> It seems to me that a large number of jobs that list a degree as a requirement don't gain much by requiring a degree, except a smaller stack of resumes to dig through.
Yes, the stack of resumes is smaller. However, I think the key here is that the employer believes the stack of resumes is likely to have a higher ratio of "good" candidates to "bad" candidates.
In economics this is called "signaling". The degree may not endow the person with any particular skill necessary to do the job, but it indicates ("signals") to the employer that the person has certain characteristics that allowed them to earn their degree.
In this case, they can show up on time, complete assignments in a timely manner and use a computer (along with some other things, I'm sure). These are (in theory) qualities that are required to earn a college degree. They are also required for almost all jobs, including fairly menial ones.
The idea is that you can create a pool that is richer in the desired characteristics by limiting it based on the signal.
So, let's pretend 50% of the general population and 75% of the college-graduate population can, say, complete assignments on time (just an example, no "lazy college student" anecdotes please). If you allow anyone to apply for a given job, 50% of the people in your hiring pool will be able to complete assignments on time. But if you limit the applicant pool to just the college graduates, you get to choose (semi-randomly in the case of job interviews) from a pool in which 50% more of the people (75%) can complete assignments on time. This means you are more likely, all else being equal, to end up with a person who can complete assignments on time.
Of course the usefulness of this sort of signal is dependent on whether the assumed characteristics are in fact required to earn a college degree, YMMV. But this is why employers require college degrees for jobs that don't seem like they should require one.
I know. My question was rhetorical, and mostly kvetching due to my fiance trying to find a job and being stymied by the fact that she doesn't have her degree yet.
The problem with those numbers is that we have no idea what % of the general population can complete an assignment in time, especially when you consider that as a hypothetical employer I don't care about the general population - I only care about the percentage of the population that's applying for my job. Again, this is all personal and highly biased, but if I had to pick a random person from my college classes and my fiance to complete some given office task, I'd give it to her, hands down. (But of course, not everyone is marrying her. Luckily for me!)
I agree that this sort of broad-strokes signalling is probably highly ineffective. However, given the difficulty of hiring, employers are likely to seize upon anything that can ease the process even slightly.
There have been interesting ideas in the tech sector, the StackOverflow jobs site for one. It would be interesting to see if anything can be done outside of tech to make hiring more effective.
Yes, the stack of resumes is smaller. However, I think the key here is that the employer believes the stack of resumes is likely to have a higher ratio of "good" candidates to "bad" candidates.
In economics this is called "signaling". The degree may not endow the person with any particular skill necessary to do the job, but it indicates ("signals") to the employer that the person has certain characteristics that allowed them to earn their degree.
In this case, they can show up on time, complete assignments in a timely manner and use a computer (along with some other things, I'm sure). These are (in theory) qualities that are required to earn a college degree. They are also required for almost all jobs, including fairly menial ones.
The idea is that you can create a pool that is richer in the desired characteristics by limiting it based on the signal.
So, let's pretend 50% of the general population and 75% of the college-graduate population can, say, complete assignments on time (just an example, no "lazy college student" anecdotes please). If you allow anyone to apply for a given job, 50% of the people in your hiring pool will be able to complete assignments on time. But if you limit the applicant pool to just the college graduates, you get to choose (semi-randomly in the case of job interviews) from a pool in which 50% more of the people (75%) can complete assignments on time. This means you are more likely, all else being equal, to end up with a person who can complete assignments on time.
Of course the usefulness of this sort of signal is dependent on whether the assumed characteristics are in fact required to earn a college degree, YMMV. But this is why employers require college degrees for jobs that don't seem like they should require one.