Romantic Priming Effects on the Social Desirability and Hireability of Self-Promoting Women

The present study examined the effects of romantic and intelligence priming on the social-desirability and hireability of self-promoting and communal female job applicants. Participants were first primed with either romantic or intelligence related images and then asked to evaluate the social-desirability and hireability of three female job applicants. These job applicants were self-promoting and competent, communal and competent, or communal and not competent. After rating the job applicants, participants were reprimed and asked to complete a scale measuring career aspiration. Results revealed that participants rated the self-promoting applicant as more hirable than the communal applicants. In contrast, the communal and competent applicant was rated more socially desirable than the self-promoting applicant. No effect of priming on participants’ career aspiration or applicants’ social-desirability or hireability was found. However, there was a marginally significant relationship between participant gender and first choice to hire.


Introduction
In the past decade women in the United States of America have made significant gains in the labor force, such as higher rates of employment and earnings. However, equality has yet to be achieved and the glass ceiling remains in many industries. In 2012, women earned only 81 cents for every dollar a man earned (U.S. Bureau of Labor Statistics, 2013). This pay gap has contributed to the feminization of poverty, meaning that women are more likely to be poor than men (U.S. Department of Commerce Economics and Statistics Administration, 2011). Moreover, the gendered workforce has kept women in lower status positions compared to men in similar occupations (U.S. Department of Commerce Economics and Statistics Administration, 2011). Women are still significantly underrepresented in higher levels of the workforce, such that women account for only 19.2% of board seats and 4.3% of CEO positions in S&P 500 companies (Catalyst, 2015).
Gender stereotypes further create significant barriers for women since they contain prescriptions that dictate how one should behave. Prescriptions for women include being warm, kind, sensitive, and friendly (Prentice & Carranza, 2002). These prescriptive beliefs about gender have yet to change despite advances made by women in the workplace (Prentice & Carranza, 2002). People tend to support the gender normative idea that women need to be more communal than men (Moss-Racusin & Rudman, 2010). Additionally, people believe that agentic traits are less desirable for women compared to men (Prentice & Carranza, 2002). Since agentic traits are associated with competence and leadership (Prentice & Carranza, 2002) men have an advantageous position. For example, agentic males are viewed as having more social skills than agentic females (Rudman & Glick, 2001).
power-seeking female candidates suffered decreased competency and communality ratings. Similarly, Zeigler-Hill and Myers (2011) showed that women (but not men) with higher self-esteem, compared to those with moderate or low self-esteem, scored lower on the warmth-trustworthiness dimension of the Partner Ideal Scales when being evaluated by heterosexual men. Additionally, Okimoto and Brescoll (2010) found that people were less likely to vote for a female politician if they perceived that the candidate was seeking power. These results suggest that gender stereotypes allow power-seeking behaviors for men, but not for women (Okimoto & Brescoll, 2010). Similarly when women's success in a male stereotyped job is clearly presented, the women are rated significantly less likable and more interpersonally hostile then men who are successful (Heilman, Wallen, Fuchs, & Tamkins, 2004). One important agentic trait is self-promotion, which involves "pointing with pride to one's accomplishments, speaking directly about one's strengths and talents, and making internal rather than external attributions for achievement" (Rudman, 1998, p. 629). Research has demonstrated self-promotion is necessary to be viewed as competent in job interview scenarios (Rudman, 1998). Rudman (1998) found that even though self-promotion is essential, women are penalized for being self-promoting in that they are deemed less likeable than gender prescription abiding women. Phelan and Rudman (2010) revealed that self-promoting women are seen as more arrogant and dominating than males who use self-promotion. Netchaeva, Kouchaki, and Sheppard (2015) showed that men were more assertive towards women who express ambitious agency (e.g., self-promotion) compared to women who express administrative agency (e.g., directness). Amanatullah and Tinsley (2013) showed that females who self-advocate for a higher salary received more negative social judgments than males. In an applicant evaluation scenario, self-promoting females were rated less likely to be interviewed and hired compared to females who did not self-promote; no effect of self-promotion on the evaluation of men was observed (Waung, Hymes, Beatty, & McAsulan, 2015). In order to be viewed as competent leaders, women need to be self-promoting, but when they express agentic traits they are sanctioned for it (Pfeffer, Fong, Cialdini, & Portnoy, 2006;Phelan & Rudman, 2010;Rudman, 1998). Economic and social penalties given to women who violate gender prescriptions are called backlash effects (Phelan, Moss-Racusin, & Rudman, 2008;Rudman, 1998). As a result of these backlash effects women are put in a dilemma; even though they need to demonstrate agency they are sanctioned for doing so (Rudman, 1998).
In addition to backlash effects, studies have shown that romantic ideologies may also be a career barrier for some women. During young adult years, women (but not men) are likely to experience goal conflict between being romantically desirable and being intelligent in Science, Technology, Engineering, and Math (STEM) (Park, Young, Troisi, & Pinkus, 2011). In Park et al.'s (2011) study, women who were primed with romantic images or overheard a romantic conversation showed less interest in STEM and being a math/science major than women who were primed with intelligent conversations or pictures. Park, Young, Eastwick, Troisi, and Streamer (2015) showed that when heterosexual women who preferred smarter romantic partners are primed with the goal of being romantically desirable, they show decreased STEM performance and interest. Additionally, Rudman and Heppen (2003) showed that women with higher levels of implicit romantic fantasies were more likely to select lower paying occupations as well as occupations with low education requirements, and also scored lower on projected income and other high-status job measures. One possible explanation for this outcome is that romantic ideologies encourage women to rely on men for economic and social gains (Rudman & Heppen, 2003). Overall, the literature suggests that women experience a conflict between being romantically desirable and pursuing career goals that imply the violation of gender prescriptions.
This study expands on the literature of backlash effects and romanticism by examining the effects of romantic priming on the social-desirability and hireability of self-promoting and communal female job applicants. Participants in this study were primed with either romantic or intelligence images and then asked to evaluate the social desirability and hireability of potential applicants for a given job description. We hypothesized that there would be no main effect for priming or participant gender on the social desirability or hireability rankings. However, a gender and priming interaction effect was predicted such that females in the romantic priming condition were expected to rate the social desirability of the self-promoting candidate lower than females in the intelligence condition. Priming was not expected to affects male participants' social desirability or hireability rankings. It was also hypothesized that females in the romantic priming condition would be less likely to select the self-promoting candidate as their first choice to hire compared to females in the intelligence condition. No main effect of gender or priming condition on participants' career aspiration was expected. However we hypothesized there would be an interaction effect, such that females in the romantic priming condition were expected to have lower career aspiration than females in the intelligence condition.

