“You Look So Beautiful… But Why Are You So Distressed?”: The Negative Effects of Appearance Compliments on the Psychological Well-being of Individuals in the Workplace

The primacy of beauty for women is frequently conveyed through language and, more particularly, through a particular class of conversational devices, compliments. Although previous research has been conducted on the impact of appearance compliments on individuals’ psychological state, research has yet to examine what happens when a compliment about the appearance clearly conveys a positive evaluation in a domain that is completely unrelated to the evaluation’s context, such as a selection interview for a job. Because the selection of a candidate should be based solely on their skills and competence, with no role for physical appearance, we purposely chose a job interview scenario to examine the psychological impact of inappropriate compliments. Through two 2-by-2 experimental studies (Ns respectively = 135 and 301), we examined whether and how anxiety and depression levels vary in women (Study 1 and Study 2) and men (Study 2) applicants who received compliments on their physical appearance, vs. their competence and had a neutral vs. positive feedback on the result of the interview. Findings of the two studies consistently showed that inappropriate compliments increase anxiety and depression levels in women but not in men, independently of the positive or neutral result of the selection interview. This research provides an important opportunity to advance the understanding of inappropriate appearance compliments on psychological well-being of the recipients in the workplace setting.


Introduction
Beauty and physical attractiveness have traditionally emerged to be among the most valuable social resources for women (Forbes et al., 2007). The primacy of beauty for women is frequently conveyed through language and, more particularly, through a specific class of conversational devices, such as compliments (Loughnan & Pacilli, 2014;Parisi & Wogan,2006). Compliments can be meant as speech acts which explicitly or implicitly attributes credit to someone other than the speaker for some "good" such as possession, characteristics, skills, etc. (Holmes, 1988). Compliments typically fall into two main broad categories -appearance and performance -and present a gendered nature: whereas appearance is more significant to compliment women, who are frequently praised for their attractiveness, performance is more important to compliment men, who are often praised for their accomplishments (Eckert & McConnell-Ginet, 2003;Rees-Miller, 2011). For instance, examining the number and type of appearance comments made and received between female and male characters in a total of 104 programs during the 1999-2000 US. prime-time season, Lauzen and Dozier (2002) showed that while female and male characters commented equally on the appearance of other characters, female characters were twice as likely to receive those comments.
Depending on the context, compliments can operate in different ways, but their primary function within interactions is to establish or reaffirm a common ground between the speaker and the recipient (Eckert & McConnell-Ginet, 2003;Wolfson & Manes, 1980). In the workplace context, compliments generally serve the persuasive function of establishing and maintaining social relationships (Rees-Miller, 2011) and empirical evidence from organizational research indicates that employees' job satisfaction is consistently correlated with receiving positive feedback and recognition (for a review, Zeb et al., 2014). Nevertheless, to date research has never tested the psychological impact of an inappropriate compliment in the workplace, i.e., a compliment that conveys a positive evaluation in a domain unrelated to the evaluation's context. Therefore, our research aims to contribute to this area of inquiry by investigating for the first time the psychological effects of receiving compliments on appearance (vs. competence) as main criterion of selection from an opposite gender recruiter during a job interview.
Generally, inappropriate compliments can be meant also as instances of microaggressions (Nadal et al., 2015). In the ethnic domain, the literature has examined this issue particularly focusing on Asian Americans who are complimented on their English skills, African American women who are praised for their exotic hair, and African Americans who are congratulated for being articulate in their language (Sue et al., 2007(Sue et al., , 2008. In the gender domain, research indicates that harassers frequently tell their victims they should be "flattered" or that "it was just a compliment" when they harass them (Taylor et al., 2018), making it challenging for the victims themselves to recognize the incident (Fitzgerald et al., 1995;Woodzicka & Lafrance, 2005).
Interesting yet inconsistent findings are also available in the domain of appearance comments and their effects on individuals' cognitive functioning and well-being. In the study conducted by Tiggemann and Boundy (2008), receiving a compliment on one's appearance had no effect on cognitive performance on a task requiring logical reasoning and spatial orientation. Yet, receiving a compliment on one's appearance decreased negative mood. In contrast, Fea and Brannon (2006) found that receiving compliments about one's appearance reduced distress and sadness in women with high self-objectification, as opposed to women with low self-objectification. Kahalon and colleagues (2018) showed that women who were asked to remind and write about past appearance-related (but not competence-related) compliments performed worse in math than women in the control/no compliment condition (Study 1) and that appearance compliments improved the mood of women with high levels of selfobjectification traits, but decreased the math performance of both women and men (Study 2). Compliments about women's weight or shape were examined by Calogero et al., (2009) to see how they affected women's body image. Self-reported data showed that women who responded positively to receiving compliments on their appearance also reported greater body awareness and dissatisfaction as a result of receiving such compliments.
In public spaces, catcalls, i.e., unwanted appearance comments from strangers, are frequently dismissed as compliments, i.e., a harmless and even flattering experience for women (di Gennaro & Ritschel, 2019;Spaccatini et al., 2019). However, research demonstrates that catcalls are associated with the fear of rape (Macmillan et al., 2000), the perceived likelihood of being a victim of gender crimes (Donnelly & Calogero, 2018), the restriction of movement (Fairchild & Rudman, 2008), and the alteration of habitual transportation routes (Livingston et al., 2015).
In the social media environment, the lack of likes or comments can have negative consequences for users; 16% of the 1,500 Facebook users surveyed by Scissors et al. (2016) reported feeling negatively when something they posted receive a low number of likes. The majority of participants in Porch's (2015) study, which focused specifically on feedback about selfies, reported feeling bad when they did not receive likes or comments. In sum, while research consistently found negative effects of appearance comments on cognitive functioning, more inconsistent results emerged about their impact on recipients' well-being.

The Present Research
Although previous research has been conducted on the impact of appearance compliments on individuals' psychological state, research has yet to examine what happens when a compliment about the appearance clearly conveys a positive evaluation in a domain that is completely unrelated to the evaluation's context. Because the selection of a candidate should be based solely on their skills and competence, with no role for physical appearance, we purposely chose a job interview scenario to examine the psychological impact of inappropriate compliments. Therefore, through two experimental studies, we aimed to determine whether and how anxiety and depression levels vary in women (Study 1 and Study 2) and men (Study 2) applicants who received compliments on their physical appearance vs. their competence.
The studies were conducted following the ethical standards defined by the Declaration of Helsinki and in conformity with the terms of the Italian Psychological Association's Code of Ethics for Psychological Research (adopted at the Italian Psychological Association General Membership Meeting on March 27, 2015; http:// www.aipass.org). In particular, before taking part in the study, participants have been informed of their rights to refuse to participate in the study or to withdraw consent to participate at any time during the study without reprisal. Informed consent was obtained from all individual participants included in the study. After they filled in the questionnaire, the participants were accurately debriefed. All the analyses were performed using SPSS and PROCESS (Hayes, 2018).

Participants and Procedure
Building on Dardenne et al., (2007), we pursued our research goals via an experimental vignette study, conducted using Limesurvey. Using a snowball procedure, starting from a coworker's social network, we recruited 135 Italian female students (M age = 23.41, SD= 2.40), who filled in an on-line questionnaire (see below), presented as a simulation of a selection interview for a paid stage in a university library.

Experimental Manipulation
All the participants red the following scenario: "Imagine to participate to a selection for a paid stage in a university library. The selection is based on your academic curriculum and on a selection interview run by Dr. Giovanni P., the manager of the library. Dr. P. begins the interview by asking about your past work experience and the days and hours you would be available to work if hired". Then, according to the experimental condition, participants were randomly assigned to one out of four scenarios, in a 2 (compliments on participants': appearance vs. competence) * 2 (result of the selection interview: neutral vs. positive) design. In the appearance and neutral result condition (n = 34), participants read that, at the end of the selection interview, Dr. P., looking them insistently, congratulated them on their physical appearance and attire. In the appearance and positive result condition (n = 37), the scenario was the same, but the following concluding sentence was added: "you are really a beautiful girl: that's why you will probably be selected". In the competence and neutral result condition (n = 36) the participants read that, at the end of the selection interview, Dr. P., looking at their curriculum, congratulated them on their competence. In the competence and positive result condition (n = 28), the following concluding sentence was added: "your academic credentials are outstanding: that's why you will probably be selected".

