In the Shadow of the Schooled Society: Feelings of Misrecognition and the Education Ladder

Over the past decades, the education system has gradually grown into a central and universal institution of society, the impact of which plays a primary role in economic and social stratiﬁcation. This stratiﬁcation, and the way this inequality is legitimated, contains serious moral judgements that favor the higher educated over the less educated. This article focuses on the socio-psychological consequences of living in such “schooled societies” for those who are more or less successful in education. We use three waves of the European Quality of Life survey with data on 65,208 individuals across 36 countries. We investigate (1) the extent to which different educational groups feel dissatisﬁed about and misrecognized by vir-tue of their education and (2) whether the centrality of the education system in society broadens the gap between educational groups in their dissatisfaction with education and feelings of misrecognition. Results show that (1) the less educated are more likely to feel misrecognized and dissatisﬁed with their education than the higher educated, and (2) in countries where education is more central, the education gap in feelings of misrecognition is substantially larger.

result of superior talent and effort (e.g. Bourdieu 1984;McNamee and Miller 2009). In this way, the social and economic hierarchies that result from educational outcomes are likely to be seen as "fair," which leaves those at the bottom of the educational ladder few opportunities for the development of a positive social identity and a sense of self-worth. The resentment that results from this experience of being banned to the margins of society are thought to constitute a fertile ground for the support for radical politics (e.g. Hall 2017, 2019).
Although the previous arguments have been put forward before, the empirical relationship between education and feelings of (mis)recognition, however, has scarcely been investigated (but see Gidron and Hall 2019). Research shows that the less educated are generally aware of their lower status position (Van Noord et al. 2019). However, to acknowledge that one has relatively low status remains vague and does not reveal the concrete meaning one gives to one's status experience. Therefore, in this paper we focus on feelings of misrecognition, that is, the extent to which people have the feeling that they do not play a meaningful role in society, that they possess a (stigmatized) identity that is looked down upon, and feel less valued than other social groups. We supplement our focus on feelings of misrecognition with dissatisfaction with education, which reflects to what extent people are dissatisfied with their own education. Together these outcomes allow us to investigate a crucial outcome of the growth of schooled societies.
We use comparative data from three waves (2007,2011,2016) of the European Quality of life surveys (Eurofound 2017; N¼65,802) that includes data on 36 European countries to empirically answer our research questions.

The (Silent) Educational Revolution
The diagnosis of the growth of the schooled society breaks with the more conventional view that represents education as an institution that merely follows the demands of other institutions (e.g., the labor market) or simply reproduces society (e.g. Baker 2014:1-19;Meyer 1977). The advent of mass education, whereby successive generations, on average, have longer schooling than their parents was accompanied by a tendency to consider education a universal right and essential for full human development. Taken together, these elements constitute the foundation of the growth of a more general culture of education which holds that "formal education is the best way to develop all humans and their capacities" (Baker 2014: 2). The concern here is an intensifying tendency (Davies and Mehta 2018: 84). Indeed, precisely because educational credentials act as a gatekeeper on the labor market, for example, credential inflation drives the further expansion of education. And precisely because people spend more time in education, education also further penetrates other realms of society creating an "increasingly thick web of school-society connections" (Davies and Mehta 2018: 84). For this reason, education is sometimes even described as a "greedy institution," which has become so dominant that the educational revolution itself has largely remained unnoticed by many scholars (see also Kuppens et al. 2018). In this paper we do not aim to document the process of the growth of the schooled society itself, but rather to focus on a crucial outcome of this process, namely, the sociopsychological consequences of living in such a society for those who were not successful (or less successful) in education: the less educated. In the next section we further discuss the particularities of education-based status processes and awareness.

The Hidden Injuries of Schooled Societies
In her work on recognition and redistribution, Fraser (2003) highlights how modern societies are characterized by both an economic hierarchy, revolving around economic inequalities and, as such, issues of redistribution, and a status hierarchy, revolving around cultural meanings, prestige, esteem, and, as such, issues of recognition. This Weberian distinction between class and status is useful in understanding the status effects of education. Education has strong economic effects on individuals, determining in large part their chances in, for example, the occupational structure (Blau and Duncan 1967). Education also affects individuals in a cultural sense: imbuing them not only with values and norms but also ideas of worthiness and respect, that is, status or recognition in the eyes of others (Lamont 2018;Ridgeway 2014), which in turn legitimize participation in meaningful societal matters, such as politics (Fraser 2003;Spruyt et al. 2018). While especially in the case of education, economic and status hierarchies are inextricably linked and mutually reinforcing factors (Layte and Whelan 2014), it is important, at least on a conceptual level, to distinguish between the economic and status effects of education. Status effects, as Fraser (1997) rightly emphasizes, can have material effects on individuals and groups, but they originate from cultural logics, rather than market or economic logics. The key question then becomes what is the cultural logic that makes education so potent for status effects?