Participants
A total of 143 students (70 male, 73 female) participated in this study and received partial credit for an introductory psychology course requirement at a private university in southern California. The mean age of the sample was 19.39 (SD = 2.19), with ages ranging from 18-34. The majority of the participants identified themselves as Caucasian (46.85%) and Asian (16.78%), followed by Latino (14.69%), Bi-Racial (13.29%), African-American (4.99%), Native Hawaiian/Pacific Islander (2.10%), and Other (1.40%).

Procedure
In order to conceal the true purpose of the experiment participants were told the study investigated what types of images college students were able to remember best. After researchers obtained informed consent from each participant, they were randomized into either a romantic or intelligence priming condition. Participants then viewed a slideshow of either 15 romantic or intelligence related images (three seconds per photo) according to their condition. The order in which photos were presented on the slideshow was randomized. Continuing with the concealment of the study's purpose participants were next asked to complete a task that required a high level of thinking in order to distract them from remembering the images. During this task, participants read a job description and three cover letters of potential job applicants and rated each applicant's social desirability and hireability. The order in which participants read the cover letters was also randomized. After reviewing each cover letter participants were asked to hire one of the applicants and select their first, second, and third choice based on the job applicants' cover letters. Participants were then reprimed with another five images (three from the previous slide show and two new images) according to their condition. The order in which photos appeared in this slide show was also randomized and images were presented for four seconds each. As part of concealing the purpose of the study, participants were told the slideshow was a memory task and were asked to specify which images they had seen before and which images they had not. Lastly, participants were asked to complete a career aspiration scale and a basic demographic questionnaire. Once all the measures were checked for completeness participants were debriefed and thanked for their time.

