Timeline for Boxplots: valid method for visualizing collinearity?
Current License: CC BY-SA 4.0
14 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 3, 2023 at 14:25 | vote | accept | Nate | ||
| S Jan 3, 2023 at 7:09 | history | suggested | Shawn Hemelstrand | CC BY-SA 4.0 |
clarified this is a regression problem and added it as a tag
|
| Jan 3, 2023 at 4:39 | review | Suggested edits | |||
| S Jan 3, 2023 at 7:09 | |||||
| Jan 3, 2023 at 4:03 | answer | added | Shawn Hemelstrand | timeline score: 5 | |
| Jan 3, 2023 at 1:15 | history | edited | Nick Cox | CC BY-SA 4.0 |
deleted 15 characters in body
|
| Jan 2, 2023 at 22:22 | comment | added | whuber♦ | That site is problematic from the outset. "As the name suggests, an independent variable should be independent. It shouldn’t have any correlation with other independent variables" is laughably wrong. The next sentence, "If collinearity exists between independent variables, the key point of regression analysis is violated," is just blatantly incorrect. The mistakes pile on from there, one wrong statement after another. Whatever concept of "regression" might motivate the writer, it certainly is not what is understood in statistics! | |
| Jan 2, 2023 at 22:22 | comment | added | Nate | Aha, got it. Thank you! | |
| Jan 2, 2023 at 22:19 | comment | added | whuber♦ | If TL depends on species, there needn't be any collinearity apart from the mathematically necessary minimum imposed by the fact that an individual belongs to a unique species. | |
| Jan 2, 2023 at 22:18 | comment | added | Nate | The quote link: stratascratch.com/blog/… | |
| Jan 2, 2023 at 22:18 | comment | added | whuber♦ | That quote expresses an interesting misunderstanding of collinearity! It's definitely not the case that changing one explanatory variable necessarily causes any of the other explanatory variables to change. | |
| Jan 2, 2023 at 22:18 | comment | added | Nate | So, if TL depends on Species (boxplots don't overlap) then TL isn't constant over all species and we may have collinearity, no? This looked like a fitting example too: stats.stackexchange.com/questions/48688/… | |
| Jan 2, 2023 at 22:16 | comment | added | Nate | @whuber, I thought this would help me: "In regression analysis, we want to isolate the influence of each independent variable to our dependent variable. This way, we can interpret the fitted coefficient of each independent variable as the mean change in the dependent variable for each 1 unit change in an independent variable while keeping the other independent variables constant. Now if we have collinearity, the key point above is no longer valid, as if we change the value of one independent variable, the other independent variables that are correlated will also change." | |
| Jan 2, 2023 at 22:12 | comment | added | whuber♦ | This sounds like someone has confused "significantly different location" with "collinear:" the concepts are completely different and boxplots can't show you a thing about collinearity. Overlap of boxplots tells you little about significant differences, either, but overlap is related to common misconceptions about overlaps of confidence intervals. | |
| Jan 2, 2023 at 21:50 | history | asked | Nate | CC BY-SA 4.0 |