Mixed women are generally more attractive

What keeps women away from professional positions? A factorial survey on the influence of the design of a job advertisement on its assessment of attractiveness


Occupational gender segregation produces and reproduces social inequality between the sexes. Against this background, this article focuses on the perception of job advertisements. If women consider individual features of a job advertisement to be more or less attractive, this can contribute to occupational gender segregation. We ask how both the linguistic design of a job advertisement and the characteristics of a job mentioned in a job advertisement influence the assessment of the attractiveness of this position for women. As part of a factorial survey, 224 female employees aged 25–40 years are presented with hypothetical job advertisements, which differ in terms of various dimensions and are to be assessed in terms of their attractiveness. The causal effects of the vignette dimensions are estimated in a multi-level model and show that flexible working hours are the most important variable. Furthermore, stereotypically feminine formulations and gender-sensitive position descriptions have a positive influence on the evaluation of a job advertisement compared to a purely masculine address. In addition, female role models are relevant.


Occupational sex segregation produces and reproduces social inequality between the sexes. Against this background, this article focuses on the perception of job advertisements. If women consider individual features of a job advertisement to be more or less attractive, this can contribute to occupational sex segregation. This article analyzes the rating of job advertisements by varying their linguistic presentation and the characteristics of the jobs mentioned in the advertisements. Employing a factorial survey, 224 female participants aged 25-40 years were asked to rate hypothetical job advertisements that differed in several dimensions. By running a mixed effects model, we identified the causal effects for each dimension. The results show that flexible working hours are the most significant factor. Furthermore, women rate advertisements as more attractive if gender-fair forms are used. Female role models are also relevant.


The German labor market is characterized by pronounced gender-specific segregation into various occupations (Hausmann and Kleinert 2014) as well as a significantly more frequent occupation of management positions by men (Holst and Friedrich 2017). This not only solidifies role models and gender stereotypes, but also produces and reproduces social inequality, since typical “women's jobs” are paid less on average (Busch 2013a) and are associated with less good career opportunities (Busch and Holst 2009; Kleinert et al. 2007). There are various explanations for the creation of gender-specific occupational patterns on the supply and demand side of the labor market. On the one hand, gender-related stereotyping or discrimination prevent access. On the other hand, different - depending on the theoretical perspective, either economic (human capital theory) or sociological (socialization theory) - mechanisms influence voting behavior and thus lead to self-selection (Anker 1997; Reskin 1993). Job advertisements form an interface between the various explanatory approaches if we assume that their design influences the interest in a job for female job seekers accordingly.

The focus of this study is the question of how both the linguistic design of a job advertisement and the characteristics of a job mentioned in a job advertisement influence the assessment of the attractiveness of this position for women. If women consider individual characteristics of a job advertisement to be more or less attractive, this can contribute to occupational gender segregation. Our focus is on three aspects that are conveyed through job advertisements. First, we ask to what extent gender-sensitive language in job advertisements, i.e. one that includes all genders, has an effect on the assessment of the attractiveness of a job advertisement. Second, we look at the applicants' characteristics mentioned in the requirement profile and analyze the extent to which stereotypical male and female formulations influence the assessment of attractiveness. Third, the influence of context characteristics of the job described (flexible working hours, managerial responsibility and female role models) is examined.

Chapman et al. (2005) in a meta-analysis with the result that the perceived fit of the applicant with the organization or the position to be filled has the greatest influence on the assessment of attractiveness, but the image of the organization or the work environment are also decisive factors. Even if the choice of words and wording in a job advertisement is varied, gender differences are evident in such a way that women in particular respond to them. Women feel more likely to be addressed by job advertisements that contain stereotypical feminine characteristics (Born and Taris 2010; Taris and Bok 1998) or words that are usually associated with women (Gaucher et al. 2011), as well as by offers which they are addressed directly with the feminine form (Bem and Bem 1973; Horvath and Sczesny 2016; Stout and Dasgupta 2011).