Priming Slide Shows
The priming slide shows consisted of 15 images (borrowed from the study of Park et al., 2011) pertaining to each condition. Previous research found that the romantic images primed participants with a want to be romantically desirable and the intelligence images primed participants with a want to be intelligent (for more information on validation see Park et al., 2011). In order to keep locations of the images consistent between each condition, six romantic priming images were excluded from the slide show because they were in outdoor settings. A new romantic priming set was created using six new images and nine images from the study of Park et al. (2011). Additionally, two new images for each condition were validated, since new images would need to be included in the memory test. The new romantic priming set was validated by a separate sample (N = 31, 58% female) where participants were asked to answer the same questions used to validate the original pictures from Park et al. (2011). These images were validated by participants rating on a Likert scale of 1 (not at all) to 9 (extremely) "How much does this image make you want to be romantically desirable" and "How much does this image make you want to be intelligent".
Results indicated that the new set of romantic images, including the two additional images to be used for the memory test, induced a greater want to be romantically desirable (M = 5.56, SD = 1.52) compared to a want to be intelligent (M = 3.54, SD = 1.42), t(29) = 6.15, p < .001, d = 1.12. Another sample of 53 participants (49% female) was used to test the intelligence images. Results indicated that the two new intelligence images resulted in a greater desire to be intelligent (M = 6.60, SD = 2.24) than romantically desirable (M = 2.09, SD = 1.57), t(52) = -14.33, p < .001, d = -1.97.

Job Description and Cover Letters
In order to test participants' opinions of communal and self-promoting women a job description (see Appendix A) and three cover letters of potential applicants were created. The job description was for a research lab coordinator position that required applicants to be both communal and self-promoting. These traits were validated using another sample (N = 38, 50% female). Participants were asked to rate two questions on a Likert scale of 0 (not at all) to 6 (very). Validation questions asked, "Communal is defined as being sensitive, warm, caring, and concerned about others. Given this definition how much do you think this job description requires candidates to be communal?" and "Self-promotion is defined as speaking with pride about one's own accomplishments, being direct about strengths and talents, and making internal attributions for achievements. Given this definition, how much do you think this job description requires candidates to be self-promoting?" A paired samples t-test indicated that the job description required applicants to be both self-promoting (M = 3.79, SD = 1.68) and communal (M = 3.71, SD = 1.39), t(37) = -.24, p = .82, d = -.04.
The cover letters used in this study (see Appendix B) were written to make the applicant seem self-promoting or communal, as well as competent or not competent. The first applicant was communal and not competent (CN), the second was self-promoting and competent (SC), and the third was communal and competent (CC). The purpose of the CN applicant was to conceal that participants were really being asked to pick between a self-promoting (SC) or communal (CC) applicant. The same sample used to validate the job description was also used to validate the cover letters. Participants answered three questions on a scale of 0 (not at all) to 6 (very). These questions included: (1) "Communal is defined as being sensitive, warm, caring, and concerned about others. Given this definition, how communal is this candidate?"; (2) "Self-promotion is defined as speaking with pride about one's own accomplishments, being direct about strengths and talents, and making internal attributions for achievements. Given this definition, how self-promoting is this candidate?"; and (3) "Given the job description, how competent do you think that this candidate is for the position?"

Hireability and Social Desirability Index
A hireability index was created to assess the general hireability of each applicant. This index consisted of four questions (see Table 1) that participants rated on a Likert scale of 0 (not at all) to 6 (very). These questions were averaged to form a mean score for each applicant, where higher numbers indicated greater hireability. Reliability for this index was strong (α = .87). Similarly, a social desirability index (α = .79) was created to examine the social desirability of each applicant. This index consisted of four questions that were rated on the same Likert scale as the hireability index, where higher scores indicated greater social desirability. Four of the eight questions used in these indices were borrowed from previous studies that examined backlash effects (Phelan et al., 2008;Rudman & Glick, 1999;Rudman & Glick, 2001 How much would you characterize this person as someone you would like to get to know better?* If you worked in the research lab how much would you want to work with this candidate? Note. *These questions were borrowed from previous studies that examined backlash effects (Phelan et al., 2008;Rudman & Glick, 1999;Rudman & Glick, 2001).

Career Aspiration
The Career Aspiration Scale (CAS) was developed by Gray and O'Brien (2007) and consists of ten items that gauge attitudes towards career goals and advancement. The CAS includes questions such as, "I plan on developing as an expert in my career field" and "I think I would like to peruse graduate training in my ijps.ccsenet.org International Journal of Psychological Studies Vol. 8, No. 4; occupational area of interest". Participants rate agreement with each item on a Likert scale of 0 (not at all true of me) to 4 (very true of me). In this study, the reliability of the scale was found to be adequate (α = .74). These ten items were averaged to create a mean score for each participant, where higher scores indicated greater career aspiration.

