Data on income for two regions (cross-section data) show large variations. Should one apply percentile transformation to run binary regression? Only one value-based variable (income) in the model.
I don't think that's necessary, you can apply the binary logistic regression
Dear Balu, you don't have to do that, unless you want to explain some particular strata or subgroup. You can enter the variable as a scalar one. However, some presuppose should be evaluated before conducting the analysis. If you fell that transforming the data will be a more informative, you could use chi-square test instead of binary logistic regression.
Regards, Thiago
dear Anthony,
I will offer you two possibilities
If you want to measure the income difference, you can compare the two averages. A simple test would indicate if the difference is significant. However, be care full of outlying points.
If you want to compare the degree of inequality on income. You can calculate the Gini index. But I am not sure there is a test to indicate that the difference of Gini index is statistically significant.
In both cases, refer to a good explanation on internet.
All the best !
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running