Beginners' question....adjusting for variables in regression

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Posts: 1
Joined: Tue Feb 25, 2014 12:24 pm

Beginners' question....adjusting for variables in regression

Postby lacuna » Tue Feb 25, 2014 12:42 pm


Not sure if this I'm wording this in a way that makes sense or not, but please bear with me - am a complete newbie to SPSS and to statistics in general really - anything involving numbers is quite daunting for me and I'm still trying to get my head around even the most basic things.
Anyway, I have a dataset to analyse and I need to do some regression analyses to see whether certain factors are predictive of various mental health symptoms.
We have baseline data, which I've already analysed, and now some Phase 2 follow-up data after some of the participants had received an intervention. However the intervention was not designed to address the specific factors/outcomes I am interested in (plus my colleagues have already done a lot of intervention vs. control analysis) so I've been asked to 'adjust for the intervention arm'. (Is this another way of saying 'controlling for the intervention'?) So basically I want to look at which factors are predictive of symptoms but I want to control for the intervention because in my particular study I am not interested in whether they had one or not, although it is possible it might have had an effect.
Does that make sense? How do I go about 'adjusting for the intervention' when doing multiple regressions?
I'm using SPSS 21 if that makes any difference.

Also, probably a silly question but I am none the wiser after googling - what exactly is the difference between 'confounders' and 'covariates'? In my study, would 'intervention' be one of these?

Many thanks in advance :)

Who is online

Users browsing this forum: No registered users and 2 guests