Tests of Differences and Nonparametric Alternatives Flashcards
REVIEW
REVIEW
When is a student t-test used?
Proper SHT for testing the mean difference between the two independent groups.
When is a paired t-test used?
Proper SHT for testing the mean differences between the two related groups.
When is a analysis of variance (ANOVA) used?
Extension of a Student’s t-test for multiple groups.
When is a repeated measures ANOVA used?
Extension of a paired t-test for multiple points in time.
When is a factorial ANOVA used?
Applied when the mean differences were compared by multiple factors.
When is a mixed ANOVA used?
Applied when the mean differences were compared by multiple factors over time.
When is a analysis of covariance (ANCOVA) used?
Applied when the mean differences were compared controlling for a confounding variable.
What are some questions to ask when deciding which statistical test to apply?
- ) Is the level of measurement continuous?
- To compare means of continuous.
- To compare proportions (percent) of categorical. - ) How many groups are to be compared?
- ) Are the data normally distributed with equal variances across groups.
- ) Are the collected data independent or matched.
NON-PARAMETRIC TESTS
NON-PARAMETRIC TESTS
When are non-parametric tests used?
When the data is not normally distributed or if the variances of the data in the groups are very different.
What are the 3 things that identifies data as parametric?
- Data level of measurement to be continuous.
- Normally distributed.
- Equal variances across groups to compare.
- __________ tests (student’s t-test, ANOVA,…) are the most powerful on parametric data.
- ____________ tests are available when data is non-parametric.
- Parametric
- Non-parametric
Non-Parametric Tests:
- ______________ test in place of the Student’s t-test.
- _________________ test in place of the paired t-test.
- ________________ test in place of ANOVA.
- ________ test in place of repeated measures ANOVA.
- Mann-Whitney U
- Wilcoxon signed-rank
- Kruskal-Wallis H
- Friedman
Non-parametric tests work on the principle of ________ the data.
ranking
Non-parametric tests work on the principle of rank ordering of scores.
- Scores are ranked from _________ to ___________.
- With the rank of __ assigned to the smallest score and _ to the highest score.
- smallest to largest
- 1, n
Non-parametric test may not be testing the null hypothesis of interest.
- Null hypotheses that can be studied using non-parametric tests tend to be very _________.
- There is not much choice for non-parametric tests.
-restrictive
- Non-parametric methods are focused on __________ testing rather than estimation.
- Can non-parametric tests calculate the confidence interval (CI)?
- significance
- No
Mann-Whitney U Test:
- A non-parametric alternative to _____________.
- Used to test the _____ difference between the ____ _________ groups.
- Student’s t-test
- mean, two independent
Wilcoxon SIgned-Rank Test:
- A non-parametric alternative to ____________.
- Used to test the _____ difference between the ____ ________ groups.
- Paired t-test
- mean, two matched
Kruskal-Wallis H Test:
- A non-parametric alternative to _____________.
- Used to test the _____ difference between the ______ or more __________ groups.
- ANOVA (analysis of variance)
- mean, 3 or more independent
Friedman Test:
- A non-parametric alternative to ________________.
- Used to test the _____ difference between the _____ or more ________ groups.
- Repeated Measures ANOVA
- mean, 3 or more related
Quiz 1:
- ) Mann-Whitney U test is used to compare (2,3 or more) sets of scores that are (independent,related).
- )The Kruskal-Wallis H test is used to compare (2,3 or more) sets of scores that are (independent,related).
- Wilcoxon signed-rank test is used to compare (2,3 or more) sets of scores that are (independent,related).
- Friedman’s test is used to compare (2,3 or more) sets of scores that are (independent,related).
- 2, independent
- 3 or more, independent
- 2, related
- 3 or more, related
Quiz 2:
-A non-parametric test was used to compare the difference between the mean alcohol consumption scores at the two institutions. The non-parametric test was used as a substitute for a Student’s t-test because the assumption of the normal distribution with the alcohol consumption scores is questionable. The mean alcohol consumption for the students at the religious institution was 11.9 (SD=27.6) drinks in the 30 days prior to the survey, which was significantly lower than 26.9 (SD=53.1) drinks per 30 days for students attending the secular university (p-value < 0.05). Which non-parametric test would they have used?
Mann-Whitney U test
What are the weaknesses of non-parametric tests?
- Less powerful tan parametric tests on parametric data.
- May not be testing the null hypothesis of interest.
- Null hypothesis that can be studied using non-parametric tests tend to be very restrictive.
- Not much choice for non-parametric tests.
- Can’t be related to the confidence interval (CI).