Introduction to Two-Way Analysis of Covariance (Two-Way ANCOVA)

This video is an introduction to the two-way analysis of covariance (two-way ANCOVA) including a description of how it is used, its elements, and the assumptions data must meet to be analyzed by the test. The assumptions of normality, homogeneity of variances, linearity, and homogeneity of regression slopes are reviewed.

[@alargeturtle]

Wow, a pleasant surprise to see Dr Grande pop up when I'm searching for stats help

[@kjartanvarbakken9825]

Hi Todd, According to the textbook Fundamentals of Biostatistics (Rosner, 2016, p. 596), the two-way ANCOVA is used when we wish to simultanously relate a Normally distributed outcome variable to two categorical variables of primary interest and, in addition, wish to control for one or more other covariates, which may be continous or categorical. Several texts (Discovering statistics using IBM SPSS by A. Fields, 2013), Wikipedia, and Penn State University Statistics (online) supports your claim of only using contious variables as covariates. However, statistics professor Rosner may be correct, and he sure is in line with information on IBM SPSS 2018. What do you think should be the reason why a categorical variable cannot be treated as a covariate in an ANCOVA model?