Measures
After the experimental manipulation, the participants responded to a five-category question in which they recalled what happened in the selection interview (response categories: "They will make me know the result of the interview", "Dr. P. congratulated me on my physical appearance", "Dr. P. congratulated me on my physical appearance and told me that, due to my body and look, I will be probably selected for the stage", "Dr. P. congratulated me on my academic credentials", and "Dr. P. congratulated me on my academic credentials and told me that, due to my curriculum, I will be probably selected for the stage"). We used this item as manipulation check. We measured anxiety via six four-category balanced items from Pedrabissi and Santinello's (1989) Italian version of Spielberger et al.'s (1983) State-Trait Anxiety Inventory (STAI), α = 0.89. Depression was measured using the Italian version of Radloff's (1977) CES-D scale, composed of 10 balanced items with four response categories, α = 0.89. These measures were previously used in Italian research on the effect of sexism (e.g., Spaccatini & Roccato,2021). These two scales' individual scores were computed by averaging the items of the scales, after recoding the con-trait items. A standard socio-demographic form followed.

Data Analyses
A preliminary contingency table between the experimental condition and participants' recall of what happened in the selection interview was used as manipulation check. Two OLS hierarchic regressions were subsequently used to predict anxiety and depression. In both the regressions, in the first model we entered two dichotomies accounting for the experimental manipulation. The first dichotomy contrasted being exposed to a compliment on physical appearance (coded as 1) to a compliment on competence (coded as -1). The second dichotomy accounted for the result of the selection interview, contrasting a positive result (coded as 1) to a neutral account (coded as -1). In the second model of the regressions, we entered the interaction between the two dichotomies. Table 1 shows that the experimental manipulation was effective: the relation between the experimental condition and its recall was strong and significant, χ 2 (16) = 277.31, p < .001, v = 0.67. Table 2 shows the results of the two hierarchic OLS regressions aimed at predicting anxiety (first six columns) and depression (second six columns) as a function of the experimental manipulation. The compliments on participants' physical appearance increased both anxiety and depression, while the result of the interview selection did not influence the dependent variables either directly or in interaction with exposure to compliments on physical appearance.

Discussion
Results of the present study provided first evidence about the negative effects of being exposed to inappropriate compliments on one's appearance in the workplace. The results of the interview selection had no effect on the dependent variables either directly or in conjunction with exposure to compliments on women physical appear-ance. Nonetheless, this study did not determine whether compliments on physical appearance in the workplace cause psychological distress differently according to the gender of the recipient. As a result of receiving compliments on their physical appearance, what happens to men's psychological distress? We addressed this open unanswered question in Study 2.

Materials and methods
Using a snowball procedure, starting from two coworkers' social networks, we recruited 301 Italian students (females = 55.1%, M age = 28.87, SD = 11.06), who participated to a study performed via the same research procedure we used in Study 1 (i.e., with an opposite gender recruiter). Ninety-one participants were randomly assigned to the appearance and neutral result condition, 75 to the appearance and positive result condition, 74 to the competence and neutral result condition, and 61 to the competence and positive result condition. Both the STAI and the CES-D showed a good reliability (α = 0.88 and α = 0.79 respectively). Like in Study 1, we pursued our research goals via two hierarchic OLS regressions, but this time we added a third model, where we entered the first-and second-order interactions between our experimental conditions and participants' gender (men = 1, women = -1). In the cells, the observed frequency and the standardized adjusted residuals (ASRs) are reported. ASRs < -2 show a significant negative association between row and column. ASRs > 2 show a significant positive association between row and column. ASR ranging from − 2 and 2 show a nonsignificant association between row and column  Note. *** p < .001

