Sunk costs and education

Lately, I’ve talked to quite a few people who graduated or a near graduating and observed the following:
About 70% of these people invested a ton of effort and time into getting great grades, doing internships at famous companies and traveling to other countries; mostly with the aim of keeping their career options open

There is nothing wrong with that behavior. It’s allows you to keep options open. However, now the sunk costs fallacy comes into play. What’s the idea behind that fallacy?

Sunk costs are costs which can’t be recovered. An example (from Wikipedia) is non-refundable movie tickets. You could either watch the movie, even if its horrible or just throw the tickets away and do something else. Rational behavior would be to do something else if this something else (for example, eating ice cream) has a higher utility than watching the horrible movie, even if you paid $50 for the movie ticket. The sunk costs shouldn’t influence your decision.
The fallacy is that people do include sunk costs instead of ignoring them. How does that apply to education/careers?

Let’s go back to our successful students. They finally graduated and have now the chance to choose the future career. What I’ve experienced is that they don’t want to throw away their past efforts and decided that they should rather choose a career in management consulting/investment banking/research instead of what they actually want to do; something that would make them happier. And all this just because they don’t want to have the feeling that they wasted some of their time.

Firstly, you don’t have wasted your time. See it this way: All your experience that you earned in this time helped you to come to the conclusions you have now.

Secondly, it’s quite normal to act like this, however not the best way. The underlying phenomena is loss-aversion. People try to avoid loss more than they try to get profit. Approximately $2 loss is equal to $1 earned.

Thirdly, this is what I call the second division effect/strategy. It comes from football leagues. There is the premiere league or first division where all the top players of a country play in. And there is the second division, where good but not great players play in.
The idea is this: You could either be a mediocre player in the premiere league or a top player in the second division. I personally be rather the top player in the second division. Interestingly, there is research which shows similar results for most of the population.
They were asked to choose one job. On the first job they would earn $50k but everybody else would earn $60k. On the other job they would earn $40k but everybody else would earn $30k. The majority choose the second job, the second division job.

All in all, the best thing is simply try to avoid to fall into this trap but when you gain more insight into yourself, you’ll probably be able to climb out of this pitfall.

Intro to Data Science (UCB)

A hour ago someone posted on hacker news about this course at UC Berkeley.
You can find the slides and the videos from last year or slides only from this year. The material looks pretty basic but covers data preparation over two weeks which is quite rare but really important.
Coming from a university that basically ignored everything which wasn’t academic, two things stand out.
Firstly, there are guest lectures from people from Google, Optimizely, Yahoo, etc and their lectures are generally quite interesting.
Secondly, the freedom in choosing the final projects is awesome. You can choose freely some data sets which interests you and play with it. There was a wide variety of data from Youtube, Last.fm, basketball to Yelp.

Generally, I think that this is a pretty good intro course into this topic. Most universities try to over-theorize such basic courses and talk in my opinion too much about maths and too little about data gathering and EDA.

Also there was a particularly good comment in the hn thread:

Someone (can’t recall the source, sorry) recently defined “data scientist” as “a data analyst who lives in California.” —baconner

#72/111: Stumbling on Happiness

What is it about?

If you want to know how we think, feel and experience happiness, Daniel Gilbert is your person. He explores how happiness is influenced and how we make decisions.

What can I learn?

Uncertainty makes you happy: You probably wouldn’t think so. I’m happier if I know that I will have my job tomorrow than being uncertain about. Actually, no. People are happier if they don’t know what will happen. If something haven’t happened yet, we often overestimate it. Therefore we feel more happy because of the opportunities that can be. You could say that we are natural optimists.

Falsification and Approval: That’s an interesting observation which you can observe in nearly every discussion. If you want a theory approved, most time a single piece of evidence is sufficient. However, if something tries to falsify your theory, you will probably demand many pieces of evidence. This is a natural behavior which is quite rational. If you would check every assumption everyday you wouldn’t doing much more.

We are very similar: About 8 of 10 people with lung cancer will die within the next five years. If you have lung cancer, you probably hope that you are in the 20% that won’t die. There are other studies, e.g. one where university professors should rate their teaching skill. About 96% rated it as above average. The worst thing is, we are actually pretty similar. Not everybody is above average. Not everybody belongs to the 20% which won’t die from lung cancer. Not everybody is a great lover and an exceptional businessman.

Conclusion

Stumbling on Happiness is pretty okay. One criticizing point is that Daniel Gilbert tries to cover too much. If you want to read more about happiness I would recommend :59 seconds. If you want to read more about decision making I would recommend reading Predictionably Irrational by Dan Ariely.