In the movie Freakonomics, based on the book by Steven Levitt and Stephen Dubner they address problems in things that have not been previously analysed. These things were analysed by using certain correlations to prove the given hypothesized causation. They addressed before this the difference between causation and correlation. Correlation is what we are given usually shown in data, causation would be why this data occurred. Now this can be tricky because correlation doesn't always mean causation, in the example the authors used polio. Because it peaked in the summer at the same time for obvious reasons as icecream sales, many scientists of the time theorized that because of this correlation that icecream caused polio. This was their example of how correlation can often mislead what the given causation is. Many of the examples the authors used they didn't specifically say what they thought the given causality was. For the "naming children" part of the movie they had data that opposed the expert on the subject leading the conclusion up to the viewer. The data they had like the identical applications for the job opposed what the sociology professor who studied the significance in names. When they analysed this topic they seemed to show multiple causations and why they might be wrong. The topic they seemed most firm in their belief of what causation connected to correlation was, was how the decision of Roe v. Wade affected the decline of crime in the late 1990's. They thought this correlation WAS causality because they had multiple pieces of evidence supporting the hypothesis and reasons for why the causality occured, it wasn't just what caused what, they were able to answer why has this caused this.
Alot of the correlations they used were innovative because for something like cheating on tests or rigged sumo wrestling there is seemingly no numbers you can look at to show a correlation. What the two authors did was they brainstormed when this cheating would occur, and then from finding those specific incentives they were able to find that suspected correlation that matched with their causation.
Freakonomics does serve as a good example to our attempt to explore the "hidden-in-plain-sight" weirdness of dominant social practices for multiple reasons. Like the chapter on names and their effects on the childs later part of life they teach us (subtly) you can choose to accept that there isn't enough evidence supporting a conclusion and essentially argue both sides. That there are certain places to look for your given correlation that will support your given causation if you look in the right places. And that you should view the world beyond what is given and appears to be the truth because the fallacy in it hasn't been brought to light.
No comments:
Post a Comment