Discrimination against members of stigmatized groups i.e., individuals who face social disapproval or negative stereotyping due to certain characteristics or attributes they possess (Goffman, 1963), is a public health issue (Bhui, 2016) that impacts both physical and mental health (Daftary et al., 2020; Soto et al., 2011; Stein et al., 2019). Discrimination in the selection contexts is particularly concerning because it negatively impacts career advancement and exacerbate wage inequalities (Heilman & Caleo, 2018; Ozeren, 2014) thus, it also has long-term financial implications.
Several stigmatized groups across different cultural contexts are the target of these unfair treatments. For instance, in the United States, 57% of black adults (Bleich et al., 2019) and 24% of LGBT+ individuals (Sears et al., 2021) report experiencing discrimination when applying for a position. In Sweden, 76% of immigrants from countries outside of Northwest Europe and North America believe that having a foreign name reduces their chances of being hired (Behtoui & Neergaard, 2009). In the United Kingdom, 11% of adults reported being rejected for a role due to their age (Workplace Discrimination Statistics in 2021, 2021).
Given the significance of discrimination, it has drawn considerable attention from governments and organizations alike. Various studies have employed a wide range of methods to explore this issue (Neumark, 2018). Over the years, governments and organizations worldwide have implemented numerous initiatives to reduce discrimination. However, the problem remains complex and has yet to be fully resolved. This ongoing challenge highlights the need for continued research to better understand discrimination, address its root causes, and develop effective solutions to reduce it.
In this thesis, I explore different angles for studying discrimination and potential ways to mitigate it that are often overlooked. First, in Chapters 1 and 2, together with my co-authors, I discuss perceived discrimination and how it can impact members of stigmatized groups. While the literature often focuses on the detrimental effects on targets, little is known about how these unfair treatments can also trigger motivation to overcome them. We build on the literature about Social identity-based impression management (Roberts, 2005) and explore how perceived discrimination could trigger the use of these additional types of impression management.
Second, the literature on discrimination often isolates a single identity, despite the fact that everyone possesses multiple identities. Isolating one identity can hinder a comprehensive understanding of the experiences of members of stigmatized groups (Kulik et al., 2007; Reid & Comas-Diaz, 1990) and reduce the effectiveness of proposed solutions to combat discrimination. To enhance this understanding, I explore how contextual cues can influence evaluations received by potential teammates at the intersections of age and gender in Chapter 3.
Together, my doctoral thesis employs several methods (e.g., field surveys, online and lab experiments, with or without incentivized decisions) and aims to contribute to the literature on discrimination in the selection process. I will now summarize these three chapters, highlight their individual contributions, and link them to each other.
Chapter 1: Perceived Discrimination in the Labor Market: Relationships with Job Seekers’ Impression Management
As mentioned above, Chapter 1 considers the perspective of members of stigmatized groups reacting to discrimination they perceive. Given that the goal of impression management (IM) is to improve how one is perceived by others (Rosenfeld et al., 2015), members of stigmatized groups may use these strategies to try to mitigate the unfair treatment they may face.
Social identity-based IM (Roberts, 2005; Roberts et al., 2014) suggests that members of stigmatized groups may use specific IM strategies rooted in their social identity as a proactive response to perceived discrimination. Social identity-based IM (SIM) is defined as “the process of strategically influencing others’ perceptions of one’s own social identity in order to form a desired impression” (Roberts, 2005, p. 694). While perceived discrimination has been proposed as a trigger that motivates the use of SIM, this linkage has rarely been tested empirically. The aim of this chapter is to fill this gap and investigate the effectiveness of these strategies.
Although traditional IM strategies, such as self-promotion, can improve hiring outcomes, they can also be risky for members of stigmatized groups, who may face backlash for behaving counter-stereotypically and could receive lower evaluations as a result (Rudman et al., 2012; Rudman & Phelan, 2008).
