Dear all,
I'm investigating perceptions of effective leadership of Germans and Indians (I want to know which leader attributes are mutually endorsed and which are not). Part of the questionnaire assessed the attitude towards cultural values of the respondent, the other part attitudes toward effective leader attributes.
I have run a questionnaire and gained N=209 responses. The questions were asked on a 7-point Likert scale and the precise questionnaire can be found here: http://globeproject.com/data/GLOBE-Phase-2-Beta-Questionnaire-2006.pdf (I only used part of the questionnaire and not all questions, as I only investigate perceptions). I hypothesize that certain elements of culture are predictors of certain perceived effective leader attributes.
Now I find it a bit difficult to assess which statistical tests make sense. I've done some research online and the information is quite ambiguous and sometimes conflicting.
Therefore, I'd be really grateful if anyone here could advise.
Thank you,
Ahmad
There is a difference between working with individual items that used Likert-style scoring, versus combining those items into a scale that would have interval-level measurement properties.
Because there have been so many questions about this issue, I have compiled a set of resources on it.:
https://www.researchgate.net/post/How_can_one_help_people_create_simple_scales_from_Likert-scored_items
Thank you very much David, I will have a look. I intend to group the individual items together.
Is it possible to use:
ANCOVA is for interval-based variables and Spearman's Rho is for ordinal variables. Which one would be more appropriate?
The reason for using Cronbach's alpha and factor analysis is to determine whether you can treat the measures as a single, interval scale. If you do treat it as interval, then there is no point in using Spearman's Rho.
I concur with Prof David Morgan suggestion that since you used interval scale , then ANCOVA is more appropriate than Spearman's Rho
I agree with Prof.David Morgan, but remember that as you analyse you interval scale variables, as you use ANCOVA, you can also refer to the values of your multiple regression results for significance confirmation. In the model of regressions, you test for the constant, the coefficients and also the whole model values of F. ANCOVA is better in this reagrd
- Badges
- Science topic
- Similar topics
- Psychology
- Social Psychology