Timeline for Best way to compare two treatment groups to a control
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 6, 2021 at 17:27 | history | edited | kjetil b halvorsen♦ | CC BY-SA 4.0 |
added 1835 characters in body
|
| Oct 5, 2021 at 5:46 | comment | added | Elasso | Also - I am wanting to know if Fertilizer types were different from one another. I am struggling to see how to determine that from this model. | |
| Oct 5, 2021 at 5:38 | comment | added | Elasso | I guess I'm just not quite understanding what "contribution to the model" means. You say "group A fertilizer does not contribute significantly to the model." The question I am trying to answer is "Does the mean weight of plants treated with Fertilizer A differ from the mean weight of plants in the control?" and same for Fertilizer B. Does this model show that? Or is there a better way to show this? | |
| Oct 5, 2021 at 4:36 | comment | added | Carl | The other p-values represent the partial probabilities of each corresponding parameter having a significant contribution to the overall model. For example, group fertilizer A does not contribute significantly to the model at p=0.1944. Try leaving it out of the model and see how that effects the significance of group fertilizer B, e.g., it may make it more significant. Generally, partial p<0.05 or at most p<0.1 are regarded as contributory. That may lower the F statistic a bit, but it is worth examining. | |
| Oct 5, 2021 at 4:18 | comment | added | Elasso | That makes sense. Can you explain to me the purpose of the individual p values for Fert A and Fert B in the table I posted then? (The last column in the coefficients table). Thanks a bunch, you've definitely helped to clear it up. I knew the model itself was significant based on the F-statistic and associated p-value, but I don't understand the relevance of the other p-values in that case. | |
| Oct 5, 2021 at 4:13 | history | answered | kjetil b halvorsen♦ | CC BY-SA 4.0 |