3 ms·

If you have two indistinguishable candidates on relevant dimensions who both passed the interview bar and are the best two candidates you have for one position,

If you have two indistinguishable candidates on relevant dimensions who both passed the interview bar and are the best two candidates you have for one position, literally just flip a coin, hire the winner, and move on.

lapcat 3y ago
> If you have two indistinguishable candidates on relevant dimensions who both passed the interview bar and are the best two candidates you have for one position, literally just flip a coin, hire the winner, and move on. Agreed. Now explain how having completed a degree is a "relevant dimension", unlike height, age, etc. You ignored my first question: "Have you considered that if it's possible for a dropout to performally equally to a degree holder in every perceivable respect, it raises doubts about the value of a degree?" My second question was actually intended to highlight the issue. What you appear to be doing with "I have evidence that the person with a degree can follow rules/policies and track towards a long-term goal with success" is reaching for one arbitrary personal characteristic and inventing an idiosyncratic, non-scientific psychological theory in order to justify discrimination. Others do the same thing with age and gender, for example (younger people and men are just better at programming, they say). You don't know the life story of these people, and there are any number of reasons for someone not completing a degree: monetary problems, family problems, a major illness or injury, natural disaster, etc. Moreover, if you select just based on degree completion, you give job candidates no opportunity to show evidence that they "can follow rules/policies and track towards a long-term goal with success" in ways other than competing a degree. It's simply a bias toward degree holders. By your own admission, you're not using the degree as evidence of job-relevant skills but rather as a psychological, "life story" factor.
- sokoloff 3y ago
  I am biased towards degree holders for programming and other tech roles. It’s not a requirement or some magic talisman of “hire immediately”, but all else equal, I’ll prefer it and, all else equal, I will use it to discriminate (in the “distinguish” or “tell one thing from another” sense). > Now explain how having completed a degree is a "relevant dimension", unlike height, age, etc. As a population, I believe that college graduates almost certainly score higher on measures of intelligence and conscientiousness than non-grads. I believe that's both true and valid for use in selecting employees for roles where intelligence and conscientiousness are useful qualities. > Have you considered that if it's possible for a dropout to [perform] equally to a degree holder in every perceivable respect, it raises doubts about the value of a degree? Of course it's possible for a non-degree holder to outperform a degree holder, or for a Mech E or English major to outperform a Comp Sci grad. That doesn’t mean I think a Comp Sci degree has no relevance or value to an SWE job. You seem to be imagining some boogeyman where “has degree” is treated as the high-order bit rather than what I’ve been consistently saying (and what I think is relevant to the article under discussion): that it’s closer to a tie-breaker (or low-order bit) and a sensible one in many regards. If someone out-interviews another for a role and clears the bar, you hire the out-performer. When you otherwise can’t tell which candidate is stronger and get down to the area where coin flips become a sensible selection mechanism, I think it's wise to choose the degree holder over using the coin. Other people might rely on height or appearance; I can't control what they do.
  - lapcat 3y ago
    > As a population, I believe that college graduates almost certainly score higher on measures of intelligence and conscientiousness than non-grads. I believe that's both true and valid for use in selecting employees. 1) You can believe whatever you want to believe, but that doesn't necessarily make it true. 2) If you're looking for IQ, then why don't you administer IQ tests to job candidates rather than looking at an extremely unreliable proxy for it? 3) You're admitting to using population statistics to evaluate individual job candidates. Do you even realize what you're saying here? That's literally prejudice. By the exact same reasoning, you could discriminate based on race, for example, e.g., Asian vs. white vs. black. The only difference between racial discimination and degree discrimination in this respect is the legality of it. Even if I granted for the sake of argument that your empirical speculation about scores and measures were true, you're just talking about averages, and many individual members of every population will be above average or below average. > You seem to be imagining some boogeyman where “has degree” is treated as the high-order bit rather than what I’ve been consistently saying (and what I think is relevant to the article under discussion): that it’s closer to a tie-breaker (or low-order bit) and a sensible one in many regards. It is often used as the "high-order bit". Like I already said a few comments ago, the quote from the article was different from your quote. But I don't think it's a sensible tie-breaker. You're taking a degree as a "life story" factor, but that's an arbitrary, singular choice of many possible life story factors. Also, you seemed to change your story a bit in the last comment, talking specifically about a comp sci degree rather than a degree in general, but there's nothing unique about comp sci vs. other degrees to indicate "can follow rules/policies and track towards a long-term goal with success" or to indicate "IQ".
    - sokoloff 3y ago
      > You're admitting to using population statistics to evaluate individual job candidates. I am, in effect, using the membership among the population of college degree holders as a tie-breaker. If you choose to judge that as indistinguishable from, or equivalent to, discrimination based on race, we have a point of disagreement that isn't likely to be resolved.
      - 3y ago

lapcat 3y ago

> If you have two indistinguishable candidates on relevant dimensions who both passed the interview bar and are the best two candidates you have for one position, literally just flip a coin, hire the winner, and move on. Agreed. Now explain how having completed a degree is a "relevant dimension", unlike height, age, etc. You ignored my first question: "Have you considered that if it's possible for a dropout to performally equally to a degree holder in every perceivable respect, it raises doubts about the value of a degree?" My second question was actually intended to highlight the issue. What you appear to be doing with "I have evidence that the person with a degree can follow rules/policies and track towards a long-term goal with success" is reaching for one arbitrary personal characteristic and inventing an idiosyncratic, non-scientific psychological theory in order to justify discrimination. Others do the same thing with age and gender, for example (younger people and men are just better at programming, they say). You don't know the life story of these people, and there are any number of reasons for someone not completing a degree: monetary problems, family problems, a major illness or injury, natural disaster, etc. Moreover, if you select just based on degree completion, you give job candidates no opportunity to show evidence that they "can follow rules/policies and track towards a long-term goal with success" in ways other than competing a degree. It's simply a bias toward degree holders. By your own admission, you're not using the degree as evidence of job-relevant skills but rather as a psychological, "life story" factor.

sokoloff 3y ago
I am biased towards degree holders for programming and other tech roles. It’s not a requirement or some magic talisman of “hire immediately”, but all else equal, I’ll prefer it and, all else equal, I will use it to discriminate (in the “distinguish” or “tell one thing from another” sense). > Now explain how having completed a degree is a "relevant dimension", unlike height, age, etc. As a population, I believe that college graduates almost certainly score higher on measures of intelligence and conscientiousness than non-grads. I believe that's both true and valid for use in selecting employees for roles where intelligence and conscientiousness are useful qualities. > Have you considered that if it's possible for a dropout to [perform] equally to a degree holder in every perceivable respect, it raises doubts about the value of a degree? Of course it's possible for a non-degree holder to outperform a degree holder, or for a Mech E or English major to outperform a Comp Sci grad. That doesn’t mean I think a Comp Sci degree has no relevance or value to an SWE job. You seem to be imagining some boogeyman where “has degree” is treated as the high-order bit rather than what I’ve been consistently saying (and what I think is relevant to the article under discussion): that it’s closer to a tie-breaker (or low-order bit) and a sensible one in many regards. If someone out-interviews another for a role and clears the bar, you hire the out-performer. When you otherwise can’t tell which candidate is stronger and get down to the area where coin flips become a sensible selection mechanism, I think it's wise to choose the degree holder over using the coin. Other people might rely on height or appearance; I can't control what they do.
- lapcat 3y ago
  > As a population, I believe that college graduates almost certainly score higher on measures of intelligence and conscientiousness than non-grads. I believe that's both true and valid for use in selecting employees. 1) You can believe whatever you want to believe, but that doesn't necessarily make it true. 2) If you're looking for IQ, then why don't you administer IQ tests to job candidates rather than looking at an extremely unreliable proxy for it? 3) You're admitting to using population statistics to evaluate individual job candidates. Do you even realize what you're saying here? That's literally prejudice. By the exact same reasoning, you could discriminate based on race, for example, e.g., Asian vs. white vs. black. The only difference between racial discimination and degree discrimination in this respect is the legality of it. Even if I granted for the sake of argument that your empirical speculation about scores and measures were true, you're just talking about averages, and many individual members of every population will be above average or below average. > You seem to be imagining some boogeyman where “has degree” is treated as the high-order bit rather than what I’ve been consistently saying (and what I think is relevant to the article under discussion): that it’s closer to a tie-breaker (or low-order bit) and a sensible one in many regards. It is often used as the "high-order bit". Like I already said a few comments ago, the quote from the article was different from your quote. But I don't think it's a sensible tie-breaker. You're taking a degree as a "life story" factor, but that's an arbitrary, singular choice of many possible life story factors. Also, you seemed to change your story a bit in the last comment, talking specifically about a comp sci degree rather than a degree in general, but there's nothing unique about comp sci vs. other degrees to indicate "can follow rules/policies and track towards a long-term goal with success" or to indicate "IQ".
  - sokoloff 3y ago
    > You're admitting to using population statistics to evaluate individual job candidates. I am, in effect, using the membership among the population of college degree holders as a tie-breaker. If you choose to judge that as indistinguishable from, or equivalent to, discrimination based on race, we have a point of disagreement that isn't likely to be resolved.
    - 3y ago