The research has:
1 IV and 3 DV's. Likert scales used yielding scores of 1-5 on Behavioural Observational Scales (BOS). n=99 with each individual being scored on all 3 DV's. Total BOS=297. 99 BOS scored for each DV. Do not need to see any relationships between the DV'S. Need to see if the IV creates an increase in score in any or all of the DV's.
If the IV also is on a Likert scale, multinomial logit, with each of DV1=2, DV1=2, etc a separate dummy variable (& one extreme obviously in the base). Stata, eg, largely automates that so you don't need to create the dummies.
Hi Chloe,
As Ted indicated, multinomial logistic regression is probably the best way to go.
I'm a fan of SPSS when learning new statistical concepts, so I've included a link to a step-by-step SPSS tutorial which might be worth watching before attempting to complete your analysis in Stata or R.
https://www.youtube.com/watch?v=RVBiD7Setms
I hope this proves helpful!
Regards,
Matt
Thank you, I was thinking pearsons r? This is why I'm confused! Also the IV is not on the Likert scale. The DV's are all independent from each other. So in essence it is more a case of IV being cause and each DV being an effect with the Likert scoring how much of an effect.
ideal would be to do multinomial logistic regression for each dependent variable
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running