Newsweek reported this week about college students paying their universities to get official credit for unpaid internships. They pretty clearly think its a bad idea. I have to agree with them but some of their reasoning was a bit off.
For example, they write,
“Unpaid internships don’t do as much for you in the job market as paid ones do. According to the 2011 Student Survey by the National Association of Colleges and Employers, paid interns spent more of their time on professional duties, while unpaid interns were given clerical ones. Sixty-one percent of paid interns working at for-profit companies received a job offer; only 38 percent of unpaid interns working at for-profit companies did. And paid interns netted higher starting salaries.”
On surface level this makes sense. If paid internships lead to better paying jobs and offers then they must be better than their nonpaying counterparts. That would be true if students were randomly assigned paid and unpaid internships, but they aren’t. Given that paid internships are probably more sought after than unpaid ones, they are able to choose the most gifted and able students. This then replicates in the job market where the more gifted and able students are more likely to get a job offer and at higher pay.
We often try and create causal connections like this but they generally aren’t true. Like I said the other day, randomness matters. I never cease to be amazed at how statistics and logic are generally misunderstood by the public, including journalists. I actually worked with a professor during graduate school collecting stories from the media where statistics were misunderstood and this is a classic example.
Thomas November 23, 2011
I have only read what you state in your blog, but to me that isn’t so much a misunderstanding of the statistics. What you say is probably true in regards to companies that offer paid internships are more selective and that translates into a higher likelihood of being hired but how is what Newsweek stated a misunderstanding? They simple make the point that unpaid internships are as good as paid ones. Furthermore, it should be the responsibility of the University that is accepting the internship to make sure that when they send a student out, the internship is actually going to help out the students. Otherwise what is the point in the first place?
Thomas November 23, 2011
I meant to say not as good paid ones.
Andrew November 23, 2011
My critique is not where they arrive, that unpaid internships aren’t as good as paid ones, but the evidence they use doesn’t prove that. The fact that unpaid internships don’t result in as many job offers or as high of salary of paid internships isn’t relevant because of what I mentioned. If a brilliant student who could have gotten a paid internship chose for some reason, or was assigned an unpaid one, they would probably be someone who was offered a job at good pay.
They are mistaking the correlation’s. The correlation isn’t between paid and unpaid internship at job offers, the correlation is between quality level and job offers which makes much more sense.
james edward November 24, 2011
Friend, how do you back up the claim (statstically) that gifted students get better paying jobs? are you taking race, gender, ‘looks’, age and other variables into account? how are you defining ‘gifted’?
Andrew November 25, 2011
I would venture to argue that if you controlled for race, gender, looks, etc that you would find students with a higher mix of EQ and IQ are more in demand than students with a lower mix of EQ and IQ. Of course more attractive people make more than their less attractive counterparts and their are definitely institutional biases against women and minorities, but my argument is that even for controlling for those factors you would that people with higher IQ and EQ would be more in demand by employers. Of course a very talented African American woman might make less than a less talented white man because of institutional biases.
But even taking that into account doesn’t really affect my arguments above. At the crux of this is the economics of employment. There are people who are more in demand than others. This can because of giftedness (some mix of IQ and EQ or special training) or because of some immoral biases. What I was arguing was that the same things that make a person in demand for an internship are the same things that makes a person in demand in the job market.