Hacker News new | comments | show | ask | jobs | submit login

I couldn't disagree more. While I've only once taken an intelligence test for a job, it was a relaxed thirty questions in thirty minutes involving pretty simple mathematics and pattern spotting, with some pretty low level word play.

I believe that this kind of intelligence test correlates well to being able to think coherently and abstractly in both numerical and non-numerical terms, and as such correlates well to being a good programmer.

As a proxy for developer performance, intelligence is strongly correlated.

A bit of street-smarts also helps with those tests:

In my one, there was clearly not enough time to do all the questions.

So I picked the questions I could easily quickly answer accurately; followed by the ones I could take and educated guess; then finally as time was running out just out and out guessed the rest. I did exceptionally well apparently :-)

Then I got the job, and worked on a typical CRUD system.

"Street-smarts"? That sounds more like school-smarts. Going for the easy ones first is standard practice when taking tests.

Hey, I followed the same strategy. I got hired too. And now I work on mainframes :X

So, on the basis of once having taken an IQ test that you found not problematic and after which they hired you, you feel good about them. That's great. Sample size of 1, obviously it must be unquestionably a good thing.

If developers aren't evaluated as people, but as units who are expected to hit some threshold of "intelligence," then that says something about the company asking you to take the test.

If everybody who tested well included everybody who was intelligent, or able to code well, you might be able to use it as a proxy. If not, you're missing out as a company.

The students who achieve the highest grades in school aren't those at the peak of the IQ range - they're the ones who actually have to put in the study time to learn the material, rather than cruising on previous knowledge. Do you want the person whose IQ score is better, or do you want the one who will do the work? If you want the better IQ score, what will you do to get them to perform when asked to do what might be regarded as drudgery?

Or are you saying that they give IQ tests to screen out the really bright bulbs, so as to isolate those who are willing to slog through whatever they're given?

> Do you want the person whose IQ score is better, or do you want the one who will do the work?

How about someone who is both smart and willing to work hard?

smart people are inclined to work smarter not harder :-)

For a while, it's the same thing.

Only on HN could someone argue that trying to select smart employees for technical jobs is a bad idea.

So the thing is, "smart" is a pretty sloppy term. Furthermore, claiming that measuring this "smartness" in a small-sample, unrelated set of tasks (a relaxed, 30 question test), and that this "smartness" on the test would transfer to "ability to write good code well and work in a team" is utter crap.

Doing well on a test is usually an indicator that you do well on that kind of test. Even good psychometrics tend to fail with calibration problems - the vast majority of these tests were written and calibrated using psychology undergraduates at western universities. That's a significant sampling problem. [0]

So I'm personally not opposed to selecting for "smart" employees. But I do think that the belief that a short test actually selects for that kind of smartness in any meaningful way is sloppy thinking.

0: http://blogs.scientificamerican.com/primate-diaries/the-weir...

I would much rather work with a team of people of average intelligence, with humility, work ethic, honest enthusiasm, empathy, etc., than with a team of brilliant people with none of those things. That goes double for technical jobs where you're building a complicated product over the long term. We're not here to solve the Putnam and go home. We're delivering value, and that involves solving a hard problem simply because there is no easier way, not because hard problems are inherently worth working on.

You people just read your own prejudices into everything.

So according to you, doing mediocre on an IQ test makes someone not smart, right?

What about people with dyslexia (such as myself)?

What about people who have test-anxiety?

They aren't arguing that hiring smart people is a bad idea, they're saying being dismissive and discriminatory is a bad idea (and illegal, in the US - violation of the ADA).

So, you think that hiring smart people is a good idea, and you think that people who do well on IQ tests are smart, and you think that people who are not smart do poorly on IQ tests, but you are adding that there are some additional smart people who don't do well on IQ tests, and you are rightfully concerned because you consider yourself to be one of them, and you think there are laws that put the burden of uncovering these people on private employers? I'm not sure that's true, but anyway, just wanted to clarify.

Out of curiosity, this is a serious question: are you a programmer and if so does dyslexia cause you problems coding? lots of library functions can have confusingly similar names and spellings for me, and I don't have any reading problems.

If I may pick a small nit. People who get good grades are "smart", people who score highly on IQ tests are "intelligent" - those characteristics are essentially orthogonal. It is generally easier for intelligent people to get smart, but they can just as easily get stupid.

Generally speaking, good developers are both smart and intelligent.

> As a proxy for developer performance, intelligence is strongly correlated.

I think sampling some code from the applicant could be a better indicator.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact