We recently ran a survey. This survey was sent to 1 million households, we received 200,000 replies.
The survey is a long survey with 1500 questions
11 different source files were used to send the survey.
We would like to prepare a profile analysis of all the files used, and identify "look alike" households that could be selected from survey file
My expectation is the following analysis work would be done
CHAID analysis for each of the data sources
LOGIT regression to identify and quality high potential households
My guess is this work will require 2-4 days of work