Family-wise error & COMPARISON Flashcards
1
Q
what is a family-wise error (FWE)?
A
The probability of making one or more false discoveries, or type I errors when performing multiple hypotheses tests.
2
Q
what is the difference between a per comparison error and a family-wise error?
A
- per comparison error: error on individual tests
- family-wise error: error on the whole set of tests
3
Q
what is priori expectation?
A
if the comparisons are planned. no need to correct for FWE
4
Q
what are post hoc comparisons?
A
if we haven’t planned our comparisons, need to correct for FWE
5
Q
what are the 3 major post hoc tests?
A
- Bonferroni correlation
- Scheffe test
- Tukey’s Honest Significant Difference Test
6
Q
What is Bonferroni correction?
A
- most conservative FWE correction
-completely accounts for FWE and ensures the FWER is below the desired level (e.g., 5%) - however with lots of comparisons it is overly conservative
- often performed on t-tests
7
Q
What is the Scheffe test?
A
- adjusts the critical F-value, using the number of treatment levels
- less conservative than Bonferroni correction
8
Q
The Tukey HSD Test
A
- uses a different statistical distribution than the F distribution to limit Type I error
- called the studentised range distribution Q, and can be used to estimate the probability of observing a range (max-min) in DV values for a given sample
9
Q
Facts about post-hoc tests
A
- they are conservative (reduce the chance of Type I errors (false positives) by greatly increasing the chance of Type II errors (false negatives))
- only very robust effects eill be detected
- null results aren’t easily interpreted
- many alternative tests have been proposed, with different trade-offs
- many post-hoc tests are available with statistical software packages such as SPSS ( see slides for how to complete)