Two elements are crucial for the effect of education on status. First, at the cultural level, the educational revolution has narrowed the definition of valuable skills and knowledge to the skills and knowledge transmitted and valued in education (Baker 2018:77). In schooled societies, academic intelligence, defined here as general higher-order cognitive skills, becomes an objective in itself and preferred over and above the specific, applied, vocational skills and knowledge that are acquired outside schools: "Academic intelligence has become such a dominating cultural construct that it shapes our ideas about general human intelligence" (Baker 2014:43).
Second, one of the defining features of modern education is that it authoritatively classifies individuals (Baker 2014;Meyer 1977). Crucially, such authority or legitimacy is not a static fact that is imposed upon people, but a practice. Indeed, several elements in how schooling works fuels a representation, which holds that educational success and failure are solely a matter of merit. Current educational practice, for example, is characterized by repeated and detailed testing, comparing, and ranking. As individuals go through the education system, at various moments they are made aware that they "belong" to one or the other level of education (Domina, Penner, and Penner 2017). In education systems with early stratification, this process of allocating and classifying people based on "tested" merit starts at the age 10 or 11 (Brunello and Checchi 2006). Increasingly, the tests used for this classifying not only focus on educational outcomes, but also pretend to measure pupils and students' scholastic potential, that is, their intelligence or learning potential. In this way, scholastic success is "essentialized," that is, represented as the result of superior and natural qualities (Swartz 2013). All this is further reinforced by the existence of (limited) social mobility in education which tends to conceal the often-large social inequalities in educational outcomes (Bourdieu 1998: 21-22). In this way, education individualizes and essentializes meanings of success and failure. It is also this feature that strongly fuels the idea that educational differences are "fair" differences (Tannock 2008). This element was already noticed by Sennett and Cobb, who, in their seminal study on the hidden injuries of social class, referred to educational credentials as "badges of ability" which are extremely hard to repudiate (Sennett and Cobb 1972: 64). These two elements render educational classification powerful and potentially very damaging for the status of the less educated.

The Inescapable Classification of Education
Another element of the potential status consequences of the growth of schooled societies is that education has long-lasting effects. Educational classification is not something that loses its relevance as soon as one has graduated and found a job. Therefore, it is reasonable to assume that the awareness of one's educational level has an enduring quality that can be found among individuals who have long since completed their educations.
First of all, the (growing) centrality of education in modern societies ensures that education is often publicly visible. Education is now the primary pathway for both social reproduction of inequality and social mobility (Hout and DiPrete 2006), which evidences the primary role education plays in labor markets. This primary role of education is compounded by the tendency to "educationalize" societal problems, where education is seen as the "universal solution" for all kinds of individual and societal problems (Depaepe and Smeyers 2008;Labaree 2008). As such, education takes a central and publicly visible position in organizing modern societies, from labor markets to people's personal lives.
Secondly, not only is education as an institution visible in the public sphere, but people are also (made) aware of what category they fall in, and additionally what the (practical) connotations are of educational categories. Long before graduation, students are aware that certain educational tracks are more prestigious than others (Spruyt, Van Droogenbroeck, and Kavadias 2015). After going through the education system, the obtained qualifications are relevant as credentials to social locations, primarily in the labor market, long after graduating (Kingston et al. 2003). In everyday life, one is regularly reminded of one's educational level so that people are, at least on a non-discursive level, aware of which behaviors or tastes one can expect from people of a certain educational level (Bennett et al. 2009;Bourdieu 1984;Stubager 2009). One concrete manifestation of the latter is found in the observation that similarity in education increases the rate of both sending and replying to initial contacts on online dating platforms (Skopek, Schulz, and Blossfeld 2011). This specific observation underscores the more general assumption that people are able to classify themselves at least in relation to these patterns of behavior -if not explicitly -as less or higher educated.
The preceding arguments conjure up the idea that in so-called "schooled societies" (1) people cannot escape from educational classification and (2) this classification entails moral judgements about people's worth and position in society. This raises the question about its consequences. Baker (2014: xii) is clear on this: "[T]he emerging schooled society takes no prisoners: increasingly one either plays the education game or risks being marginalized with a wounded self-image." In this study we focus on that outcome of schooled societies by assessing the relationship between people's educational position and two indicators of low social status: feelings of misrecognition and dissatisfaction with one's education.

R E S E A R C H S T R A T E G Y
Regarding the empirical research, this paper has two key aims. First, we are interested in feelings of misrecognition or dissatisfaction with education among educational groups as a form of education-based status awareness. We aim to demonstrate that the effects of education on feelings of low status cannot fully be reduced to (the lack of) economic resources associated with specific educational levels. Second, we aim to introduce a dynamic element in the analysis. The growth of schooled societies is a long-term process. There are no data that allow us to empirically study such long-term within-country evolutions. Therefore, we search for systematic between-country variation. More specifically, we assess whether the strength of the relationship between education and feelings of low status differs between countries in terms of the share of the higher educated in a country. Specifically, we assess whether education is more divisive in countries with a larger share of the higher educated and whether education is thus more central.

Indicators of Low Status
We focus on two indicators of feelings of low status. First, feelings of misrecognition is a general indicator of the awareness of low status. "(Mis)recognition" is a term that is commonly used to mean two different things. In the Bourdieusian sense, misrecognition means the extent to which individuals are (un)able to recognize how much their social position and practice are produced by societal structures, due to the naturalization of these societal structures (i.e., a form of false consciousness). However, in another sense misrecognition refers to the fact that people perceive misrecognition of their (cultural) worth, commonly associated with the work of Taylor, Fraser, and Honneth (Fraser and Honneth 2003;Honneth 1996;Taylor 1994), and this is the way we use the term. We conceptualize feelings of misrecognition as the extent to which people have the feeling that they do not play a meaningful role in society, possess a (stigmatized) identity that is looked down upon in society, and feel less valued than those in other social groups. We assume that individuals are aware of their (lower) status position, although they need not be necessarily cognizant of all social processes that are directly responsible for it. Hence, with this conceptualization we, importantly, do not prioritize one cause of low status over another (e.g., low status due to a lower income, or low status due to a lower education). However, we can determine how strongly these feelings correlate with different individual characteristics, and hence determine to what extent these feelings are related to education.
In order to allow us to focus more precisely on education, we also look at individuals' dissatisfaction with their education. The awareness among people of educational classification that accompanies the status differences between educational groups makes it likely that when people are aware of their (lower) status position that education is (partly) responsible for this position. Individuals are thus relatively likely to make a connection between their educational level and their status position. Hence, we posit that those who feel that they lack recognition in society are more likely to state that they are dissatisfied with their education. In this way, dissatisfaction with education measures a sense of worth, or self-esteem, due to education. To be sure, dissatisfaction with education can of course arise from more sources than the lower status associated with a lower educational level, such as a low quality of education or prospects on the labor market. At the same time, it is reasonable to assume that the low cultural value associated with being less educated will affect people's dissatisfaction with their education. As such, we treat dissatisfaction as an important indicator of education-specific awareness of low status, even though the dissatisfaction with education could arise from other factors as well. While misrecognition taps the awareness of low status in the eyes of others, dissatisfaction with education assesses how people themselves think about their education. In this paper we consider feelings of misrecognition and dissatisfaction with one's education as two (indirect or subtle vs. more direct and blatant) measures of an experienced low (education-based) status.
In studying the consequences of experienced low educational status, we will take indicators of the strictly economic situation of respondents into account. Education has both material and cultural or status effects. These material effects can, in their turn, also affect one's status: low income is itself also a stigmatized identity (Easterbrook et al. 2019). Given our focus on the drawbacks of the schooled society, however, it is important to assess whether education has additional status effects, besides its economic effects, on feelings of misrecognition.

Cross-national Variation in the Centrality of Education
Our approach posits that the educational classification gains its power largely due to the central role education plays in modern societies (Baker 2014;Meyer 1977). To the extent that education can be considered a global institution (Baker and LeTendre 2005), we expect to find a strong general pattern. That said, it is likely that the effect of education on feelings of misrecognition is moderated by the centrality of education in a country (Van Noord et al. 2019). For instance, as educational expansion increases the marginalization of the less educated and alters educational norms, it may increase the stigmatization of the less educated as incapable (Solga 2002). Or conversely, the greater visibility of the higher education system (and simultaneously, greater visibility of higher educated people) may lead to a norm of being higher educated, rather than an exception. There is no single particular way to measure the extent of the centrality of education, as it consists of multiple dimensions. Many of these dimensions are difficult to catch in a (quantitative) indicator, especially one that can be used for cross-national research. However, previous research has indicated that the share of those who are higher educated in the population can function as a good proxy for the centrality of education in societies (Baker and LeTendre 2005;Van Noord et al. 2019). With this measure of share of higher educated, we assess whether the strength of the relationship between education and feelings of misrecognition or dissatisfaction with education is stronger in countries that can be characterized more strongly as "schooled societies."

Data and Measures
To answer our research questions, we rely on data from the European Quality of Life Survey (EQLS, Eurofound 2017) which is a European wide, nationally representative survey, held roughly every four In the Shadow of the Schooled Society 5 years (most recent wave: 2016). Since several variables of interest are not used in the 2003 wave, we exclude this wave, and only use data from the last three waves: 20072012. This gives 18,586 respondents in 2007, 24,470 in 2012, and 22,746 in 2016. The total number of respondents is 65,802, after listwise deletion. As some countries are only present in one wave, there are 36 unique countries in the dataset with 98 year-country combinations (henceforth called samples). For the analyses that include macro level variables, we have, due to limited data availability, only 70 samples across 25 countries.
Our main three measures of interest are education, feelings of misrecognition, and dissatisfaction with education. Education is operationalized into three levels (lower, middle, and higher) based on ISCED levels: 0-2 for less educated, 3-4 for middle educated, 5-8 for higher educated (these levels correspond to primary, secondary, or tertiary levels of education). Feelings of misrecognition are based on two items, listed below. The original battery of items in the EQLS also contains a third item focusing on feelings of being looked down upon. While this is an important component of misrecognition (Layte and Whelan 2014), the item wording focuses explicitly on "job situation or income." We leave this item out of the scale in order to not have the scale focus on specific reasons for feelings of misrecognition. 1 "I feel left out of society" "I feel that the value of what I do is not recognized by others" The scale is an average score of the two items, each tapping into a related but different component of misrecognition (i.e., a personal difficulty of perceiving one's place in society as meaningful vs. an assessment of the judgement of others). We then z-score standardized this measure, so that higher scores reflect stronger feelings of misrecognition. Since this correlation between both items (Pearson's r: 0.384) is rather low, we have also run all analyses with the items separately (see Tables A3 and A4 in the online appendix). These analyses show there are no substantial differences between the two items. Therefore, we present only the results of the feelings of misrecognition scale with both items.
Dissatisfaction with education is based on a single item: "Could you please tell me on a scale of 1 to 10 how satisfied you are with each of the following items, where 1 means you are very dissatisfied and 10 means you are very satisfied? 'Your education.'" This item is z-score standardized and recoded so that higher scores indicate a stronger dissatisfaction with education. Dissatisfaction with education correlates 0.204 with feelings of misrecognition.
Beyond these variables of interest, we have two sets of individual-level control variables. The first are the demographic controls of gender and age. Gender is a dichotomous variable where 0 means male and 1 female. Age is a continuous variable running from 25-75. We removed all respondents under the age of 25 as they are likely to have not finished their education yet. The second set of controls are economic controls intended for the analyses that aim at distinguishing between cultural and economic bases of status and recognition. These economic controls consist of income and objective material deprivation. Income was measured continuously and was corrected for differences in price levels (PPP euros) in the original dataset. We recoded this variable into sixteen equal-frequency (counted within country-wave samples) categories to deal with outlier/extremely high values. This variable is included also as a quadratic term in our models to deal with non-linear effects of income. Objective material deprivation consists of three scales measuring whether respondents are able to afford certain basic purchases and the quality or luxury of their living accommodation. The questions for these scales are listed below. The full list of items per question is listed in the online appendix (Table A5). The response categories for the items are all dichotomous: yes/no. All three scales range from 0-1 and are coded in such a way that higher scores reflect more objective material deprivation (Cronbach's alpha's: 0.64; 0.84; 0.60). The scales are entered separately in the analyses.
"Has your household been in arrears at any time during the past 12 months, that is, unable to pay as scheduled any of the following?" (Two items) "There are some things that many people cannot afford, even if they would like them. For each of the following things on this list, can I just check whether your household can afford it if you want it?" (Six items) "Do you have any of the following problems with your accommodation?" (Six items) We have also added contextual variables for the sample level (that is, added for each wave within each country). First of all, the share of the higher educated. This variable is used as an explanatory variable for the analyses seeking to explain differences between countries. Share of higher educated is used here as a proxy of the centrality of education in society (Baker 2014;cf. Van Noord et al. 2019).
