Sage Journals: Discover world-class research

Abstract

The authors present results from a multifactorial survey experiment conducted on a sample of Polish working-age adults querying their evaluations of just pay for men and women. The experiment was designed to investigate whether gender pay gaps are considered fair and to what extent they can be explained by status-based processes. Respondents rated the fairness of the earnings of hypothetical men and women with different ages and parenthood statuses in 21 occupations. The occupations differed in prestige and gender composition. The authors find that perceived just earnings for women are marginally lower than they are for otherwise identical men, regardless of respondents’ characteristics. Just gender pay gaps are largest among incumbents of high-prestige occupations and in male-dominated and gender-balanced occupations. The authors also find that motherhood penalties are not acceptable. In fact, respondents view both men and women with more than two children as deserving of a parenthood bonus.

Keywords

gender beliefs just earnings gender pay gaps

Introduction

Popular perceptions of fair pay contribute to the stability of inequality (Buchel, Luijkx, and Achterberg 2021; Jost, Banaji, and Nosek 2004). Some scholars (Auspurg, Hinz, and Sauer 2017; Thébaud and Pedulla 2016) have argued that stalled progress in closing gender pay gaps (GPGs)—with such gaps defined as differences between men and women in their actual earnings—is due in part to the persistence of gendered beliefs (Ridgeway 2011) as well as to the tendency to justify existing inequalities (Sauer 2020). People need to make sense of disparities around them, and they are therefore predisposed to perceiving the status quo as “good, fair, natural, desirable, and even inevitable” (Jost et al. 2004:887). The alignment of the distribution of rewards with normative expectations solidifies the social order and its perception as just, proper, and valid (Hegtvedt, Johnson, and Watson 2016). Actual pay disparities are, therefore, often treated as expected and fair inequalities, regardless of whether those inequalities are based on ascribed or achieved statuses (Auspurg et al. 2017; Jasso, Shelly, and Webster 2019; Sauer 2020). In short, people tend to justify existing inequalities on the basis of various types of available information.

Studies of just earnings provide mixed results regarding the magnitude of and mechanisms behind just GPGs (JGPGs), which are understood here as pay differences between men and women that are considered fair, acceptable, or appropriate. Depending on factors including the characteristics of respondents, measurement methods, country, and time, some studies have found significant JGPGs favoring men (Auspurg et al. 2017; Jasso and Webster 1997), whereas others have found no gaps (Jasso and Webster 1999) or even gaps favoring women (Shamon and Dülmer 2014; for general reviews, see Jasso et al. 2019; Sauer 2020). These inconsistencies may to some degree be explained away by cohort and country effects; the studies we cite were conducted in different countries and welfare regimes (e.g., Germany, the United States), at different times (e.g., the 1970s, 1990s, and 2010s), and on different types of respondents (e.g., samples of students and general population). Yet we argue that such discrepancies between the results of different studies may also be attributed in part to differences in the characteristics of the occupations that might strengthen (or attenuate) gender effects. Thus, in our study we aimed to investigate the degree to which GPGs are treated as fair and appropriate—that is, the size of JGPGs—with a particular focus on the moderating role of gender segregation and occupational prestige. To our knowledge, although studies of just pay evaluations sometimes include occupational prestige as a control variable (e.g., Sauer 2020) or the interaction between gender and occupational gender composition (e.g., Auspurg et al. 2017), they do not explore in depth the role of occupational characteristics, such as prestige or gender composition, in shaping JGPGs. This omission may be contributing to the conflicting results.

Our experiments were conducted in Poland, which is an interesting context for studying JGPGs. The gap in the employment rate between Polish men and women is among the largest of all European Union (EU) countries with male employment rates approximately 15 percent higher than those of women (Eurostat 2022). In addition, gender-based occupational segregation in Poland has increased in recent years (Hurley et al. 2021), and gender traditionalism is relatively high compared with that in Western Europe and even in some Central and Eastern European countries (Andringa, Nieuwenhuis, and Van Gerven 2015; Weziak-Bialowolska 2015). Yet the unadjusted GPG in Poland is one of the narrowest among EU members; at the time of our experiment (conducted in 2017), the Polish gap amounted to 7 percent (Eurostat 2023). Studies also estimate moderate motherhood wage penalties and relatively higher fatherhood bonuses (Cukrowska-Torzewska and Lovasz 2016).

Several factors may help explain this constellation of observations. One such factor may be the policies regarding maternity and parental leaves. Fully paid maternity leave has been guaranteed in Poland since 1976, although the duration of leave has varied over time (with a minimum of 20 weeks for a single birth). Women are also entitled to partially paid, job-protected parental leave that can extend the postnatal break to one year (which is the norm for new mothers) and job-protected unpaid leave that can be extended up to 35 months (1 month is guaranteed for the other parent). In 2010, paternity leave and parental leave sharing options were introduced in Poland, although the uptake of parental and paternity leaves by men is minimal (van Belle 2016). It has been argued that long maternity and parental leaves have negative effects on both GPGs and on women’s employment because of, among other things, loss of human capital by mothers, occupational sorting on the basis of gender and family plans, and employers’ reluctance to hire and promote women and mothers (Budig, Misra, and Boeckmann 2016; Hipp 2020). Investigating how gender, parenthood, and different job characteristics affect the extent to which respondents accept pay inequalities in such a setting is therefore important for policy as well as for developing sociological theory.

Our contribution to the sociology of gender inequalities is therefore threefold. First, our study serves as a test of generalizability of claims regarding the relationship between gender, age, and just rewards. Such relationship has not been experimentally established in Poland. We hypothesize that women will be seen as deserving lower pay than equally qualified men and that the size of this JGPG will be larger among older vs. younger workers; our results support these hypotheses. Second, by studying the extent to which the effect of gender on just rewards expectations is moderated by the characteristics of occupations, we provide new insights regarding the mechanisms underlying status-based reward differentials. We predict and demonstrate that JGPG are largest among incumbents of high-prestige occupations and in male-dominated and gender-balanced occupations. Finally, we conduct our experiment in Poland, a country with policies that promote traditional gender roles and push women out of the labor market. Such policies, in turn, serve to reinforce status beliefs about women and assign childrearing responsibilities to mothers.

We begin our argument with an overview of the research on the existence and sources of GPGs. We then present the theoretical framework of the status organizing processes guiding our hypotheses regarding JGPGs (Berger et al. 1998; Ridgeway 2011). Finally, we describe our methods and the results of our tests. We conclude by specifying the theoretical implications of our study as well as its limitations.

Existence and Sources of GPGs

Women earn less than men across the world. Even after controlling for a host of direct and indirect productivity measures and job characteristics, pay differences are significant and substantial (Blau and Kahn 2017; Hurley et al. 2021; Ponthieux and Meurs 2015). GPGs persist despite the proliferation of policies aimed at promoting gender equality. In the EU, numerous directives and initiatives, and the growing presence of women (including highly educated women) in labor markets, have not caused gaps to narrow as quickly as was expected. In 2019, the average unadjusted GPG in EU-27 countries was 14.1 percent and had not changed significantly in a decade. GPGs are wider among employees with more human capital (i.e., older workers with tertiary or posttertiary education and/or longer job tenure) and among persons on full-time and permanent contracts. Those patterns are usually interpreted as indicative of the cumulative effect of gender and parenthood-related career interruptions (Hurley et al. 2021; Mari and Luijkx 2020).

