2

Background

Currently, many companies hiring software development engineers use a whiteboard algorithms interview technique. This involves asking the job candidate to code a solution to an algorithm question on a whiteboard while the hiring manager and other team members watch and critique.

This interview technique has long been criticized as making the candidate uncomfortable, as ineffective, and perhaps at least a little demeaning. Many in the industry believe it is little more than a hazing ritual. See the following links for more information:

Why is hiring broken? It starts at the whiteboard.

Tech Hiring Has Always Been Broken. Here’s How I Survived it for Decades.

!Whiteboard: A better way to hire the best developers

So, even though I and many others don't believe in the effectiveness of whiteboard interviews, I have supported the right of companies and individuals to legally use them, at least until a few weeks ago when I happened across a university study of whiteboard interviews.

Does Stress Impact Technical Interview Performance Study

This is a study out of North Carolina University conducted last year. An article titled You're testing them wrong: Whiteboard coding interviews are 'anti-women psychological stress examinations', published in July of last year by the Register says about the study:

As a consequence, whiteboard interviews may fail to assess coder competency. Rather, the researchers argue, they measure how well job candidates handle anxiety.

The researchers found that stress hinders interview performance, with participants in the traditional technical interview exhibiting higher cognitive load, lower scores, and higher stress levels. In essence, social anxiety took otherwise qualified job candidates out of the running because of the circumstances of the interview.

What's more, whiteboard technical interviews appear to favor men over women.

'We also observed that no women successfully solved the problem in the public setting, whereas all women solved it correctly in the private setting,' the paper says.

This is a link to the study. There is also another article published by the university that summarizes the results of the study.

Currently, just 14% of software developers in the US are women. This study suggests that one of the primary reasons for this poor representation is because of sex discrimination.

So, is this new study enough to encourage legal experts representing large corporations to at least worry about sex discrimination lawsuits from women that have failed whiteboard interviews? If you are a corporate attorney and it was your job to minimize legal risks to that company, would you recommend to human resources that whiteboard interviews are now too legally risky and the company should avoid them, or do you believe instead that it is nothing to worry about and it is okay to do business as usual?

5
  • 1
    But development is a social endeavor. I would argue that most developers should not be hired unless they can write an essay (in English) on the spot. The biggest difficulty that developers face is communicating their ideas to each other. Technical problems are largely a non-issue once you reach a certain minimal level of proficiency. It's also not at all clear why you think that man, on average, communicate better than women. Traditional view is quite the contrary. – grovkin Feb 5 at 4:51
  • 1
    The question is not about communication it is about command performance and anxiety/confidence. Confidence can be false confidence and performance under a spotlight may or may not be a valid job criteria – George White Feb 5 at 4:57
  • @GeorgeWhite interacting in such an environment is part of verbal communication skill set. – grovkin Feb 5 at 4:59
  • A test of their verbal skill set would be a more valid test if the topic were not sprung on the applicant as a puzzle. I don't personally have an opinion on puzzle testing under pressure as a valid or not valid data point in evaluating someone for any specific job. It might be for some and not for others. – George White Feb 5 at 5:05
  • 2
    This should not be closed as opinion-based. Answers can indicate what factors have made other teats and qualifications permitted or forbidden as gender discrimination. They can alaso refer to general legal principles in this are, including the degree to which disparate-impact analysis is accented by courts. This is objective enough for this site. This Q should be reopened. – David Siegel Feb 5 at 15:34
6

Barrister to instructing solicitor: “Am I working for the plaintiff or defendant?”

Plaintiff

Your honour, this excellent study and the expert testimony you have heard from the authors, clearly demonstrates that whiteboard testing is sexual discrimination by another name. The clear correlation between the gender of the interviewee and their ability to perform this task, which is unrelated to the job and is actually a test of how good the person is at handling performance anxiety, is being used to prejudicially and unlawfully screen out female applicants.

Defendant

Your honour, this discredited paper and the inconsistent testimony of the author clearly demonstrated that the methodology was flawed and the conclusions are unsupportable. When pressed, it turned out that the number of females encompassed by the catch-all word “all” was 6 - hardly a sample big enough to draw any meaningful conclusions from. Further, it turned out that the number of men who also improved was unknown because they didn’t test them! So we have the authors of a study with gender discrimination in its very methodology drawing asinine conclusions from it about gender discrimination!

It may be that whiteboard testing is better at finding competent actors than programmers and this may make it a terrible recruitment tool. However, my client can run their businesses in whatever inefficient way they choose so long as it isn’t unlawful discrimination - and this isn’t that.

0

Are companies at sexual discrimination risk for using whiteboard interviews during the job interview?

No. The conclusion that a criterion for hiring is suboptimal or defective does not mean it is unlawful. For it to constitute sexual discrimination, a male candidate would have to perform worse than female candidates in the interview & tests and still be hired in lieu of those more competent females.

This study suggests that one of the primary reasons for this poor representation is because of sex discrimination.

That suggestion seemingly ignores that the supply of workforce in software development is predominantly male. This is palpable, in part, from the significantly lower percentage of female enrollment in IT (or related) college/university degrees.

Some proponents of theories of feedback loops and systemic reinforcement of stereotypes would posit that the collective perception of "sexual discrimination in the IT labor market" reinforces the gender gap in IT education/enrollment (that is, that the rationale of "why bother enrolling if only males will be hired" discourages females from enrolling at all). However, that perception falls short of proving that a company indulges in sexual discrimination by selecting candidates via whiteboard interviews.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.