Social media has changed many aspects of life. For instance, it has enabled friends to easily communicate with each other and helped governments share a variety of alerts, from weather emergencies to pandemics. However, the use of social media to evaluate job applicants has received very little attention. Understanding the use of social media in hiring is challenging because many businesspeople, when asked, claim they don’t use social media (e.g., Facebook) in the hiring process. Yet, answers to anonymous surveys suggest that as many as 70% of recruiters use social media to assess potential employees with Facebook being one of the two most frequently mentioned sources (along with LinkedIn). We focus our research on Facebook, which could be problematic because it was not designed to provide data to support hiring or facilitate professional interactions. So we wondered what types of information recruiters see on Facebook and how that information may influence their evaluation of applicants.

In a recent article, we collected data from two samples of job seekers to assess what types of information were available on their Facebook sites. Our first sample was a group of 266 job seekers who mirror the age distribution of the U.S. workforce. Our second sample was 140 graduating college students who were applying for jobs. We categorised information from these job seekers’ Facebook sites using a structured approach with multiple members of our research team coding information on each site. We found a great deal of information that should concern both human resources professionals, as well as job seekers whose social media information is being assessed.

Our first category of information involved information that is of concern to governmental regulatory agencies (and is often legally prohibited from being used for employment related decisions).  We found that national origin was identifiable for 56% of individuals in our workforce representative sample and 66.4% in our college sample.  There also was information on disability status in 7.1% of our workforce sample, but only .7% in our college sample.  Religion was identifiable in 41.4% and 57.9%, respectively, in the two samples.

Second, we coded other information of a personal nature available on Facebook. Sexual orientation was routinely available (58.6% and 79.3% for our workforce and college samples, respectively), as was marital status (57.9% and 73.6%, respectively). The presence of children was available in 48.5% of our workforce sample, but only 24.3% of our college sample. Political views were available for 26.3% of our workforce sample and 55% of our college sample. In our college sample, pictures were available to code attractiveness for 90.7% of job seekers and obesity for 85.7% of the sample.

Third, we coded other information that was personal in nature, but that organisations might view as helpful when screening job applicants. Sexually related behaviour was available for 15% and 23.6% of job seekers in our samples and gambling for 11.3% and 4.3% of our samples. Alcohol use was available (25.6% and 55%, respectively), as was tobacco use (8.3% and 2.1%, respectively). Non-prescription drug use was available for 7.1% and 5.7% of our samples. Overall, organisations can access a wide variety of information on job seekers’ Facebook sites. The problem is that much of this information is either (a) prohibited by equal employment laws and/or (b) occurs outside of work and, thus, may be irrelevant to how applicants would do on the job if hired. Furthermore, some of this information is highly personal, and it might be difficult for hiring officials to “unsee” it (i.e., forget once they have been exposed to it).

In another study, we asked recruiters to review the Facebook sites of our college sample and rate the hireability of each job seeker. We found that several of the categories of information were related to recruiters’ ratings. For example, women were rated more highly than men, and individuals who were married, engaged, or in a relationship were rated higher than singles. Individuals who posted their religious views/affiliation were rated lower than those who did not.  Individuals with information on alcohol use, drug use, or sexual behaviour were also rated lower than individuals without such information. Thus, there is evidence that a number of types of social media information mattered for recruiter ratings of hireability.

Pre-employment assessments should demonstrate that applicants who score well on the tests, interviews, etc. also perform well once on the job. To see whether assessments of Facebook meet this criterion, we followed the individuals from our college sample for roughly six months and had their supervisors rate their level of job performance. The data suggested that there was no correlation between Facebook assessments and later job performance.  That is, recruiter ratings of Facebook sites were unrelated to job performance.  Moreover, we found some evidence that using Facebook to screen applicants could adversely affect African-American and Hispanic applicants. There is also evidence, from other sources, that job applicants believe Facebook assessments are not related to job performance and that organisations that assess this information are invading their privacy.

All told, social media such as Facebook appear to be a Pandora’s box for companies that choose to assess this information when making hiring decisions. Facebook, for example, contains information that is likely prohibited from being used in employment decisions or is irrelevant to job performance. At the same time, evidence suggests that recruiter judgments based on job seekers’ social media information (a) are not related to future job performance, (b) might result in adverse impact against minority applicants, and (c) are considered by some applicants to be an invasion of privacy. It is hard to imagine a more troublesome box being opened in pre-employment selection.

♣♣♣

Notes:

• This blog post is based on What’s on job seekers’ social media sites? A content analysis and effects of structure on recruiter judgments and predictive validity, in the Journal of Applied Psychology.
• The post expresses the views of its author(s), not the position of the CBI, LSE Business Review or the London School of Economics.
• Featured image by Glen Carrie on Unsplash
• When you leave a comment, you’re agreeing to our Comment Policy

Liwen Zhang is a lecturer (equivalent to assistant professor) at the University of New South Wales, Sydney. She earned her Ph.D. degree from The Florida State University. Her research focuses on organisational staffing and human capital resources. In the area of staffing, she is interested in understanding how individuals seek jobs and how organisations attract and select these individuals. In the area of human capital resources, she is interested in how human capital resources emerge and how they contribute to organisational performance.

Chad Van Iddekinge is the Bank of America professor of management at The Florida State University, where he directs the organisational behaviour and human resources doctoral program. His research focuses on how organisations make staffing decisions and how those decisions affect job applicants and the quality and diversity of a firm’s workforce. Professor Iddekinge is affiliated with a number of professional associations. He is a past associate editor of Personnel Psychology and currently serves on the editorial boards of Journal of Applied Psychology, Journal of Management, and Personnel Psychology.

John Arnold is a doctoral candidate at The Florida State University and begins as an assistant professor of management at the University of Missouri this summer. His research addresses questions related to how organisations make hiring decisions, including the methods organisations use to evaluate job applicants, the types of employees organisations recruit, and how organisational decisions affect who applies to jobs in the first place.

Philip Roth is Trevillian distinguished professor of management at Clemson University. Phil’s research interests involve employee selection, political affiliation, and social media in organisations. Phil is past chair of the research methods division of the Academy of Management and has served eight years on the executive committee of the human resources division of the Academy of Management. He earned his PhD from the University of Houston.

Filip Lievens is Lee Kong Chian professor of human resources at the Lee Kong Chian School of Business of Singapore Management University. He is also an honorary professor at the University of Cape Town and a visiting professor at Ghent University, Belgium. His main interests deal with talent acquisition, talent assessment, and adaptability. As an internationally recognised scholar, he has published over 50 papers in top-tier journals. He is affiliated with a number of professional associations.

Stephen E. Lanivich is an assistant professor of entrepreneurship and management at the University of Memphis. His research interests include the entrepreneurial mindset and cognitions, entrepreneurs’ perceptions of resources, and opportunity recognition and fit. His work has appeared in leading academic journals and has been featured in The Frontiers of Entrepreneurship.

Samantha Jordan is a Ph.D. candidate in organisational behaviour and human resource management at The Florida State University. Samantha’s research interests include selection/recruitment, social influence and organisational politics, and work-family dynamics.

Print Friendly