Hiring: evaluating collegiality and "fit" without bias?
August 23, 2017 4:15 AM   Subscribe

I'm a member of a small research group that is making some hiring decisions and I've been asked for my feedback on a few candidates. Could someone point me to some resources/best practices about how to evaluate whether or not someone seems like they'd be a good colleague, without imposing a notion of "a good fit" that could be unfair (especially to under-represented minorities)?

We're a small group, and negative interpersonal relationships can not only be stressful but can damage group productivity. So, like many similar groups, we want to be careful with hiring. We don't want to be stuck with a colleague who is inflexible, incurious about others' work, unfriendly, a poor communicator, territorial, defensive, arrogant, excessively competitive, disingenuous, or whatever. On the other hand, I'm also aware of the literature on how selecting for social "fit" can really involve selecting for unconscious biases or can be otherwise unfair, and I'm a (gay) cis, white, middle-class, able-bodied, allistic male.

My question is, what are some good "best practices" for fairly evaluating people on the dimension of collegiality, which is kind of nebulous but also important? Some types of bias are obvious to me (e.g., accent, age, non-work activities), but I'm sure there are also many potential biases I'm not considering. One suggestion I found is to write out a rubric for my own evaluations. That seems like a good idea, but I'm having trouble articulating what exactly it should or shouldn't include, so that it captures the right things while still being fair.

For reference, I'm in an academic lab in STEM. These interviews have already happened, but advice for how to conduct future interviews to minimize bias would also be welcome, especially because I'm hoping to start my own group at some point. (Anon because it's about my workplace, which is probably guessable from my profile.)
posted by anonymous to Work & Money (11 answers total) 26 users marked this as a favorite
 
Looking forward to more answers on this myself. Just a quick thought for now: A rubric is a good idea and then going over a list of known biases/stereotypes and see whether your expectations and assessments are constant across demographics. For example, stereotypically, women are supposed to be more nurturing, deferential, friendly, helpful, etc. Check whether you need to counter that bias. Check the phrasing and thus interpretation of assessments, was the male candidate actually confident and assertive and the female candidate overbearing and aggressive or did they exhibit very similar behavior? There have been studies showing that women have to publish much more than men to be viewed as having equal scientific productivity (Wolt et al.), what is the case here? Other studies show that women are perceived to talk overly much even before reaching their share of time/person. Did anyone think the fat person was not so sharp or lazy, is that based on facts? Did the person with an accent get labeled as hard to communicate with, what led to that? Are male candidates evaluated on potential and female candidates on proven track record? If the topic of family comes up, is that seen as a bonus for men (stable environment and support) and a detriment for women (will be distracted by obligations and de-prioritize research)? Basically, list all the biased stuff you see around you and read about and then actively question evaluations point-by-point.
posted by meijusa at 5:09 AM on August 23, 2017 [6 favorites]


The way my company does this is to figure out what's important to us (in our case, based around the company values and anything specific to the department/team that's hiring) and then ask very specific questions around that area. One thing we care about is getting work done in teams with lots of collaboration, so the questions are "tell me about a time" format to fit that theme, and the hiring guidelines have a bit of detail on what a good answer vs a red flag answer would sound like.

This way you're specifically filtering for the qualities you want to complement your team, which hopefully should be distinct from things that you might be ruling out if you were looking more broadly for people who are like us/people we think would fit well into the team for reasons we can't articulate. The questions should be job- and attitude- specific enough that they're about the person as an employee and not really anything to do with their gender/sexuality/religion/health status etc.

We've also reviewed our hiring guidelines recently with diversity in mind, and ended up making some changes based on this (e.g. there was a section about looking for eye contact at interview, which isn't actually necessary as long as the person is a good fit in the job-specific areas, and could have been unfairly filtering out candidates on the spectrum).
posted by terretu at 5:12 AM on August 23, 2017 [5 favorites]


Also Microsoft have free online training on unconscious bias. It's more about recognising this in the workplace than about hiring, but it only takes 30mins or so to complete and it's got some useful refreshers on this stuff.
posted by terretu at 5:15 AM on August 23, 2017 [2 favorites]


I think you're trying to do something almost impossible. If you're evaluating candidates on intangibles rather than concrete skills, keeping bias out of it is super hard.

