This question already has an answer here:

so i am proposing to conduct a study on indirect contact as an intervention for stigma reduction in a police population. My IV is the intervention with two levels (intervention group and control group) and my DV's are a attitudinal measure, intergroup anger and a behavioural measure. I will be conducting a pre - post - follow up test and I am planning to use a repeated measures ANOVA and an ANCOVA so I can measure covariates such as (years in duty, previous contact etc). I am struggling to get this straight in my head as I confused if I can use covariates as a measure of moderator effects? Also slightly confused how I would introduce the covariates within the proposal?

Any help or guidance would be greatly appreciated


marked as duplicate by AdamO, kjetil b halvorsen, Stephan Kolassa, Peter Flom Dec 20 '17 at 13:03

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • $\begingroup$ who are the subjects ? police persons ? male or female ? what is your stigma scale? more specific and details could be helpful in answering your questions ? $\endgroup$ – Subhash C. Davar Dec 16 '17 at 1:51
  • $\begingroup$ The subjects are trained police officers, male and female. And i will be using attitudinal scales to measure stigma (CAMI scale - community attitudes towards mental illness and the RIBS scale - reported and intentional behaviour scale) $\endgroup$ – Laura Dec 17 '17 at 15:45
  • $\begingroup$ State your problem in brief including literaure review? Justify your proposed hypotheses in the light of your objectives. Indicate your operational plan for data collection including Scales etc to be used and show data analysis plan . The issues raised by you can be addresed more meaningfully once you clear about the hypotheses and objectives of your study. $\endgroup$ – Subhash C. Davar Dec 18 '17 at 4:16
  • $\begingroup$ I think the key term for which you are searching is mixed effects model. There are many posts on this site which may be relevant to you. It is straightforward to introduce moderator variables as covariates into the mixed model. $\endgroup$ – mdewey Dec 18 '17 at 16:01