Hi,

I'm interested in the possibility of using cognitive testing as part of my hiring process. I think that using a data-driven approach to hire for good intelligence and cognition may be helpful.

a) Any general feedback on this concept?

b) Does anyone have any experience in conducting (as opposed to receiving) such tests in the context of hiring potential staff?

c) Any particular recommendations of providers?

d) I'd imagine being a very low-volume user of such services - are there many free or open source tests that would be accessible to someone who's not formally trained in cognition/testing/psychology?

To (a) - it's a great idea and there's a strong correlation between IQ and job performance.

In practical terms it will result in hiring fewer non-Asian minorities, so if you're in a bat**** crazy state where diversity in hiring matters, be aware of that and throw a bone somehow to these people.

In general on hiring, Joel Spolsky (programming blogger who went on to develop StackExchange and others) wrote a great article about what you look for in an employee: smart and gets things done. Everything else is irrelevant for technical jobs, and if either of those are missing, they're a "no hire". Here's the (very long) blog with some great thoughts about hiring related to intelligence and ability - it's programming related but there are a lot of insights in there.

Thanks. I'm likely hiring for various customer service roles with very specific language skills (eg, small Eastern European languages) so demographic diversity is not really much of an issue here. As you'd appreciate, the ethnic diversity of people with native Bulgarian language speaking skills (for example) is obviously very limited.

If it's customer service, why base it on cognitive performance as opposed to building a specific team from specific personality types? I.e. Clifton Strengths finder.

Hiring high quality, capable people seems pretty obvious - I would assume very customer service is a place where a much higher IQ makes little to no difference.

I disagree completely. Hence, my interest in learning more about about how to properly evaluate this in a job interview process.

I'm surprised you're going to the trouble of IQ tests for customer service roles.

That said, the data is pretty clear that higher IQ is well correlated with job success across a broad variety of jobs (not just cognitively demanding professional roles), so it probably makes a lot of sense.

Are you looking for Ladino's?

Omnia group is pretty good, though I'm not entirely sure they do exactly what you need. They are one to look into though

I don’t know if the literature is readily available in public sphere but there is a lot of data in the call center/customer space space about expected performance and wonderlic/whatever test scores. You basically just can’t be too stupid but too smart is no good either.

I have seen wonderlic cutoffs as low as 12 (roughly IQ of 80, 10th percentile) and as high as 16 (~90 IQ). This was over 5 years ago. I wouldn’t be surprised if cutoffs are higher now because customer service software is increasingly complicated and there are hoards of now nearly accentless well educated/smart Indian English speakers happy to work for 15-30 an hour.

I don't know what this means, or even whether you're being sarcastic.
Thanks - I found them at https://www.omniagroup.com
We're not employing a huge chunk of people, so every hire counts. If we can hire better, then that's helpful!

Ladino is Judaeo-Spanish, but evolved with whatever language the natives spoke where jews settled after being exiled, like Bulgaria or Romania in eastern europe. Part curiosity, part sarcasm

For my super low-budget purposes, I found this to be helpful: https://openpsychometrics.org

Thanks y'all for your contributions and feedback (more welcome always!)

Woah... first of all, I think you're going to be just mildly disappointed in the cognitive abilities of people looking for eastern european customer service jobs.

This is why some employers don't require a degree, it's not that they don't prefer it, they are just being realistic. Likewise with drug testing, if you drug test but then aren't prepared to fire employees that fail it then why do it.

I would focus on language skills and EQ, give them a few generic scenarios and have them respond to it. Testing IQ of someone who want to work in a call center with a headset on is a horrible idea.

And... be careful about doing this for higher level jobs as well. Those tests tend to filter out people who have spent time prepping for those kinds of tests. Plenty of very smart people wouldn't be able to construct a good bb defense range if they've never played poker before but plenty of idiots on 2+2 are more than capable of that and that's definitely a cognitive test of sorts.

The companies that are famous for tests like this have created an entire sub industry of prep classes to get a job at McKinsey, Uber or Google. These tests may have been a good measurement of IQ 30 years ago, but these days with all the prep work available online you're really just testing to see if they've prepared for questions like this.

You're just testing for pedantry and prepardeness at this point. Prepare a brief test that basically mirrors what they'd actually be doing in the job and talk to them about their process that led them here to this interview. Hearing someone explain how they found out about the job, what they were looking for, etc etc will tell you a lot about their thought process.

You can hire an Indian developer for a fraction of that... I'm shocked you'd pay that much for customer service

That’s what Americans get paid.

