Which of the following is an accurate description of Simpson's paradox?



When groups of data are aggregated, an association can get stronger because of a confounding variable. That confounding variable is usually the number of observations in different groups of data.

When groups of data are combined, an association can get stronger because of a lurking variable. That lurking variable is usually the number of observations in the different groups of data.

When groups of data are separated, an association can get stronger because of a lurking variable. That lurking variable is usually the number of observations in the different groups of data.

When separate groups of data are combined, an association can reverse direction because of a lurking variable that was lost when the different groups of data were lumped together.

----

For the table below which of the following are true?
I. The sum of the values of all the conditional distributions must be 1.

II. Temperature and crime rate appear to be related (the warmer the temperature, the higher the crime rate).

III. The conditional distribution for Normal Crime Rate is roughly similar to the marginal distribution Temperature.

Crime Rate
________Below___ Normal____ Above
Temp. __Below 12 , 8 , 5
___ ___ Normal 35 , 41 , 24
______ Above 4 , 7 , 14


A. I only
B. II only
C. II and III only
D. III only
E. I and III only

2 answers

http://www.google.com/search?client=safari&rls=en&q=Simpson's+paradox&ie=UTF-8&oe=UTF-8

What table?
The table is supposed to be that ugly fail that says "Crime Rate" at the top, "below," "normal," and average beneath it, etc.