# Large non-significant effect in a generalized linear mixed model

In my experiment I measure accuracy (0-1) in two groups, on two different types of stimuli. I use the generalized linear mixed model because I think this is the only model that can deal with binary DVs and with nested data (trials within the same participant).

SPSS syntax:

GENLINMIXED
/DATA_STRUCTURE SUBJECTS=id REPEATED_MEASURES=trial COVARIANCE_TYPE=AR1
/FIELDS TARGET=accuracy TRIALS=NONE OFFSET=NONE

• I know that genlinmixed only reluctantly gives tables of parameter estimates and their standard errors, but you could possibly get that information for us? It is almost impossible to answer as is, I think. If you are at all familiar with other statistical packages (e.g., R, SAS, Stata) I would encourage you to try one of those. The syntax for this type of model is much easier, and you get standard output that can be readily shared rather than "a red line instead of a gold one". – Joshua Jul 18 '12 at 2:12