Blog

Bias in Hiring: A Quick History (Part One of Four)

Jun 10, 2022

Hiring used to be so easy. Scan a resume, do an interview or two, and find a candidate who was a “good fit.” Any manager would tell you that their gut knew who the right person for the job was.

The problem was that this process was rife with unconscious (and sometimes conscious) bias. People tend to more easily connect with others who are like them – be that gender, race, socioeconomic status, education, or their favorite football team.

How cognitive bias impacts the hiring process

In the 1970’s, psychologists Daniel Kahneman and Amos Tversky defined the concept of cognitive bias in their research into human decision-making. Cognitive bias – also described as unconscious bias – can manifest in various ways. Here are some examples of how it often plays out in the hiring process.

Confirmation bias is the tendency to selectively search for or interpret information in a way that confirms one’s preconceptions or hypotheses. Once a person – like a hiring manager – makes a decision, consciously or unconsciously, they will often ignore contrary evidence that doesn’t support the decision.

Mere exposure effect is where people prefer things merely because they are familiar with them. As an example, an internal candidate is competing with an external candidate for a job opening. The internal candidate is familiar, you bump into her all the time in the cafeteria. You don’t really know her work but she is familiar. The external candidate is not.

In-group bias is when people give preferential treatment to others they perceive as members of their group. We feel most socially comfortable with people who look like us, or are similar in age, class, or other ways. For instance, we “click” with someone who shares the same family status, hobbies or regional origin. This causes bias, and that increased comfort level can lead to interviews where an “in-group” candidate has a greater opportunity to speak, share, and demonstrate capabilities.

Are written tests the solution?

To counteract these unconscious biases, organizational psychologists and HR professionals sought out methods of measurement that were more objective. Most common were written tests in one of three categories: 1) intelligence or cognitive ability tests, (2) personality tests, and (3) interest inventories. But written tests used for candidate selection came under fire because they were generally not validated, and many were considered discriminatory. And while a test can be valid, it’s not useful if it’s not matched properly to the job or indicative of job performance.

In the 1990’s, HR pursued other approaches to reduce bias, such as interviewer training and structured interviews. The latter – also referred to as behavior-based interviews – uses competencies as the foundation. After identifying the key skills required for the job, interview questions are developed that ask the candidate to supply specific examples from their work experience that demonstrate that skill. This was a step in the right direction. The format structures the interview to focus on the job competencies. However, the rogue variable, as always, is the human asking and listening to the answers. Any examples of cognitive bias mentioned above could quickly and easily come into play.

The evolution of data analytics

Advances in artificial intelligence have dramatically changed our world in the past 20 years. Few of us even realize how much AI is used to support our daily activities on our computers and smartphones. AI and, more specifically, data analytics, now empower us to collect huge amounts of data and mine and organize that information for insight and decision-making.

In the early days of 2000 and data analytics, Modern Hire scientists began testing for predictors of job performance.

Over time, as selection science has grown and AI has developed, tools for reducing bias in hiring have become more robust. TheBias in hiring has attracted lots of attention lately (and some might say, finally!), but it’s always been with us. use of data, and the use of analytics to find patterns in the data, is essential to predicting job performance. And, after all, that is the goal of a successful job hire.

Organizations today need hiring software that is not just efficient, effective, and engaging but also ethical. Science-driven, yes, but ethical, based on unbiased data that is validated and proven to predict job performance. Ultimately, your hiring platform should help identify best-fit candidates in a fast, consistent, fair, and engaging process.

If we think of unbiased hiring as the pinnacle, then early attempts have been like hiking up the mountain on foot – HR could only make so much progress. Some of the advanced science and technology available today is like the cable car, getting you closer to the goal faster. The cable car also takes everyone consistently to the same place and makes the same pinnacle possible for everyone to reach.