[zencycler]

Now read this very carefully to see if YOU can recognize the misleading methodology (not surprising) employed by this survey, in order to get the result that they wanted:

"...Alabama and Arkansas have the second and third highest divorce rates in the U.S., at 13 per 1000 people per year while New Jersey and Massachusetts, more liberal states, are two of the lowest at 6 and 7 per 1000 people per year."

Okay, did you spot it? If not, here is a hint:

at 13 per 1000 people
at 6 and 7 per 1000 people

That's right, they are measuring the divorce rate by taking the number of divorces per the number of PEOPLE in the population. Any good scientist knows that to reach conclusions from from a data set like this, you need to remove all the "variables". Clearly, in reaching the conclusion (that they wanted to reach) about conservatives they did not take into account that people in conservative states are far more likely go get married. More marriages, mean more divorces!

In other words, in these fractions 13/1000 vs. 6/1000 or 7/1000, the denominator should not be PEOPLE (1000) they should be MARRIAGES!

I am almost willing to bet that if this survey was done correctly, than the result would be something like 13/300 in the red state vs. 6/100 in the blue state, (or something like that). But of course, since that would shoe (correctly) the higher actual divorce rate in the blue states, they opted to use the flawed method to get the result that they wanted.

[zencycler]

Wish I could edit AFTER posting: "that would shoe (correctly) = "that would show (correctly)

[MSF BU]

If the analysis were done county by county (or town in the case of New England) I am sure the results would be very different. There are very, very liberal voting counties in Alabama.

[TurboZamboni]

Is this my old college statistics prof?
(good catch)