I do not understand your DV so I don't know at this point. can you please give examples of what a couple of observations on the DV might look like. It is fairly clear that you do NOT want multiple linear regression. I would recommend the attached notes and the book at this link:

https://www.amazon.com/Fundamentals-Biostatistics-Bernard-Rosner/dp/130526892X/ref=sr_1_1?keywords=bernard+rosner&qid=1557947853&s=books&sr=1-1

as helpful resources. Best, David Booth

At the first thought, I think your DV is not a count data. Could you possibly provide a physical interpretation of your DV.

Your question is not totally clear to me. If your dv are 7 different types of food then you are to run an anova. Poisson regression a would be appropriate if your dv is one food and how many times this food was consumed, 0 to 6, or something like that

Hi,

I wanted to use child dietary diversity ( that indicates how many food groups the child has consumed : min- 0, max -7 ) .. For this variable, a dietary recall was done to check which groups child has consumed.. vitamin A ( yes or no) ..together each child's dietary score is compiled out of 7 groups.. Eg: few got a score of 2 of 7..

Please let me know which analysis is suitable..

Since your dv is nominal you want to fit a loglineat model. The equivalent of an anova model with a discrete dv