ANOVA Dunnett's Post-hoc Analysis

Valentinos

I was wondering which one of the many Dunnett's Post hoc analysis in ANOVA is the one used by JASP. The reason I'm asking is that T3 and C are more suited for violation of the variance's homogeneity assumptions.

JohnnyB

Hi @Valentinos,

The Dunnett test that JASP offers, is the type where all levels are compared to a control group (i.e., "Dunnett’s Many-to-One Comparisons Test"). This is based on the multcomp package (specifically, the glht and mcp functions). The versions of the test you mention, if I understand correctly, refer to the type of test where unequal variances are assumed. In JASP, we do offer such a test, under the Games-Howell test.

Kind regards,

Johnny

Valentinos

Hi @JohnnyB and thanks for your reply,

I wasn't referring to Games-Howell test, that I know It's a different one.

Dunnett produced different post-hoc tests, I'll point to the literature if you want, in the while I check for the R package used if I could find more information.

Thanks again