In this study, a factorial survey is used to simulate a job search in which the women surveyed are presented with hypothetical job advertisements that differ both in terms of their linguistic design and other characteristics of the job described. The women surveyed then rate the attractiveness of the job advertisement. Factorial surveys offer the advantage that the causal effect on the attractiveness assessment can be determined for each feature of the job advertisement. By varying both the contextual factors of the job and the linguistic design of individual features in this study, we can determine their influence on the assessment of the attractiveness of job advertisements.

In the following we explain how job advertisements influence the perception of a job and to what extent certain aspects of a job advertisement can affect how well women assess themselves for the job in question. We identify factors that make a position appear more “typical for men” - and thus supposedly less suitable for women - and derive hypotheses that relate these factors to the assessment of the attractiveness of a job advertisement. Then we present our experimental design and give an example of a job advertisement as presented to the women interviewed for evaluation. Furthermore, we provide an overview of the data collection, our quality assurance measures and the social statistical composition of the data set. Finally, we present the results of our multilevel model and embed them in the context of occupational gender segregation and social inequality.

Theory and hypotheses

Job advertisements aim to recruit suitable and qualified job seekers. Companies use job advertisements to present their organization and to describe the tasks of the job and the desired profile of the applicants. Job advertisements influence how job seekers perceive a job and whether they perceive themselves as suitable (Chapman et al. 2005; Rafaeli and Oliver 1998). Basically, we assume that the way in which job advertisements and their specific characteristics are formulated has an influence on whether women perceive this position as suitable and therefore attractive.

According to the lack-of-fit model (Heilman 1983), people estimate their suitability for a professional field or a specific job on the basis of the perceived requirements of this position and perceived personal characteristics. Gender stereotypes play a decisive role here. If women perceive a poor fit with typical male positions, this can result in less identification with the position as well as the consideration of a change of position (Peters et al. 2012). How “typical for men” a position appears is determined, among other things, by the formulation of the job advertisement (Bem and Bem 1973; Gaucher et al. 2011; Horvath and Sczesny 2016; Stout and Dasgupta 2011), the characteristics in the requirement profile (Born and Taris 2010; Taris and Bok 1998) as well as the discrepancy to typical female role expectations (Eagly and Wood 1991).

In the following, we will focus on women and their perception of gender-sensitive language, stereotypical formulations and contextual features of the job, so that we can then derive hypotheses regarding the attractiveness of a job advertisement. In this context, a high assessment of attractiveness is interpreted as an indicator of a high degree of fit.

Gender-sensitive language in job advertisements

Current studies show that the grammatical system of a language influences the perception of the world (Boroditsky et al. 2003). Thus, in addition to its function as a communication medium, language seems to be involved as a structure-building element in the development of fundamental psychological processes of perception and cognition. This assumption is also followed by feminist language criticism, which denounces the linguistic disadvantage of women. According to her, an asymmetrical linguistic treatment of women and men can be found both in language use and in the language system (Samel 2000). The linguistic means with which linguistic acts are carried out - such as the omission of both names - as well as linguistic norms in the sense that, for example, feminine forms are often derived from masculine and masculine pronouns can refer to all persons are criticized. As a result, gender stereotypes are formed through language, integrated into perception and thinking, and gender hierarchies are maintained. For example, a country comparison showed that when other influencing factors such as geographical location or culture are controlled, the linguistic system correlates with the degree of gender inequality in a country (Prewitt-Freilino et al. 2012).

An asymmetry between the sexes is created in particular by the androcentric use of language, according to which masculine personal designations are neutral and can also be used as a substitute for designating women - and all other non-male persons (Samel 2000). The masculine used generically stages the man as the norm and all others as a deviation, so that the man is always allowed to feel that he is being addressed exclusively and the rest of them are faced with the question of whether they are meant or not (Frank 1992) .

