Can someone please help me out?

#1
Hey guys,

I have a problem that probaly one of you can help me with. I am using IBM SPSS 24 and got a dataset which I am using for my research internship. Okay so i have a dependent variable called underweight that is dichotomous (Yes/No), however i want to transform this into a continous variable (percentage 0-100). The dataset is from a Survey of Households and each case is a member under 5 years old of the household for this variable. I ultimately want to see wheter the differences in underweight between clusters, region and climatic zones in the country I'm doing my research on (Benin) significantly differs. Now I can ofcourse use crosstabs and examine the % per climatic zones (3), regions (12) and clusters (448) however i want to use them in a T-test so i need to have them as a variable. Does anybody know a trick how to get this done? thank you in advance!
 
#3
hmm okay but as for now I only want to see if the differences between the groups are significant, thats not the purpose of a logistic regression rigth?
 
#4
Yes, you can. You can test for differences between climatic zones, regions and clusters.

Of course a more simple method would be to just do chi-squared tests, or tests about proportions.