What I might try to do is give up on looking for fit, because you're inevitably going to prefer the demographically familiar candidate on the intangibles, and limit yourself to indicators of non-fit. If you've got a concrete reason to think someone would be a personality problem, note it, think about whether it might be the result of demographic bias, and only consider it if you're sure it isn't.
posted by LizardBreath at 6:07 AM on August 23, 2017 [13 favorites]


It's not that hard to come up with a set of "tell me about a time when" questions that, if asked consistently, will tell you what you need to know.

Think about questions like "tell me about a time when" --

- you had a conflict with a work peer and what you did to resolve it
- you had a difference of opinion from your manager about what ought to be done
- someone around you wasn't holding up their end of a group project

&c.

Try not to have "ideal answers" in mind when you ask these. Just see what comes up.

The problematic "fit" evaluations, in my experience (at a company which was famously selective and culture-heavy) have to do with questions like "what do you do for fun?" or giving points to work-irrelevant stuff like having been a competitive swimmer in college, or docking people for having gone to a second tier school even if the reason they went there was that they got a full ride scholarship. I saw a lot of this when I was an in house recruiter, and the results were predictably ageist, classist, and sexist.
posted by fingersandtoes at 7:41 AM on August 23, 2017 [14 favorites]


To add to the examples above:

"Tell me about your best experience working on a team."
"What role do you usually fill on a team project?"
"What would your ideal team look like?"

These all look for ability to work on a team rather than a "fit".

In my experience, usually I don't hear an answer and think, "Wow, that's exactly what I'm looking for." Usually good answers are basic, "we overcame a challenge with our powers combined and it was great." Instead what I find is that those questions easily bring up folks who are not great team players.

Examples:
* One person's favorite team experience was working as a liason for two teams. It was working by themselves to aid the teams rather than being a part of either team.
* Many, many people who have never worked on a team at all. They might have a team story, but you can suss out that their idea of collaboration is working alone but reporting to the same supervisors.
* Sometimes a person's story is about trying to make a team when everyone around them wanted to work alone. They have no actual team experience, but this shows that they wanted it so badly they went against the grain to try and make one.
* Some people's team experiences show them repeatedly leaning on someone else for help but never letting anyone lean on them.

This article in Model View Culture talks about the problems for trying to hire for "fit" and provides examples and links to further resources.
posted by tofu_crouton at 8:51 AM on August 23, 2017 [3 favorites]


I would also add that you take extra time after the interview to examine your assumptions about candidates who do not resemble the existing team before you reject them. Whatever you reasons are for rejecting a minority candidate, take extra time to mull over if there might be cultural reasons for their answer. Take for example someone who comes across as sounding too harsh with clients. Is it maybe that they are coming from an industry where that's common? If someone sounds like they don't bond with their team on a personal level, consider if maybe it's because they have children and couldn't participate in afterwork activities. In some cases, you won't have the information to make these guesses, but in some cases you will.
posted by tofu_crouton at 8:55 AM on August 23, 2017


We don't want to be stuck with a colleague who is inflexible, incurious about others' work, unfriendly, a poor communicator, territorial, defensive, arrogant, excessively competitive, disingenuous, or whatever.

I can see what you're getting at, but that seems difficult to define and identify, and probably kind of excessive.

For instance, why 'friendly'? What does that mean, really? Someone who's not default hostile? Would it be OK if someone was polite and pleasant but just didn't necessarily want to be friends with their coworkers? Because people who have less in common with your group might not necessarily want to be friends with them, which is one of the major ways these homogeneous cultures get perpetuated.

What does 'inflexible' mean? Are you talking about rule cops or people who really, really have to pick their kids up by 6?

By poor communicator, do you mean people who have a hard time being understood, or people who don't bother to try? Because the latter would be fine to filter by if you could, but there are many reasons, some entirely the fault of the listeners, that someone might have a hard time being understood. Apart from the obvious things like people whose speech, neurology, or cultural background is outside the norm in your workplace, there's a whole lot of unconscious bias that comes into how people parse what someone is saying based on their perceptions of that person. It is very, very common for people to unconsciously 'stupid up' or selectively dismiss things others say based on their preconceived notions about the speaker's demographic. It happens to women all the time, and while ultimately, the communication fails, the problem is the listener, not the speaker.

Which brings us to being territorial and defensive and that type of thing. There are lots of things that default majority types can afford to be chill about that others can't. You ask one of the ten white guys in the room to do some kind of scut work or something, it's different from if you ask the only woman or PoC or something. Someone who has been a minority in their workplace has likely been mistaken for a secretary or a janitor many times, and have often had to come up with a default response to that. Similarly, women in male dominated professions, for example, have their ideas and their actual work taken over or just outright stolen with alarming frequency. Men will have some stories about that, but women have ten times more. It would follow that women would be likely to develop some defensive and territorial behaviors that men don't.

If you're interviewing people who have some work experience, it is very likely that people who've been a minority in their profession have experienced more hostile behaviors in the workplace, and be on guard against it, which can come across as defensive, inflexible, excessively competitive, territorial, or whatever.

If you have a fairly homogeneous bunch of employees, your list of ideal candidate qualities will often reflect that, and it may be unreasonable to expect someone else to come in and fit into that culture.

What it comes down to is that fostering a healthy and inclusive workplace culture is your job. I don't know how well you can really filter for personality issues without filtering people based on their personal experience. What you can do, however, is stay on top of your workplace's culture. Hire people with the right professional qualifications, and keep an eye on how people interact and how they treat each other. Discourage negative behaviors, encourage positive ones, and ensure that everyone treats everyone else fairly, and when they don't, address it quickly and decisively.

No matter what you do, you risk hiring someone who is disruptive, hostile, and undermining. I don't know that that sort of thing can be filtered for effectively, as those types of people are often also pretty deceptive and know how to say the right things to the right people. I really think your best bet is to deemphasize 'culture fit' expectations in the hiring process and work on building the culture you want through effective management.
posted by ernielundquist at 9:47 AM on August 23, 2017 [8 favorites]


Too late for this round - but something that I'm trying to implement at my company is redacting all personal information from application materials. So anything that could give a clue to gender, race, economic background like names, addresses, etc. removed from resumes and coverletters by a staff person not involved in the hiring process.

This will not help with interviews but could avoid the problem of excluding whole categories of applicants due to bias.
posted by sol at 10:37 AM on August 23, 2017 [1 favorite]


You might also review your interview process to see if it allows personality to come through. An in-person group interview of 10 is different than a group phone interview or a series of individual interviews or a skype interview. One of the best group phone interviews I've ever been part of was led by someone who did a great job of making it seem like a conversation rather than checklist of questions.
posted by beaning at 11:18 AM on August 23, 2017


Not precisely the same, but I'm in Seattle and recently went to a great training by the Nonprofit Assistance Center on applying equity principles to grant proposal review. The training actually covered a lot of what you're asking about, and I wouldn't at all be surprised if they could offer training or resources on hiring. (And for what it's worth, I work for local government, not a nonprofit, and they were happy to work with us.) If you happen to be around here, I highly recommend.
posted by centrifugal at 1:11 PM on August 23, 2017 [2 favorites]


« Older What do I do with a sudden influx of cash due to...   |   Should I disclose my PTSD and Generalized Anxiety... Newer »
This thread is closed to new comments.