1
$\begingroup$

We have performed analyses by dividing the participants into 6 groups. The attribution of the groups was based on two variables: the Body Mass Index (BMI) and Metabolic syndrome (MetS):

  • if BMI<25 and MetS=0 then group=1
  • if BMI<25 and MetS=1 then group=2
  • if BMI>30 and MetS=0 then group=3
  • if BMI>30 and MetS=1 then group=4 etc...

Next, the variable 'group' was predicted by baseline variables. Can we call this a stratified analysis? or should it be named as 'a combined analysis', though I'm not very familiar with the latter!

$\endgroup$
1
$\begingroup$

The term stratified analysis is usually reserved for an analysis where you divide the data-set into strata using one of your variables and then perform the same analysis within each stratum. So from your description that is definitely not what you did. The term combined analysis is not one with which I am familiar but it sounds more like a meta-analysis which is not what you did either.

What you did will normally be reported using the name of the model you actually used, so if you used a multinomial logistic regression that is what you would report (for example, I am not saying you have to do that).

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.