10	.17 (.11)	.06	.07 (.08)
Constant	.21	3.82	.26	4.41	−.03	5.15	−.10	4.27	−.10	5.53

Note: Standard errors shown in parentheses. †p < .10. *p < .01. **p < .01. ***p < .001.

Manipulation Checks

As shown in Table 3, participants in the high-inequality conditions believed that inequality was higher than those in the low-inequality conditions (d = .57, p < .001), and those in the low-mobility conditions perceived slightly more inequality than those in the high-mobility conditions (d = .13, p < .05). The mobility treatment had a very large effect on perceived mobility, with participants in the low-mobility conditions perceiving much less mobility than those in the high-mobility conditions (d = −.98, p < .001). Mobility perceptions were not significantly affected by the inequality treatment (d = .07, p = ns). We conclude that the treatments affected perceptions as intended and that participants’ beliefs about mobility were more sensitive to our treatment than their beliefs about inequality. This may signify that participants had stronger preconceived notions about inequality than about social mobility, and so the latter attitudes are more easily influenced. We return to this point later in our discussion of the ancillary condition.

Independent versus Interaction Effects

We tested whether the inequality and mobility treatments interacted for each of our dependent variables, and as in experiment 1, they did not. As shown in Appendix C, all inequality × immobility interaction terms were nonsignificant and descriptively small (|d| < .20 for all). Again, this suggests that mobility perceptions matter on their own, not because they moderate the effects of perceived inequality. This provides support for hypotheses 1A and 1B. Tables 3 and 4 thus provide main effects of condition on our manipulation checks and dependent measures.

Effects of Political Orientation

Direct and interactive effects of our continuous measure of political orientation are provided in Appendix D. As in experiment 1, political orientation strongly influenced participants’ views. Conservatives tended to perceive less inequality and more mobility than liberals, were less concerned about inequality and mobility, believed the economy to be fairer and more meritocratic, and were less likely to prefer government actions (mean |d| = .44, p < .001 for all). The low-mobility treatment decreased perceived mobility slightly more among conservatives than liberals (immobility × conservatism: d = −.11, b = −.09, SE = .05, p < .05) but did not eliminate the effect among liberals. Importantly, political orientation did not interact with treatment for any other variable. Thus, as in experiment 1, we find that our treatments had largely the same effects for both liberals and conservatives.

Judgments of Inequality and Mobility

Participants were more likely to agree that inequality was too high in the high-inequality conditions (d = .29, p < .001), but the mobility treatment did not significantly affect this attitude (d = .08, p = ns). Both treatments increased participants’ tendency to say that mobility is too low, though the mobility treatment had a larger effect (inequality treatment: d = .17, p < .01; immobility treatment: d = .45, p < .001).

If this is due to treatment effects on how much inequality and mobility participants perceive, rather than simply the salience of the issue, then perceived inequality and mobility should mediate these effects. We conducted mediation analyses (Appendix F) using the khb package in Stata (Breen, Karlson, and Holm 2013) to test whether these effects were driven specifically by changes in perceived inequality and mobility. We found that they were: treatment effects on perceived inequality and mobility led to changes in participants’ evaluation of inequality as too high and mobility as too low.

Perceptions of Economic Fairness and Meritocracy

As shown in Table 4, both high inequality and low mobility reduced beliefs that the American economy is fair (inequality treatment: d = −.17, p < .01; immobility treatment: d = −.24, p < .001) and meritocratic (inequality treatment: d = −.13, p < .05; immobility treatment: d = −.39, p < .001) relative to low inequality and high mobility. The two treatments had similar-sized effects on perceived economic fairness, but low mobility had a greater effect on perceived meritocracy than did high inequality.

Again, we conducted mediation analyses (Appendix F) to test whether the two treatment effects were mediated by effects on perceived inequality and mobility. We found evidence of mediation for both treatments: perceived inequality played a larger role in mediating treatment effects on perceptions of economic fairness, but perceptions of inequality and perceptions of mobility mediated effects on perceived meritocracy at nearly equal (standardized) rates. The overall larger effects of the mobility treatment on perceived economic fairness and meritocracy is more properly attributed to the mobility treatment’s larger effect on mobility perceptions. In other words, our mobility treatment had larger effects on attitudes about fairness and meritocracy because it was easier to change perceptions of mobility than inequality.

Government Actions

As in experiment 1, treatment effects were smaller for preferences for government action (Table 4). Indeed, the effects observed in experiment 2 are even smaller than those in experiment 1, and none were significant at the p < .05 level. The lack of effects is somewhat surprising, as we found treatment effects on perceptions of economic fairness and meritocracy, and perceptions of economic and meritocracy, in turn, predicted support for government action on both inequality (perceived fairness: d = −.72, p < .001; perceived meritocracy: d = −.61, p < .001) and mobility (perceived fairness: d = −.57, p < .001; perceived meritocracy: d = −.46, p < .001). Despite this, the total effects of treatment were quite small and failed to reach statistical significance. Thus, we did not conduct formal mediation analyses.

Ancillary Condition

To measure subjective assessments of the levels of inequality and mobility, we administered the manipulation check items from experiment 2 to a separate sample of 252 MTurk workers (n = 230 after excluding participants who failed attention and comprehension checks).

Participants rated inequality as very high (mean = 6.03, SD = .96). Indeed, the overwhelming majority (98.3 percent) of participants rated inequality above the midpoint on our Likert-type items. They had less extreme views about mobility: the average mobility rating was 3.08 (SD = 1.26). This is on the lower side of moderate mobility, and the higher standard deviation for mobility than inequality perceptions (1.26 vs. .96, p for difference < .001) indicates that there was less consensus about the level of mobility than there was about the extent of inequality. As in the ancillary condition of experiment 1, conservatives perceived less inequality and more mobility than liberals (perceived inequality: d = −.45, b = −.23, SE = .03, df = 229, p < .001; perceived mobility: d = .39, b = .26, SE = .04, df = 229, p < .001).

Thus, participants in the ancillary conditions of experiments 1 and 2 had a clear understanding that inequality in the United States is very high, but their mobility perceptions were more optimistic, with many participants perceiving moderate or even high levels of mobility. This supports our suggestion that the overall larger effects of our mobility treatment are due to less pessimistic views about the level of mobility than inequality among the participants in our sample.

Experiment 2 Discussion

The results of experiment 2 are highly consistent with those of experiment 1. Specifically, the inequality and mobility treatments had only independent, not interactive, effects. We found no significant interactions, and all interaction terms were descriptively small. Thus, we found no support for the argument underlying hypothesis 2, which predicts that participants would be more willing to accept high inequality as long as mobility is high. Rather, both mobility and inequality mattered independent of one another, with the mobility treatment having larger effects than the inequality treatment.

Although attitudes about meritocracy and fairness were influenced by our manipulations, we again found limited effects on preferences for government action. Given that policy preferences are so strongly tied to other aspects of political ideology (Alesina et al. 2018; Ashok et al. 2015; Bullock, Williams, Limbert 2003; Cech 2017; Kluegel and Smith 1986; McCall 2013), it may be that these attitudes are simply more difficult to affect in an experiment, and they may require either much more prolonged exposure to—or direct personal experience with—rising inequality or immobility.

Discussion

When people judge whether the American dream is still alive, and whether the American economic system is functioning as it should, we might expect them to pay attention to both how unequal the United States is and how easily people are able to climb its economic ranks. But perceived inequality and mobility are not independent beliefs, making it difficult for social scientists to determine whether (and how) beliefs about the level of inequality and mobility combine to influence attitudes about fairness and meritocracy.

To our knowledge, this is the first investigation to explicitly disentangle the causal effects of perceived inequality and mobility by experimentally manipulating each separately. By doing so, we were able to assess the independent and interactive effects of each on beliefs about the United States as a meritocracy, concerns about inequality and mobility, and support for government actions that may remedy inequality and immobility.

Overall, both higher perceived inequality and lower perceived mobility increased participants’ concerns about economic fairness, and these effects were independent rather than interactive. Thus, our findings suggest that mobility is important in its own right, independent of the level of perceived inequality. Our study also adds to a literature showing mixed or null effects of these perceptions on specific policy preferences (Alesina et al. 2018; Kuziemko et al. 2015; McCall et al. 2017). The smaller effects on policy attitudes suggest that increased concern does not directly translate into preferences for specific remedies, perhaps because participants are uncertain of the best remedies for these issues (McCall 2013; McCall and Kenworthy 2009) or because policy preferences (and government action in general) are so strongly tied to established political ideologies (Ashok et al. 2015; Cech 2017; Eriksson and Simpson 2012; Pedersen and Mutz 2018). It is also quite possible that change in policy preferences is more likely to come from change in individuals’ material circumstances (Owens and Pedulla 2014) than from their knowledge about the state of inequality and mobility.

Although both treatments reduced the belief that the American economy is fair and meritocratic, mobility tended to have a greater effect. One reason for this might be greater consensus among our participants that inequality is high. This is consistent with research showing that Americans tend to see inequality as too high (McCall 2013; Osberg and Smeeding 2006; Pedersen and Mutz 2018). In contrast, they tend to overestimate the level of social mobility and are relatively satisfied with their chances of getting ahead (Alesina et al. 2018; Davidai and Gilovich 2015, 2018; Gallup 2018; Kraus and Tan 2015). Our ancillary conditions that measured, rather than manipulated, perceptions of inequality and mobility provide further evidence: using both quantitative and subjective measurement techniques, we found that participants thought inequality was very high, but tended to be more optimistic about mobility. Furthermore, there was strong consensus on the level of perceived inequality, with the vast majority of participants thinking that inequality is high or very high, but less consensus on the level of mobility.

Importantly, we found very similar effects whether we manipulated mobility in absolute (experiment 1) or relative (experiment 2) terms. This suggests that people may care more about whether there are opportunities to move up than whether that movement is relative or absolute. Indeed, this would be consistent with the views of social scientists, who interpret low mobility of either type as signaling a breakdown in the equity of the American economy (Alesina et al 2018; Chetty et al 2017; Davidai 2018). But future research should look more systematically at how laypeople distinguish between absolute and relative mobility.

More generally, results from these ancillary conditions suggest that informing participants of the high levels of inequality that exist in the United States is unlikely to change their sense of the economy and the extent to which it is fair, as they already understand that inequality is very high. On the other hand, participants’ views of social mobility were more positive. Coupled with our findings that inequality and mobility perceptions both have independent effects on beliefs about fairness and meritocracy, these ancillary conditions suggest that efforts to communicate to Americans that the nation’s economic meritocracy has broken down should emphasize low levels of mobility over high levels of inequality. By extension, our findings suggest that large-scale changes in concerns about economic fairness in the future are more likely to result from heightened awareness of the low social mobility rates in the U.S. (Chetty et al. 2017; Corak 2016; Putnam 2016) than from changes in beliefs about inequality.

Supplemental Material

SRD-20-0031.R1_supplementary_information_FINAL – Supplemental material for Judgments of Economic Fairness Are Based More on Perceived Economic Mobility Than Perceived Inequality

Supplemental material, SRD-20-0031.R1_supplementary_information_FINAL for Judgments of Economic Fairness Are Based More on Perceived Economic Mobility Than Perceived Inequality by Nicholas Heiserman, Brent Simpson and Robb Willer in Socius

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a grant from the National Science Foundation (SES-11626056) to Brent Simpson.

ORCID iD

Nicholas Heiserman

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Nicholas Heiserman is a PhD candidate in sociology at the University of South Carolina. His research includes work on attitudes about inequality, intersectionality and stereotype content, and group processes. His work has been published in the American Sociological Review and Social Psychology Quarterly.

Brent Simpson is a professor and chair of sociology at the University of South Carolina. His recent research has appeared in the American Journal of Sociology, American Sociological Review, Annual Review of Sociology, the Journal of Personality and Social Psychology, and Proceedings of the National Academy of Sciences.

Robb Willer is a professor of sociology, psychology, and organizational behavior and the director of the Polarization and Social Change Laboratory at Stanford University. He studies a variety of topics, including the social psychology of contemporary American politics. His research has appeared in American Sociological Review, American Journal of Sociology, the Journal of Personality and Social Psychology, Proceedings of the National Academy of Sciences, and Nature Human Behaviour.

References

Alesina

Alberto

La Ferrara

Eliana

. 2005. “Preferences for Redistribution in the Land of Opportunities.” Journal of Public Economics 89(5–6):897–931.

