Many of us have taken personality quizzes to determine which “Game of Thrones” or “Stranger Things” character we are. Personality assessments have a more elevated role in talent recruitment. But, how accurate are they?
Older personality tests aren’t very accurate, research indicates. As technology develops, however, personality assessments are improving. “Current research is continually working to combine data sets to give us the most power to detect biases,” says Eric Sydell, vice president of research and innovation at Shaker International. “Over time, this intelligence is leading to assessments that are both less biased than historical psychological measures and more predictive of job performance and turnover.”
Here are three ways that technology is shaping the development of personality assessments.
New developments in personality assessments are designed to curb dishonesty. Accuracy errors in assessments often stem from a tendency by candidates to answer questions in ways they think prospective employers want them to answer. New assessments offer two positive extremes, creating more honest answers.
“Each extreme is paired with items that are equally socially desirable,” says Evan Theys, North American team lead of global products for Aon's Talent Assessment Solutions. “This ultimately creates more accurate personality profiles for candidates, and gets results that are more reflective of them and how they are most likely to behave within the workforce.”
Technology is eliminating the repetitiveness of assessments and can reduce the behaviors that cloud true personality characteristics. New assessments are responsive and adaptive to the candidate’s answers. This software makes the process more personal and offers improved, less redundant tasks.
This means the answers are more accurate and the assessments are less tedious for candidates than in the past. “Computer-adaptive testing increases the candidate experience with assessments because they're shorter,” Theys says. “Candidates don't have to deal with all of these repetitive items over and over again.”
Video assessments allow personality traits to surface more organically through conversation. Employers can ask leading questions in a recorded interview, which can move the candidate to reveal personality characteristics. These responses are then mined by the software to build a personality profile.
“One of the most promising uses of technology in personality assessments is using deep learning to score unstructured responses in the form of video, audio and text,” Sydell says. “Previously, the field’s ability to make sense of such data was limited to keyword searches and other rudimentary techniques. Now deep learning can evaluate massive amounts of information and use it to predict outcomes with a degree of accuracy far greater than previous techniques.”
In the future, Theys says, social media will play a role in profile building. “We’ll be able to leverage available data that people provide on their LinkedIn and other social pages to build comprehensive personality profiles,” he says. For now, though, there are too many concerns with privacy to be worked out before that data can be utilized.