We investigate the relationship between the perception of discrimination and the use of IM on LinkedIn, today’s most widely used professional social media platform, through an online survey conducted with U.S. job seekers currently using this platform. Our findings show that perceived discrimination is positively related to the increased use of different IM strategies, suggesting a potential role of perceived discrimination in triggering the use of IM (Neel et al., 2013; Roberts, 2005). Importantly, this relationship remains significant even when controlling for personality traits, excluding an alternative explanation often proposed to explain the use of IM (Bourdage et al., 2018). The use of IM was, in turn, related to better job-related outcomes for job seekers, highlighting the potential of IM to improve job-related outcomes.
One of the main contributions of Chapter 1 was to establish the relationship between perceived discrimination and the use of IM on LinkedIn through a field survey with real job seekers. However, this methodology also has its drawbacks. The results are based on self-reported data and do not allow for the establishment of a causal impact of perceived discrimination on the use of IM on LinkedIn. Chapter 2 was designed to address this gap.
Chapter 2: Navigating Discrimination: How Job Seekers Proactively Use Impression Management on LinkedIn
Building on the findings of Chapter 1, my co-authors and I designed an experiment to causally test our results. In Chapter 1, we investigated the impact of perceived discrimination without specifying any targeted identity. For Chapter 2, we focused on "youngism," which refers to unfair treatment for being too young (Duncan & Loretto, 2004; Francioli & North, 2021). This form of ageism is often overlooked in both research and practice (Schmitz et al., 2023).
However, younger adults are often perceived as entitled, arrogant, lazy, and inexperienced (Finkelstein et al., 2013; Francioli & North, 2021), and they receive the least favorable sentiments among all age groups (Francioli et al., 2024).
For Chapter 2, we developed and pretested a manipulation to increase the salience of youngism discrimination. To move away from self-reported data and measure actual behaviors on LinkedIn, we created a platform that mimicked the professional social media environment. Participants were randomly assigned to one of the two conditions: either high or low discrimination salience for being too young and then asked to create their profiles on our platform. For Chapter 2, we employed this design in a lab experiment with Swiss university students (Study 1), and online with a U.S. sample (Study 2).
Contrary to our expectations, results from Study 1 showed that discrimination salience increased the use of IM only for younger men, not younger women. Younger men in the high discrimination salience condition invested more effort when creating their profiles (e.g., included more sections in their profiles), used more IM (e.g., mentioned more soft skills), and as a result, received better evaluations from independent raters. However, these results were not found among younger women. These findings indicate an intersectional effect, where the intersections of age and gender shape reactions to perceived discrimination. Intersectional studies (Cole, 2009) suggest that younger women and younger men might face different forms of ageism, possibly because women already anticipate gender discrimination. Thus, women might already increase their IM usage even when youngism discrimination was not salient. An alternative explanation could be that young men may feel more confident and therefore invest less effort compared to young women when discrimination is not salient.
We aimed to replicate and extend these results in Study 2 with individuals of all ages. We found no significant effect of discrimination salience on IM used, nor did we find an interaction with gender, despite the fact that young participants reported higher perceived discrimination when facing high youngism discrimination salience. The differences in results between Study 1 and Study 2 suggest that the impact of perceived discrimination on IM may be more complex than initially hypothesized. These differences could be explained by varying cultural contexts between the U.S. and Switzerland, particularly in terms of individualism and long-term orientation (Country Comparison Graphs Country Comparison Graphs, 2014; Hofstede, 1984). Another alternative explanation could be the difference in sample composition between Study 1 (i.e., students) and Study 2 (i.e., general population). Finally, the artificial setting of our experiment could also explain these results. Participants had no incentives to construct a better profile and, therefore, may have felt less motivated to use IM.
In sum, Chapter 2 contributes to the literature by highlighting the nuanced role of perceived discrimination in shaping the behaviors of job seekers from stigmatized groups. However, the findings underscore the context-dependence of these links and suggest that IM can be a tool to combat discrimination, but solutions should be tailored to consider cultural context and the multiple identities to which the target individuals may belong.