Evidence that having children has a negative effect on women’s wages and career prospects is ubiquitous. Mothers in the United States (e.g., England et al. 2016), Western Europe, Central Europe (e.g., Cukrowska-Torzewska and Matysiak 2020), Eastern Europe (e.g., Nizalova, Sliusarenko, and Shpak 2016), and Australia (e.g., Cooke 2014) suffer wage and occupational prestige losses (Abendroth, Huffman, and Treas 2014). The size of motherhood wage penalties (i.e., wage losses incurred by mothers compared with women with no children) vary between countries and welfare and tax regimes (Budig et al. 2016), as well as by income and occupational prestige (England et al. 2016), number of children, and occupation (Cukrowska-Torzewska and Matysiak 2020). In general, mothers incur an earnings penalty, whereas men (excluding those taking up paternity leaves; Rege and Solli 2013) accrue fatherhood premiums and bonuses (wage increases among fathers compared with men with no children).

Gender- and parenthood-based pay gaps are interrelated. Both demand and supply side explanations posit that shared beliefs regarding “ideal” or “preferred” worker attributes affect the job-matching process (Ridgeway 2011:101–109). Biases related to stereotypical masculine or feminine traits and motivations, as well as gendered expectations regarding parental involvement, contribute to actual or perceived differences in, or loss of, human capital by potential mothers, mothers, and women in general (Combet and Oesch 2019; Hipp 2020; Mari and Luijkx 2020). Studies show that there are discrepancies in hiring offers for mothers and young nonmothers in countries with more traditional gender roles (e.g., Spain; González, Cortina, and Rodríguez 2019) and/or those living under conservative welfare systems (e.g., Germany; Hipp 2020).

In addition, there is evidence of voluntary and involuntary gender sorting into occupations on the basis of actual or perceived compatibility with gendered definitions of skills and childcare responsibilities (Thébaud and Charles 2018). Consequently, there is an overrepresentation of women in lower paying occupations. Furthermore, earnings decrease in occupations with a growing female presence (Mandel 2013, 2018), lending more support to the argument of gender-based perception of work. In other words, GPGs are, to a large extent, explained by occupational gender segregation (Cukrowska-Torzewska and Matysiak 2020). The latter, in turn, has been linked to gendered definitions of skills, gender essentialism (Levanon and Grusky 2016) and gender-based devaluation of labor in female-dominated occupations (Levanon, England, and Allison 2009; Murphy and Oesch 2015). All these processes—that is, gender-based occupational labor devaluation, gender sorting into occupations, and perceived mismatch between family and work responsibilities—are driven in large part by evaluative beliefs regarding men and women’s qualities, skills, motivations, and suitable roles that also underlie perceptions of the fairness of pay.

Theory and Hypotheses

Status, Reward Expectations, and JGPGs

Evaluative gender beliefs are a type of status belief. Status beliefs refer to shared evaluations of the abilities, traits, and societal worth of particular social categories, including not only gender, but also age, ethnicity, and other characteristics. They affect relative expectations of performance and of rewards’ allocation in work-related contexts (Berger et al. 1998). Gender is a characteristic that people automatically use for categorization, and it universally activates status beliefs (Ridgeway 2011). As long as individuals are aware of the “gender frame,” regardless of whether they endorse it personally, they judge themselves and others (and expect others to judge them and others) according to the stereotypes contained within this frame (Correll et al. 2020; Ridgeway 2011; Ridgeway, Korn, and Williams 2022). This means that both men and women tend to associate men with certain, more highly valued, traits and abilities. They form generally higher performance expectations for men compared with women, and evaluate men as more competent, capable, and praiseworthy.

Such disparate expectations shape perceptions of fair pay, which in turn can be manifested in patterns of behaviors (e.g., decisions about promotions and raises) that generate and perpetuate inequality. More specifically, gender as a status characteristic is linked to general and specific performance expectations that translate into job opportunities, chances to contribute at the workplace, and evaluations of contributions and performance. Information regarding a person’s “productivity” or achievements is perceived through the lens of their gender. Persons assessing the abilities of male and female employees, whether those assessors are managers, other employees, or independent actors, will often apply stricter standards of competence to members of devalued groups (e.g., women and mothers; Correll et al. 2020; Foschi, Ndobo, and Faure 2019; Fuegen et al. 2004; Ridgeway et al. 2022; Thébaud 2015). They will, in other words, require more evidence of skills and general fitness for the job and tolerate less evidence of incompetence.

As the allocation of rewards ought to be consistent with status expectations, persons with lower status (women) will be seen as deserving of lower earnings than similarly qualified individuals with higher status (men) (Auspurg et al. 2017; Berger et al. 1998).

Hypothesis 1: Both male and female respondents will assign lower just rewards to the vignette profiles (VPs) of women compared with those of men.

Age is another characteristic that people automatically use to categorize people, and age beliefs are also a type of status belief that affect performance expectations and evaluations and perceptions of just rewards (Kelley, Soboroff, and Lovaglia 2017). In just earnings research, the status value of age has received only indirect attention. However, Jasso and Webster (1999) showed that gender inequality in just pay stems in part from unequal evaluations of the value assigned to men and women’s job experience, which they measure using age as a proxy.

Age beliefs disadvantage older individuals of both genders in task settings but the application of double standards for competence will be more detrimental to evaluations of just earnings of older women as compared with older men. That is, assuming that both age and gender activate status beliefs related to the level of competence, if men and women have the same work experience, a woman’s age will be treated as less indicative of a given level of competence (as a harsher standard of competence is applied to women) than it will for her male colleagues (the higher status group). Research also indicates that status beliefs influence expectations of who should occupy the role of authority (Ridgeway et al. 2022) and that disparities in the representation of men and women in workplace authority positions are substantial. These disparities manifest well before the transition to parenthood and increase with age regardless of parenthood status and education level (see Figure 3 in Stojmenovska and England 2021).

This is because harsher ability standards for devalued groups are more likely to be activated in higher stakes situations, for example, final hiring decisions rather than decisions to shortlist candidates (Biernat and Fuegen 2001), decisions to hire full-time rather than part-time employees (Mari and Luijkx 2020), or decisions to hire or promote those with more rather than less experience (as the former will typically require higher wages). For example, one resume study of age-based discrimination in Poland demonstrated that younger applicants (28 years old) had a higher likelihood of receiving a callback than did older applicants (52 years old), with the younger men being the most preferred candidates among the four age and gender combinations and older women the least preferred (Baszczak et al. 2021). In short, older age—being a status characteristic and acting as a proxy for years of labor market experience—activates harsher standards for women. Consequently, the perceived JGPG will wider among older employees than it will be among younger employees.

Hypothesis 2: JGPGs will be wider among older VPs then among younger ones.

The status value of gender may also be amplified by expectations linked to motherhood and fatherhood roles (Mari and Luijkx 2020). Almost half of Polish European Social Survey respondents¹ indicated that it is essential for men and women to have children. Insofar as gender and gender-specific parenthood status beliefs are interrelated, disentangling their effects on performance and reward expectations poses a challenge. On one hand, descriptive and prescriptive gender beliefs assign caregiving and communal traits to women, regardless of whether they have children (e.g., Prentice and Carranza 2002). In conservative gender regimes such as Switzerland’s, actual GPGs between young men and women are significant before the onset of family-related division of household labor, even controlling for skills, job characteristics, and attitudes toward work and family (Combet and Oesch 2019). These gaps persist throughout the professional career.