It measures to what extent the institution of education has taken a central role in society. Specifically, it denotes the percentage of individuals between the ages of 25-65 who have attained a tertiary diploma/degree (OECD 2019).
We also add macro level control variables. These are GDP per capita, general government total expenditure (henceforth government expenses; IMF 2018), and income inequality (Gini index; World Bank 2019). Earlier research has demonstrated that income inequality affects status anxiety levels of different income groups (Layte and Whelan 2014); hence, we use these economic control variables in the analyses seeking to explain country variance in feelings of misrecognition and dissatisfaction with education. Table A1 in the online appendix lists all data sources of these contextual (control) variables. Additionally, Table 1 lists the descriptive statistics for all variables and Table 2 gives an overview of the scales.

R E S U L T S
The presentation of the results is organized in two subsections. First, we assess the relationship between education and our two dependent variables: feelings of misrecognition and dissatisfaction with education. We do this with and without taking into account indicators of respondents' material/financial position to assess whether the level of education explains variation over and above those explained by the more purely economic indicators. Second, we explore the variation between countries and try to explain this by the share of higher educated. All multilevel models presented consist of three levels, with individuals nested in waves, which are nested in countries (Schmidt-Catran and Fairbrother 2016).

The Added Explanatory Power of Education
Theoretical accounts of education have distinguished between not only the potential economic but also the social status effects of education. We are primarily interested in the extent to which education is indeed a basis for social status, as it is through their position in the status hierarchy that individuals are likely to experience feelings of misrecognition. We ran multilevel regressions with feelings of misrecognition and dissatisfaction with education as the dependent variables. The models proceed stepwise. Model 1 contains only education, model 2 adds the demographic control variables age and gender, and model 3 adds the two economic variables: income and the three objective material deprivation scales. Table 3 shows the results of the analyses with feelings of misrecognition as the dependent variable. Table 4 presents the results for dissatisfaction with education. Figure 1 graphically presents the first three models. Model 1 in Table 3 reveals that education is indeed negatively related to feelings of misrecognition, with the higher educated being less likely to feel misrecognized than the less educated. Model 2 shows that, although age has a negative relationship with feelings of misrecognition In the Shadow of the Schooled Society 7 (i.e., the young report stronger feelings of misrecognition), the point estimate of education is not substantially affected by controlling for age. In contrast, when in model 3 the economic variables are entered, the coefficient of higher education on feelings of misrecognition decreases to about a third of the coefficient in model 2 (from b ¼ -0.389 to b ¼ -0.120). In this way, the part of education that we can attribute to the economic correlates of education is slightly more than two times larger than the "non-economic" part of education. The economic variables themselves have large effects; hence, people's feelings of misrecognition are also strongly tied to their economic position. 2 The same analyses are also done for dissatisfaction with education (Table 4). The results do not entirely mirror each other. Again, model 1 shows that education has a negative relationship with dissatisfaction with education. Unsurprisingly, less educated people are less satisfied with their education than those with a higher education. Model 2 shows that these results are robust for the control variables gender and age, although they indicate that women are less satisfied with their education than  Note: Spearman correlation for feelings of misrecognition is the Spearman correlation between the two feelings of misrecognition items. OMD is objective material deprivation. Though the Spearman correlations between these scales are listed, only the separate scales are used in the analyses, not the OMD meta-scale. men, and older people are more satisfied. Adding the four economic control variables barely affects the relationship between education and dissatisfaction with education (model 3). Though all four are strongly related to dissatisfaction with education, including them in the model decreases the relationship of education only by about a tenth. Low economic position does positively affect people's dissatisfaction with their education, suggesting that people recognize the role education plays in their lower economic position. This, however, does not affect the difference between educational groups. Dissatisfaction with education seems to be primarily driven by an understanding of the meaning of people's education per se, rather than what it achieves in economic markets. Indeed, while the size of education's effect on feelings of misrecognition is much smaller than the size of the effect of the economic controls, when it comes to dissatisfaction with education, the effect size of education is larger than of the economic controls.

