1
$\begingroup$

I'm running a glmer and have a few questions regarding how to interpret the output and how to report it:

it's not clear to me what the main effects are given that they are all in reference to a base category, so probably it is useless to talk about a 'main effect' of X with an associated p-value.

I would like to run further contrasts to the ones report it in the summary output, can I use emmeans()?

If I get the p-values that I'm interested in with emmeans, should I report these p-values together with the associated estimate? and how should I report this estimate? is it a Beta?

This is my code:

prob_stat = glmer(cbind(freq_correct, count-freq_correct) ~ 0+impulsivity*ITI +(1|rat_ID), family ="binomial", data= day_1) summary(prob_stat)

MainContrasts<- emmeans(prob_stat, pairwise~ ITI | impulsivity)

enter image description here

Improve this question
$\endgroup$
4
  • $\begingroup$ related to this: stats.stackexchange.com/questions/87412/… $\endgroup$
    – Chiara Toschi
    Commented Jul 22, 2020 at 14:37
  • $\begingroup$ Please don't put picture of output into your posts. Just post the actual output with ``` above and below and it will be formatted nicely $\endgroup$
    – Robert Long
    Commented Jul 22, 2020 at 16:14
  • $\begingroup$ Also why doesn't you model have a global intercept ? And please explain your study design and research question(s) $\endgroup$
    – Robert Long
    Commented Jul 22, 2020 at 16:22
  • $\begingroup$ Thank you! it doesn't have an intercept because I specified 0 in the coding, if you remove that impulsivityHI becomes the intercept. research question and design: I have 36 rats some are highly impulsive (HI), some are low-impulsive (LI) and some are MIDs. I ran them on two separate days on a task that presents a cue after 4 variable inter-trial intervals (ITIs), these intervals are: 3000ms, 5000ms, 7000ms and 9000ms. I'm looking at the probability of making a correct response (DV) as a function of the impulsivity phenotype (3 levels); Day (2 levels) and the ITI (4 levels). Thank you! $\endgroup$
    – Chiara Toschi
    Commented Jul 22, 2020 at 20:47

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.