I have textual data that i need to classify into categories that i have created. The total data that i have is about 50k records and i have categorized each of these records into 8 categories. I need to run and ANOVA or Regression to see if the text categories have any influence on a continuous variable that is also a part of my analysis. However, i am unable to run an ANOVA or Regression since records are mapped to multiple categories. I understand that this violates the assumption of independence. Is there someway by which i can uniquely categorize all the text data into unique single categories and even if they happen to overlap SPSS should be able to re-categorize into another group which is a combination of my preset 8 categories. Would really appreciate if somebody could help me.
Thanks & regards