All time references are in CEST
Perceptions of Inequality and Justice 2
|Session Organisers|| Ms Jule Adriaans (Bielefeld University)
Dr Sandra Bohmann (Socio-Economic Panel Study at DIW Berlin)
Professor Stefan Liebig (Socio-Economic Panel Study at DIW Berlin)
Mr Matteo Targa (Socio-Economic Panel Study at DIW Berlin)
|Time||Tuesday 18 July, 14:00 - 15:30|
Reducing inequalities in life chances and outcomes is identified as one of the key societal challenges of today and the economic turmoil experienced as a consequence of the pandemic as well as the ongoing war in Ukraine have exacerbated questions of social inequality and social justice across Europe.
One important contribution that the social sciences have made and continue to make in this debate, is highlighting the importance of subjective evaluations in understanding the persistence of inequalities as well as the proposed far-reaching consequences for well-being and social cohesion. Research continues to show that individuals misperceive inequality, evaluate inequalities in terms of justice, and hold normative beliefs that legitimize inequalities – all of which help to understand why inequalities persist, why inequalities do not necessarily translate into adverse consequences, and how individuals will react to policies aiming to address inequalities.
National and international surveys offer rich data for studying the determinants and consequences of such subjective perspectives on inequality. For example, Round 9 of the European Social Survey (2018/2019) featured a module on “Justice and Fairness in Europe”, the International Social Survey Programme fielded the fifth iteration of its “Social Inequality” module in 2019, and the German Socio-Economic Panel Study included a questionnaire module on social inequality in 2021. All of which extend beyond a narrow scope of income and wealth inequality but also cover issues of social mobility, preferences for distributive principles, life chances, and political procedural justice, allowing for a comprehensive account of perceptions of inequality and justice.
We are inviting contributions that address methodological questions with respect to survey measures and survey-embedded experiments that capture attitudes towards inequality as well as substantive applications of survey research that shed light on the determinants and consequences of subjective perspectives on inequality and justice.
Keywords: Social inequality; Justice; Perceptions; Survey research
Dr Hye Won Kwon (University of Turku) - Presenting Author
Professor Jani Erola (University of Turku)
Amidst the evidence of intergenerational mobility no longer increasing, today’s adults report stronger beliefs in the importance of hard work to get ahead in life than adults in the past. To understand this puzzling discrepancy between objective reality and subjective beliefs, we suggest studying two dimensions of meritocratic beliefs—the perception of meritocracy and support for meritocracy. While the former is about one’s understanding of how the system works (i.e., rewards are distributed according to one’s effort in society), the latter captures one’s beliefs about how the reward system should work (i.e., rewards should be distributed according to one’s effort in society). Despite consistent evidence that equal access to levels of education is an important mechanism for changing and maintaining social mobility in societies, past work documented mixed evidence about how one’s own education relates to the perception of and support for meritocracy. We suggest this inconsistency could partly relate to differing levels of access to tertiary education in the country. We conducted a multilevel analysis using data from the ISSP 2009 Social Inequality module. We found the perception of meritocracy and the support for meritocracy are linked to a college degree in the opposite direction, indicating that the college educated show stronger perceptions of meritocracy but weaker support for meritocracy than the non-college educated. Our findings support the moderating role of access to higher education: the gap between the college educated and the non-college educated in support for meritocracy is larger in countries with broader access to higher education than in countries with limited access to higher education. The findings of our study demonstrated the important but under-examined role of access to higher education as a contextual factor in understanding people’s meritocratic beliefs across countries.
Mr Gonzalo Franetovic (University of Milan) - Presenting Author
Social inequality is an important field of research not only in material matters. Indeed, a big portion of academic publications has focused on the way that people perceive, explain and are concerned about these inequalities. However, attitudes towards inequalities have tended to be explained mainly from an individual or country-level perspective, thereby neglecting possibly important meso-level factors, such as people's social relationships. Reacting to this, new approaches have emphasized the role of social segregation patterns in the social environments in which people develop their lives. From these perspectives, individuals exposed to more diverse socioeconomic social interactions tend to explain more inequalities by structuralist beliefs, such as coming from a wealthy family or people’s race, rather than individualist ones, like hard work or knowing the right people. However, to date, few studies have empirically tested these claims.
This research addresses this gap with the following research question: To what extent does the subjective socioeconomic segregation of people's everyday interactions affect their attitudes towards inequality in contemporary societies? To answer this, I use data from the 2019 ISSP Module of Social Inequality for 23 countries. Three hypotheses will be tested: the more socioeconomically heterogeneous the everyday interactions of individuals, the greater their perception of inequality (H1), the higher their structuralist beliefs (H2), and the lower their individualist beliefs (H3) about inequality. Linear regression models with country fixed-effects will be used, controlling for individual variables that the literature recognizes as influential. Interaction effects between socioeconomic heterogeneity and the objective socioeconomic position of individuals will be estimated to reveal differences in the associations across the population. Differentiated models by world regions will be conducted to check if the relationships are more prevalent in some societies than others.
Preliminary results show that higher everyday heterogeneity is associated with a greater perception of inequality, an increase in structuralist explanations of inequality, but also an increase in individualist inequality beliefs. This suggests that exposure to diversity in social environments plays an important role in increasing people's awareness of inequality, as well as of the multiplicity of factors that determine it.
Professor Juan Carlos Castillo (Universidad de Chile) - Presenting Author
Mr Julio Iturra (University of Bremen)
Mr Kevin Carrasco (Universidad de Chile)
The perception of economic inequality is a growing research area in survey studies. Although there are different operationalizations and measurement options, perceived salary gaps are widely used as they offer several advantages in terms of large variance which allows group comparisons. The items used to build the perceived salary gap are direct questions about the perceived salary for a low-status occupation (i.e. factory line worker) and a high-status occupation (i.e. manager), which are included in several waves of surveys as ISSP. The present paper delves into an issue barely considered so far when using this kind of items, namely the gender of the occupations. The English version of these items is gender-neutral, which is characteristic of this language by using the pronoun “the”. Therefore, firstly we do not know whether respondents attribute a gender to the occupations when answering the questions. For instance, when asked about a “manager”, are respondents thinking about a man or a woman? Would it be a different response if gender were specified? Secondly, these items need to mention a specific gender in many languages that do not have the neutral option, and by default, a male gender is used (as for instance “El gerente” / he-manager in Spanish). In this sense, it could be argued that most of what we know so far is about male perceived gaps Using data from the Desiguales Survey in Chile (2017, N=2,613) we compare the perceived salaries for four different occupations where gender was specified and randomly assigned: half of the respondents were asked about male occupations and the other half about female occupations. Preliminary results show that women perceived lower salaries than men in general and, when asked about women’s occupations, the perceived salary is even lower.
Miss Daria Szafran (University of Mannheim) - Presenting Author
Dr Ruben Bach (University of Mannheim)
The increasing popularity of algorithms in allocating resources and services in both private industry and public administration have sparked discussions about consequences for inequality, fairness, and justice in contemporary societies. Previous research has shown that the use of automated decision making (ADM) systems in high-stake scenarios such as child protection services or the legal justice system might potentially lead to adverse societal outcomes such as systematic discrimination against certain social groups. To understand such unwanted consequences of ADM on fairness, justice, and discrimination, scholars have proposed to evaluate and mitigate biases in automated decision making processes using a series of technical fairness concepts and metrics. What remains unclear, however, is whether these technical fairness definitions correspond to subjective fairness perceptions of citizens that are eventually affected subjects of ADM systems. This study aims at closing this gap by analyzing open-ended answers from the German Internet Panel (Wave 54, July 2021), a probability-based longitudinal online survey. Respondents evaluated hypothetical scenarios describing the use of ADM systems in HR, criminal justice and banking contexts on a four-point scale ranging from “Not at all fair” to “Very fair”. Subsequently, they were asked to justify their fairness evaluation providing an open-ended answer. Those answers are inductively coded using a qualitative approach. The themes we identify based on respondents’ individual understanding of fairness are then used to construct a general fairness framework in the domain of ADM. Our empirical results and the fairness framework contribute to research on the societal acceptance of algorithms and provide insights into real-world fairness perceptions of citizens.
Professor Katrin Auspurg (1Department of Sociology, LMU (University of Munich)) - Presenting Author
Mr Paul Hufe (University of Bristol)
Mr Andreas Peichl (ifo Center for Macroeconomics and Surveys, ifo Institute)
Ms Laila Schmitt (1Department of Sociology, LMU (University of Munich))
Mr Marc Stoeckli (ifo Center for Macroeconomics and Surveys, ifo Institute)
We present a multifactorial survey experiment to measure perceptions of actual and just labor market income differentials. Questions the module is designed to answer include: To what extent do individuals believe that income differences are caused by base wages or by returns to characteristics that individuals typically have in different occupations? To what extent do (different) individuals believe that wage differentials are caused by meritocratic principles (human capital, labor input) or ascriptive characteristics (e.g., gender)? And to what extent are these perceptions consistent with their notions of justice? Answering these questions is essential for research on human capital investments and redistributive preferences, for example. Individuals may underinvest in human capital if they perceive lower returns, or they may percieve inequalities more equitable or less equitable due to specific (mis)perceptions.
Previous research has relied on abstract item questions. For example, respondents were asked to indicate whether they perceived inequalities were too low or too high, or what a realistic or fair income would be for people described only by their occupation. These measurements yielded inconsistent results, possibly because respondents had to fill in missing information. In addition, abstract questions did not allow examining underlying (mis)perceptions (e.g., beliefs about base levels vs. returns).
Our survey experiment tries to overcome these limitations. It was implemented in the Socio-Economic Panel Innovation Sample in 2019. Respondents evaluated 3 short descriptions (vignettes) of hypothetical individuals differing on experimentally varied dimensions including occupation, working hours, educational, gender, region (Eastern/Western Germany) and income, resulting in approx. 5,600 vignette evaluations. The vignettes were assessed for both plausible and fair income to construct individual indicators that measure perception biases (actual returns vs. beliefs on returns) and fairness experience (beliefs on returns vs. beliefs on fair returns). We will present methodological and substantive results.