Hello,

I am using the MIXED command in SPSS for an analysis of EEG ERP data. The analysis is complete, and I have the results, except that for the post-hoc pairwise comparisons, I would like to report not just the p-, but also the t-value. As far as I understand, the t-value is not shown in the output and I cannot find a toggle to include it...

Following, I paste my output of interest and my syntax (in case needed).

Output example tables are attached as a file below (numbers greyed out since not relevant).

I don't see any t-value output. I'm not 100% sure, but I would suppose the Mean Difference (I-J) is the difference between the estimated marginal means of the compared pairs, while t should be T = d /SEd, where d is the mean difference of all di = yi − xi of two sets of paired data points X and Y (which in my case would correspond to the levels "Poor" and "Typical" of the Factor "group") and SEd is the standard error of the mean difference.

So can I use the Mean Difference (I-J) and the standard error shown in the pairwise comparison output to calculate the t-value, i.e. (Mean Difference) / (Std. Error), would that be correct?

Any feedback on whether this is the correct way to get the t-value and whether there is a more direct way to output the t-value would be greatly appreciated!

Syntax of interest:

MIXED value BY group hemisphere cond WITH rias_nix_T1 CBCL_Aufmerksamkeitsstoerung_T

/CRITERIA=CIN(95) MXITER(100) MXSTEP(10) SCORING(1) SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) LCONVERGE(0, ABSOLUTE) PCONVERGE(0.000001, ABSOLUTE)

/FIXED= rias_nix_T1 CBCL_Aufmerksamkeitsstoerung_T cond group hemisphere cond*group cond*hemisphere hemisphere*group cond*hemisphere*group| SSTYPE(3)

/METHOD=REML

/PRINT=DESCRIPTIVES SOLUTION TESTCOV

/RANDOM=INTERCEPT | SUBJECT(subject) COVTYPE(VC)

/EMMEANS=TABLES(OVERALL)

/EMMEANS=TABLES(cond) COMPARE ADJ(Bonferroni)

/EMMEANS=TABLES(group) COMPARE ADJ(Bonferroni)

/EMMEANS=TABLES(hemisphere) COMPARE ADJ(Bonferroni)

/EMMEANS=TABLES(cond*group) COMPARE(group) ADJ(Bonferroni)

/EMMEANS=TABLES(cond*group) COMPARE(cond) ADJ(Bonferroni)

/EMMEANS=TABLES(cond*hemisphere) COMPARE(hemisphere) ADJ(Bonferroni)

/EMMEANS=TABLES(cond*hemisphere) COMPARE(cond) ADJ(Bonferroni)

/EMMEANS=TABLES(hemisphere*group) COMPARE(group) ADJ(Bonferroni)

/EMMEANS=TABLES(hemisphere*group) COMPARE(hemisphere) ADJ(Bonferroni)

/EMMEANS=TABLES(cond*group*hemisphere) COMPARE(hemisphere) ADJ(Bonferroni)

/EMMEANS=TABLES(cond*group*hemisphere) COMPARE(group) ADJ(Bonferroni)

/EMMEANS=TABLES(cond*group*hemisphere) COMPARE(cond) ADJ(Bonferroni).