Various research results reflect this assumption and show that supposedly neutral forms, such as the generic masculine, are spontaneously and immediately associated with men as typical references (for an overview see Stahlberg et al. 2007). The notional inclusion of women in personal references in the generic masculine is less and the use of gender-sensitive forms leads to more frequent mentions of women (Heise 2000; Stahlberg and Sczesny 2001). These results can also be found in the labor market context. For example, the participants in an experimental study only consider women to be just as suitable for a high management position as men despite having the same qualifications if this position is described with a male and female title (Horvath and Sczesny 2016). In job advertisements, addressing women with masculine personal pronouns (Stout and Dasgupta 2011) or male job titles (Bem and Bem 1973) negatively affects their interest in applying. The androcentric language use thus also affects the professional and career plans of women.

The androcentric language does not make it clear that women are addressed in the same way as men and, moreover, ignores women as participants in the conversation or as potential incumbents (Samel 2000). We therefore suspect that women feel differently addressed by different forms of position designation in job advertisements and consequently perceive themselves as being differently suitable for a position. Building on these considerations, we derive the following hypothesis:

H 1

Women rate job postings with a gender-sensitive title as more attractive than those with a masculine title.

Stereotypical formulations in job advertisements

Agency and communion are two fundamental dimensions of social information processing and behavioral orientation that can be traced back to Bakan (1966) (for an overview see Abele and Wojciszke 2014). While communion refers to the integration of the individual in the community and is expressed in cooperation, friendship and the striving for social integration, Agency emphasizes the autonomy of an individual and manifests itself in self-assertion, assertiveness and the striving for self-realization.

The gender-specific association of agent behavioral orientations with men and communal behavioral orientations with women (Bakan 1966), the origin of which Eagly and Wood (1991) identify in particular the gender-specific division of labor and the gender role expectations based on it, is of importance for the present study. Gender-specific association means that certain behaviors are more likely to be attributed to women or to men and are therefore mentally linked to a gender. According to role theory, people tend to behave consistently with their respective gender roles (Eagly and Wood 1991), which is why women are particularly well suited to positions for which stereotypically feminine characteristics are required (Heilman 1983). Stereotypes can be conveyed in job advertisements via the requirement profile in a direct or indirect way. Directly by demanding agentic or communal characteristics (Born and Taris 2010; Taris and Bok 1998), and indirectly by demanding a supposedly neutral characteristic - a characteristic that is typically ascribed neither to women nor to men, but which is gender-specific through its formulation Evokes associations (Gaucher et al. 2011). There is empirical support for both forms, which is why the following hypotheses are derived:

H 2

Women rate job advertisements with a stereotypically female quality in the job profile as more attractive compared to those with a stereotypically male quality.

H 3

Women rate job advertisements with a wording typically associated with women in the job profile as more attractive compared to those with wording typically associated with men.

Contextual features of the job posting

According to the gendered organization concept, the gender of a person systematically determines the distribution of income, tasks and positions (Acker 1990). The social separation of the areas of life production and reproduction established in the course of industrialization is firmly embedded in organizational structures. The ideal member of the world of work is fully available to the company and has no other obligations outside of gainful employment (Williams 2000). This norm can be met primarily by men who have the housework and family work done by another person - usually a woman. According to role theory, this division of women and men into different areas of life is the origin of gender role expectations, which in turn influence the social behavior of women and men in different ways (Eagly and Wood 1991). Such traditional employment and working time patterns are still widespread (Wanger 2015), which is why female life courses still seem to be strongly characterized by the compatibility of family and work.

The compatibility of family and work determines not only the hours worked, but also the career plans of women. People who align their lives with gainful employment are better suited to positions of responsibility than people who have to split up their commitment (Acker 1990). A high availability of time is one of the characteristics of a management position and has a significant influence on career opportunities (Holst and Friedrich 2017). With regard to the division of time, it can be seen that women are less likely than men to occupy positions that involve long working hours and thus make it more difficult to combine family and work (Busch 2013a, 2013b; Cha 2013). Women seem to attach great importance (or have to attach) to flexible organization of their working hours in order to be able to do justice to the additional reproductive work that has to be performed.

