I hope you are able to help me. I'm not familiar with RStudio but have to use it for my MA thesis on uptalk in reality TV.
I did post hoc t-tests with emmeans and have the output. But I don't know which values indicate the likelihood of the factors I have tested. So my question is: Which of the values I put in bold stands for probability?
Hi @AngieBee, it is not clear what you are testing for, what is your model, is it a linear regression or? From the table none of the values are probability - you have a standard error, a lower and upper confidence bound (or limit I guess). But since we can't see what tests you are doing and on what model, it is difficult to say more.
Thanks for your reply. I'm using linear mixed effects Models (glmer) to find out which characteristics are significant predictors of a particular linguistic feature. These characteristics have sub-categorizations. I have now determined which characteristics are significant predictors and would like to know which sub-categories trigger the feature more than other do, so I'm using t-tests with emmeans.
My problem is that I don't know how to deal with the output..does a higher emmean value refer to a higher occurrence of the feature and a lower emmean refer to a lower occurrence? At least this would be in line with my results...
emmeans has several detailed "vignettes" (long-form documentation essays) bundled with the package that can help you understand what it does and how to use it. Have you seen any of these yet? From within R, you can find them with the command, browseVignettes(package="emmeans"), or view them directly from the package page on CRAN.