What are the assumptions that I need to consider when running a mixed design ANVOA? Let's say I have the following data set:

                 |       Scenario 1       |     Scenario 2         |
                 |Trial 1|Trial 2| Trial 3|Trial 1|Trial 2| Trial 3|
              S1 | ...
 Condition 1  S2 | ...
              S3 |
              S5 |
 Condition 2  S6 |
              S7 |

Where Condition x is the between-subject variable, S1, S2, ... are the subjects, and Scenarios are within-subject variables, where each Scenario has some nested Trials.

Using lmer I can describe this model by: lmer(value~Condition*Scenario + (1+Scenario|Player) + (1|Scenario/Trial).

Taking a more traditional approach I can collapse over Trials so I have the a mean value for each Scenario.

I think I should do the following:

  1. I should check Homoskedasticity per Scenario/Trial
  2. Check for the absence of influential data points

Is there something I am missing?

  • $\begingroup$ I do not think that Heteroskedasticity has anything to do with ANOVA $\endgroup$ – Subhash C. Davar May 22 '14 at 15:01
  • $\begingroup$ I was following Bodo Winter's tutorial. $\endgroup$ – Pio May 22 '14 at 16:19

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.