Timeline for Time spent in an activity as an independent variable
Current License: CC BY-SA 3.0
16 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 3, 2019 at 19:40 | comment | added | Gavin M. Jones | I'm curious how the interpretation of b1 (weeks breastfeeding) would change if the indicator variable for b2 "non_breastfeeding" was reversed, so that it represented "breastfeeding" (1=breastfeeding, 0=not breastfeeding). Would b1 still be able to be interpreted as a linear effect for those that breastfeed? What other issues would this cause? | |
| Jun 3, 2019 at 7:32 | comment | added | Maarten Buis | @GavinM.Jones I never thought of the need to name it or to cite this: it is just a straightforward application of continuous and indicator variables. Consequently I don't have a good reference for you. The closes thing I could quickly dig up is Treiman, D. J. (2009): Quantitative Data Analysis. Doing Social Research to Test Ideas. San Francisco: Jossey-Bass. , chapter 7 discussed something similar. The model contains a constant. | |
| May 30, 2019 at 19:44 | comment | added | Gavin M. Jones | Apologies for resurrecting an old post, but this is a fantastic answer by @MaartenBuis and I've got three quick questions. First, is there a "name" for this approach that you know of? Second, is anyone aware of a textbook or other cite-able reference document that describes this approach in greater detail? It seems very useful and must be described somewhere, I would think. Third, I wasn't sure based on the notation in Maarten's answer, but is there also an intercept (b0) in this model? Thanks! | |
| Jun 25, 2018 at 19:49 | comment | added | Maarten Buis | I answered that already in my previous comment. So apparently something was unclear, but without knowing what was unclear in that answer there is little I can do. | |
| Jun 25, 2018 at 18:53 | comment | added | dwhdai | @MaartenBuis If 12 weeks is assigned to the non-breastfeeders, wouldn't this influence the parameter estimate for $\beta_1$ (the continuous portion)? | |
| Apr 19, 2013 at 13:05 | comment | added | D L Dahly | Ok, I find that convincing given that it's something I can check for myself. Many thanks. | |
| Apr 19, 2013 at 12:28 | history | edited | Maarten Buis | CC BY-SA 3.0 | added 1 characters in body |
| Apr 19, 2013 at 12:21 | comment | added | Maarten Buis | @DLDahly That is correct, the estimate of B1 will not change when you change the time-value for the "non-breastfeeders". In fact, B2 is the only parameter that will change when you change the time-value for the "non-breastfeeders", all others will remain unchanged. | |
| Apr 19, 2013 at 12:07 | comment | added | Andy W | Very nice response Maarten. Here is a similar question/answer on the site that shows a similar situation in including an independent variable that only pertains to a particular subgroup. | |
| Apr 19, 2013 at 12:05 | comment | added | D L Dahly | Ok, that's very helpful. Let me ask one more quick follow-up...if I am understanding you correctly, then the estimated value for B1 should be the same regardless of what time-value I give the B2=1 people. Is that right? | |
| Apr 19, 2013 at 12:00 | vote | accept | D L Dahly | ||
| Apr 19, 2013 at 11:18 | history | edited | Maarten Buis | CC BY-SA 3.0 | added 3 characters in body |
| Apr 19, 2013 at 9:16 | comment | added | Maarten Buis | @DLDahly I have edited my answer to deal with these doubts | |
| Apr 19, 2013 at 9:15 | history | edited | Maarten Buis | CC BY-SA 3.0 | added 2531 characters in body |
| Apr 18, 2013 at 11:12 | comment | added | D L Dahly | I appreciate the answer (and the others), but I'm having a hard time accepting it. If I include a 1:0, and the continuous time variable, I still have to assign the non-breast-feeders a value for time (or else they drop for a missing co-variate). Even conditional on the 1:0 variable, I don't see how including the non-breast-feeders as time=0 doesn't affect the regression coefficient. Perhaps also adding the product interaction term between the two would make more sense? | |
| Apr 17, 2013 at 8:05 | history | answered | Maarten Buis | CC BY-SA 3.0 |