From these arguments we conclude, firstly, that women prefer positions with flexible working time models and, secondly, that they avoid time-consuming management positions. We derive the following hypotheses:

H 4

Women rate job advertisements with a flexible working time model as more attractive than those with a fixed working time model.

H 5

Women rate job advertisements without intended management responsibility as more attractive compared to those with intended management responsibility.

In addition to the scope and flexibility in the organization of working hours, other factors also determine whether women perceive certain positions as suitable for themselves or not. In order to increase the proportion of women in the organization and in management positions, companies can use targeted measures to promote women. The use of female role models in internal and external communication, for example, has proven to be suitable for this (Lutz 2018). A person's self-concept can be positively influenced by contact with successful individuals from the same social group. In addition, the feeling of belonging to the area in which the role model is successful is strengthened and the motivation to get involved in it increases (Dasgupta 2011). By confronting female executives, gender role expectations can be broken so that women feel more part of the corporate environment. Studies show, for example, that female role models have a positive influence on the self-assessment of women (Asgari et al. 2012) or their striving for a position with managerial responsibility (Hoyt and Simon 2011). Even the depiction of a woman on a job advertisement shows a positive effect on self-assessed suitability (Bosak and Sczesny 2008). With female role models, women seem to feel that they belong to a company or a position in it, which is why the following hypothesis is derived:

H 6

Women rate job postings with a female role model as more attractive compared to those without a female role model.

Data and methodology

To test our hypotheses, we use a factorial survey. With this experimental method we can identify the causal effects of the linguistic characteristics and the context characteristics on the assessment of the attractiveness of a job advertisement. As part of a factorial survey, the respondents are presented with concrete descriptions of the situation (vignettes), which are to be evaluated according to a specific criterion. Different characteristics (dimensions) are varied independently of one another so that the effect of the individual vignette dimensions on the assessment of the respondents can be determined (Auspurg and Hinz 2015). Specifically, we simulate a job search in which the variations in the job advertisements correspond to the differences in real job offers. In order to make the evaluation situation as realistic as possible, the vignettes are designed in such a way that all important elements of a standard job advertisement according to Moser and Sende (2014) are contained either directly in the vignettes themselves or in the preceding general introductory text. The vignettes are supplemented by a questionnaire which, in addition to information on socio-demographic data, also contains questions about the importance of the dimensions depicted in the vignettes.

Structure of the vignettes

In total, six dimensions are varied in each vignette: the job title, the requirement profile with both a stereotypically male or female characteristic and a neutral characteristic, the formulation of which is typically associated with women or men, as well as the working time model, career opportunities and the presence of female role models . In the following, the individual dimensions and their respective characteristics are presented in more detail; an overview is provided in Table 1.

The position description dimension includes a total of five variants of addressing: "Employees", "Employees (m / f)", "Employees", "Employees" and "Employees". Compared to the purely male form of “employee”, the generic reading of “employee (m / f)” is supported by the addition in brackets, but because the masculine address as “employee” remains unchanged, the focus remains on men.Footnote 1 The participle form “employees” is grammatically neutral and therefore neither directly addresses one gender nor excludes others. In addition to other options for gendering, the gender asterisk was chosen in the form of “employees” because it is becoming more and more common at the moment - for example, the members of the Bündnis 90 / Die Grünen party officially agreed on this form (Kohlmaier 2015) - and therefore its mode of operation is of particular practical interest.