Systematic Between-country Variance in Feelings of Misrecognition
The preceding section focused on the general pattern of the education-misrecognition relation. In this subsection we explore variation between countries, and, as such, include a more dynamic element in the analysis that refers to the long-term (cultural) developments that the thesis of the growth of schooled societies departs from. The country-level intercepts in Tables 3 and 4 indicate that both feelings of misrecognition and dissatisfaction with education contain significant variance at the country and sample level. We now aim to explain this variance. The share of higher educated is used as a proxy for the centrality of education in society. Table 5 presents the predicted means and simple effects of the two models that each include an interaction term with education and share of higher educated. Model 1 has feelings of misrecognition as its dependent variable, model 2 dissatisfaction with education. The results of the full multilevel model can be found in Table A2 in the online appendix. All these models control for the basic demographic variables of age, gender, income, and the three objective material deprivation scales and also GDP per capita, income inequality, and government expenses.
The results show that the share of higher educated indeed moderates the relationship of education with feelings of misrecognition: the higher the share of higher educated, the larger the difference between the higher and less educated in feelings of misrecognition (see also figure 2). In samples with a small share of higher educated (one standard deviation below the mean; 20.2 percent), this difference equals -0.017 (p¼0.509): the mean feeling of misrecognition of the higher educated is 0.017 lower than for the less educated. This difference in countries with a larger share of higher educated (one standard deviation above the mean; 36.8 percent) is -0.234 (p<0.001), again higher educated have a lower mean for feelings of misrecognition than the less educated. The difference between these two coefficients equals -0.218 (p<0.001). This means that the difference between the higher and less educated is 0.218 standard deviations larger in samples with a large share of higher educated, compared to samples with a small share. 3 This increase in difference between lower and higher educated is roughly equally due to decreased feelings of misrecognition of the higher educated and increased feelings of misrecognition of the less educated. The share of higher educated does not moderate the relationship between education and dissatisfaction with education (p¼0.908). Educational differences in dissatisfaction with education do not significantly differ in samples with either a small or a large share of higher educated. So, while the share of higher educated has a large impact on the difference between the higher and the less educated in feelings of misrecognition, this is not the case for dissatisfaction with education.

Exploratory and Robustness Analyses
While we treat our data as essentially cross-sectional, we do have multiple waves over time at our disposal. This would allow us to distinguish between short-term and long-term changes by leveraging the fact that our data contain both changes over time (nine years) within a country and differences between countries, which partly reflect changes over a longer time period. Therefore, we disaggregate the macro-level variables, such as the share of higher educated, in order to distinguish within-country In the Shadow of the Schooled Society 11 changes from between-country differences. We do treat this analysis as exploratory, since the evolution of societies towards schooled societies is a long-term trend dating back to at least the 1960's, and for some countries even around 1870 (Meyer, Ramirez, and Soysal 1992). As such, this cultural change is probably not linear and less detectable on the short-term. However, looking at the shortterm might tell us something about the current dynamics. Technically, distinguishing within-country from between-country differences is done by centering macro-level variables at the country mean.  Feelings of misrecognition of the three educational groups, across share of higher educatedNote: Shaded areas denote 95% confidence intervals.
Thus, we have country mean-centered all macro-level variables, and included the country mean as a separate variable in the analysis (Astell-Burt and Feng 2018; Bell, Jones, and Fairbrother 2018;Thijs, Keim, and Geerlings 2019).We have done this for both our primary macro-level variable (share of higher educated) and for our macro-level control variables. In this section we also test the robustness of our analyses by not only interacting education with the share of higher educated, but also with the macro-level control variables.
Firstly, the disaggregated share of higher educated variables indicate that both the country mean and the country mean-centered variables significantly moderate the relationship of educational level with feelings of misrecognition (see Table 6, Model 1). So, both short-term changes over time and absolute differences between countries are associated with a widening of the misrecognition gap between the less and higher educated. To further test the robustness of the interaction effects we also add the other control variables as interaction terms with education. In this way, we can test whether the interaction between share of higher educated is not actually due to interactions of education with other macro-level variables. We have also disaggregated these control variables into country meancentered and country mean variables. Model 2 includes these interaction terms. The significant moderation of the effect of higher educated on feelings of misrecognition by country mean-centered share of higher educated disappears. Further investigation showed that it was the inclusion of the country mean-centered GDP per capita moderation that caused the country mean-centered share of higher educated moderation to lose its significance. This means that even though short-term country-level increases in the share of higher educated are related to an increase in the education effect on misrecognition, the moderating effect of an increase in the share of higher educated cannot be separated from the moderating effect of an increase in GDP per capita. This is perhaps not surprising, given the limited number of countries. Indeed, none of the country mean-centered variables or interactions were significant, suggesting that this model simply had a lack of statistical power due to the combination of few countries and several correlated predictors. In any case, and consistent with the main analyses presented earlier, differences between countries in share of higher educated are still robustly associated with a larger misrecognition gap between the higher and less educated, and this gap is significantly larger in countries with more higher educated.

