I need some guidance on selecting the appropriate statistical methods for my research. I am investigating lone violent extremists by subtype (far right, far left, Islamic and single-issue) on 4 dichotomous variables (mental health, substance use, possessing radicalized family members, and possession radicalized friends). I am interested in assessing differences on these variables between lone extremist subtypes. Suggestions? Chi-square? ANOVA? I should I also add I am working with archival data via the PIRUS database.
You have to ask your supervisor. ANOVA and Chi2 are two simple methods, they may only be used in a small part of a dissertation for a Ph.D.
Hi Tamara. I'm afraid I don't see how you can get any meaning out of a dichotomous variable on mental health and substance use. You should select validated measures with continuous scales that can differentiate between high and low severity. Best John Kåre
Hello,
I agree with the previous response. If you use mental health as an indicator, then you should combine it with others because it is very non-specific in its current form. Given that you have several predictors of radicalization, you may be better off performing a latent variable model which gives you a way to establish profiles of individual using imperfect indicators.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA