Social Media In Pre-Employment Screening: When Did Personal Judgemental Prejudices Replace Validity?

It is clearly evident that social media is woven into our everyday lives and technology enables immediacy in terms of sharing personal information on line. Our everyday social activities are in the public arena, warts and all, and consequently open to scrutiny. However, how pervasive is this scrutiny in organisational recruitment? Legislation is in place to guard against discriminatory practices, however, it is perhaps inevitable that recruiters would be tempted to examine the digital lives being led by potential recruits to their organisations. It is an easy opportunity to attempt to get an insight into the character of the individual who has applied for a post; an attempt at enhancing cultural fit decision-making. However, this approach can be highly problematic. From a Business Psychology perspective it can be argued that recruitment is about ‘accounting for performance,’ that is when we measure potential performance, either via interview, psychometrics or live observational centres we are accounting for how any given individual will perform in a role. We may account for 17% at interview through to 65% with assessment centres. Whatever strategy is used there is always the possibility that the baby will be thrown out with the bath water. If we have hiring errors of judgement that is based on valid decision-making strategies, what do organisations hope to achieve by trawling through an individuals personal on-line space as a last bastion of feed into appointment? This divisive, prejudicial practice is not only ill thought out at best, it is irrelevant in terms of performance predictive validity. It makes hiring organisations the custodians of everyday life comment and behaviour, with the almost Orwellian self appointed authority to make value judgements on what is right and what is wrong.

First, lets examine the psychology of posting on-line. Research suggests that personality is an important factor to consider when investigating the causes and consequences of people’s engagement with social media. Further, the images we present in the virtual world may not necessarily reflect who we are in real life. Do you take selfies? Dr Terri Apter, psychology lecturer at Cambridge University, says taking selfies is all about people trying to figure out who they are and project this to other people. “It’s a kind of self-definition,” says Dr Apter. “We all like the idea of being sort of in control of our image and getting attention, being noticed, being part of the culture.”

People want to control the image projected and this image will vary with context, just as in real life. We all have distinct friend, family and work personas. If we examine older adults, their profiles will often feature their wives and children. The profiles of university students feature what they believe is the most interesting part of their lives, and this will vary enormously. Pictures of drinking and partying are interesting to certain peer groups. They can provide cues on what to wear, where to go, and how to act, young adults look to their peers to see what are the best parties and activities, which are illustrated on Facebook/MySpace. I personally don’t have a single picture posted of myself in the library. Other postings may also serve to communicate the importance of particular relationships because these bonds may provide security regarding an individuals self worth.

Social media lends all users a “public” persona, and when users try to present themselves in a way that matches how they want to be seen, this creates potential problems. Part of the problem is that the norms of one community aren’t the norms of another. So when we produce an ‘out of work persona’ that aligns to the cultural nuances of any given external group, and this represents a difference in terms of how we behave in the workplace, this can lead to clashes on social media, that is then interpreted by others.

It is likely that the most common Facebook regrets revolve around sensitive topics like alcohol, sex, politics, religion or “emotional content.” Often, the sources of these regrets are unintended consequences or unintended audiences. And, such postings are the ones most likely to be utilised when assessing an individual’s suitability for a job post. Uninhibited behaviour on-line is a gift for certain personalities in organisations. Nevertheless, it is one thing wanting to know a little more about the views, motivations and lifestyles of job seekers; it is another thing to take a prejudicial leap of faith regarding that individual’s cultural fit and performance potential. This is further compounded when delegated to a third party to undertake such activity, where there is high potential for identity mistakes to be made. And, even wider judgemental bias can occur through delegated authority.

As a Principal Business Psychologist I often evaluate the validity of recruitment measures in order to determine the extent to which selection tools can predict job performance. Measures have different types of validity that capture different qualities. There are three major types of validity: content validity, construct validity, and criterion validity.

Content validity refers to how comprehensively the measure assesses the underlying construct that it claims to assess. Construct validity refers to whether the measure accurately assesses the underlying construct that it claims to assess. Criterion validity examines how well the construct correlates with one’s behaviour in the real world across multiple situations and manifestations. For instance, does the measure adequately capture the construct (e.g., innovation) as it presents in real life (e.g., time management, planning and organising, leadership, etc.)?

The reliability of a measure refers to whether the measure gets repeatable results. Will the recruitment and selection processes that a company uses work every time they need to hire someone, or just once? If their processes get good results every time, those measures can be said to be reliable.

On the basis of the above, I have to ask about the validity and reliability of social media screening. What construct does it purport to investigate, how well does it measure that construct, how well do any conclusions drawn correlate with required behaviour, and finally how consistent are the results?

The fact remains that screening must be done with caution, so that recruiting managers do not inadvertently act in manner that could be considered discriminatory, or unfair in other ways.

In common with other paperwork associated with the recruitment process, any personal data gathered during the recruitment/screening process should be handled and retained in accordance with any policy guidance on record keeping.

Organisations need to guarantee that:

  • the same restrictions apply to online checks as they do to all other aspects of the recruitment process;
  • personal data should only be accessed if it is relevant to suitability for the role;
  • only the absolutely necessary personal information that is relevant to the job should be collected;
  • social media searches should not be used as a personal fishing exercise;
  • reasonable steps should be taken to ensure the accuracy of any personal details accessed online;
  • a distinction should be drawn between the use of social media for mainly private purposes and for mainly professional purposes, i.e. viewing LinkedIn is acceptable, viewing Facebook is not;
  • information that is in the public domain regarding someone’s professional profile can be used;
  • before online searches are conducted, applicants should be advised that information about them might be gathered in this way;
  • applicants should be given an opportunity to respond to any adverse findings from online searches, where they may be considered in the decision-making process.

Only then can job applicants be reassured that it is their skill, motivational and personality drivers that are being evaluated, not their life choices. The latter has no place in legitimate, fair and open hiring decisions.