Say I have a two group pre test / post test design. The IV is attendance or non attendance at an educational course. The DV is ordinal from 0 - 5 / Strongly Agree - Strongly Disagree
I have been reading about the benefits of ANCOVA - would it's use be appropriate here and if not, what would be?
An ANCOVA is simply a multiple regression with no interaction term between the IV and the covariate. You can perform a proportional odds regression given your ordinal outcome.
The ANCOVA is useful if there is balance on the pre-test across both groups and/or you have random assignment to both groups. See Knapp & Schafer (2009) for further discussion of this. Also, Miller and Chapman (2001) provide a very good discussion of ANCOVA.
Knapp, T. R., & Schafer, W. D. (2009). From Gain Score t to ANCOVA F (and vice versa). Practical Assessment, Research & Evaluation, 14(6). Retrieved from pareonline.net/getvn.asp?v=14&n=6
Miller, G. A., & Chapman, J. P. (2001). Misunderstanding analysis of covariance. Journal of Abnormal Psychology, 110(1), 40–48. https://doi.org/10.1037/0021-843X.110.1.40
