I can't give you specific reasons, but my intuition when I've run into something similar is that the packages are computing differently because of either a default variable that's been specified for you, or because the underlying computation is different. So perhaps ezANOVA is doing something fishy somewhere? I suddenly get values much closer to that of SPSS (and the descriptive means) If I perform the ANOVA using the AFEX package: results <- aov_ez(ĮMeans <- emmeans(results, ~ group | session | cue | flanker)
Keep in mind that the rest of the statistics and the descriptive means áre equal between SPSS and R. So even with equal group sizes in both conditions, I end up with quite different means. Since the group sizes are unbalanced, I also redid the analyses above after making the groups of equal size. Am I mistaken in this? And if so, how come the two give different results? It's been my understanding that the estimated marginal means should be the same as the descriptive means in this case, as I have not included any covariates. I run into descrepancies as compared to the Estimated Marginal Means table from the SPSS GLM output (for the same interactions), eg: However, when I try to get to the estimated marginal means, using the emmeans package: eMeans <- emmeans(results$aov, ~ group | session | cue | flanker) Group(1) - Session(1) - Cue(center) - Flanker(1) Which again line up perfectly with the descriptives from SPSS, e.g. I also ask for the means per cell using: descMeans <- ezStats( WSDESIGN=session cue flanker session*cue session*flanker cue*flanker session*cue*flanker EMMEANS=TABLES(group*session*cue*flanker) WSFACTOR=session 2 Polynomial cue 3 Polynomial flanker 2 Polynomial Rt.2.no.congruent rt.2.no.incongruent rt.2.ngruent rt.2.spatial.incongruent BY group Rt.1.ngruent rt.1.spatial.incongruent rt.2.ngruent rt.2.center.incongruent In SPSS I use the following GLM: GLM rt.1.ngruent rt.1.center.incongruent rt.1.no.congruent rt.1.no.incongruent In R I'm using the ezAnova package to do the mixed-model anova: results <- ezANOVA( Unequal number of participants per group level (25,25,23) Within = Session (2 levels), Cue (3 levels), Flanker (2 levels) Since I'm not confident in my R skills yet, I use the exact same dataset in SPSS to compare my results. I recently switched from SPSS to R, and I'm now trying to do a mixed-model ANOVA. So this is a bit of a hail mary, but I'm hoping someone here has encountered this before.
0 kommentar(er)
