I've initially used a GLMM on a set of data and I found that one of my predictor variables (categorical) was the only one in the top models explaining the response variable.
One reviewer asked us to use a second test to show the differences among the means of the different levels of this variable. We did so, but we found means did not differ.
I know a linear model and an Anova test quite different things, but I'm not confident how to interpret these results and/or how to argue with the reviewer.
Could anyone give a helping hand? Thanks in advance!
2 things,: 1. Assuming you did the regression and ANOVA correctly there should be no difference between the two. 2. The reviewer knew this and asked you to see if in the
the ANOVA there was a difference among the levels of the categorical variable. You did as asked and found no difference between them. Thus you have done the same test two different ways and got the same results both ways. The x variables are telling you that they are not explaining the y variable in any way that statisics can confirm. I would suggest that you not argue with the reviewer in this case. This,of course, is assuming that I understand your question in the first place. Best wishes, David Booth
See these links:
https://www.theanalysisfactor.com/why-anova-and-linear-regression-are-the-same-analysis/
https://stats.stackexchange.com/questions/28938/why-do-linear-regression-and-anova-give-different-p-value-in-case-of-consideri
https://www.researchgate.net/post/I_am_getting_different_results_from_one_way_ANOVA_and_General_Linear_Model_when_using_the_same_variables_Does_someone_have_any_insight
https://www.researchgate.net/post/Linear_regression_vs_ANOVA_in_R--Different_results
https://www.statsimprove.com/en/what-is-the-difference-between-anova-and-regression-and-which-one-to-choose/
I may recommend you test the data with a regression model only including the single significant categorical variable. If the variable is still found significant, then you may need to check the ANOVA test results. In SAS, the ANOVA test is given along with regression models.
If the variable is not significant. Then it may suggest that some variables in your before regression model cause this issue (e.g., the correlation between independent variable?).
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA