I am looking for binary outcome "speeding ticket" (1/0)   I have a categorical variable "bright color" which is categorized into 1, 2, 3 and 0.   I will run a chi-sq with speeding ticket*bright color    On the other hand I make three categorical variables using the same information from bright colors 1, 2, and 3. Each bright color is made into a separate variable: bright color 1 (1/0) bright color 2 (1/0) bright color 3 (1/0)   Then I will run a chi=sq with each variable such as: bright color 1*speeding ticket bright color 2*speeding ticket bright color 3*speeding ticket   Will there be a correlation between running a chi-sq testing bright color as a single variable  vs. speeding ticket with categories 1, 2, 3 and on the other hand running 3 chi-sq testing bright colors as individual variables vs. speeding ticker(bright color1, bright color 2, bright color 3)??   If I find a significant association when running each color as a separate variable, should I expect to find a significant association when running all the colors as one variable with multiple categories?   Thank You     ... View more

  Why is that when I categorize a variable in logistic regression by making it binary at the 75th percentile cutoff, it makes Variable 2 which was previously significant into non-significant. Then, when I change the categorization to binary while using an outlier number much greater than the 75th percentile as the cut off , Variable 2 then becomes significant again?   For example   1) model event1= variable 1(continuous), variable 2(categorical)  - variable 1 is significant, variable 2 is significant   2) model event1= variable 1 (categorical at 75th percentile), variable 2(categorical)    - variable 1 is significant, variable 2 becomes non-significant   3) model event1= variable1 (categorical at outlier point, much greater than 75th percentile), variable 2(categorical)  - variable 1 is significant, variable 2 is again significant ... View more

I am assessing for outcome "eventX" with survey data.   One variables, "diseaseX" has an association of p=0.023 on univariate chi square.    When placed in the multivariate regression model with multiple other variables, it has a lower p-value of 0.0003. SAS does not give any messages about correlation and the model has convergence.    If this is due to some kind association where one variable reenforces another (forgot what thats called), then how can I find which variable it is? Otherwise, how can I deal with this, is it ok to leave the variable in the model, if there is an association?   Please explain. Thanks  ... View more