On the other hand, being a mother is perceived as especially incongruent with being an “ideal worker” (Benard and Correll 2010; Correll, Benard, and Paik 2007; Ridgeway and Correll 2004; Williams 2021). Motherhood is generally believed to be inconsistent with being career-oriented, as well as with human capital investments and accumulation (Correll et al. 2007; Fuegen et al. 2004). Fatherhood, in contrast, activates beliefs about breadwinning roles (Thébaud and Pedulla 2016). Mothers are held to stricter standards of performance, whereas fathers are perceived as especially motivated to work hard. Parenthood expectations, in other words, are normative and gender-specific; they interact with expectations regarding gender and—when parenthood status becomes salient—accentuate gendered beliefs about competences and motivation in work settings (Thébaud and Pedulla 2016).

Hypothesis 3: JGPGs will be wider among VPs of mothers and fathers than among nonparent men and women.

Status Incongruity and JGPGs

Our final two hypotheses relate to the potential importance, discussed in the introduction, of the role of occupational characteristics in shaping JGPGs. In this context, it bears noting that status-based theories, on which all of our hypotheses are based, belong to the incongruity approach to explaining gender inequities (Manzi 2019). According to incongruity models, the greater the perceived mismatch between gender and the skills viewed as crucial to a job, the more inequality in pay will be tolerated. The assumed incongruity (between gender and a given job) helps explain why women and mothers (but not men and fathers) are seen as especially incompatible with the ideal-worker model. It also helps explain why larger actual gender pay inequalities have been observed at the higher end of the occupational prestige scale (where perceived job commitment is especially valued) and in male-dominated jobs (in which incongruity on the basis of gender essentialism is especially salient).

More specifically, gender carries information not only about the general worth of being a man versus being a woman, and the general abilities, competences, dispositions, motivations, associated with these categories, but also about specific abilities and competences that are relevant to performing certain types of jobs (i.e., stereotypically masculine or feminine) (cf. Thébaud and Charles 2018). Depending on whether an occupation is gender balanced or gender segregated, the gender of the person whose earnings are being evaluated will activate diffuse or specific performance expectations, respectively. In the case of the gender-segregated occupations, gender will additionally activate specific performance expectations and hence will have a stronger impact on reward expectations (Ridgeway 2011; Tak, Correll, and Soule 2019).

Being a woman in low-prestige and female-dominated occupations, however, will confer less advantage (in terms of perceived just earnings) over men in those same jobs than that granted a man in high-prestige, male-dominated, and gender-balanced occupations over their female counterparts in those jobs. For instance, women who are childcare workers will be seen as deserving moderately higher earnings than men who are childcare workers, but men who are CEOs will be seen as deserving far more than women who are CEOs.

Hypothesis 4: JGPGs will be wider in high-prestige occupations than in low- or medium-prestige occupations.

Hypothesis 5: JGPGs will be wider in male-dominated occupations than in gender-balanced occupations and wider in gender-balanced than in female-dominated occupations.

Finally, we do not formulate separate hypotheses regarding the impact of level of education or years of schooling. In Poland, educational and certification requirements for many occupations are highly regulated, especially at the higher end of the prestige spectrum. Including information about formal level of education alongside information about occupation or occupational tenure would therefore be redundant and would produce too many implausible combinations in our vignettes. In analytical terms, such variables would be collinear with occupational prestige and minimum experience.

Methods

Vignette Design

A multifactorial (vignette) experiment was conducted in 2017 in Poland on a nationwide sample of men and women aged 20 to 64. Each vignette described a fictitious person in terms of their gender, age, occupation, years of on-the-job experience, and number of children. The respondent’s task was to assess how appropriate the randomly assigned amount of earnings was for the person described in the vignette (Auspurg and Hinz 2015; Jasso and Webster 1999; Wallander 2009).

The vignettes were preceded by a general description of the task as well as of the “employee” population. The purpose of this instruction was to reduce unobserved heterogeneity of the vignettes. That is, we explained that all VPs are full-time employees, live in a medium-sized town, have always been positively evaluated by their supervisors, and were never unemployed.

Vignettes were created by crossing levels of four characteristics: gender, age, occupation, and parenthood status. Ages of VPs varied between 30 and 65 years, in 5-year increments. Manipulating parenthood status posed a challenge, as respondents indicated during pretests that explicitly specifying “childlessness” made them aware of the purpose of the study. Steiner, Atzmüller, and Su (2016:61) emphasize that when respondents guess the objective of the study, they may adjust their evaluations on the basis of social desirability or political correctness (e.g., “equal pay for equal work”). However, the degree to which status information is available for processing or is relevant to the situation determines if it will be activated (Berger et al. 1998). This means that for motherhood and fatherhood to have an impact on justice evaluations in a vignette study, that study needs to visibly delineate parents from nonparents. Following other experimental designs testing the effects of motherhood (e.g., Bygren, Erlandsson, and Gähler 2017; Correll et al. 2007; Mari and Luijkx 2020), we therefore split our vignettes into “no information” about the family status, parent of one child, parent of two or more children. In the latter two categories of vignettes, we made parenthood status salient by explicitly using the terms “mother” or “father.” The category of two or more children was further subdivided into individuals with two, three, or four children. Vignettes in this category were assigned two children with probability 0.50, three children with probability 0.35, and four children with probability 0.15. That way, our vignettes looked more “realistic,” as describing individuals as having “two or more children” might have appeared artificial to respondents.

Occupational titles included in the vignettes were selected from among the occupations listed in the Polish Sociological Classification of Occupations (PSCO) in such a way as to cover the whole spectrum of occupational prestige. PSCO uses similar methodology and occupational dimensions as the U.S. National Opinion Research Center or the International Standard Classification of Occupations (ISCO). As its main purpose is to serve as a sociological classification relevant in Poland rather than simply a statistical one, however, it adds additional dimensions of, for instance, educational requirements and ownership of workplace. Each occupation in the PSCO is assigned a prestige score varying between 0 and 100 (cf. Domański, Sawiński, and Słomczyński 2009). We first grouped all the occupations in three broad categories of prestige: (1) high, comprising all occupations with a prestige score of 70 or higher; (2) low, comprising all occupations with prestige score lower than 40; and (3) medium, comprising all other occupations. Then, from each prestige category we selected seven occupational titles to include in the vignettes, choosing them in a manner that allows to minimize the within category and maximize the between category differences. The occupations used in the vignettes and their corresponding prestige scores (in parentheses) are as follows: professor of early education (97.3), anesthesiologist (87.8), actor in a theatre (86.9), radio news presenter (81.2), plane pilot (80.6), information technology security specialist (77.5), artistic director in an ad agency (72.7), veterinary technician (66.1), librarian (61.3), train dispatcher (60.9), nurse (55.9), customs officer (54.9), real estate agent (51.4), accountant (50.9), cashier in a supermarket (27.6), mailperson (27.5), waitperson (25.7), hairdresser (21.2), telemarketer (18.0), janitor (10.4), and cloakroom attendant (8.6). The information about the prestige score was not included in the vignette.

This means that with 2 gender categories, 8 age levels, 21 occupational titles, and 3 levels of the parenthood status, we had a total of 1,008 possible combinations. As some of these combinations (e.g., a 30-year-old full professor) were improbable, we removed them from the vignette population. Removing them reduced the total number of available vignettes to 996. As it would be impossible for a single respondent to evaluate all of them, a stratified random sample of vignettes was drawn for each respondent separately. That is, the vignette population was first stratified by occupation and, subsequently, 1 vignette was picked at random from each stratum, resulting in a sample of 21 vignettes comprising the full spectrum of occupational prestige.