D I S C U S S I O N A N D C O N C L U S I O N
This paper engages with the literature that studies the (moral) implications of the growth of schooled societies (Baker 2014;Meyer 1977;Solga 2002;Tannock 2008;Van Noord et al. 2019) and how these societal changes affect education as a source of identity and status (Spruyt and Kuppens 2015;Stubager 2009). To that end, we explored feelings of misrecognition and dissatisfaction with education among educational groups in Europe. Both outcomes tap into different aspects of how education is related to status. We used the European Quality of Life Survey (Eurofound 2017) to investigate multiple questions.
First, we found that, when controlling simultaneously for four indicators of economic resources, education still affects feelings of misrecognition, though the effect size was relatively small. The effect size was much larger for dissatisfaction with education, probably because a more direct measure of low status due to education is naturally more related to education. The relationship between education and dissatisfaction with education was only barely influenced by economic resources, hence people's dissatisfaction with education seems to be predominantly status related. An important note on the comparisons, however, is that it is difficult, if not impossible, to entirely disentangle the status effects from the material effects of education. Indeed, their entanglement means that both are likely to be affected by the other at every step of the way. Our results do imply that interpreting education as a simple class indicator underestimates the centrality of education in contemporary societies and is, in this sense, misguided. Indeed, in analogy with Baker's (2014) distinction one could argue that rather than being a secondary indicator for social class, in schooled societies people's education level In the Shadow of the Schooled Society 13 has become a primary indicator of people's overall social position that reconciles both cultural and economic status aspects.
Second, we explored cross-national variation in the relationship between education and feelings of misrecognition and dissatisfaction with education. In modern society education has taken such a central role that these societies can be called "schooled societies" (Baker 2014). Though the "educationalization" of society is a multi-layered process, which goes beyond the simple expansion of education, the share of the higher educated is a good measure of the extent of this transformation as the expansion of higher education is one of the primary ways in which education attains such a central role. We found that the share of higher educated across countries is associated with a substantial increase in difference between higher and less educated in feelings of misrecognition. No such pattern is found for dissatisfaction with education, however. Regardless of the share of higher educated, the differences between educational groups in dissatisfaction stay the same. The differences in dissatisfaction between the higher and less educated are already large, and unexplained by economic factors; hence, this can clearly be seen as a status indicator that is already predominantly and autonomously determined by education. Perhaps this explains why a further expansion of education does not increase differences between higher and less educated. For feelings of misrecognition, however, we arrive at a similar conclusion as Van Noord et al. (2019: 669): "as education becomes more pervasive and widely shared, rather than leveling social differences, ironically it also becomes more distinctive and diagnostic in distinguishing people along group lines." In this paper, we expand on Van Noord et al. (2019), who investigated people's perception of their social rank and its relation to educational level. The authors show that while people are aware of their lower status position, education's stratifying role in society is generally seen as legitimate. We expand on this by focusing on the extent to which people are satisfied with their current position. When combined, both studies suggest that while the less educated are generally aware of their lower status position and express personal dissatisfaction with that position, this does not, as Van Noord et al. (2019) demonstrate, immediately lead to a questioning of the legitimacy of the education system. The importance of education in determining who gets ahead in society is still experienced as something legitimate. Against this background, it is remarkable that dissatisfaction with individuals' educational level, in this study, displays considerably more disagreement between educational groups than Van Noord et al.'s (2019) measure of the legitimacy of the education system. While we do not know of any studies that combine individuals' dissatisfaction with their own educational attainment with dissatisfaction with the education system as a whole, this does raise the question of whether feelings of lower status will lead people to dissatisfaction with education, precisely when education is seen as legitimate. In this way, dissatisfaction with one's own education can be seen as a strategy of individual mobility (Ellemers, Knippenberg, and Wilke 1990;Tajfel and Turner 1979): one problematizes one's own education rather than the (source of) inequality itself. In general, our measure of dissatisfaction with education raises questions on the genesis of this dissatisfaction. We posit that this contains a strong status component, because as the less educated are more aware of the low status generally accorded to a lower educational level, they would become more dissatisfied with their education. However, this dissatisfaction could also have other sources. So, further research could focus on how people come to be dissatisfied with their education and to what extent this can be seen as cultural.
