5 ms·

I’m very skeptical that someone with an IQ of 80 could be employed as a software developer. I think other explanations are more likely. Your coworker could be j

by chroma 5y ago

I’m very skeptical that someone with an IQ of 80 could be employed as a software developer. I think other explanations are more likely. Your coworker could be joking or lying. He could be mistaken (eg: interpreting a percentile as an IQ score). The test could have been mis-graded. Or maybe he was just having a really bad day. If he took an IQ test again, I seriously doubt he’d get such a low result. A person with an IQ of 80 usually has trouble functioning in society. It can be difficult to teach them concepts like alphabetization or hypotheticals. For someone with such cognitive impairments to become a software developer beggars belief.

vbsteven 5y ago
It could also be a partial IQ score of 80. IQ tests are divided in 4 key areas each scored separately. If all 4 scores are similar they can be averaged to get the final tested score. If they are not it’s called a “gap” and the average score is meaningless. With gap results it’s not uncommon to see 2 low areas scoring 80s and 2 high areas 120 or even 130+. “A developer declaring a tested IQ of 80” could be a gap case and somewhere along the way the nuance got lost in translation.
- sudosysgen 5y ago
  Not necessarily. They could also score in the 70s in 2 areas, score 115 in another, and 90 in the last one, and get a score in the 80s. Beyond that, IQ is not even predictive enough of job performance for a standard deviation to be a death sentence.
sudosysgen 5y ago
Come on, IQ 85 is only one standard deviation below the average, and correlates at around .4 with job performance. That means you'd expect someone with IQ 85 to be on average less than a standard deviation under the average at job performance. It's definitely possible for someone with IQ85 or IQ80 to be better at programming than the average IQ100 programmer, and it's not even that unlikely. Let's not overstate this.
- ggreer 5y ago
  The average software developer does not have an average IQ. Look at the figures at the end of this paper[1]. Even professions like nursing have IQs averaging close to 110. "Science and engineering occupations" are around 115 average, with nobody below 90. If you look at college majors, you'll get similar results. Our occupations and interests cause us to self-segregate into cognitive strata. For the same reason you don't know many Trump voters or young earth creationists (despite the fact that they're 1/3rd of the US population), you probably don't know many dumb people (or many super-geniuses). Individual differences in cognitive ability are far greater than our personal experience indicates. 1. https://www.gwern.net/docs/iq/2002-hauser.pdf https://www.gwern.net/docs/iq/2002-hauser.pdf
  - sudosysgen 5y ago
    According to the paper you linked, right around 20% of people in IT, materials engineers, and even people in the natural sciences have an IQ in the 80s or lower. (under 90). Wouldn't you believe that, that's only slightly less than the general population which is at 25%, or right around what I predicted above :) The median IQ difference between college professors and service workers is not even a full standard deviation. Beyond that, I said average IQ100 programmer, of which there are many that can do a perfectly serviceable job in many sectors. Individual differences in cognitive ability are great, yes, but they are not even close to being perfectly represented in IQ. I don't know many Trump voters or young earth creationists because of where I live, but I know quite a bit of their relative equivalent where I am right now. Yes, there are a lot of differences, but I know enough young-earth-creationist and far-right engineers/stem people to know that it's not that simple. Anyways, I think I made my point clear, IQ is not a very good predictor of job performance, you can definitely be a programmer with IQ85, and a great many are.
    - ggreer 5y ago
      You misread the figures. The bottom of each bar is the 10th percentile. If you draw a horizontal line, "Natural science - physical, life, and math" has its 10th percentile around 92.

by chroma 5y ago

vbsteven 5y ago

It could also be a partial IQ score of 80. IQ tests are divided in 4 key areas each scored separately. If all 4 scores are similar they can be averaged to get the final tested score. If they are not it’s called a “gap” and the average score is meaningless. With gap results it’s not uncommon to see 2 low areas scoring 80s and 2 high areas 120 or even 130+. “A developer declaring a tested IQ of 80” could be a gap case and somewhere along the way the nuance got lost in translation.

sudosysgen 5y ago
Not necessarily. They could also score in the 70s in 2 areas, score 115 in another, and 90 in the last one, and get a score in the 80s. Beyond that, IQ is not even predictive enough of job performance for a standard deviation to be a death sentence.

sudosysgen 5y ago

Come on, IQ 85 is only one standard deviation below the average, and correlates at around .4 with job performance. That means you'd expect someone with IQ 85 to be on average less than a standard deviation under the average at job performance. It's definitely possible for someone with IQ85 or IQ80 to be better at programming than the average IQ100 programmer, and it's not even that unlikely. Let's not overstate this.

ggreer 5y ago
The average software developer does not have an average IQ. Look at the figures at the end of this paper[1]. Even professions like nursing have IQs averaging close to 110. "Science and engineering occupations" are around 115 average, with nobody below 90. If you look at college majors, you'll get similar results. Our occupations and interests cause us to self-segregate into cognitive strata. For the same reason you don't know many Trump voters or young earth creationists (despite the fact that they're 1/3rd of the US population), you probably don't know many dumb people (or many super-geniuses). Individual differences in cognitive ability are far greater than our personal experience indicates. 1. https://www.gwern.net/docs/iq/2002-hauser.pdf https://www.gwern.net/docs/iq/2002-hauser.pdf
- sudosysgen 5y ago
  According to the paper you linked, right around 20% of people in IT, materials engineers, and even people in the natural sciences have an IQ in the 80s or lower. (under 90). Wouldn't you believe that, that's only slightly less than the general population which is at 25%, or right around what I predicted above :) The median IQ difference between college professors and service workers is not even a full standard deviation. Beyond that, I said average IQ100 programmer, of which there are many that can do a perfectly serviceable job in many sectors. Individual differences in cognitive ability are great, yes, but they are not even close to being perfectly represented in IQ. I don't know many Trump voters or young earth creationists because of where I live, but I know quite a bit of their relative equivalent where I am right now. Yes, there are a lot of differences, but I know enough young-earth-creationist and far-right engineers/stem people to know that it's not that simple. Anyways, I think I made my point clear, IQ is not a very good predictor of job performance, you can definitely be a programmer with IQ85, and a great many are.
  - ggreer 5y ago
    You misread the figures. The bottom of each bar is the 10th percentile. If you draw a horizontal line, "Natural science - physical, life, and math" has its 10th percentile around 92.