For the requirement profile, the choice of the two stereotypes and the neutral characteristic is based on frequently used formulations in job advertisements. Based on the distinction between Communion and Agency by Bakan (1966), “team-oriented collaboration and the ability to cooperate” is set as a stereotypical female and “independent way of working and assertiveness” as a stereotypical male characteristic. The choice of gender-associated formulations is based on a corresponding study by Gaucher et al. (2011), who were able to identify a comprehensive list of words with clear gender associations and used them in a follow-up study to describe a mixed-gender profession. In order to vary the vignettes, the formulations “motivate” were chosen as female and “challenge” as associated with male.

Within the context of the context characteristics, a distinction is made in the working time model dimension between “flextime” and “fixed working hours” and with regard to the career opportunities dimension, the assumption of future managerial responsibility varies between “optional” and “planned”. The internal existence of a female role model is operationalized via the farewell greeting in the vignette, with either the team leader "Stefanie Müller" or the team leader "Christian Schmidt" appearing as the subject. According to a list published by the Society for the German Language (no year), Stefanie and Christian are the most popular first names among today's mid-30s; According to the German Digital Family Dictionary (Brandmüller and Peschke 2019; Kunz 2019), Müller and Schmidt are the two most common surnames in Germany.

In total, a universe of 160 vignettes results, which corresponds to the number of all possible combinations of vignette dimensions (5 · 2 · 2 · 2 · 2 · 2). None of the combinations has to be excluded for reasons of plausibility, which is why all 160 vignettes are taken into account in a full survey. These are randomly distributed over 20 different decks, so that each deck contains eight vignettes. According to Auspurg and Hinz (2015), with a maximum number of ten vignettes per person, fatigue and learning effects can be avoided, which is why this division seems appropriate. All respondents were given a random deck consisting of eight job advertisements, which should be assessed with regard to their attractiveness on the basis of the description. An eleven-point scale was available for this information. The individual vignettes were shown one after the other so that the respondents did not create a ranking. The following table 2 shows an example of the vignette from which the interviewed women are expected to be the most attractive.

In addition to the varying dimensions in the vignettes, other relevant characteristics of a job advertisement are kept constant in the introductory text. The respondents should assume that they have the necessary specialist skills for the positions offered, that the jobs are open-ended, offer reasonable conditions and do not require a change of residence.

Operationalization of further variables

In addition to the varying vignette dimensions as predictors, other respondent characteristics are also taken into account in the analyzes. In particular, we collect the respondents' assessments and attitudes towards various characteristics of professional work. We ask the importance of various dimensions of professional work directly (for an overview of the questions, see Fig. 1 in the appendix) by asking the respondents to indicate how important they are on a scale from 1 “not at all important” to 5 “very important” the respective dimensions are. In doing so, we ask about the importance of gender-sensitive addressing by managers as well as by colleagues. In addition, the respondents are asked to indicate how important they are to flexible working hours, good opportunities for advancement and various principles of equal treatment (balanced treatment of everyone in terms of remuneration and recognition, avoidance of preferential treatment and promotion, prevention of discrimination).

Finally, we collect further characteristics of the respondents. In addition to the socio-demographic information on gender, age, place of birth, marital status and number of one's own children, information on current employment and the highest vocational qualification are also taken into account. Satisfaction with the current position is measured using a five-point scale; the daily working hours differentiate between “full-day” for working hours of over 30 hours per week and “half-day” for 10 to less than 30 hours per week. For the highest vocational qualification, the breakdown into vocational or non-vocational secondary education, bachelor's and master's degrees is chosen.

Data collection and social statistical composition of the data set

The data was collected online from the end of February to the beginning of March 2019 using the Norstat Panel (www.norstat.de). Even if a representative sample of the population cannot be drawn with a commercial panel, the internal validity of the results is given by the experimental design in the sense of the randomized assignment of the vignette decks to the respondents (Czymara and Schmidt-Catran 2016). No bivariate correlations can be found between the vignette dimensions, which is why the experimental factors can be viewed as independent of one another and the experimental design has a high internal validity in this regard (see Table 4 in the appendix).

To ensure that only people who are relevant to the subject of the experiment take part in the study, the respondents were invited by email according to certain criteria. All respondents are employed full-time. In addition, only people without personnel responsibility were taken into account, so that there is a fit with the advertised position as a team member. The age is set at 25–40 years, a phase in which a job change can be seen as likely.

Since it is of particular importance that the respondents read the instructions of the experimental design carefully and conscientiously assess each job advertisement, we use the example of Stumpf et al. (2020) two quality assurance measures. First, we add an additional profile to the series of vignettes and ask the respondents to tick a certain scale value in the description. If the respondent does not tick the desired scale value, the survey is ended immediately. Second, all respondents who required less than 80 seconds to process all vignettes were excluded from the analysis. A short processing time can be interpreted as an indication that the vignettes are not carefully read and evaluated. With these measures, we rule out possible biases caused by people who do not process the survey carefully and ensure that the data only contain responses from respondents who take the survey seriously.

The dataset includes a total of 224 women. A look at their socio-statistical composition shows that they are on average 32 years old and almost exclusively born in Germany (95%). More than half of the respondents are single (61%), a third are married (33%) and the majority are childless (80%). 7% cited non-professional and 57% vocational secondary education as the highest vocational training qualification, 15% have a university degree at bachelor’s level and 21% at master’s level or higher. The respondents work full-time (99%) mostly in permanent contracts (85%) and are on average more satisfied than dissatisfied with their current employment.


We describe the results of the vignette analysis below. The multivariate analysis is intended to estimate the influence of the various dimensions of the job advertisement on the assessment of attractiveness for women, whereby the coefficients are used to test the respective hypotheses. A multi-level model is used for this, which takes into account the hierarchical data structure resulting from the respondent and vignette level.Footnote 2 A positive sign of the coefficients shown in Table 3 indicates that the advertised position is rated as more attractive. Taking into account various control variablesFootnote 3 there are no significant differences with regard to the effect sizes or directions of the vignette dimensions. Since the conclusions do not change when the control variables are included, we only present those results of the analysis in Table 3 that only contain the vignette dimensions themselves for reasons of better readability.

We start with the question to what extent a gender-sensitive language influences the assessment of attractiveness. With regard to the formulation of the position designation, it can be seen that women perceive gender-sensitive forms as more attractive overall. Compared to job advertisements with the purely masculine title of "employee", all other formulations have a positive sign. Offers with the designations “employee (male / female)” and “employee” are rated 0.30 and 0.29 points more attractive by the women surveyed on the scale of the dependent variable, which ranges from 1 to 11. Both formulations are statistically significant. The terms “employees” and “employees” also have a positive sign (0.21 and 0.14 points), but are not statistically significant. The first hypothesis can thus be at least partially confirmed.

With regard to the stereotypical formulations in job advertisements, a statistically significant effect on the assessment of the attractiveness of a job advertisement can be seen, especially for the stereotypical female characteristic. Women rate job postings with communal characteristics 0.18 points more attractive than those with agentic characteristics. In this specific case, preference is given to a position that requires team-oriented collaboration and the ability to cooperate compared to an independent way of working and assertiveness. With regard to the formulation typically associated with women, the results show a trend. Women rate job advertisements as more attractive if the female-associated phrase “motivate” is used in contrast to the male-associated phrase “challenge”. However, the difference of 0.15 points is not statistically significant, although it is at the limit (tValue = 1.85). Thus the second hypothesis is accepted, the third has to be rejected.

Within the context of the context characteristics, both the working time model and the presence of a female role model have a positive influence on the assessment of attractiveness. Compared to fixed working hours, job advertisements with flexible working hours in the form of flexitime are rated 0.95 points more attractive. The level of the coefficient indicates the particular relevance of flexible working time models. In addition, the reference to a female role model in the job advertisement has a positive effect on their assessment of attractiveness (0.19 points). However, the prospect of career opportunities and the related timely assumption of managerial responsibility have no statistically significant effect. The fourth and sixth hypotheses are therefore accepted, the fifth rejected.

