Any one who can give me hint on how to determine the cut off point for p-value of bivariable logistic regression to take the variables to multivariable regression?
That's a very poor method of selection because it just doesn't work. Here's two papers on the subject. First is about method doesn't work and second one is about how you might deal with this in a much better way.
Best wishes, David Boot
As David said, "univariable screening" of candidate predictors is a poor method for determining what variables to include in a multivariable model. Frank Harrell mentions it specifically in his Author Checklist:
- https://discourse.datamethods.org/t/author-checklist/3407
Look for the heading Use of stepwise variable selection. HTH.
It depends on your research field. If you medical field, for example working on effectiveness of a medicine on human (which is a very sensitive issue and needs to be precisely determined), it should be very low (0.01 or less). If you are working social science, the cut off value for p is usually 0.05. You can go through papers of your relevant fields where multivariate regression has been used and check what was cut off p-value for those papers.
The adjusted odds ratio is simply calculated in the multivariate logistic model where you include several variables simultaneously (two or more independent variables). This means that the adjusted odds ratio in the multivariate model is calculated for one independent variable by controlling for other independent variables in their reference value. Whereas, in the univariate model we calculate the crude odds ratio to study the causality between a single independent variable and our dichotomous variable (dependent variable). Moreover, the univariate model is built before the multivariate model because from the univariate model and based on the probability (p-value
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size