Basically what the title says. I have run a logistic regression with an ordinal independent variable (scale from 1-7) and get an odds-ratio of 2. I am unsure of how I interpret this? Do I find the odds of being in the "yes category" of the binary dependent variable given that the independent variable is 1 and then report that it doubles with each increase in the independent variable?
$\begingroup$
$\endgroup$
1
asked Jun 6, 2017 at 10:26
-
$\begingroup$ Do you have other variables? Or do you have a 2 by 7 table? $\endgroup$– David SmithJun 6, 2017 at 17:29
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
It depends on how you coded it/entered it into your statistical model. If you entered the independent variable as a continuous variable (i.e. just entering the values 1-7), your interpretation is correct: the odds ratio of 2 indicates that the odds of the dependent variable being 1/'yes' increases by a factor of 2 for every point above zero.
If however, you would indicate that the independent variable (the scale from 1-7) is a categorical one (by creating binary dummy indicator variables for each level except a reference level), you would obtain the odds ratios of six of the seven possible values compared to the reference level. The advantage over the above method is that categorizing this way you do not assume a linear or proportional effect along the scale. Do note this costs 6 degrees of freedom instead of 1, and therefore could require more data to obtain viable results.