The occupations were also classified as male-dominated, female-dominated, or balanced on the basis of data about the proportion of women working in those occupations.² To simplify our tests, the first category (male-dominated) is constituted by occupations with fewer than 30 percent women, and the second category (female-dominated) by occupations with more than 70 percent women. All other occupations were sorted into the third category (balanced). This information about the gender composition or prestige of the occupation was not included in the vignette. Vignettes included only the names of the occupations. However, both the level of prestige and gender composition were included in the analysis, as explained in the next section.

Apart from the four characteristics listed above, individual vignettes also included information about on-the-job experience. That information was added to the vignettes after the vignette population was created. Job experience was calculated by adding a small amount of random variation to a linear transformation of age, the transformation being age minus years of education. Note that, as previously explained, in Poland, mothers are eligible for a maximum of 35 months of job-protected parental leave. During this period, their employment contracts remain valid, allowing them to resume their duties once the leave concludes. Consequently, even a three-year leave does not create a career gap or signify unemployment. Unless a woman explicitly states her parental status or did not have a standard employment contract when she became a mother, her employment history does not suggest any interruptions related to childcare responsibilities.

Respondent’s Task

Respondents were tasked with evaluating the fairness of six net (take-home) per-month earnings amounts for a given vignette. In Poland, net salaries are universally understood to be independent of the number of children employees have. Child tax credits are given out by the government at the end of the tax year, and so although they do affect one’s net earnings, salaries are always interpreted to mean net earnings directly from one’s employer. In addition, child tax credits in Poland are very modest. The amounts were in the Polish currency, listed in a random order. For each vignette, they were drawn from a larger set comprising 20 different amounts. That is, we first created a distribution of earnings matching the properties of the actual income distribution in Poland in terms of mean and variance. Then, in the next step, we drew a set of 20 distinct amounts from that distribution. Finally, a randomly selected subset of these 20 amounts was assigned to each vignette separately.

We used the “indirect method” with multiple amounts per vignette, introduced by Jasso and Webster (1999), on the basis of the assumption that it may be difficult for people to know exactly how much a given person in each occupation should be paid. Despite this uncertainty, they should be able to assess how close a specific amount comes to what they believe would be a fair pay for that person. Thus, by asking how close a series of hypothetical earnings amounts come to what a respondent thinks is a just pay for a vignette, we can estimate the just pay with a greater precision than if we asked about a single amount.

Respondents’ answers were coded on an 11-point scale ranging from −5 (“much too low”) to 5 (“much too high”), with the midpoint of 0 labeled as “appropriate.” In Poland, since the rise of the populist Law and Justice party, the concepts of justice and fairness are highly politicized, so we decided to use a more neutral term, which should not have affected the evaluations (Adriaans et al. 2021).

The example of a vignette is provided in Figure 1. Each screen included one vignette and six amounts, which respondent was asked to drag and drop in the evaluative category corresponding to their judgements. Consequently, although respondents were asked to provide a substantial number of evaluations, the task was not too exhausting for them. They encountered only 21 screens and were tasked with a straightforward arrangement of the amounts on a scale for each vignette and screen. In the Supplemental Material, we provide information about average time needed to evaluate each vignette and vignette sets and reliability checks.

Figure 1.

An example of a vignette used in the study.

The responses are assumed to represent respondents’ estimations regarding how much the specific amount departs from what would be appropriate given the characteristics of the VP. We used the responses to estimate the amount that the respondent considered appropriate for the given vignette (i.e., “just earnings”). The just earnings are specific to each respondent-vignette pair. In our study, there were 2,001 respondents. With 21 vignettes per respondent, the number of all respondent-vignette pairs in that study was 42,021. However, a preliminary inspection of the data revealed that some respondents gave illogical or inconsistent responses in some cases. We classified a set of responses by a given respondent to a given vignette as illogical if (1) the respondent evaluated all amounts assigned to a given vignette as equally appropriate or (2) there were contradictions in how the respondent evaluated amounts assigned to the vignette. After these illogical responses were removed, we were left with a total of 38,183 vignette-respondent pairs. More details about this procedure and reliability checks are provided in the Supplemental Material.

Respondent Sample

Participants were registered members of an Internet panel run by Kantar, one of the leading marketing and opinion research companies in Poland. At the time of the experiment, this panel consisted of 91,000 registered users whose sociodemographic data were verified in a two-step process. The panel reflects the population structure of Poland thanks to purposeful recruitment of groups underrepresented among Internet users. The sample of panel participants was not probabilistic but was designed to match the composition of the Polish working-age population in terms of age, gender, education, and geographical location. Respondents were recruited to this study in a way allowing for fulfilling the demographic quota. For participating, respondents were rewarded according to the panel membership rules. The vignettes were a stand-alone study and not one of many modules in recurring data collection (e.g., omnibus research). The descriptive statistics regarding the vignettes’ population and respondents’ sample are presented in Table A1 in the Supplemental Material.

Analytical Method

To test our hypotheses, we regress the log of just earnings against vignette characteristics. The details of the procedure for estimating just earnings from justice evaluations provided by the respondents are given in the Supplemental Material. We use the logarithmic transformation of just earnings, rather than the untransformed amounts, because the latter has a highly skewed distribution, with the skew equal to 3.9.

The independent variables in the models include vignette characteristics and interactions of those characteristics. Age is coded as a binary variable equal to 1 for those older than 40 years and 0 for those younger than 40 years. In Poland, fewer than 5 percent of births are among women 40 years of age or older (GUS [Statistics Poland] 2016). To simplify our analyses, and assuming that 40 is perceived as an age after which the likelihood of taking maternity and parenthood-related leaves decreases, we divided our employee profiles into two age groups (up to 40 and older than 40).

Parenthood status is a categorical variable that distinguishes VPs with children from those for whom the number of children is “unspecified” (see above); the latter category is the reference category in the regression models. Profiles with children were further coded as those with one child or two children and more than two children. In Poland, since the 1990s, the fertility rate is approximately 1.5, which means that an average woman has either one or two children. We refer to the no-information profiles as “nonparents.”

The proportion of women in an occupation is represented by three categories, male dominated, female dominated, and gender balanced, as specified in the previous section. Occupational prestige is fit as a binary variable coded 1 for occupations of high prestige (i.e., with a PSCO prestige score of 70 or higher) and 0 otherwise.

The simplest model we consider — our baseline model — has the following form:

\begin{array}{l} y_{i j} = α_{0 j} + β_{1} femal e_{i j} + β_{2} ag e_{i j} + β_{3} prestig e_{i j} \\ + β_{4} childre n_{i j} + β_{5} typ e_{i j} + ϵ_{i j}, \end{array}

(1)

where $y_{i j}$ is the natural logarithm of just earnings for VP i in the eyes of respondent j, $ϵ_{i j}$ is the error term, and $α_{0 j}$ is the respondent-specific intercept. As each respondent was asked to evaluate a number of vignettes, our data can be said to form a two-level structure, in which the just earnings amounts are level 1 observations and respondents are level 2 observations. Consequently, we fit a random-intercept model, which allows the intercept to vary across respondents. That is, the intercept in equation (1) can be expressed as $α_{0 j} = β_{0} + u_{0 j},$ where $β_{0}$ is the intercept averaged across all respondents and $u_{0 j}$ is the error term, or random effect, associated with respondent $j$ . After substitutions, our baseline model changes to:

\begin{array}{l} y_{i j} = β_{0} + β_{1} femal e_{i j} + β_{2} ag e_{i j} + β_{3} prestig e_{i j} \\ + β_{4} childre n_{i j} + β_{5} typ e_{i j} + u_{0 j} + ϵ_{i j} . \end{array}

Given the logarithmic transformation applied to our data, all results are expressed on the log scale. The JGPG, or the GPG our respondents believe would be fair, is expressed as the difference between the estimated log of just earnings of women and the estimated log of just earnings of men. The wider the pay gap (i.e., the less women deserve in comparison with men, according to our respondents), the more the estimated pay gap departs from zero in the negative direction.

