Timeline for Why is the two-sample test giving me inconsistent results?
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 8, 2021 at 10:39 | comment | added | Scortchi♦ | @Klik: No - I did have in mind a two-stage nested design, with batch a random effect nested within the fixed effect build; but more because that's a natural way of viewing the results you presented in your question than because it's necessarily a good approach given your goals, constraints, & costs (about which I'd be mostly guessing). | |
| Jun 7, 2021 at 18:58 | comment | added | Klik |
Thank you for following up with my question - I appreciate the time. I'm glad you pointed out how the sample size scales the results for t-statistics. I had given those scaling terms some attention but wasn't sure what other ways I could approach this. I will follow up by investigating mixed-models - very helpful point. On that note, do you have any favorite mixed-model approaches? I'm implementing this through SQL queries so part of my consideration is ease of implementation.
|
|
| Jun 7, 2021 at 18:48 | vote | accept | Klik | ||
| Jun 7, 2021 at 11:20 | history | edited | Scortchi♦ | CC BY-SA 4.0 |
added 693 characters in body
|
| Jun 7, 2021 at 10:10 | history | answered | Scortchi♦ | CC BY-SA 4.0 |