Timeline for Count data as an independent variable in OLS- using a dummy variable+ the variable linearly to account for skewness
Current License: CC BY-SA 3.0
4 events
when toggle format | what | by | license | comment | |
---|---|---|---|---|---|
Jul 20, 2013 at 22:00 | vote | accept | Dimitrie Mihes | ||
Jul 20, 2013 at 21:59 | comment | added | Dimitrie Mihes | 0 is the true value, as no. of articles is a proxy for issue salience. A negative value would have no substantive interpretation (it cannot decrease issue awareness, only not provide any).Concerning the 2nd question, I assume linearity mainly because previous literature models it this way, although I came across some who recommend logging it. I don't see this as a viable method, as I end up with a lot of missing values. Moreover, by looking at the above threads, adding a constant isn't an option either. Using the sqrt seems to approach a linear relationship, but still leaves the 0s. | |
Jul 20, 2013 at 1:16 | history | edited | Peter Ellis | CC BY-SA 3.0 |
added 15 characters in body
|
Jul 20, 2013 at 0:23 | history | answered | Peter Ellis | CC BY-SA 3.0 |