Multinomial logistic regression categorial scale question

I have data about political participation where respondents answers questions 7 (yes or no) that indicate some form of political participation. for example: did you contact a politican or government in the last12 months (1. yes

I have created categories out of those 7 questions: supportive participation(2questions), protest participation(2questions), contact participation(1 question) and group participation.(2 questions)

Now, I want to place all those 4 categories into 1 categorie: political participation.
So making it a nominal variabel that I can use for multinomial logistic regression.

How do I do this?

I want to see how poverty is linked to the participation on all those different forms of political paritcipation
I think you have to combine your scales. In your case, all scales measure participation so I would say just make a sum to become a new score/ category, being 'political participation'. Then you can make 2 groups within this new category to make it nominal, for instance respondents with average 'no'-score and respondents with average 'yes'-score. (Or don't sum up and devide each category in 'average-no' and 'average-yes' to measure separately)
Hope this was helpfull...