The baseline model can be used to test hypothesis 1, but it is not sufficient to test all the hypotheses we propose. To test the other hypotheses, we thus extend the baseline model by adding to it interaction terms corresponding to the effects specified in a given hypothesis. To illustrate, hypothesis 2 states that the JGPG is wider in the older age group than in the younger one. This hypothesis predicts, in essence, that the size of the vignette’s gender effect varies by vignette’s age category. To test this hypothesis, we extend the baseline model by including an interaction term between the dummy for being female and the dummy for being in the older age group.

Thus, the model used to test hypothesis 2 has the following form:

\begin{array}{l} y_{i j} = β_{0} + β_{1} femal e_{i j} + β_{2} ag e_{i j} + β_{3} prestig e_{i j} + β_{4} childre n_{i j} \\ + β_{5} typ e_{i j} + β_{6} ag e_{i j} \times f e m a l e_{i j} + ϵ_{i j} + u_{0 j} . \end{array}

The only difference between this model and the baseline is the inclusion of the interaction between age and gender. On the basis of this model, the estimated (log of) just earnings of a woman in the younger age group can be obtained by setting female_ij = 1 and age_ij = 0:

β_{0} + β_{1} + β_{3} prestige + β_{4} children + β_{5} type .

Similarly, the estimated (log of) just earnings of a man in the younger age group can be obtained by setting female_ij = 0 and age_ij = 0:

β_{0} + β_{3} prestige + β_{4} children + β_{5} type .

The difference between the former and the latter formula is equal to $β_{1},$ and it corresponds to the (log of) the just gender gap in the younger age group. In a similar way, we can obtain model-predicted formulas for the (log of) just earnings of a woman and a man in the older age group by setting female_ij = 1, age_ij = 1 and female_ij = 0, age_ij = 1, respectively. Following the same steps as those described earlier, the estimated (log of) JGPG in the older age group can be shown to be equal to $β_{1} + β_{6} .$ On the log scale, hypothesis 2 predicts the JGPG in the older age group to depart further away from zero in the negative direction than the JGPG in the younger age group. Formally, then, hypothesis 2 implies that

β_{1} + β_{6} < β_{1},

which amounts to $β_{6} < 0 .$ In other words, hypothesis 2 is supported when the interaction between being female and being in the older age group is negative. Other hypotheses are treated in a similar way. The results from the various regression models we estimate are shown in Table 1. These results are the basis for the tests of our hypotheses presented in the next section.

Table 1.

Parameter Estimates and Fit Statistics for Mixed-Effects Linear Models Used to Test Hypotheses.

	Model 1	Model 2	Model 3	Model 4	Model 5
Intercept	8.441*** (.007)	8.438*** (.007)	8.441*** (.007)	8.439*** (.007)	8.442*** (.008)
Female	−.014*** (.001)	−.009*** (.002)	−.015*** (.002)	−.011*** (.002)	−.016*** (.003)
Older than 40 years^a	.092*** (.001)	.096*** (.002)	.091*** (.001)	.092*** (.001)	.092*** (.001)
High prestige^b	.263*** (.002)	.263*** (.002)	.263*** (.002)	.268*** (.002)	.263*** (.002)
One or two children^c	.001 (.001)	.001 (.001)	.003 (.002)	.001 (.001)	.001 (.001)
Three or more children^c	.041*** (.002)	.041*** (.002)	.033*** (.003)	.041*** (.002)	.041*** (.002)
Gender balanced^d	−.158*** (.002)	−.158*** (.002)	−.158*** (.002)	−.158*** (.002)	−.156*** (.003)
Female dominated^d	−.080*** (.002)	−.080*** (.002)	−.080*** (.002)	−.080*** (.002)	−.088*** (.003)
Female × older than 40 years		−.008** (.003)
Female × one or two children			−.005 (.003)
Female × three or more children			.017*** (.004)
Female × high prestige				−.010*** (.003)
Female × balanced					−.004 (.004)
Female × female dominated					.016*** (.004)
AIC	−41,436.9	−41,433.3	−41,446.1	−41,438.9	−41,458.5
BIC	−41,351.4	−41,339.2	−41,343.5	−41,344.9	−41,355.9
Log likelihood	20,728.5	20,727.6	20,735.1	20,730.5	20,741.3
Number of observations	38,183	38,183	38,183	38,183	38,183
Number of groups: respid	1,964	1,964	1,964	1,964	1,964
Var: respid (intercept)	.094	.094	.094	.094	.094
Var: residual	.015	.015	.015	.015	.015

Note: AIC = Akaike information criterion; BIC = Bayesian information criterion; respid = respondent.

Reference category: up to 40 years.

Reference category: low- or medium-prestige occupation.

Reference category: no children or unspecified.

Reference category: male-dominated occupation.

p < .01. ***p < .001.

Results

According to hypothesis 1, our respondents view women as less deserving of earnings than men with the same characteristics. With gender represented as a dummy variable coded 1 for female, hypothesis 1 is supported if the main effect of gender in the baseline model is negative. This is precisely what we find (see Table 1). The magnitude of this effect may not be very large in absolute terms—with women seen by our respondents as deserving earnings 1.5 percent lower than those of comparable men—but its sign is consistent with the prediction. We also conducted a split ballot study with two vignette populations that were comparable in all respects except for job experience. That is, vignettes in one population had 5 years more experience, on average, than vignettes in the other population. The estimated JGPGs for these two populations are 1.6 percent and 2.4 percent, respectively, similar to the ones we find in the present study.

Note, however, that hypothesis 1 does not just say that female VPs are seen as deserving lower earnings, on average, than otherwise similar male VPs. It also predicts that they are seen this way by both male and female observers. The latter prediction would be falsified if we found out that male and female respondents differ in their expectations for just earnings for male and female VPs. One way to test the prediction, then, is to split our sample by respondent’s gender and fit the relevant regression model again to each subsample separately. The main effect of being female is estimated at −0.015 in the female subsample (with the corresponding 95 percent interval ranging from −0.019 to −0.011) and at −0.014 (95 percent confidence interval −0.018 to −0.011) in the male subsample.³ Thus, the estimates are negative, as expected, and similar in magnitude in both subsamples, consistent with hypothesis 1. We can thus conclude that the hypothesis is supported by our data.

As regards hypothesis 2, it is supported if the interaction term between the dummy for being female and the dummy for being in the older age group is negative, as explained above. The results are consistent with this prediction: the interaction is negative and statistically significant, implying that the JGPG in the older group is indeed wider than in the younger group, albeit modestly so. Note that this result captures the gendered effect of age net of parental status. It also bears reiterating that this hypothesis and its test refer to situations in which men and women of different ages, regardless of their parenthood status, are full-time workers with the same number of years of employment.

Hypothesis 3 implies that the interaction between being female and parenthood status is negative. Recall that parenthood status is represented in our analysis by two variables: a dummy for having one or two children and a dummy for having more than two children. The interaction between the former dummy and being female is not statistically significant (see model 3 in Table 1). The interaction between being female and having more than two children, in turn, is statistically significant — but it is positive, contrary to what hypothesis 3 predicts. Thus, not only do our results not support hypothesis 3, but we find that both male and female parents of more than two children are viewed as meriting higher earnings.

