Hello All,
In SPSS, the Mauchley's test does not provide any result if the sample size is less than the repeated measurement count.
I have a sample size of 19 and a repeated measures of 22.
My Mauchley test's value results are as 'p=.' (just a dot) and chi square values = '.' (just a dot; no number is displayed).
I observed that:
(1) If my repeated measurement is greater than sample size (eg. Repeated measurement = 22 and sample size = 19), the Mauchley's test fails to give a value for chi-squared and p-value, and displays only a '".". However, I could see a value for Greenhouse-Geisser and Huynh-Feldt.
(2) if I include only 5 repeated measurement in my analysis for sample size = 19 (i.e repeated measurements < sample size), Mauchley;s test gives a value for both chi square and p-value.
I have attached the mauchley test results for the first case.
In my first case, should I assume that the sphericity has been violated and go for a correction?
Thanks
Sri
Two concerns:
1-Mauchly's test is not a good indicator of sphericity.
2-Not all repeated measures exhibit sphericity type covariances. And usually it is not the case. So a better model should be preferred.
I recommend using mixed models or latent growth curve modeling for your data.
This same basic question was discussed on the SPSS mailing list recently. See the link below. HTH.
http://spssx-discussion.1045642.n5.nabble.com/Undefined-Mauchly-s-Test-td5733212.html
Two concerns:
1-Mauchly's test is not a good indicator of sphericity.
2-Not all repeated measures exhibit sphericity type covariances. And usually it is not the case. So a better model should be preferred.
I recommend using mixed models or latent growth curve modeling for your data.
One shouldn't be using the Mauchley test (which is not very robust or very powerful).It also addresses the wrong issue. The question is not whether sphericity is violated, but how severely. If you have more than 2 levels there is nearly always some departure from spheicity. Epsilon estimates are measures of the deviation and values close to 1 indicate little or no problem, while values below 1 indicate progressively more severe departures. A sensible option is to rely on one of the corrections whenever epsilon estimates are say below 0.9 or 0.95.
For pure repeated measures models the corrections are known to work well. More complex models can be helpful if you have mixed measures.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS