Great topic- technical comment, I think instead of bargraphs by category, you can try showing that information by adding a "consideration=1" dummy and running it in regression. If you are curious about gender differences, you can add an interaction "consideration*gender". This can give you the average effect and tell you if it is significant or not. We can discuss more after fall break when we talk about advanced regression in class but let me know if you have questions. In terms of your findings, I first thought there seem to be no age or gender differences but upon looking at it again, some categories have changed on the x axis in terms of ranking, so focusing on why those might be the case would be helpful. It would probably help to run a logit model here.
Great topic- technical comment, I think instead of bargraphs by category, you can try showing that information by adding a "consideration=1" dummy and running it in regression. If you are curious about gender differences, you can add an interaction "consideration*gender". This can give you the average effect and tell you if it is significant or not. We can discuss more after fall break when we talk about advanced regression in class but let me know if you have questions. In terms of your findings, I first thought there seem to be no age or gender differences but upon looking at it again, some categories have changed on the x axis in terms of ranking, so focusing on why those might be the case would be helpful. It would probably help to run a logit model here.