Alesina

Alberto

Stantcheva

Stefanie

Teso

Edoardo

. 2018. “Intergenerational Mobility and Preferences for Redistribu-tion.” American Economic Review 108(2):521–54.

Ashok

Vivekinan

Kuziemko

Ilyana

Washington

Ebonya

. 2015. “Support for Redistribution in an Age of Rising Inequality: New Stylized Facts and Some Tentative Explanations.” NBER Working Paper No. 21529. Cambridge, MA: National Bureau of Economic Research.

Benabou

Roland

Efe A.

2001. “Social Mobility and the Demand for Redistribution: The POUM Hypothesis.” Quarterly Journal of Economics 116(2):447–87.

Bobo

Lawrence

. 1991. “Social Responsibility, Individualism, and Redistributive Policies.” Sociological Forum 6(1):71–92.

Breen

Richard

Karlson

Kristian Bernt

Holm

Anders

. 2013. “Total, Direct, and Indirect Effects in Logit and Probit Models.” Sociological Research and Methods 42(2):164–91.

Bullock

Heather E.

Williams

Wendy R.

Limbert

Wendy M.

2003. “Predicting Support for Welfare Policies: The Impact of Attributions and Beliefs about Inequality.” Journal of Poverty 7(3):35–56.

Buhrmester

Michael

Kwang

Tracy

Gosling

Samuel D.

2011. “Amazon’s Mechanical Turk: A New Source of Inexpensive, Yet High-Quality, Data?” Perspectives on Psychological Science 6(1):3–5.

Cech

Erin A.

2017. “Rugged Meritocratists: The Role of Overt Bias and the Meritocratic Ideology in Trump Supporters’ Opposition to Social Justice Efforts.” Socius 3:1–20.

10.

Chambers

John R.

Swan

Lawton K.

Heesacker

Martin

. 2014. “Better Off Than We Know: Distorted Perceptions of Incomes and Income Inequality in America.” Psychological Science 25(2):613–18.

11.

Chambers

John R.

Swan

Lawton K.

Heesacker

Martin

. 2015. “Perceptions of U.S. Social Mobility Are Divided (and Distorted) along Ideological Lines.” Psychological Science 26(4):413–23.

12.

Cheng

Siwei

Wen

Fangqi

. 2019. “Americans Overestimate the Intergenerational Persistence in Income Ranks.” Proceedings of the National Academy of Sciences 116(28):13909–14.

13.

Chetty

Raj

Grusky

David

Hell

Maximilian

Hendren

Nathaniel

Manduca

Robert

Narang

Jimmy

. 2017. “The Fading American Dream: Trends in Absolute Income Mobility since 1940.” Science 356(6336):398–406.

14.

Clifford

Scott

Jewell

Ryan M.

Waggoner

Philip D.

2015. “Are Samples Drawn from Mechanical Turk Valid for Research on Political Ideology?” Research and Politics 2(4):1–9.

15.

Cohen

Jacob

. 1988. Statistical Power for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum.

16.

Coppock

Alexander

Leeper

Thomas J.

Mullinix

Kevin J.

2018. “Generalizability of Heterogeneous Treatment Effect Estimates Across Samples.” Proceedings of the National Academy of Sciences 115(49):12441–46.

17.

Corak

Miles

. 2016. “Inequality from Generation to Generation: The United States in Comparison.” Working Paper No. 9929. Bonn, Germany: Institute for the Study of Labor.

18.

Côté

Stéphane

House

Julian

Willer

Robb

. 2015. “High Economic Inequality Leads Higher-Income Individuals to Be Less Generous.” Proceedings of the National Academy of Sciences 112(5):15838–43.

19.

Dawtry

Rael J.

Sutton

Robbie M.

Sibley

Chris G.

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.

20.

Davidai

Shai

. 2018. “Why Do Americans Believe in Economic Mobility? Economic Inequality, External Attributions of Wealth and Poverty, and the Belief in Economic Mobility.” Journal of Experimental Social Psychology 79:138–48.

21.

Davidai

Shai

Gilovich

Thomas

. 2015. “Building a More Mobile America—One Income Quintile at a Time.” Perspectives on Psychological Science 10(1):60–71.

22.

Davidai

Shai

Gilovich

Thomas

. 2018. “How Should We Think about Americans’ Beliefs about Economic Mobility?” Judgment and Decision Making 13(3):297–308.

23.

Day

Martin V.

Fiske

Susan T.

2017. “Movin’ on Up? How Perceptions of Social Mobility Affect Our Willingness to Defend the System.” Social Psychological and Personality Science 8(3):267–74.

24.

Della Fave

Richard L.

1980. “The Meek Shall Not Inherit the Earth: Self-Evaluation and the Legitimacy of Stratification.” American Sociological Review 45(6):955–71.

25.

Eriksson

Kimmo

Simpson

Brent

. 2012. “What Do Americans Know about Inequality? It Depends on How You Ask Them.” Judgment and Decision Making 7(6):741–45.

26.

Evans

M.D.R.

Kelley

Jonathan

Peoples

Clayton D.

2010. “Justifications of Inequality: The Normative Basis of Pay Differentials in 31 Nations.” Social Science Quarterly 91(5):1405–31.

27.

Gallup. 2018. “Americans’ Views on Economic Mobility and Economic Inequality in the U.S. (Trends).” Retrieved September 11, 2020. https://news.gallup.com/poll/228980/americans-views-economic-mobility-economic-inequality-trends.aspx.

28.

Hargittai

Ester

Shaw

Aaron

. 2020. “Comparing Internet Experiences and Prosociality in Amazon Mechanical Turk and Population-Based Survey Samples. Socius. Retrieved September 11, 2020. https://journals.sagepub.com/doi/10.1177/2378023119889834.

29.

Hauser

Oliver P.

Norton

Michael I.

2017. “(Mis)perceptions of Inequality.” Current Opinion in Psychology 18(1):21–25.

30.

Heiserman

Nicholas

Simpson

Brent

. 2017. “Higher Inequality Increases the Gap in the Perceived Merit of the Rich and Poor” Social Psychology Quarterly 80(3):243–53.

31.

Hunt

Matthew O.

Bullock

Heather E.

2016. “Ideologies and Beliefs about Poverty.” Pp. 93–116 in The Oxford Handbook of the Social Science of Poverty, edited by Brady

Burton

L. M.

New York: Oxford University Press.

32.

Kiatpongsan

Sorapop

Norton

Michael I.

2014. “How Much (More) Should CEOs Make?” Perspectives on Psychological Science 9(6):587–93.

33.

Kluegel

James R.

Smith

Eliot R.

1986. Beliefs about Inequality: Americans’ Views of What Is and What Ought to Be. New Brunswick, NJ: Transaction.

34.

Kraus

Michael W.

Tan

Jacinth J. X.

2015. “Americans Overestimate Social Class Mobility.” Journal of Experimental Social Psychology 58:101–11.

35.

Kuziemko

Ilyana

Norton

Michael I.

Saez

Emmanuel

Stantcheva

Stefanie

. 2015. “How Elastic Are Preferences for Redistribution? Evidence from Randomized Survey Experiments.” American Economic Review 105(4):1478–1508.

36.

Lepianka

Dorota

van Oorschot

Wim

Gelissen

John

. 2009. “Popular Explanations of Poverty: A Critical Discussion of Empirical Research.” Journal of Social Policy 38(3):421–39.

37.

McCoy

Shannon K.

Major

Brenda

. 2007. “Priming Meritocracy and the Psychological Justification of Inequality.” Journal of Experimental Social Psychology 43(3):341–51.

38.

McCall

Leslie

. 2013. The Undeserving Rich: American Beliefs about Inequality, Opportunity, and Redistribution. Cambridge, UK: Cambridge University Press.

39.

McCall

Leslie

Burk

Derek

Laperrière

Marie

Richeson

Jennifer A.

2017. “Exposure to Rising Inequality Shapes Americans’ Opportunity Beliefs and Policy Support.” Pro-ceedings of the National Academy of Science 114(36):9593–98.

40.

McCall

Leslie

Kenworthy

Lane

. 2009. “Americans’ Social Policy Preferences in the Era of Rising Inequality.” Perspectives on Politics 7(3):459–84.

41.

Mijs

Jonathan J. B

. Forthcoming. “The Paradox of Inequality: Income Inequality and Belief in Meritocracy Go Hand in Hand.” Socio-Economic Review.

42.

