Flashcards
15. repeated-measures designs
Cards in this deck (8)
- 1the assumption of sphericitythe variation within conditions is similar and no two conditions are any more dependent than any other two
- 2what if the variances of e.g. two out of three conditions are roughly equal, but the third is not?you can assume local circularity/local sphericity
- 3how many conditions do you need for the assumption of sphericity to be relevant?at least 3. it's not relevant if you're comparing only two means.
- 4what's the effect of violating the assumption of sphericity?it creates a loss of power and an F-statistic that doesn't have the distribution that it's supposed to have (an F-distribution)
- 5what to do if the assumption of sphericity is violated?look at Greenhouse-Geisser's estimate or Huynch-Feldt estimate
they adjust the DF for the affected F-statistics, which changes the p-value but not the test statistic - 6what's the subdivision of the total SS in a repeated-measures design?1. between-participant variation
2. within-participant variation
a. SSmodel (effect of experiment)
b. SSresidual - 7what to do when the Greenhouse-Geisser correction and the Huynh-Feldt correction give different p-values/conclusions about the significance?average the two and base your conclusions on that p-value
- 8what can you use the multivariate test statistic (MANOVA) for?it doesn't assume sphericity. you can find the results in the table 'multivariate tests'
Information
- Description
- ssr exam 2 Flashcards on 15. repeated-measures designs, created by Sabine - on 18/10/2020.
- Created by
- Sabine -
- Views
- 0
- Rating
More on Factorial and Repeated-Measures Designs