Appendix: Correlation Between Unemployment Levels and Tea Party Membership?

Aug 28, 2014, 8:05
56394
Rate this item
(0 votes)

Correlation Between Unemployment Levels and Tea Party Membership?

19 October 2010

{jb_dropcap}A{/jb_dropcap}n IREHR analysis of Tea Party online membership and unemployment data demonstrates that there is very little if any relationship between unemployment and Tea Party membership. To look for a correlation between Tea Party membership and unemployment rates, we examined the unemployment rate data for all 372 cities available from the Bureau of Labor Statistics for January 2010 (around the highest level of recent unemployment rates) with the online membership data for 1776 Tea Party, FreedomWorks Tea Party, ResistNet, Tea Party Nation, and Tea Party Patriots for the same period compiled by IREHR.[280]

 

The correlation coefficient (r) provides a measure of how strongly related two variables are and in which way they vary together. These values could potentially range from 1 for a very strong positive correlation (when unemployment goes up, Tea Party membership goes right up with it) and -1 for a very strong negative relationship (when unemployment goes up, Tea Party membership goes down). Statisticians usually call any correlation above 0.80 very strong and below 0.19 very weak to non-existent.

In this case, the correlation coefficient between unemployment and the percent of Tea Party members in a city is 0.083.[ii] This indicates a very weak to non-existent relationship. In effect, knowing the unemployment rate of a city tells us next to nothing about whether there will be a higher or lower level of Tea Party online membership in a city.

This relationship is also not statistically significant - that is there is a greater than 1 in 20 chance (1/13.5) that there is not even any association between these two variables at all. When this is the case, statisticians and social scientists generally conclude that they do not have strong enough evidence to conclude that these variables are related.

It is often said that correlation is not causation. This is true. Just because two things are correlated, doesn’t mean that one causes the other. However, correlation is a necessary part of causation. if A is said to cause B, then A and B must at least be correlated - a change in one must be associated with a change in the other. This data - the most comprehensive available on Tea Party online membership - provides no convincing evidence of a correlation between unemployment and membership. As such, it provides no convincing evidence that unemployment causes Tea Party online membership.

Last modified on Friday, 09 September 2011 20:51
Charles Tanner Jr

Chuck Tanner is a co-coordinator of Borderlands Research and Education. Borderlands is committed to using strategic research to support indigenous sovereignty and treaty rights and environmental justice.

Login to post comments

Related items