Newport

Frank

. 2015. “Americans Continue to Say U.S. Wealth Distribution Is Unfair.” Gallup. Retrieved September 11, 2020. http://news.gallup.com/poll/182987/americans-continue-say-wealth-distribution-unfair.aspx.

43.

Norton

Michael I.

Ariely

Dan

. 2011. “Building a Better America—One Wealth Quintile at a Time.” Perspectives on Psychological Science 6(1):9–12.

44.

OECD (Organisation for Economic Co-operation and Development). 2016. “OECD Income Distribution Database (IDD): Gini, Poverty, Income, Methods and Concepts.” Retrieved September 11, 2020. https://www.oecd.org/social/income-distribution-database.htm.

45.

Osberg

Lars

Smeeding

Timothy

. 2006. “‘Fair’ Inequality? Attitudes toward Pay Differentials: The United States in Comparative Perspective.” American Sociological Review 71(3):450–73.

46.

Owens

Lindsay A.

Pedulla

David S.

2014. “Material Welfare and Changing Political Preferences: The Case of Support for Redistributive Social Policies.” Social Forces 92(3):1087–1113.

47.

Paolacci

Gabriele

Chandler

Jesse

. 2014. “Inside the Turk: Understanding Mechanical Turk as a Participant Pool.” Current Directions in Psychological Science 23(3):184–88.

48.

Pedersen

Rasmus

Mutz

Diana C.

2018. “Attitudes toward Economic Inequality: The Illusory Agreement.” Political Science Research and Methods 7(4):835–51.

49.

The Pew Charitable Trusts. 2012. “Pursuing the American Dream: Economic Mobility Across Generations.” Retrieved September 11, 2020. https://www.pewtrusts.org/~/media/legacy/uploadedfiles/wwwpewtrustsorg/reports/economic_mobility/pursuingamericandreampdf.pdf.

50.

Piketty

Thomas

Saez

Emmanuel

. 2014. “Inequality in the Long Run.” Science 344(6186):838–43.

51.

Putnam

Robert D.

2016. Our Kids: The American Dream in Crisis. New York: Simon & Schuster.

52.

Reynolds

Jeremy

Xian

. 2014. “Perceptions of Meritocracy in the Land of Opportunity.” Research in Social Stratification and Mobility 36:121–37.

53.

Saez

Emmanuel

Zucman

Gabriel

. 2016. “Wealth Inequality in the United States since 1913.” Quarterly Journal of Economics 131(2):519–78.

54.

Schröder

Martin

. 2017. “Is Income Inequality Related to Tolerance for Inequality?” Social Justice Research 30(1):23–47.

55.

Semega

Jessica

Kollar

Melissa

Creamer

John

Mohanty

Abinash

. 2019. “Income and Poverty in the United States: 2018.” U.S. Census Bureau Current Population Reports P60-266. Retrieved September 11, 2020. https://www.census.gov/content/dam/Census/library/publications/2019/demo/p60-266.pdf.

56.

Shariff

Azim F.

Wiwad

Dylan

Aknin

Lara B.

2016. “Income Mobility Breeds Tolerance for Income Inequality: Cross-National and Experimental Evidence.” Perspectives on Psychological Science 11(3):373–80.

57.

Solt

Frederick

Yue

Hudson

Kevan

Song

Yungmin

Dong “Enrico”

. 2016. “Economic Inequality and Belief in Meritocracy in the United States.” Research and Politics 3(4):1–7.

58.

Starmans

Christina

Sheskin

Mark

Bloom

Paul

. 2017. “Why People Prefer Unequal Societies.” Nature Human Behaviour 1(4):0082.

59.

Swan

Lawton K.

Chambers

John R.

Heesacker

Martin

Nero

Sondre S.

2017. “How Should We Measure Americans’ Perceptions of Socio-economic Mobility?” Judgment and Decision Making 12(5):507–15.

60.

Weinberg

Jill

Freese

Jeremy

McElhattan

David

. 2014. “Comparing Data Characteristics and Results of an Online Factorial Survey between a Population-Based and a Crowdsource-Recruited Sample.” Sociological Science 1:292–310.

61.

WID.WORLD. 2018. “World Inequality Database.” Retrieved September 11, 2020. http://wid.world.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.28 MB