Timeline for With two related variables, eg, religion and religiosity, how do I transform them into one variable for regression?
Current License: CC BY-SA 3.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 25, 2018 at 0:21 | answer | added | Dave Harris | timeline score: -3 | |
| Apr 25, 2018 at 0:20 | comment | added | Alexis | @DaveHarris I am with MarkWhite: your tone is reading hostile. Your comment "I am trying to be helpful. I am not trying to be rude. I am being matter of fact. There is no emotional content involved on my side. It is how you are reading it." suggests you might benefit from today's xkcd. You tone is coming through in your imputations of meanings the OP has not written. | |
| Apr 25, 2018 at 0:00 | answer | added | Mark White | timeline score: 4 | |
| Apr 24, 2018 at 23:58 | review | Close votes | |||
| Apr 25, 2018 at 11:50 | |||||
| Apr 24, 2018 at 23:50 | comment | added | Mark White | @DaveHarris The OP specifically mentioned dummy-coding the categorical variable, which would not treat Muslim as literally three times as much as Christianity. Please follow the guidelines for not being rude and moreover very much unhelpful to the OP. | |
| Apr 24, 2018 at 23:38 | comment | added | Mark White | @DaveHarris you could explain this in a nicer tone and actually provide some constructive feedback. As far as categorical variables go, it is very common to represent them as zero-or-one dummy codes, so you very much can multiply them together. If what they are trying to do doesn’t make sense, explain in an answer and in a way that makes people feel as if they want to actually return to this forum for help in the future. | |
| Apr 24, 2018 at 23:25 | history | asked | Sophie | CC BY-SA 3.0 |