According to hypothesis 4, the JGPG is expected to be wider in high-prestige occupations than in low- or medium-prestige ones. To test this hypothesis, we fit an extension of the baseline model in which we include an interaction between the dummy for being female and the dummy for working in a high-prestige occupation. Hypothesis 4 is supported if the interaction’s coefficient is negative. This is indeed the case: the estimated JGPG is wider in high-prestige occupations than in low- and medium-prestige occupations combined.

On the basis of hypothesis 5, we expect the JGPG to be the widest in male-dominated occupations and the narrowest in female-dominated ones. To test this prediction, we extend the baseline model by including an interaction between gender and the categories measuring the proportion of women in an occupation (i.e., male dominated, gender balanced, and female dominated). With male-dominated occupations being the reference category in the regression model, the hypothesis is supported if the interactions are positive. Our findings are consistent with this prediction in the sense that the model-predicted JGPG for female-dominated occupations is indeed narrower than the JGPG for male-dominated occupations. As regards gender-balanced occupations, however, the model-predicted JGPG is not significantly different from those for the male occupations.

Except for hypothesis 3, our results are consistent with our predictions. The reported effect sizes are small, however. In large samples such as ours, even very small effects can prove statistically significant. This is certainly a reason to treat them with some caution. We discuss these points further in the next section.

Discussion and Conclusions

In this study we evaluate, first, whether gender disparities in earnings are considered just by Polish working-age respondents. Status-based theories predict that gender beliefs and gender bias in performance expectations affect not only labor market opportunities, but also perceptions of how rewards should be distributed in evaluative settings. This prediction has not been previously evaluated in Poland.

The results of our vignette experiment provide some support for the claim that GPGs are perceived as just because of status-based mechanisms of bias against women. Specifically, our primary hypothesis—that estimated just earnings for women will be lower than they are for otherwise identical men—is supported by our empirical results; both male and female Polish respondents view such a gap as appropriate. Moreover, consistent with our prediction, respondents judge as appropriate wider GPGs among older employees (VPs) than among younger ones. This observation further supports a claim that status-based gender bias is pervasive, not limited to childbearing age, and linked to the application of double standards for competence.

The second goal of our study was to explore the extent to which the effect of gender on just rewards expectations is moderated by the characteristics of occupations, namely occupational prestige and the degree of gender balance. As discussed above, previous studies of JGPGs have not considered in-depth the role of occupational prestige or gender segregation as contextual or moderating variables. We hypothesized that JGPGs will be the widest in high-prestige and male-dominated occupations. This is exactly what we observed. The gaps, as predicted, are also wider in gender-balanced occupations than in female-dominated ones.

JGPGs in gender-balanced occupations, in contrast, are similar to those observed in male-dominated occupations. This is somewhat inconsistent with our predictions, which posit that JGPGs will be wider in male-dominated than in gender-balanced occupations. One possible explanation for this result is that the gender composition of an occupation does not perfectly align with its gendered task typification. Gender may play an organizing role in any situation where nothing prevents it from having such an effect. That is, information is treated as task-relevant unless there is specific information to the contrary. In both male-dominated occupations (e.g., commercial airline pilot) and gender-balanced occupations demanding masculine traits (e.g., director in an advertising agency), being a man may confer specific expectation advantages.

An even simpler explanation, one which also points to a limitation of our study, is methodological in nature. Because of data availability, we used information about gender composition at a more aggregate (three-digit) level of occupational classification. As “extreme segregation” occurs at the most detailed occupational level, our measure of occupational gender composition may not capture segregation sufficiently to reveal the differences in JGPGs between gender-balanced and male-dominated occupations (Levanon and Grusky 2016). Notwithstanding, our study indicates that gender-specific characteristics of occupations indeed constitute an important variable moderating the effects of gender beliefs on just earnings evaluations.

Although the JGPGs we report vary under hypothesized conditions, they are small in magnitude. Note, however, that in Poland the unadjusted GPG is among the lowest in the EU (about 7 percent at the time of the experiment). Perceptions of how rewards should be distributed are linked to actual inequalities. Although we cannot assume that actual GPGs constitute a ceiling for JGPGs, there is evidence of a positive association between the size of actual and justly perceived gender pay differences (Sauer 2020). Moreover, our test was also carried out using the most conservative parameters, including vignettes that implied full-time employment, no career breaks, and universally positive performance evaluations of all (hypothetical) employees. The purpose was to capture the “pure” effect of gender beliefs, but this likely served to reduce the magnitude of our estimates.

Finally, our results indicate that Polish respondents generally consider salaries too low, regardless of vignette characteristics. The mean monthly salary considered just by our respondents is almost 60 percent higher than the mean actual salary. Some gender effects may thus be masked by respondents’ apparent use of need-based rules for evaluating just earnings (Shamon and Dülmer 2014). A similar situation was observed in Ukraine, where a general perception of salaries being too low obscured patterns of the application of different justice principles (Gatskova 2013). Future studies in settings where typical salaries are considered too low might further understanding of the relationship between aggregate and between-subgroup perceptions of just pay.

The third and final goal of this study was to test the proposition that in Poland, a country with traditional gender roles and family-related policies, status beliefs about women are intertwined with those about mothers. We hypothesized that motherhood and fatherhood statuses will accentuate gender differences, and that JGPGs will be wider among mothers and fathers than among men and women. Our results are inconsistent with this hypothesis. According to our respondents, otherwise identical mothers and fathers both should receive higher earnings than female and male employee profiles without children. This might be linked to historically high participation of Polish mothers in the labor market during the “communist era” because of the policy of full-employment and a relatively well-developed network of childcare facilities under that regime (Mikucka 2013). Although the postcommunist transformation was a setback for women’s labor participation and state support for working mothers, the “Polish working mother” myth might have persisted. That is, mothers may not be afforded special status, as working mothers are perceived as the norm. Investigating the content of gender and motherhood stereotypes should be the next step in explaining determinants of fair pay beliefs in Poland.

We examined, however, whether it might have been also due to our choice not to state explicitly that vignettes had no children, which we did because our pretests indicated that such information made the purpose of the experiment obvious to respondents. Our supplemental analyses did not support this possibility, but we of course cannot reject it. Similarly, it is possible that it is not being a mother or a father or number of children (as we assumed), but rather the age of the dependent children and/or partnership status, that determines whether gender-specific parenthood beliefs affect just pay evaluations.

That our results generally, albeit weakly, supported our hypotheses regarding GPGs and largely failed to do so for our sole prediction about parenthood gaps is an important contribution to research on determinants of stability of gender pay inequalities. Moreover, our results indicate that occupational characteristics play an important role in just earnings evaluations.

Supplemental Material

sj-docx-1-srd-10.1177_23780231241227158 – Supplemental material for Status and Just Gender Pay Gaps: Results of a Vignette Study

Supplemental material, sj-docx-1-srd-10.1177_23780231241227158 for Status and Just Gender Pay Gaps: Results of a Vignette Study by Kinga Wysieńska-Di Carlo and Zbigniew Karpiński in Socius

Footnotes

Acknowledgements

We thank the participants of the 2018 annual meeting of the American Sociological Association in Philadelphia and the 2019 Motherhood Penalty Seminar in Warsaw for their invaluable input regarding the research design and analysis. We are especially grateful to Matthew Di Carlo and Marta Kołczyńska for their comments on the earlier versions of this article, which helped to improve it greatly. We would also like to express our gratitude to the two anonymous reviewers for their thoughtful comments and suggestions.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Polish National Science Center (grant 2015/19/B/HS6/03169).