Discussion and outlook

This study examines whether the characteristics of a job that can be conveyed in the context of a job advertisement influence their assessment of the attractiveness of women. The focus is on the linguistic design of the position description, the characteristics required by applicants and the context characteristics (flexible working hours, managerial responsibility and female role models) of the job described. With the help of a factorial survey, a job search with 224 female employees is simulated. These are supposed to assess hypothetical job advertisements in terms of their attractiveness, whereby the individual job advertisements differ with regard to the use of gender-sensitive language and stereotypical formulations as well as the design of the contextual features.Based on the experimental design, the causal effect on the attractiveness assessment can be determined for each dimension of the job advertisement.

In general, the results show that women evaluate job advertisements depending on the selected address in the job title as well as with regard to their requirements and context characteristics. In addition to the working conditions themselves - in the form of working time regulations - the design of a job advertisement, in terms of gender-sensitive language and gender stereotypical formulations, also has an influence on the assessment of the attractiveness of job advertisements. In detail, the results of the present study show that women tend to rate job offers with gender-sensitive job titles (“employees” and “employees”) as more attractive than those with purely masculine positions (“employees”). However, the positive effect of gender-sensitive formulations cannot be clearly demonstrated for every formulation alternative. The two alternative formulations “employees” and “employees” are not rated as more attractive compared to the purely male position designation. With a view to the requirement criteria, the women surveyed rate job advertisements as particularly attractive if stereotypical female characteristics are required in the job profile. In contrast to the results of Gaucher et al. (2011) it makes no difference for the women questioned in the present study whether a formulation typically associated with women is used in the requirement profile compared to a formulation typically associated with men. The context characteristics of job advertisements are characterized by the degree of flexibility in the organization of working hours and a reference to a female role model in the work environment. As the traditional patterns of employment and working hours suggest (Wanger 2015), the results of the present study also show that women are particularly important in the organization of flexible working hours. Although flexible working hours play a major role for women, they do not seem to be actively avoiding a potentially time-consuming career path. The results at least show that it makes no difference to your assessment of the attractiveness of a job advertisement whether or not a position offers excellent career opportunities with prompt assumption of managerial responsibility. In addition, women rate advertisements for positions in which a woman already works in a similar or higher position for the company better than offers without such a reference to a female role model. Similar patterns of professional selection processes are also shown by other studies that emphasize the importance of structural features for professional gender segregation (including Busch 2013b; Damelang and Ebensperger 2020; Leuze and Strauss 2016).

The analysis of the individual vignette dimensions shows that jobs with flexible working hours in particular are rated as particularly attractive. The particular importance of flexible working hours can be explained by the fact that the working time model has a direct influence on the compatibility of family and work. However, our results can also prove that with the design of a job advertisement - in the form of the use or non-use of gender-sensitive language and gender stereotypical formulations - factors that are independent of the context and content of a job also significantly influence the attractiveness of jobs and the perceived fit . For women, gender-inclusive language can have positive effects on career plans, whereas stereotypically masculine formulations and the generically used masculine may keep them away from jobs and positions.

In addition to the undisputed importance of flexible working hours, the results also suggest that language is involved in the (re-) production of social inequality through the influence it has on occupational gender segregation. The positive effects of gender-sensitive language and stereotypical female formulations found in our study can be explained by the perception of a better fit with the requirements of the position, which, according to Carless (2005), means that women are more likely to consider accepting the job offer in question. The formulations used in job advertisements draw a line between professions that tend to appeal to women and those that tend not to appeal to women. In this sense, language can express a lack of equality of opportunity, which manifests itself in the fact that women are less likely to choose certain occupations. If this linguistic boundary runs between occupations with high and low income, high and low social prestige, or high and low degree of responsibility, women have less chance of choosing well-paid, highly regarded and responsible occupations. This consolidates gender stereotypes and continues to classify people into the categorical system of “gender”. Gender as a central principle of social differentiation will therefore remain just as much as the social inequality associated with occupational gender segregation.

