Social Media & Selection

"Information is not knowledge"

Albert Einstein

Social media platforms such as Facebook or LinkedIn have been described as a new way for organizations to obtain information about job applicants. A manager can browse through applicants' profiles, assess their qualifications, and evaluate whether they would fit with the job and the organization. Numerous surveys have shown that many hiring managers already engage in such practices. And, in some cases, they may decide to reject a candidate based on what they have found online. 

While practitioners are often quickly adapting to the new opportunities offered by social media, research is generally lagging behind. My own research aims at better understanding applicants' behaviors and strategies on social media, how managers use social media for screening or assessing applicants, and and the value of social media as novel selection methods.

One way job social media can impact applicants chances to get a job is when they post content that could be seen as  faux pas (i.e., posts seen as inappropriate by employers - e.g., about drugs, excessive partying, etc.). In a study, I examined how employers' use of social media in selection, but also applicants' internet and social media use and their personality would influence the likelihood of faux pas postings. For instance, applicants engage in less faux pas postings when they realize that a high proportion of employers use social media in selection, but their behavior also depends on their level of extraversion and general online behaviors (see article here).

Job applicants can also engage in strategic behavior on social media to make a good impression on potential employers. In a recent book chapter, Julia Levashina (Kent State University, OH) and I developed a typology of impression management tactics that applicants can use online. We also surveyed a number of young graduates to illustrate how they think about what to include on their profile, how to present it in a way that would impress employers, and how to strategically hide or remove information that might hurt their candidacy (see chapter here). 

In a recent article, a team of graduate students and I developed and validated a measure of applicant impression management on Facebook. We found that applicants use of such tactics on Facebook was associated with some personality traits (e.g., honesty-humility) as well as job search self-efficacy.

The Facebook Impression Management Scale (FIMS) is available here.

If in theory social media can be used to assess applicants' qualification, it is also important to better understand what qualifications can be assessed on which platform. In a study conducted with Adrian Bangerter (University of Neuchatel, Switzerland), we examined differences between personal and professional social media. 

We  found that manager/recruiters and potential applicants (students and graduates) both perceive professional platforms like LinkedIn as a potential antecedent of Person-Job fit information (e.g., about skills, abilities, competencies). In contrast, both perceived personal platforms like Facebook as a potential antecedent of Person-Organization fit information (e.g., about personality, values). Moreover, when evaluating the same Facebook profile, recruiters and potential applicants focused on different sections of the profile. But they tend to infer similar personality traits (see article here).

More recently, a team of graduate students and I examined job seekers' attitudes towards employers' use of cyber-vetting, for instance how they viewed the validity, justice, and privacy invasion associated with such practices (see article here). We developed and validated a measure of such attitudes. In addition, we found that attitudes largely differed between social media platforms (e.g., more positive for LinkedIn, less positive for Instagram).

The Attitudes Towards Cybervetting (ATC) scale is also available here.

See also our recent examination of ATC in China, focusing on social media like WeChat, QQ, and Weibo, here.

An initial series of studies, done together with Julia Levashina (Kent State University, OH), involves examining the potential of LinkedIn as a selection tool. We conducted a series of studies looking at hiring manager's perceptions of LinkedIn (vs. more traditional tools like resumes of job interviews) to assess applicants, skills, abilities, or personality. We also empirically explored how reliable and how valid LinkedIn-based assessment could be, if such assessments could lead to adverse impact, or whether structuring the process could improve its effectiveness.

Overall, we found that LinkedIn can be an effective tool to assess applicants' cognitive abilities, their level of extraversion, as well as some specific skills like leadership or communication (see article here).

More recently, in a follow-up study done in collaboration with Rhea Stronach (see article here), we recruited 154 experienced North American workers, asked them to complete a series of tests/measures, and to give us access to their LinkedIn profiles. We then asked two groups of raters (200 hiring professionals and 6 graduate students) to review their profiles, asses their personality, cognitive ability, and the likelihood that they would be “good organizational citizens” (i.e., OCB), and provide hiring recommendations. Finally, we obtained automated assessments of personality from LinkedIn profiles using the language-based tool "Receptiviti". 

We found that human raters generally agreed in their assessments of personality traits, cognitive ability, OCB likelihood, and hiring recommendations. Such ratings were associated with limited adverse impact, based on race, gender, or age. Human raters’ assessments were (somewhat) consistent with self-reports or test scores from our “applicants” only cognitive ability and honesty-humility (but not any other traits).

Interestingly, ratings of personality traits (especially extraversion and conscientiousness) were positively associated with the Receptiviti automated ratings. In other words, human raters who reviewed the LinkedIn profiles assessed those traits of “applicants” quite similarly to automated assessments performed on the textual content of the profiles.