Data Analysis
Results from Shapiro-Wilk tests indicated that all social desirability and hireability scores for each candidate were not normally distributed (p < .05), so Wilcoxon Signed-Ranks tests were used to analyze the data. In order to use an analysis of variance (ANOVA) to test for main and interaction effects between gender and priming condition two new normally distributed variables were created: (1) the difference in hireability score between the SC candidate and the CC candidate; and (2) the difference in the social-desirability score between the SC candidate and the CC candidate. Additionally, an ANOVA was used to examine main and interaction effects of priming condition and gender on participants' career aspiration.

First Choice to Hire
A Chi-square analysis revealed a marginally significant relationship between gender and first choice to hire χ2 (2, N = 143) = 5.48, p = .06, in which males more frequently selected the CN candidate as their first choice to hire than females (see Figure 1). However, Chi-square analysis found no significant (p > .05) effects of priming on first choice to hire.

Career Aspiration
Results from an ANOVA showed that female participants' career aspiration (M = 2.81, SD = .60) did not significantly differ (p > .05) from males (M = 2.96, SD = .58). Similarly, no significant differences in career aspiration between the romantic (M = 2.88, SD = .62) and intelligence (M = 2.90, SD = .57) priming conditions was found. Results also showed no significant (p > .05) interaction effects of participant gender and priming on career aspiration.

Discussion
This study explored the effects of romantic and intelligence priming on the social desirability and hireability of self-promoting and communal female job applicants. Results did not support our hypothesis that females in the romantic priming condition would be more likely to select the SC candidate as their first choice to hire. However, participants did rate the SC candidate as more hirable than the CC and CN candidates. In contrast, the CC candidate was rated more socially desirable than the SC and CN candidates. Additionally, results showed there was a marginally significant relationship between participant gender and first choice to hire. The present study also examined the effects of romantic and intelligence priming on career aspiration. Results showed no main or interaction effects of participant gender or priming condition on career aspiration. Therefore, results did not confirm our hypothesis that females in the romantic priming condition would have lower career aspiration than females in the intelligence priming condition.
In job hiring scenarios criteria is often shifted to disadvantage self-promoting women so that their success is undermined and their supposed deficit in social skills is overemphasized (Phelan et al., 2008). However, this shift in hiring criteria is not observed when the applicant is male, even though he possesses the same social deficits (Phelan et al., 2008). These changes in criteria and other backlash effects significantly disadvantage self-promoting women. Even when women abide by gender prescriptions and are more communal than self-promoting, their social skills are undervalued and their lack of self-promotion and other agentic traits are overemphasized (Phelan et al., 2008).
The appearance of confidence seems to be positively related to many individual outcomes in organizations such as hiring and promotion decisions, but gender differences continue to persist. For example, research conducted by Guillén, Mayo, and Karelaia (2016) found social attraction, or being liked by colleagues, moderates the competence-confidence gender gap. While for men competence is sufficient to appear confident, competent women must be liked to reap the benefits of their competence. So much so that being disliked turns competence into a liability for women. Paradoxically, more competent women appear less confident.
Research by Heilman and Okimoto (2007) suggests that women are penalized for success in male gender-typed areas of the workforce because their success implies they lack communality. However, when information about a successful woman in a traditionally masculine domain is supplemented with evidence that she is also communal, negative reactions are reduced (Heilman & Okimoto, 2007). Similarly, Brescoll and Uhlmann (2008) revealed that when a female expressed anger in a professional setting she received a lower status conferral, salary, and competence ratings compared to a male who also expressed anger. However, when a female provided an explanation for her anger she received higher status conferral and salary compared to a female target that did not mention an explanation for her anger. Together these results suggest that women can mitigate the negative effect of violating gender stereotypes.
Previous research has demonstrated that women are just as likely as men to elicit backlash on self-promoting and/or agentic women (Parks-Stamm, Heilman, & Hearns, 2008;Rudman, 1998;Rudman & Kilianski, 2000). Our study aligned with previous literature since the self-promoting candidate received backlash in that she was rated less socially desirable than the qualified communal candidate. Even though the communal candidate was socially desirable she was not rated as hirable as the self-promoting candidate. This supports previous literature that has shown self-promotion is needed to be viewed as competent (Rudman, 1998). The backlash that self-promoting women receive can significantly impact their careers. Heilman et al. (2004) showed that an employee's likability affected participants' recommendations for the employee's salary level and whether or not they wanted to have them as a manager. Heilman et al. (2004) thus demonstrated that while competency is important, so is likability. In their meta-analysis, Williams and Tiedens (2016) found that when dominance is explicit (but not implicit) women are rated as less likable than dominate men. However, dominant targets were rated as more competent than non-dominant targets independent of target gender. Results from the meta-analysis demonstrated that dominant behavior had a greater negative effect on women's downstream outcomes (e.g., hireability) compared to men's. Williams and Tiedens (2016) suggest that these downstream outcomes follow patterns of likability ratings and not competence ratings. Some research has suggested that women are aware of the penalties for violating gender prescriptions, such that in one study women anticipated greater relational problems when being in a leadership position (i.e., CEO, political leader, director of a science research center) compared to men (Lips, 2001). Interestingly, results from the present study showed that while male and female participants selected the CC candidate in equal proportions, they differed in that men chose the SC candidate less than females and more often selected the CN candidate. These results suggest that some males may prefer a communal and unqualified candidate to a self-promoting and qualified candidate. This finding may imply that in some cases, abiding by communal gender-prescriptions is more important for females than demonstrating agency and capability.
In addition to confirming previous findings on backlash effects, the present study examined the effects of romantic versus intelligence priming on the social desirability and hireability of female applicants. Results revealed no significant effects of priming condition on the social desirability or hireability scores of the applicants. These results suggest that romantic ideologies may not provoke participants to further sanction female applicants who violate gender prescriptions. Additionally, results showed that romantic priming had no effect on participants' career aspiration. These results do not necessarily conflict with previous studies that have shown the negative influence of romanticism on women's career aspiration. The Career Aspiration Scale may not have been as sensitive to changes in career goals as previous studies' measures (e.g., expected income).