ORCID iDs

Kinga Wysieńska-Di Carlo

Zbigniew Karpiński

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Kinga Wysieńska-Di Carlo is a senior research fellow in the Research Group on Comparative Analysis of Social Inequality of the Institute of Philosophy and Sociology at the Polish Academy of Sciences. Her research interests include status and legitimacy processes, social stratification and inequality, international migration, and discrimination. She has a PhD in sociology from Jagiellonian University in Krakow, Poland.

Zbigniew Karpiński is an associate professor in the Institute of Philosophy and Sociology at the Polish Academy of Sciences and a research officer in European Commission’s Joint Research Centre in Seville. His research interests include educational stratification, meritocracy, the reproduction of social inequality, and formal modeling of social processes. He has a PhD in sociology from the Institute of Philosophy and Sociology at the Polish Academy of Sciences in Warsaw.

References

#Abendroth

Anja-Kristin

Huffman

Matt L.

Treas

Judith

. 2014. “The Parity Penalty in Life Course Perspective: Motherhood and Occupational Status in 13 European Countries.” American Sociological Review 79(5):993–1014.

#Adriaans

Jule

Liebig

Stefan

Sabbagh

Clara

Jasso

Guillermina

. 2021. “What’s in a Word? Just vs. Fair vs. Appropriate Earnings for Self and Others.” Social Justice Research 34(4):397–427.

#Andringa

Wouter

Nieuwenhuis

Rense

Gerven

Minna Van

. 2015. “Women’s Working Hours: The Interplay between Gender Role Attitudes, Motherhood, and Public Childcare Support in 23 European Countries.” 35(9/10):582–99.

#Auspurg

Katrin

Hinz

Thomas

. 2015. Factorial Survey Experiments. Thousand Oaks, CA: Sage.

#Auspurg

Katrin

Hinz

Thomas

Sauer

Carsten

. 2017. “Why Should Women Get Less? Evidence on the Gender Pay Gap from Multifactorial Survey Experiments.” American Sociological Review 82(1):179–210.

#Baszczak

Łukasz

Trojanowska

Maja

Wincewicz-Price

Agnieszka

Zyzik

Radosław

. 2021. “Dyskryminacja ze Względu na Wiek na Polskim Rynku Pracy—Diagnoza” [“Age-Based Discrimination on the Polish Labor Market—A Diagnosis”]. Warsaw, Poland: Polski Instytut Ekonomiczny.

#Benard

Stephen

Correll

Shelley J.

. 2010. “Normative Discrimination and the Motherhood Penalty.” Gender & Society 24(5):616–46.

#Berger

Joseph

Fisek

M. Hamit

Norman

Robert Z.

Wagner

David G.

. 1998. “Formation of Reward Expectations in Status Situations.” Pp. 121–54 in Status, Power and Legitimacy, edited by Berger

Zelditch

Jr . New Brunswick, NJ: Transaction.

#Biernat

Monica

Fuegen

Kathleen

. 2001. “Shifting Standards and the Evaluation of Competence: Complexity in Gender-Based Judgment and Decision Making.” Journal of Social Issues 57(4):707–724.

10.

#Blau

Francine D.

Kahn

Lawrence M.

. 2017. “The Gender Wage Gap: Extent, Trends, and Explanations.” Journal of Economic Literature 55(3):789–865.

11.

#Buchel

Ondrej

Luijkx

Ruud

Achterberg

Peter

. 2021. “Objective and Subjective Socioeconomic Status as Sources of Status-Legitimacy Effect and Legitimation of Income Inequality.” Political Psychology 42(3):463–81.

12.

#Budig

Michelle J.

Joya

Misra

Irene

Boeckmann

. 2016. “Work–Family Policy Trade-Offs for Mothers? Unpacking the Cross-National Variation in Motherhood Earnings Penalties.” Work and Occupations 43(2):119–77.

13.

#Bygren

Magnus

Erlandsson

Anni

Gähler

Michael

. 2017. “Do Employers Prefer Fathers? Evidence from a Field Experiment Testing the Gender by Parenthood Interaction Effect on Callbacks to Job Applications.” European Sociological Review 33(3):337–48.

14.

#Combet

Benita

Oesch

Daniel

. 2019. “The Gender Wage Gap Opens Long before Motherhood. Panel Evidence on Early Careers in Switzerland.” European Sociological Review 35(3):332–45.

15.

#Cooke

Lynn Prince

. 2014. “Gendered Parenthood Penalties and Premiums across the Earnings Distribution in Australia, the United Kingdom, and the United States.” European Sociological Review 30(3):360–72.

16.

#Correll

Shelley J.

Stephen

Benard

Paik

. 2007. “Getting a Job: Is There a Motherhood Penalty?” American Journal of Sociology 112(5):1297–1339.

17.

#Correll

Shelley J.

Weisshaar

Katherine R.

Wynn

Alison T.

Wehner

Joanne Delfino

. 2020. “Inside the Black Box of Organizational Life: The Gendered Language of Performance Assessment.” American Sociological Review 85(6):1022–50.

18.

#Cukrowska-Torzewska

Ewa

Lovasz

Anna

. 2016. “Are Children Driving the Gender Wage Gap?” Economics of Transition and Institutional Change 24(2):259–97. doi:10.1111/ecot.12090.

19.

#Cukrowska-Torzewska

Ewa

Matysiak

Anna

. 2020. “The Motherhood Wage Penalty: A Meta-analysis.” Social Science Research 88–89:102416.

20.

#Domański

Henryk

Sawiński

Zbigniew

Słomczyński

Kazimierz M.

. 2009. Sociological Tools Measuring Occupations. Warsaw, Poland: IFiS.

21.

#England

Paula

Bearak

Jonathan

Budig

Michelle J.

Hodges

Melissa J.

. 2016. “Do Highly Paid, Highly Skilled Women Experience the Largest Motherhood Penalty?” American Sociological Review 81(6):1161–89.

22.

#Eurostat. 2023. “Gender Statistics: Statistics Explained.” Retrieved January 29, 2023. https://ec.europa.eu/eurostat/databrowser/view/EARN_GR_GPGR2/default/table?lang=en

23.

#Foschi

Martha

Ndobo

André

Faure

Alice

. 2019. “Assessing and Blocking Double Standards for Competence.” Pp. 19–45 in Advances in Group Processes, Vol. 36, edited by Thye

S. R.

Lawler. Bingley

E. J.

, UK: Emerald.

24.

#Fuegen

Kathleen

Biernat

Monica

Haines

Elizabeth

Deaux

Kay

. 2004. “Mothers and Fathers in the Workplace: How Gender and Parental Status Influence Judgments of Job-Related Competence.” Journal of Social Issues 60(4):737–54.

25.

#Gatskova

Kseniia

. 2013. “Distributive Justice Attitudes in Ukraine: Need, Desert or Social Minimum?” Communist and Post-communist Studies 46:227–41.

26.

#González

M. José

Cortina

Clara

Rodríguez

Jorge

. 2019. “The Role of Gender Stereotypes in Hiring: A Field Experiment.” European Sociological Review 35(2):187–204.

27.

#GUS [Statistics Poland]. 2016. “Małżeństwa oraz Dzietność w Polsce” [“Marriage and Birth Statistics in Poland”]. Retrieved July 26, 2022. https://stat.gov.pl/files/gfx/portalinformacyjny/pl/defaultaktualnosci/5468/23/1/1/malzenstwa_i_dzietnosc_w_polsce.pdf.