Results
A contingency table showed that the relation between the experimental condition and its recall was strong and significant, χ 2 (12) = 355.41, p < .001, v = 0.63 (see Table 3). Thus, our experimental manipulation proved to be effective. Table 4 shows the results of two hierarchic moderated OLS regressions aimed to predict anxiety (first 9 columns) and depression (second 9 columns). In terms of the effects of the experimental manipulation, the obtained results corroborated those of Study 1: The compliments on candidates' physical appearance increased anxiety and depression, while the result of the selection interview did not influence the dependent variables either directly, or in interaction with exposure to compliments on physical appearance. Interestingly, female gender was positively associated with anxiety and depression. Moreover, the interaction between female gender and exposure to compliments on physical appearance was significantly linked with both dependent variables. Two simple slope analyses showed that exposure to compliments on physical appearance significantly increased anxiety, simple slope = 0.43, SE = 0.05, p < .001, and depression, simple slope = 0.24, SE = 0.05, p < .001 among women, but not among men, simple slope = 0.10, SE = 0.05, p = .06 and simple slope = 0.02, SE = 0.04, p = .52, respectively. The other two-way interactions and the three-way interaction were not significantly linked with the outcomes.

Discussion
The results of this study nicely confirmed what was found in Study 1, showing an interaction effect between being a woman and being exposed to compliments on one's physical appearance in a job interview conducted by a recruiter of the opposite gender. Additionally, only women experienced a significant increase in anxiety and depression after being exposed to compliments on their physical appearance while men were unaffected.

General Discussion
Language may constitute a crucial means of perpetuation of gender inequality (Menegatti & Rubini, 2017). Overtly negative labels, such as derogatory ones, have been the focus of the majority of research on the discriminatory effects of language (Carnaghi & Maass, 2008;Greenberg & Pyszczynski, 1985) with less emphasis on seemingly positive language, such as compliments. Not only are compliments a powerful linguistic device, but they are real social actions (Eckert & McConnell-Ginet, 2003). As Knapp and colleagues argue (1984), "the compliment is a speech act worthy of study because it is ubiquitous, valued and problematic" (p. 12). Compliments present a peculiar gendered nature; women indeed receive more comments (compliments and criticisms) about their appearance than men (Eckert & McConnell-Ginet, 2003;Swim et al., 2001) and appearance compliments are frequently intended as chivalrous gestures (Moffitt & Szymanski, 2011). However, their potential inappropriateness is often difficult to detect, resulting in the perception of rudeness or the fear of being perceived as rude if the recipient does not appreciate the compliment.
The experimental research presented here -comprised of two studies -provides one of the first investigations into how inappropriate compliments based on appearance impacts on individuals' well-being in a workplace setting. In both studies, participants were asked to imagine competing for a paid stage audition in the university library, with the recruiter, i.e., the library manager, explaining that the selection would have been based on their academic curriculum and the evaluation of the interview. Controlling for the (positive vs. neutral) outcome of the selection, participants in the experimental condition read that the recruiter complimented them on their appearance and attire (vs. competence) at the conclusion of the selection interview. In the first study, we found that compliments on physical appearance (but not compliment on competence) of a male recruiter induce psychological distress in women. To gain a deeper understanding of this phenomenon, Study 2 examined the research goals from Study 1 considering a sample that included men, interviewed by a female recruiter. The results confirmed the findings of Study 1 showing that exposure to physical appearance compliments significantly increase anxiety and depression in women. Interestingly, this effect was not observed for men. This result is in line with previous research by Kahalon and colleagues (2018, Study 1) who showed that regardless of the context (workplace vs. park), men who imagined receiving an appearance comment viewed sexualized comments less negatively than women. Since men are traditionally accustomed to not being judged on the basis of their physical appearance for the purpose of gaining social and professional recognition

Note.
AN: Compliments on participants' physical appearance + neutral result. AP: Compliments on participants' physical appearance + positive result. CN: Compliments on participants' academic credentials + neutral result. CP: Compliments on participants' academic credentials + positive result. In the cells, the observed frequency and the standardized adjusted residuals (ASRs) are reported. ASRs < -2 show a significant negative association between row and column. ASRs > 2 show a significant positive association between row and column. ASR ranging from − 2 and 2 show a nonsignificant association between row and column  .01 (Eckert & McConnell-Ginet, 2003;Swim et al., 2001), it is plausible that receiving a comment (even if it is inappropriate) about their appearance does not pose a threat to their social value and is therefore not detrimental to their well-being. In line with this, previous research by Jackson and colleagues (1987) has shown that men place a lower value on all attributes of appearance, except height. When a person compliments another, a series of subsequent social actions are triggered. When someone's appearance is praised, it implies not only that the speaker has a favorable opinion of that person's appearance but that the recipient's physical appearance is an integral part of the image they are attempting to project (Eckert & McConnell-Ginet, 2003). Although some compliments may appear to be uplifting for women, they simultaneously remind women that they are being evaluated solely on their appearance (Calogero et al., 2009). Focusing on the effects of receiving a (inappropriate) compliment during the job selection process helps illustrate how seemingly marginal or even positive aspects can be rather detrimental in the short term to women's psychological health and in the long term their ability to place themselves effectively in the workplace. Being loaded with cultural values, inappropriate compliments on appearance can contribute to reinforcing traditional social norms because not only do they communicate what a particular person values, but also, and more crucially, what the society values as a whole (Knapp et al., 1984).
Despite the novelty of this contribution, some limitations to the present research need to be acknowledged. First, we adopted a snowball sampling strategy by spreading the link to the questionnaires online. Although this recruiting technique allowed us to reach sufficient sample sizes quickly and thus is a frequently used data collection strategy (e.g., Spaccatini et al., 2019;Spaccatini & Roccato, 2021), it presents some limitations. As suggested by Thornton and colleagues (2016), indeed, recruiting participants online through social networks often results in an unbalanced nonrepresentative sample and uncontrolled survey administration. Thus, future studies should be carried out in a more controlled setting (e.g., simulating a selection interview in the lab) and adopting different recruitment strategies. Collecting data in the lab should be an advantage in terms of the assessment of our dependent variables too. Indeed, alongside with self-reported measures, it could allow to assess physiological reactions to the experimental manipulation. Moreover, more complex and articulated communication including a compliment (e.g., 'you have a great look', 'you are beautiful', 'your dress fits your body so well'…) could be provided to the participants, and the semantic range of the content of the compliment could be extended to other inappropriate domains (e.g., participants' outfit or clothing choices). In this extension, a control group, exposed to a neutral message, could be added. Second, we showed that women experienced more psychological distress when exposed to appearance compliments, however we do not know whether their individual characteristics could moderate this pattern. Previous research demonstrated that ideological standpoint (i.e., system justification beliefs in Pacilli et al., 2019; ambivalent sexism in Spaccatini & Roccato,2021) buffers the negative effect of exposure to sexism during selection interview on psychological distress. In a similar vein, not all women would be affected by exposure to appearance compliments in the workplace to the same extent. Thus, future research is needed to understand whether and how individual variables, such as ideological standpoints, would moderate women's reactions to appearance compliments. Additionally, in both studies we presented a scenario in which the compliment was made by a recruiter whose gender was opposite to that of the participant. Future studies could manipulate the gender of the recruiter, to explore whether it would affect the reaction to appearance compliments in interaction with participants' gender, while controlling their sexual orientation.
In spite of these limitations, this research has both theoretical and practical strong points. From the methodological point of view, according to the literature on this topic, its vignette approach allowed us to maximize the balance between the internal and the external validity of our experiment (Aguinis & Bradley, 2014;McInroy & Beer, 2022). Obviously, a field replication of this study could be interesting in order to increase the ecological validity of the research. On a theoretical level, it provided an important opportunity to advance the understanding of inappropriate appearance compliments on psychological well-being of the recipients in the workplace setting. On the practical level, it provided significant insights to contrast gender discrimination in the workplace. Indeed, despite the great progress that has been made over the past decade, there is still a long way to go to achieve gender equality in the workplace. According to the Eurostat (2020) data on the current situation in the workplace in Europe, compared to men women are disadvantaged as concerns their employment rates (61% vs. 76%), career opportunities (in Europe, only 33% of managers are women) and wage gap (14.8% in favor of men). In Italy, the context in which we conducted this research, the data are consistent with this picture: compared with men, women are still underrepresented in the workplace, are employed in less prestigious work positions and earn less. As appearance compliments from strangers proved to influence women's perceptions of social environments, transforming public spaces into hostile environments for women, our research show that the same phenomenon occurs in the workplace. In order to increase women's motivation to participate in the workplace, it is crucial to combat the tendency to view women's values based on their pleasant physical appearance, i.e., as aesthetic and therefore inactive, decorative objects upon which to make comments and judgments.