Chapter 3: Context Matters: The Role of Technological Tool Descriptions in Age and Gender Intersectional Discrimination
Results from Study 1 in Chapter 2 reveal an interesting intersectional difference, with only younger men reacting to discrimination salience. In Chapter 3, I continue to explore intersectionality, focusing specifically on the intersections between age and gender. These two intersections are particularly relevant because a significant portion of the population could be at risk of discrimination based on gender. Ageism is also an important discrimination to study due to population aging, which is expected to have a major impact on the workplace (White et al., 2018), especially for older women whose participation in the workforce continues to grow (Toossi & Morisi, 2017).
Despite its major contributions to the study of discrimination, research about intersectionality often reveals mixed findings. Two alternative hypotheses are competing: Double Jeopardy, which suggests that people with multiple subordinate identities suffer more discrimination (Beal, 2008), and Intersectional Invisibility, which proposes that people with multiple subordinate identities suffer less discrimination (Purdie-Vaughns & Eibach, 2008). One potential explanation for these conflicting results is that stereotypes may depend on context (Casper et al., 2010; Kornadt et al., 2013).
For this project, I investigated the following question: How does the description made by a company of a technological tool used in the workplace impact intersectional discrimination based on age and gender when evaluating potential teammates? To answer this question, I employed a between-subjects experimental design based on vignettes. Participants were asked to assume the role of a manager selecting a potential teammate. I manipulated the context by describing the technological tool used in the workplace as either a communication tool or a high- tech tool (Comunello et al., 2017). Additionally, participants evaluated either a man or a woman, who was either young (below 30) or older (above 50).
A question emerging from the discussion in Chapter 2 was the impact of incentives on participants' responses. The effect of incentives in research on discrimination is equivocal, and the literature on this topic remains scarce. Therefore, I applied the design above for Chapter 3 in two separate studies, one with incentives and one without.
The results show substantial differences between Study 1 and Study 2. In Study 1 (without incentives), the only significant effect was that older potential teammates received lower evaluations than younger ones. However, when participants were incentivized to predict the evaluations given by others (Study 2), significant interactions emerged: older individuals and women were rated more positively when the tool was described as a communication tool compared to a high-tech tool, supporting a context-dependent view of stereotypes. However, across our studies, no intersectional effects were found, providing no evidence to support conclusions about the intersectionality of age and gender.
Chapter 3 makes two important contributions. First, it shows that contextual cues can influence evaluations received by members of stigmatized groups (women and older people). Carefully designing job ads could be a potential tool to decrease discrimination. Even if the effect is small, these initiatives are highly cost-effective, and given the number of members of stigmatized groups, the overall improvements could be substantial.
Second, the inclusion of incentives in Study 2 led to considerably different results compared to Study 1. One potential reason for these differences is the varying perspectives taken by the participants. In Study 1, participants reported their personal opinions. In contrast, in Study 2, when incentivized to predict the responses of others, participants focused on second-order normative beliefs (i.e., beliefs about others' beliefs). These can be interpreted as two different perspectives. However, both are relevant because, in the evaluation and selection context, it is likely that recruiters and managers use both. My results emphasize the need for future research to investigate these two different angles.
Conclusion and Perspectives
Understanding and fighting against discrimination in the selection contexts is crucial. These economic disadvantages significantly affect an individual's quality of life, influencing their access to essential resources, opportunities for personal and professional development, and overall well-being throughout their lifetime. Moreover, reduced financial stability and slower career progression can weaken a person's position at work, limiting their influence and ability to make meaningful changes related to discrimination in their professional environment.
In sum, this thesis employs a multifaceted approach to studying discrimination in selection contexts, exploring both the perception of discrimination and the strategic responses it elicits from members of stigmatized groups. By addressing discrimination faced by members of stigmatized groups, this research not only contributes to the existing literature but also provides insights that could inform more equitable hiring practices. Importantly, this work explores two potential solutions to combat discrimination: the use of Social identity-based Impression Management by individuals to navigate and mitigate unfair treatment, and the impact of contextual cues, such as in job descriptions, that can influence evaluations and reduce discriminatory outcomes.
Additionally, my thesis highlights the complexity of discrimination, showing how it is influenced by cultural contexts and the multiple identities held by members of stigmatized groups. These findings underscore the need for further research to continue addressing these issues through the lens of intersectionality, with the ultimate aim of creating a fairer selection process for everyone.