Timeline for Interpretation of Interaction when the results in one of the two groups is not statistically significant
Current License: CC BY-SA 4.0
22 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 19, 2023 at 15:00 | vote | accept | user89547235 | ||
| S Jan 19, 2023 at 13:03 | history | bounty ended | CommunityBot | ||
| S Jan 19, 2023 at 13:03 | history | notice removed | CommunityBot | ||
| Jan 14, 2023 at 14:46 | comment | added | Sextus Empiricus | Ah, I missed that, the model is for a hazard ratio, which is zero for males without treatment. | |
| Jan 14, 2023 at 14:40 | comment | added | EdM | @SextusEmpiricus Cox models don't have intercepts. The non-parametric baseline hazard serves that purpose. Under the proportional hazards assumption, regression coefficients represent differences in log-hazard that are constant in time regardless of the underlying baseline hazard over time. | |
| Jan 14, 2023 at 8:17 | comment | added | Sextus Empiricus | I am actually also wondering about the missing intercept in the table. | |
| Jan 14, 2023 at 8:09 | comment | added | Sextus Empiricus | This before last sentence "However, how can one interpret the non-statistically significant results observed for females while the effect remained statistically significant in males?" It is unclear which numbers or coefficients in the previous text are used to state that there is a non-significant result for females and a significant result for males. | |
| Jan 14, 2023 at 8:02 | comment | added | Sextus Empiricus | @EdM That might be to the case (and you explained it nicely how that can work), but I would like the OP to clarify this. I understood that those two values are significant as it shows in the text two intervals that indicate significance (although the intervals are not clearly defined); and I thought that it might be about to the coefficient for sexFemale which has a 0.2834 p-value. | |
| Jan 14, 2023 at 3:01 | history | tweeted | twitter.com/StackStats/status/1614095256269590529 | ||
| Jan 14, 2023 at 1:52 | answer | added | Graham Wright | timeline score: 1 | |
| Jan 13, 2023 at 22:39 | comment | added | EdM | @SextusEmpiricus for females, I took the OP to mean the sum of the coefficient for A (for males as the reference category) and the A:female interaction coefficient, for the estimated net association of A with outcome in females. | |
| Jan 13, 2023 at 22:35 | comment | added | Sextus Empiricus | "However, how can one interpret the non-statistically significant results observed for females, while the effect remained statistically significant in males? " What does this mean? How do you differentiate between 'for males' and 'for females'? | |
| Jan 13, 2023 at 22:21 | answer | added | EdM | timeline score: 3 | |
| S Jan 11, 2023 at 12:01 | history | bounty started | user89547235 | ||
| S Jan 11, 2023 at 12:01 | history | notice added | user89547235 | Canonical answer required | |
| Jan 9, 2023 at 16:45 | comment | added | user89547235 | Yes, it was. I am rather interested in the interpretation of such results from a general point of view rather than in the specific scenario, as this (non-statistically significant interaction with different statistical significance in the two groups) is one common situation, so that would be particularly interesting | |
| Jan 9, 2023 at 16:31 | comment | added | EdM | Your calculation of the A*sex=Female coefficient is based on the sum of the A:sexFemale and the A coefficients. In estimating the CI, did you take into account the covariance between those coefficients or just use their individual standard errors? Please edit the question to show those details. You might still get a "statistically insignificant" result but it's important to start with knowing whether the CI were calculated properly. | |
| Jan 9, 2023 at 15:18 | comment | added | user89547235 | @EdM done, I have added more details are requested, hopefully my question is now understandable | |
| Jan 9, 2023 at 15:17 | history | edited | user89547235 | CC BY-SA 4.0 | Added more details are requested |
| Jan 9, 2023 at 14:35 | history | edited | EdM | edited tags | |
| Jan 9, 2023 at 14:34 | comment | added | EdM | Please provide more details on the Cox model that you used. With a dichotomous predictor like sex one typically only has a single "main"-effect coefficient and a single "interaction" effect with a treatment. Without seeing the model as written, it's hard to know how to interpret the separate A*sex=Female and A*sex=Male coefficients. Please provide that information by editing the question, as comments are easy to overlook and can be deleted. | |
| Jan 9, 2023 at 11:52 | history | asked | user89547235 | CC BY-SA 4.0 |