I have a number of rankings of countries based on various indices. For example, HDI, happiness, religiosity, etc.
The basic idea is to check whether any of these are associated. For example, whether high HDI indicates high happiness levels, etc.
Any suggestion is welcome. I am using SPSS primarily, and Deducer and RapidMiner sparingly.
There are basic tools eg Spearman's rank correlation. On the other hand, you may recourse to more sophisticated approaches involving latent variables with probit or logit links.
Providing your data meets the assumptions, start with crosstabulations and chi square and go to bivariate correlation from there if you just want to see the direction and strength of the association. Otherwise, use ordinal regression. Hope it helps in some way
covariance analysis can be used among relationship in these variables.
Spearman's Rho and ordinal regression are what I had in mind. The doubts arose because the data here is not survey responses which I am used to. This is in the form of country names and indices, lists with 50 entries each.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS