3 ms·

The bigger problem is employers insist on people having college degrees when they don't need one. Likewise, employers insist on people having college degrees wh

The bigger problem is employers insist on people having college degrees when they don't need one. Likewise, employers insist on people having college degrees when maybe different forms of education could be just as good like certifications. This ends up forcing everyone into going to college just to get a degree for the hell of it. That's the problem. IMO, I think there is going to be a huge market for software companies that can adequately verify someone's aptitude. If I am an employer, I want the best people I can find in my specific industry. I don't care if they have a degree or not. The problem is how do you find these people in the masses of people out there. If software was created that could evaluate people's skills directly, then that would be huge.

dgoldstein0 7y ago
This phenomenon is called degree inflation there's been a bunch of articles about it, such as https://www.forbes.com/sites/prestoncooper2/2018/01/08/employers-demanding-college-degrees-weaken-the-economy/ https://www.forbes.com/sites/prestoncooper2/2018/01/08/emplo... As far as software being able to judge people's aptitude... this is certainly interesting and I'm pretty sure I know some startups dabbling in it. The general problem though is hard - it's basically automating interviewing in some form. Which is easier if you let humans' write the questions & evaluation methods, but of course is then only as good as what you are screening for.
dragonwriter 7y ago
> IMO, I think there is going to be a huge market for software companies that can adequately verify someone's aptitude. Measuring aptitude isn't hard and we've got firms that have been doing it quite well for decades with high predictive power on future success in the domain for which aptitiude is tested. But employers don't care about aptitude (capacity to learn a skill), because we aren't in a system of lifetime employment. They care about present skill. And, it's uolely we are going to find good tests for that that don't involve lots of samples of applying the skill, which for simple tasks that can be repeated dozens of times an hour is easy and already readily available, for complex abstract processes that involve a mixture of lots of small task skills integrated through abstract reasoning, a quick easy test would be a breakthrough. But I think there is good reason to think that that is fundamentally intractable, the issue not being in the testing method but the nature of the phenomenon being tested.

dgoldstein0 7y ago

This phenomenon is called degree inflation there's been a bunch of articles about it, such as https://www.forbes.com/sites/prestoncooper2/2018/01/08/employers-demanding-college-degrees-weaken-the-economy/ https://www.forbes.com/sites/prestoncooper2/2018/01/08/emplo... As far as software being able to judge people's aptitude... this is certainly interesting and I'm pretty sure I know some startups dabbling in it. The general problem though is hard - it's basically automating interviewing in some form. Which is easier if you let humans' write the questions & evaluation methods, but of course is then only as good as what you are screening for.

dragonwriter 7y ago

> IMO, I think there is going to be a huge market for software companies that can adequately verify someone's aptitude. Measuring aptitude isn't hard and we've got firms that have been doing it quite well for decades with high predictive power on future success in the domain for which aptitiude is tested. But employers don't care about aptitude (capacity to learn a skill), because we aren't in a system of lifetime employment. They care about present skill. And, it's uolely we are going to find good tests for that that don't involve lots of samples of applying the skill, which for simple tasks that can be repeated dozens of times an hour is easy and already readily available, for complex abstract processes that involve a mixture of lots of small task skills integrated through abstract reasoning, a quick easy test would be a breakthrough. But I think there is good reason to think that that is fundamentally intractable, the issue not being in the testing method but the nature of the phenomenon being tested.