Indians get more like 2-15, sometimes lower. It’s a huge range I know but quality also varies as much. If you can speak English with an accent that Americans/Brits can understand easily, navigate Salesforce and/or proprietary customer service menus (a lot of which are abominable), and deal with cranky customers, you can get a full time job at a higher end call center paying about 10k usd a year (+-2k or so).

If you have additional skills on top of that, you’re starting to look at jobs working for multinationals directly doing back office work remotely. (Think tax returns and compliance for European/American companies and internal IT departments).

The only thing I’d add is the significance of how much people actually want the job. A If you choose the most qualified person for a job that requires a lot in training you’ll be having to eat that cost again sooner rather than later. You need the right combo of competence and desperation.

I think you'd be surprised. But either way, I just want an extra data point to make more data-driven decisions.

I'm not really worried about people studying IQ tests to try to game my effort here, that's not a credible risk.

The challenge is that a number of candidates have no direct experience, so judging them on past activities/work history is challenging.

I don't have a good way to measure future potential (especially amongst staff who are often relatively young in their career) so this seems like a good additional piece of data to consider.
I hear where you're coming from, and appreciate this issue.

Toothy Nazied up this thread in record time.

If you really want to do this, just ask for standardized test scores. They correlate well enough and you'll save lots of money.

IIRC, in the US, it is illegal to do so as it will be considered racial discrimination by default unless you can demonstrate the necessity of the cognitive test and what you are testing in relation to the job responsibilities. An example would be police IQ tests to filter out anyone above 115 IQ due to their internal studies about higher IQ people being unable to make decisions within the time required by a police officer acting in normal duties. Even so, that still had to be litigated out.

In other countries it's probably okay.

Bolded is an interesting tidbit I'd never heard. They really want animals out there acting on their primal instincts. Reminds me of intelligence being vetted out of football players

Normally I'd agree with this 100%

But I imagine anyone who's applying for a job to talk to angry Bulgarians over the phone in a calming manner and still has a boss issuing IQ tests to boot will be desperate AF for some work

You missed my point, which was it's not really a good measurement. Especially so if their job is customer service. It truly is as arbitrary as asking for a BB defense range.

I am with the other guy, this is a bad job nobody wants, hire the guy with the least prospects who will be able to do it. You don't want someone leaving you in 6 months for another place with better prospects.

As you said yourself, it's a small team, just like you want to make the most out of each hire - each talented and ambitious hire will see that it's a small team and thus not much upward trajectory unless they leave. Someone may need any job today but will still keep looking for a better one even after they accept your offer. This is especially true once after a few months they realize that just because you work in a small company means you'll grow and expand along with it (if that even happens).

I've been a cofounder at a number of startups. A big part of hiring isn't just finding the all stars but knowing where to just get something reliable for the lesser roles. Something stable that won't leave you needing to do the hiring and training process all over in January. Finding the most intelligent young guy for a dead end position is a mistake you'll be repeating several times a year until you hire 43 year old Bob who just wants to slowly pay off his mortage and feel smart getting the $100 questions right on Jeopardy. Bob will stay with you until retirement.

This job is about making customers happy, that's not something that requires a high IQ. It requires a people person with basic intelligence.

rickroll,

Your assumptions and characterisations of the job (and applicants!) are entirely inaccurate, and they form the premise of a generally unhelpful post as a consequence. That said, I'm grateful for your effort and honest point-of-view.

Of course there are other (and more important!) factors than IQ. I was just interested in obtaining more data to make better decisions about hiring.

I work with VCs and advise a lot of startups, everyone seems to think they are the best employer ever and their company is the best company to work for in the world and yet a good 2/3 of the major things that go wrong at small companies is when a member of the 20 man team quits and everyone has to scramble and cover until they find a replacement. They never see it coming, that company is their baby and have a difficult time separating their own obvious interest in the company and it's success from the lower level employees viewpoint. They are generally someone who was only there for the $60k a year and then that incentive vaporized when they got an offer for $65k that had a better commute. So I'm a big advocate of hiring for reliability when it's not a high level position.

This is 100% why so much salary is tied up in unvested stock at startups, because when someone leaves a small team it can cripple the company. As you hire him he's a luxury, but 4 months from now once he's integrated he's going to become a necessity.

IQs, ambition and career jumping are all a bit correlated.

If it's not someone who is going to personally change the direction of the company, I'm not too worried about whether or not they can solve a rubik's cube - yet another thing where the solving factor isn't IQ but rather filtering those who bothered to google how to do it.

I didn't mean any contention or ill feelings or anything. It also seems I was wrong to underestimate your objectivity over the quality of both the position and the applicants. It's harder to communicate when not in person. Without that visual cue like a smile everything can come off unpleasant. I'm sorry if it came off that way.

Good luck

-Adam