# How can a categorical variable where respondents can choose more than one response be used as a predictor in multiple regression?

I have conducted a survey which asks respondents to indicate the informational sources regarding a product. There are 8 information sources (e.g., internet, newspaper, brochure and so on) and people can choose more than one source.

I want to use information source a predictor variable in a multiple regression. However, because people can choose more than one information source, it does not seem possible to to treat it as a single categorical variable.

How can a categorical variable where respondents can choose more than one response be used as a predictor in multiple regression?

Use dichotomous indicators (often referred to as dummy variables) to represent the items within this one question. For example, a variable called internet, and then a variable called newspaper, so on so forth. If a person picked both, they got a 1 in each. If a person only picked newspaper, then enter 0 for internet and then 1 for newspaper.
• If you think the effects are not additive (and you have enough data), you might try creating a binary internet only, newspaper only, and internet and newspaper variables. You can then see if the effects of each channel reinforce each other. Sep 26 '13 at 17:39