Maybe It'll Taste Good This Year...

Here's some random stuff related to inductive arguments. First, here's a video of Lewis Black describing his failure to reason inductively every year around Halloween:

Also, here's a stick figure comic about scientists' impulse to get a big enough sample.

Inductive & Abductive Args

Here are the answers to the handout on inductive arguments and abductive arguments (inferences to the best explanation) that we did as group work in class.

Inductive Args
1) The three people I talked to at The Roots concert told me they hated the opening act Talib Kweli. Therefore, nobody at the concert liked the opening act.

This argument is overall bad because of the small sample size. We don't know exactly how many people went to the concert. Still, given what we know about concerts and the popularity of The Roots, we can probably safely conclude that the crowd was at least in the hundreds. A sample of 3 given this probable overall population is too small.

2) Every time I’ve seen a rolling billiard ball hit a stationary billiard ball, the stationary ball starts moving. So the next time I roll one billiard ball into another, the stationary one will move when hit.

This argument is overall good. Again, we don't know the exact numbers on this. We don't even know who the person making the argument is. Still, if we start with the assumption that the person making this argument is typical, we can probably safely conclude that she or he has watched or played pool a decent amount. So our sample is probably hundreds or thousands of billiard ball collisions.

Also, each billiard ball collision is fairly representative of pool ball collisions in general. This seems to be a good example of the principle of "You've seen it once, you've seen them all."

3) Brandon Rush averaged around 13 points a game the past three years playing college basketball for Kansas. So I expect him to average 13 points a game when he plays in the NBA this year.

This argument is overall bad. The sample is actually large enough: Rush played close to 40 games each season in college. However, the sample of college game performance is not representative of NBA performance. Since there are better players and tougher competition in the NBA than there are in college, most players do not perform as well statistically in the pros as they did in college.

Abductive Args
1) In a recent study, 100% of those who took a new birth control pill didn’t get pregnant. Only males participated in the study. Thus, the birth control pill must be very effective.

This argument is overall bad. Concluding ing that this pill effectively prevents pregnancy is not the best explanation of the evidence we have. One big background assumption we have is that males do not get pregnant. Hypothesizing that the participants didn't get pregnant because they are male is a much better explanation of the evidence, since it matches our expectations more.

(Scientists are actually developing a male birth control pill. But this, of course, prevents men from getting their female sexual partners pregnant. There isn't a need for something that prevents the men themselves from getting pregnant... or is there?)

How Many Are Those Doggies In the Window?

The only semi-reliable source I was able to find on the total population of dogs on Earth places the number at around 400 million. That book, however, doesn't explain how it arrived at that figure, and it does warn that the estimate could be off fairly significantly.

Here's a map the top ten pet dog populations by country (although there's no source on this map's estimates, so caveat emptor):


Also, some dogs do not bark. That's why, in the inductive argument we discussed in class, I was careful to conclude only that "Most dogs bark," not "All dogs bark."

One more random fun fact: they used 217 dalmations to film the live-action remake of 101 Dalmations.

Hey, this is too much focus on dogs for my taste.