Limitations and Future Research
It should be noted that no exploratory or confirmatory factor analysis was conducted to test the validity of both the social desirability and hireability indices. While some questions used in the indices were taken from published studies (Phelan et al., 2008;Rudman & Glick, 1999;Rudman & Glick, 2001), the additional items added to the index would benefit from such data analyses. The age of the sample was fairly young and should be expanded upon. We cannot say whether the same results would be observed in an older population that has more experience in the workforce.
The present study only included applicants high in either self-promotion or communality. Heilman and Okimoto (2007) demonstrated that when information about self-promoting females also includes information that they are communal negative reactions are minimized. Future studies should explore the hireability and social desirability ratings of applicants that are high in both self-promotion and communality. As well, future research should explore differences in females' social desirability and hireability in both gender incongruent and congruent positions. Lastly, further research should examine how self-promoting women are treated in the workplace in order to better understand the longitudinal effects of backlash.

Self-Promoting Competent (SC) Candidate
To whom it may concern, My name is Jessica N. Newman and I am currently a junior at Azusa Pacific University. I am interested in the Research Lab Coordinator position because I feel that it would further my skills in psychological sciences. Based on my numerous qualifications I can assure you that I am more than capable of being the Research Lab Coordinator.
I have a thorough understanding of how to use SPSS and Excel to organize and analyze data. In addition, I have been selected to be a teaching assistant for multiple upper division psychology courses. As a teaching assistant, I have worked alongside fulltime faculty to develop course curriculum and train students to write appropriately for psychological sciences. As well, I was awarded the Trustee's Scholarship (cumulative GPA 3.9) and have consistently been honored on the Dean's List.
My distinctive research qualifications include leading a study on the placebo effect and increased endurance in my research methods course. To expand upon my study from research methods, I collaborated with Dr. Williams in a research practicum setting. Results from this collaboration were presented at an APA conference in poster format. Following this presentation, I submitted an article to a psychological journal.
Given my qualifications I would be an excellent fit for the Research Lab Coordinator Position. If you need any further information please contact me at jnewman1010@apu.edu or at (626)  I have participated in a research practicum and research methods class experiment. In research practicum, our team assisted a professor on their study concerning child development. During my research methods class we ran an experiment to examine the association between stress and relationship satisfaction. As a research team we presented the results at an APA conference and also worked together to submit an article to a psychological journal.
While being a part of these research teams I was exposed to using both SPSS and Excel. Additionally, I aid professors as a teaching assistant for upper division psychology classes. As a teaching assistant I have organized and led study groups to prepare students for exams and have helped the professors with grading. I have also been a consistent member of the Dean's list and have received the Trustee's Scholarship (GPA: 3.9).
I would love to meet the research team members who are currently a part of the lab, so that I could see if I would be a good addition to the research lab team. I look forward to further discussing this opportunity with you. If you have any questions or concerns, please feel free to contact me by e-mail at mjacobs1010@apu.edu or by phone at (626) 555-7384.
Thank you for your time,

Copyrights
Copyright for this article is retained by the author(s), with first publication rights granted to the journal.
This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/).