It seems like access to someone’s social media profiles would be a huge advantage to recruiters. Potentially, you get to see a side of the candidate they don’t always show in a job interview, and you could spot red flags and avoid problems later.

But a new study by Harvard Business Review reveals some pitfalls as well: often the data on social media profiles reveals the answers to questions an interviewer can’t legally ask. That can lead to unconscious bias for or against the candidate coming into play.

The real question is, “Does using social media as part of the screening process help us hire better people?” Here are some of the findings, and some answers.

Hiring Goes Social

Many employers now ask for access to social media profiles or even for candidates to share them during interviews. A lot of candidates simply agree and allow it. According to a 2018 CareerBuilder survey, 70% of employers visit social media profiles as a part of their screening process, and 54% have rejected candidates because of what they found there.

The HBR survey looked at 266 social media profiles and what they revealed. Most reveal the answer to off-limits questions:

• Gender, Race, and Ethnicity (100% of profiles)
• Disabilities (7%)
• Pregnancy Status (3%)
• Sexual Orientation (59%)
• Political Views (21%)
• Religious Affiliation (41%)

Many contained profanity and references to gambling, drinking, and even drug use. But there is a big difference between what people do at work and what they do on their own time, and by “social media screening” employers may focus on the wrong things.

Legal Landmines

This can unleash some legal landmines. Social media screening that reveals much of this information is something that employers cannot take into account when they make hiring decisions. But telling someone to ignore something they saw on social media and actually getting them to do it is something else altogether.

In fact, as part of their study, HBR had 39 recruiters look at 140 of the profiles, selected randomly, and had them rate the candidate based on what they saw. They found that recruiters did pay attention to relevant job data, but also appeared to be swayed by things like relationship status, gender, and religion.

• Married or those in a relationship rated higher than those who were single.
• Women won out over men.
• Those who expressed religious views scored lower.

This is largely due to unconscious bias, something that, while we can talk about it, is harder to overcome than we think.

Unconscious Bias

One thing we don’t know from this study is the demographic of the recruiters who looked at the profiles. But it doesn’t really matter. Unconscious bias exists in everything we do. Humans, left unchecked, tend to hire people who look, act, and think like them. In fact, the old adage about first impressions is a form of unconscious bias that skips looking at all the facts in any given situation.

Just telling someone they have unconscious bias is not enough. It takes training and discipline to overcome, and even then, no system is perfect. In fact, recent studies showed that Amazon’s AI hiring tool, built to avoid bias, showed bias instead, and the company was forced to shut it down. That’s because even AI is trained by the data fed into it.

Eliminating it is not a simple matter, and clearly, social media screening tends to introduce more data that can trigger unconscious bias rather than less.

In that case, what difference does it make? Does social media screening help companies hire better people?

Hiring Better People

For the final part of their study, HBR got supervisor ratings for 81 of the candidates in the second part of the study after 6 to 12 months in employment. They also surveyed the employees to see if they intended to stay in their current jobs. Then they showed those Facebook profiles to a new set of recruiters to evaluate this.

But this time, they separated the recruiters into two groups. First, they just asked for an evaluation of the profiles. The second group got some special instructions about what to look for, what they should ignore, and to be mindful of bias.

Neither group’s evaluations predicted job performance or job longevity accurately. Even with detailed instructions, social media screening was no better in either group of recruiters.

Final Takeaways

So should you use social media screening as a part of your hiring process? The data says it is likely not a good idea. There’s very little benefit to companies, and because of unconscious bias, any job recruiter will evaluate everything they see on a social profile, even if they are just looking for red flags or relevant data.

Another consideration is that candidates often “clean up” their profiles, including posts from their friends, comments, and also will often tighten their privacy settings when job hunting. Many will have two accounts, one public and one personal, and will be very careful about what they post in a public forum.

Social media is also often a “highlight reel” and just the fun stuff about a person’s life, not a glimpse at who they are at work.

There are alternatives, and maybe in the future through AI and Machine Learning, we will be able to use social media profiles as a tool for hiring and managing employees once they are onboarded. Another technique that can help is resume partitioning. But until we have better solutions, avoiding poring over a candidate’s social media profiles is probably best for most hiring managers.

It’s one more way to step away from unconscious bias, and simply hire the best person for the job.

‍