28.

#Hegtvedt

Karen A.

Johnson

Cathryn

Watson

Lesley

. 2016. “Social Dynamics of Legitimacy and Justice.” Pp. 425–44 in Handbook of Social Justice Theory and Research, edited by Sabbagh

Schmitt

. New York: Springer.

29.

#Hipp

Lena

. 2020. “Do Hiring Practices Penalize Women and Benefit Men for Having Children? Experimental Evidence from Germany.” European Sociological Review 36(2):250–64.

30.

#Hurley

John

Grubanov-Boskovic

Sara

Bisello

Martina

Vacas

Carlos

Fana

Marta

Fernández-Macías.

Enrique

2021. “European Jobs Monitor 2021: Gender Gaps and the Employment Structure.” Luxembourg: Publications Office of the European Union.

31.

#Jasso

Guillermina

Shelly

Robert

Webster

Murray

. 2019. “How Impartial Are the Observers of Justice Theory?” Social Science Research 79:226–46.

32.

#Jasso

Guillermina

Webster

Murray

Jr.

1997. “Double Standards in Just Earnings for Male and Female Workers.” Social Psychology Quarterly 60(1):66–78.

33.

#Jasso

Guillermina

Webster

Murray

Jr.

1999. “Assessing the Gender Gap in Just Earnings and Its Underlying Mechanisms.” Social Psychology Quarterly 62(4):367–80.

34.

#Jost

John T.

Banaji

Mahzarin R.

Nosek

Brian A.

. 2004. “A Decade of System Justification Theory: Accumulated Evidence of Conscious and Unconscious Bolstering of the Status Quo.” Political Psychology 25(6):881–919.

35.

#Kelley

Christopher P.

Soboroff

Shane D.

Lovaglia

Michael J.

. 2017. “The Status Value of Age.” Social Science Research 66:22–31.

36.

#Levanon

Asaf

England

Paula

Allison

Paul

. 2009. “Occupational Feminization and Pay: Assessing Causal Dynamics Using 1950–2000 U.S. Census Data.” Social Forces 88(2):865–91.

37.

#Levanon

Asaf

Grusky

David B.

. 2016. “The Persistence of Extreme Gender Segregation in the Twenty-First Century.” American Journal of Sociology 122(2):573–619.

38.

#Mandel

Hadas

. 2013. “Up the Down Staircase: Women’s Upward Mobility and the Wage Penalty for Occupational Feminization, 1970–2007.” Social Forces 91(4):1183–1207.

39.

#Mandel

Hadas

. 2018. “A Second Look at the Process of Occupational Feminization and Pay Reduction in Occupations.” Demography 55(2):669–90.

40.

#Manzi

Francesca

. 2019. “Are the Processes Underlying Discrimination the Same for Women and Men? A Critical Review of Congruity Models of Gender Discrimination.” Frontiers in Psychology 10:469.

41.

#Mari

Gabriele

Luijkx

Ruud

. 2020. “Gender, Parenthood, and Hiring Intentions in Sex-Typical Jobs: Insights from a Survey Experiment.” Research in Social Stratification and Mobility 65:100464.

42.

#Mikucka

Małgorzata

. 2013. “The Transition to Insecurity: Employment Dynamics and Its Sociodemographic Differentiation.” International Journal of Sociology 42(4):71–99.

43.

#Murphy

Emily

Oesch

Daniel

. 2015. “The Feminization of Occupations and Change in Wages: A Panel Analysis of Britain, Germany and Switzerland.” SOEPpaper No. 731. Social Science Research Network. Retrieved from https://ssrn.com/abstract=2564612.

44.

#Nizalova

Olena Y.

Sliusarenko

Tamara

Shpak

Solomiya

. 2016. “The Motherhood Wage Penalty in Times of Transition.” Journal of Comparative Economics 44(1):56–75.

45.

#Ponthieux

Sophie

Meurs

Dominique

. 2015. “Gender Inequality.” Pp. 981–1146 in Handbook of Income Distribution, Vol. 2, edited by Atkinson

A. B.

Bourguignon

. New York: Elsevier.

46.

#Prentice

Deborah A.

Carranza

Erica

. 2002. “What Women Should Be, Shouldn’t Be, Are Allowed to Be, and Don’t Have to Be: The Contents of Prescriptive Gender Stereotypes.” Psychology of Women Quarterly 26(4):269–81.

47.

#Rege

Mari

Solli

Ingeborg F.

. 2013. “The Impact of Paternity Leave on Fathers’ Future Earnings.” Demography 50(6):2255–77.

48.

#Ridgeway

Cecilia L.

2011. Framed by Gender. New York: Oxford University Press.

49.

#Ridgeway

Cecilia L.

Correll

Shelley J.

. 2004. “Motherhood as a Status Characteristic.” Journal of Social Issues 60(4):683–700.

50.

#Ridgeway

Cecilia L.

Korn

Rachel M.

Williams

Joan C.

. 2022. “Documenting the Routine Burden of Devalued Difference in the Professional Workplace.” Gender & Society 36(5):627–51.

51.

#Sauer

Carsten

. 2020. “Gender Bias in Justice Evaluations of Earnings: Evidence from Three Survey Experiments.” Frontiers in Sociology 5:22.

52.

#Shamon

Hawal

Dülmer

Hermann

. 2014. “Raising the Question on ‘Who Should Get What?’ Again: On the Importance of Ideal and Existential Standards.” Social Justice Research 27(3):340–68.

53.

#Steiner

Peter

Atzmüller

Christiane

Dan

. 2016. “Designing Valid and Reliable Vignette Experiments for Survey Research: A Case Study on the Fair Gender Income Gap.” Journal of Methods and Measurement in the Social Sciences 7(2):52–94.

54.

#Stojmenovska

Dragana

England

Paula

. 2021. “Parenthood and the Gender Gap in Workplace Authority.” European Sociological Review 37(4):626–40.

55.

#Tak

Elise

Correll

Shelley J.

Soule

Sarah A.

. 2019. “Gender Inequality in Product Markets: When and How Status Beliefs Transfer to Products.” Social Forces 98(2):548–77.

56.

#Thébaud

Sarah

. 2015. “Status Beliefs and the Spirit of Capitalism: Accounting for Gender Biases in Entrepreneurship and Innovation.” Social Forces 94(1):61–86.

57.

#Thébaud

Sarah

Charles

Maria

. 2018. “Segregation, Stereotypes, and STEM.” Social Sciences 7(7):111.

58.

#Thébaud

Sarah

Pedulla

David S.

. 2016. “Masculinity and the Stalled Revolution: How Gender Ideologies and Norms Shape Young Men’s Responses to Work–Family Policies.” Gender & Society 30(4):590–617.

59.

#van Belle

Janna

. 2016. “Paternity and Parental Leave Policies across the European Union.” Santa Monica, CA: RAND.

60.

#Wallander

Lisa

. 2009. “25 Years of Factorial Surveys in Sociology: A Review.” Social Science Research 38(3):505–20.

61.

#Weziak-Bialowolska

Dorota

. 2015. “Differences in Gender Norms between Countries: Are They Valid? The Issue of Measurement Invariance.” European Journal of Population 31(1):51–76.

62.

#Williams

Joan C.

2021. Bias Interrupted: Creating Inclusion for Real and for Good. Boston: Harvard Business Review Press.

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

0.25 MB