Our analyses distinguishing between the economic and non-economic effects at least suggest that this dissatisfaction is not simply the result of different economic outlooks of the less and higher educated. We have focused on the broader categories of less, middle, and higher educated, corresponding to primary, secondary, and tertiary education. There is reason to believe that due to educational expansion of higher education, distinction processes could start to emerge within higher education. With increased access to higher education, the need for economically more privileged families and individuals to distinguish themselves leads them to secure "quantitatively similar but qualitatively superior educational credentials" (Torche 2011:768). Investigating this dimension of "horizontal stratification" (Gerber and Cheung 2008) requires data on, among others, fields of study and, more importantly, the prestige of schools/universities which are not often collected in many large-scale surveys that also include subjective social status measures. The way this qualitative or horizontal stratification could play a role in feelings of misrecognition also relates to whether such feelings are mostly the result of concrete experiences by those with lower social status rather than the positive experiences with higher status.
At the same time, it is important to note that the current processes described here are not without resistance. The "intensifying logic" described earlier in the theoretical section is also accompanied with a "logic of resistance": "an aggressive skepticism of the cultural authority of schooled professionals and the cognitive authority of schooled credentials" (Davies and Mehta 2018:84). This resistance is visible in the variation of dependence on the educational prestige of the most selective degrees and institutions across different kinds of industries in their selection of elites (Brint et al. 2020;Brint and Yoshikawa 2017) and also in challenges of populists in the political arena against educational elites and "posttruth"-discourses (Davies and Mehta 2018; Yl€ a-Anttila 2018). In any case, the implications of all these mechanisms are potentially strong, both when related to education but also with respect to feelings of misrecognition more generally. Recent research into the effects of subjective social status shows strong effects on various outcomes and behaviors, including health (Adler et al. 2000;D'Hooge, Achterberg, and Reeskens 2018) and attitudes towards the self and society (Brown-Iannuzzi et al. 2014;Payne 2017). Potentially, effects of perceived misrecognition might be larger than subjective social status. Feelings of misrecognition are not just about the subjective positioning in a status hierarchy; they also focus on the negative and experienced consequences of low status-that is, stigmatization and discrimination-and concrete experiences of being denied respect (Lamont 2018;Lamont, Beljean, and Clair 2014). In this way, with feelings of misrecognition we might be primarily capturing the negative experiences with low status. These matters also demonstrate that it is important to distinguish between different forms of status experience, and, hence, different indicators of social status.
The effects of misrecognition on attitudes towards democracy and politics can be large. Misrecognition, according to Fraser (2000:23), comes about, in large part, through "institutionalized patterns of cultural value that constitute one as relatively unworthy of respect or self-esteem" that consequently impede parity of participation in politics and institutions in general. A "status-based cultural conflict" (cf. Noordzij, Van der Waal, and De Koster 2019) between higher and lower educated is then indeed likely to underlie attitudes towards (political) institutions, with feelings of misrecognition playing a potentially key role. As the difference in misrecognition between the higher and less educated was larger in countries with a larger share of higher educated, it bears remembering that these status differences are not static and might increase, even alongside processes that are often seen as positive (such as the educational expansion to larger groups of individuals who previously did not have access to higher education).
In this paper we have focused on the socio-psychological consequences of living in a schooled society for those who were not successful (or less successful) in education. It should, however, be stressed that as (1) the growth of schooled society is an intensifying process, and (2) educational credentials are a positional good, the effects of living in a schooled society are unlikely to be limited to the less educated. In fact, Markovits (2019: xx) shows how the increasing importance of education combined with the privatization of education (and the associated growing importance of the reputation of the schools and universities), leads to a horse race that starts to divide the middle class from the elite. Among the latter it instills a "collective anxiety" driven by fear of not measuring up. In this way, education-based meritocracy ". . . simultaneously excludes most people and damages the few that it admits." All in all, then, this paper shows that education and feelings about one's education are relevant in questions of status and recognition. In contemporary "schooled societies" (cf. Baker 2014) which tend to "educationalize" social problems (cf. Labaree 2008): "[E]verybody knows and everybody knows that everybody else knows that education rules in modern society" (emphasis added) (Kingston et al. 2003: 55). Our research suggests that this awareness has important implications for people's subjective status position and feelings of recognition, especially in countries where the share of higher educated is larger. Indeed, the great irony is that although education is regularly presented as the "great equalizer," that representation itself may become an independent element in the reproduction of educational differentials in status and recognition.