Conditions for significance tests?
My class is reviewing for the AP exam, and tomorrow we're having a quiz on conditions for all significance tests. Unfortunately, I don't have my book with me. I do remember most of the conditions, I just need someone to help fill in the blanks.
SRS, Normality (n>30). Is there anything else? Also, is there a difference in conditions between one sample and two sample z-tests?
SRS, Normality (If n<15, no skewness or outliers; if 15<n<30 no outliers, can be relative skewness, if n>30 no outliers). Again, am I missing anything? Are the conditions different for matched pairs test and two sample test?
3. 1 Prop z test:
SRS, population > 10n, n(p-hat) > 10, n(1-p-hat) > 10
4. 2 prop z test:
SRS, N >10n; n1p1 ≥ 5; n1(1-p1) ≥ 5; n2p2 ≥ 5; n2(1-p2) ≥ 5
5. Chi squared tests:
SRS, All expected counts are at least 1; No more than 20% of the expected counts are less than 5. (Are there any differences in the conditions for chi squared tests of goodness of fit, assoc/ind, homogeneity?)
That's all I can remember. What am I missing?