In order to be able to derive recommendations for action in practice, the first question that arises is to what extent the assessment of the attractiveness of a job advertisement can serve as an indicator of actual application behavior. Since validation studies find a close connection between the decisions in the vignette experiment and those in reality (e.g. Hainmüller et al. 2015; Petzold and Wolbring 2019), this seems to be the case. Differences in the appraisal of attractiveness of a job advertisement (re-) produce occupational gender segregation, since it is more difficult for women to identify with a job and they may therefore refrain from pursuing a certain career path. Therefore, companies should check job advertisements with regard to their effect on women and formulate them carefully. Flexible working hours proved to be of particular relevance, which is why companies should enable appropriate working hours and should refer to this in their job advertisements. In order to arouse greater interest in a position in women, this can be explicitly addressed, and characteristics typically associated with men in the position description should be avoided. It can also be helpful to make women who are already working in the organization visible so that they can act as role models for potential applicants.


  1. 1.

    Due to the experimental logic, diverse is not listed in brackets.

  2. 2.

    Since the residuals of level 2 (respondents) are slightly skewed to the left, we carried out robustness tests in which quadratic and cubic transformations of the dependent variables were applied, which led to very comparable results overall.

  3. 3.

    We have integrated the following control variables into the model: importance of gender-sensitive approach by managers, importance of gender-sensitive approach by colleagues, importance of flexible working hours, importance of good opportunities for advancement, importance of balanced treatment of everyone in terms of remuneration and recognition, importance of avoiding preferential treatment in recruitment and promotion, Importance of preventing discrimination, age, place of birth, marital status, children, highest professional qualification, fixed term, satisfaction with current position and the position (from 1–8) in which the vignettes were displayed.


  1. Abele, Andrea E., and Bogdan Wojciszke. 2014. Communal and Agentic Content in Social Cognition: A Dual Perspective Model. Advances in Experimental Social Psychology 50:195–255.

    Google Scholar

  2. Acker, Joan. 1990. Hierarchies, Jobs, Bodies: A Theory of Gendered Organizations. Gender and Society 4(2):139–158.

    Article Google Scholar

  3. Anchor, Richard. 1997. Theories of occupational segregation by sex: An overview. International Labor Review 136(3):315–339.

    Google Scholar

  4. Asgari, Shaki, Nilanjana Dasgupta and Jane G. Stout. 2012. When Do Counterstereotypic Ingroup Members Inspire Versus Deflate? The Effect of Successful Professional Women on Young Women’s Leadership Self-Concept. Personality and Social Psychology Bulletin 38(3):370–383.

    Article Google Scholar

  5. Auspurg, Katrin, and Thomas Hinz. 2015. Factorial Survey Experiments. Thousand Oaks, CA: Sage.

    Book Google Scholar

  6. Bakan, David. 1966. The duality of human existence. An essay on psychology and religion. Chicago: Rand McNally.

    Google Scholar

  7. Bem, Sandra L., and Daryl J. Bem. 1973. Does Sex-Biased Job Advertising “Aid and Abet” Sex Discrimination? Journal of Applied Social Psychology 3(1):6–18.

    Article Google Scholar

  8. Born, Marise Ph., And Toon W. Taris. 2010. The Impact of the Wording of Employment Advertisements on Students' Inclination to Apply for a Job. The Journal of Social Psychology 150(5):485–502.

    Article Google Scholar

  9. Boroditsky, Lera, Lauren A. Schmidt, and Webb Phillips. 2003. Sex, Syntax, and Semantics. In Language in Mind. Advances in the Study of Language and Thought, Eds. Dedre Gentner and Susan Goldin-Meadow, 61-79. Cambridge, Massachusetts: The MIT Press.