Non-Standardized Interview Prompts Can Lead to Bias

All job interviews are not created equal when HR diverts from a standardized process, creating the potential for bias in favor of or against a candidate, according to recent academic research from Canada.

While HR professionals tend to use some forms of structured interviews by preparing questions ahead of time, more than three-fourths use their own prompts and probes to give a nudge to candidates struggling to answer an interview question, according to findings from 301 Canadian HR professionals.

“The net effect of that ends up looking as though the person from the visible minority does not give very strong answers when in fact they may have been given fewer opportunities to say what they actually know,” says lead researcher Sheldene K. Simola, assistant professor of the business administration program at Trent University in Ontario, Canada.

Simola and fellow researchers Simon Taggar and Geoffrey W. Smith conducted their research in 2007, concentrating on professionals in Canada to see how real-world practice aligned with the Canadian Human Rights Tribunal, a body of anti-discrimination law. HR professionals in Canada not only conduct job interviews but sometimes take on full responsibility for making a hire.

Those surveyed are “a relatively well-seasoned group of interviewers” but don’t always follow guidelines set out by Human Rights Tribunal and field researchers, Simola and her colleagues found.

What’s needed is more emphasis on standardizing the interview process, formal training of interviewers, and stressing the importance of job analysis for developing accurate job descriptions, they write in The Employment Selection Interview.

Among the 78 percent of HR professionals using non-standardized prompts and probes with job candidates, 13 percent always add questions during the interview, 25 percent do so most of the time, and 37 percent sometimes do so.

Making up their own rather than using standardized prompts or probes “creates quite an opportunity for bias to enter into the interview process even if that is not what the interviewer is intending to do and even if the interviewer may be quite committed to having a fair interview,” Simola told SHRM Online.

Researchers found that only 12.6 percent of HR professionals surveyed use rating scales to evaluate candidates’ answers during job interviews.

While their research was limited to Canada, the findings can apply to HR professionals beyond Canada’s borders, according to Simola. She and her colleagues believe that their findings are relevant to union shops, where grievances about the selection interview might be referred to arbitration.

“When the same interview questions have been administered to all candidates in a structured and objective way, the interview is more likely to be judged as fair and the grievance more likely to be dismissed,” she said in a press release.

“However, when interviewers stray from a structured and objective process,” she added, “potential unfairness exists and the grievance is more likely to be upheld.”

A structured interview is multifaceted and should include behavioral interview questions, she told SHRM Online.

HR professionals looking for more-thorough answers from candidates might want to bear in mind a more recent study, released March 2008 from Purdue University’s Krannert School of Management.

More than 90 percent of job candidates are less than truthful during an interview, and follow-up questions actually increase the number of false answers a candidate gives, it found.

“Our findings suggest that [follow-up questions and probes] may actually increase faking because it tells the candidate what the interviewer is interested in,” Michael Campion, professor of management, said in a press release.

Purdue researchers developed an “Interview Faking Behavior” scale of different ways candidates give answers aimed at making a good impression.

Specific questions about how the applicant handled situations in their previous jobs are less susceptible to false answers than questions about hypothetical situations, researchers found.

Kathy Gurchiek is associate editor for HR News. She can be reached at kgurchiek@shrm.org.