by DrSteveB
DailyKos
Fri Oct 19, 2007 at 04:47:52 AM PDT
As an epidemiologist I am trained to analyze risk. One of the types of analysis we use in studying a cohort, such as the House of Representatives, is Relative Risk.
Statistics proves that the vote was not bipartisan as some Repuglican apologists in the media are trying to claim.
Relative Risk is:
The proportion of diseased people amongst those exposed to the relevant risk factor divided by the proportion of diseased people amongst those not exposed to the risk factor. This should be used in those cohort studies where those with and without disease are followed to observe which individuals become diseased.
In this case, the cohort is the House of Representatives. The Relative Risk (RR) is the risk of an event (developing a disease; voting against SCHIP) relative to exposure (being a Repuglican). Relative risk is a ratio of the probability of the event occurring in the exposed group versus the control (non-exposed) group, or in this case the probability of the voting against SCHIP for Repuglicans compared to Democrats.
In epidemiology we typically try to reduce the analysis to what is called a “2 by 2 table”
|
OUTCOME
|
||
| EXPOSURE | Disease+ | NonDisease- |
| Exposure + | A | B |
| NonExposure – | C | D |
| Relative Risk (or Risk Ratio) = (a / (a+b)) / (c / (c+d)) | ||
Therefore:
|
Against SCHIP
|
For SCHIP
|
|
| Republican |
154
|
44
|
| Democrat |
2
|
229
|
Running the statistical calculations, we find that the Relative Risk = 89.8, with a 95% confidence interval from 22.6 to 357.8; P-value < 0.0001.
What this means is that the probability of voting against SCHIP is about 90 times greater if the Representative is a Repuglican then if they are a Democrat.
Got that?
The probability of a Republican voting against SCHIP are 90 times greater than the probability of a Democrat voting against SCHIP. That is NOT bipartisan!
And yes, it is highly statistically significant, with the chance that this would occur by random chance is less than 1 in 10,000. If we were to re-run the experiment 100 times, 95 times the relative risk would be between 22 and 357.
Just how bad is this? Well the relative risk of lung cancer and smoking is 11, and is often cited as one of the highest we see in medicine.
The vote was not bipartisan.
Republicans are a very toxic exposure.
The way to prevent this (the cure) is elect a lot more Democrats.