×

Our Turn: The trouble with whiteboard interviews



Last modified: Tuesday, December 22, 2015
In a week when Secretary of Defense Ash Carter announced that all combat jobs in the military are now open to women, signifying the openness of one of America’s most traditional institutions to radical change, an alumna of Dartmouth College interviewed for a Bay Area tech position. The alumna has a vivacious personality and many accomplishments, including a graduate degree in video game design. She has coding and design experience, and has created her own complete games for release. She also brought prior technology industry experience to the table.

But the alumna had a surprise in store for her. She encountered what has become an industry standard: the whiteboard algorithm interview. Well prepared to describe her experience with game design and assembling productive teams, she was put off by this type of interview, which fails to test even the coding experience done solo at a neutral computer screen. Instead, she faced a performance test to “show off” her coding chops in front of an all-male panel.

The trend toward “whiteboard interviews” constitutes a form of hazing, bound to attract only certain personalities and demographics to tech jobs. Members of stereotyped groups such as women and people of color are particularly vulnerable to the use of this interview technique in fields dominated by white males. This vulnerability is backed by years of research in psychology but is currently ignored by tech companies.

Whiteboard coding interviews have the character of an oral exam, something those of us with Ph.D.s have experienced and almost all graduate students dread. On-the-spot technical performance by a woman in front of a group of men is scientifically predicted to lower performance.

Social psychologist Claude Steele found years ago that achievement gaps are tied to situations where people can be reminded of stereotypes: We actually seem to make prevailing stereotypes come true, whether we like it or not. This phenomenon works across many situations. For example, white men reminded of their “whiteness” end up proving the stereotype that “white men can’t jump.” In a math classroom, mentioning “women” as a group, even unrelated to a test, will unconsciously remind women of stereotypes about women and math and will lower their test scores. As a result, underrepresented groups in particular are made to feel “unworthy” of the time taken to be interviewed by the interview process.

In contrast with the lone academic model, itself an anachronism in an age when researchers collaborate across the globe, successful design teams incorporate diversity across the spectrum. Diverse teams in business only strengthen them: as “7 Habits” guru Stephen Covey noted, “Strength lies in differences, not in similarities.” Furthermore, diversity has been shown to improve the way people think, benefiting both the minority and the majority.

The highly male-dominated field of video game creation is not unique in discouraging women to rise professionally. Similar barriers are seen in the physical sciences and engineering where women remain a minority among practicing professionals, despite their numbers rising in academic programs.

What underlies this disconnect between attracting women to technical fields and seeing them succeed professionally is employment barriers once they enter the marketplace. To bring the perspective and talent of the broadest group of consumers to the table, interview practices need to simulate the interactive workplace experience of the most successful tech companies. A whiteboard exam tests what examiners already know. They should be looking for what they don’t know.



(Mary Hudson is the Eleanor and A. Kelvin Smith Professor of Physics and Astronomy at Dartmouth College. Mary Flanagan is the Sherman Fairchild Distinguished Professor in Digital Humanities at Dartmouth College.)