I am using Data which contain multi- collinearity between variables. the r-square was very low as well. When we aggregate the data using:
SPSS- Data- Aggregate,
the R- square become better, and the there become no more multi-collinearity between variables. the question is, can I use the new Data (aggregated data) instead of my original data and get the same results related to relationships between variables and R-square ?
Thank you,
Anas
One of the solutions to remedy multicollinearity is using factor analysis to transform correlated variables into independent factors. This is in somehow is type of aggregations.
Thank you very much, but I wonder, do you mean that because the sample size will decrease, or there are other reasons? Ajit kumar Roy
Matthew Kerry
Technically it is possible. But you are changing your statistical unit. That means you are talking about someone else. The aggregated measures are good measures of what? How your research questions are affected by this?
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS