Re:The Other Money Problem
by mathattack on Friday December 12, 2014 @12:58AM
(#8738775)
Another source that disagrees...
http://fivethirtyeight.com/features/economists-arent-as-nonp...
It is very hard to not introduce bias into social sciences. This is also why psychology and software engineering studies are hard to trust. It is more pervasive in economics when there are rewards to be linked to a given worldview. (You can make a lot of side money as an expert witness if you have written a lot of papers either solidly pro or against anti-trust legislation)
The source you're linking is mentioned in the first 3 words of the article he linked. The MotherJones.com article is a direct response to the article you linked.
I'm not sure what the units are there in that graph, but when trying to scale the vertical and horizontal of the graph to have 1 unit vs 1 unit, it seems to have a larger looking slope to me. Still smaller than one though.
I may have made mistakes in resizing it though.
Re:The Other Money Problem
by yummyfajitas on Friday December 12, 2014 @03:25AM
(#8739043)
The X and Y axes do not have the same units, so scaling them this way is meaningless. The key point is that while there line has slope, the spread of the data is much larger.
I.e., the ideology of the study's author gives you very little information about the numerical values in the study.
http://www.motherjones.com/kevin-drum/2014/12/economists-are...