PARAMETRIC AND NON PARAMETRIC TESTS Flashcards
What are the four major parametric tets?
Z test, T test, ANOVA and Two way ANOVA
A tets used when the population is greater than 30
Z test
A test when the population is greater than 10
T test
A test determining the significant differences between means of three or more independent variables
One way ANOVA
A test compares the mean differfences between groups that have been SPLIT into two independent variables (called factors) on the dependent variable
Two way ANOVA
What are the situations which you are required to use parametric tests?
- If the data involves nominal and ordinal measurement.
- Data do not satisfy assumptions underlying parametric tests
- If the variance is extremely high
What are the three major examples of nonparametric tests?
Chi square, Binomial test and Sign tests
A nonparemetric test used to know the difference of the distribution of the categorical variables from one another
Chi square
Two kinds of Chi Square test
- Chi square for independence
2. Chi square goodness of fit
A type of chi square which u want to know if there is a significant difference of the two categorical variables from a single population
Chi square for independence
A type of chi square which u want to know if one sample data is consistent from a distribution of categorical variables
Chi square goodness of fit
It is a nonparemetric test which is an exact test of statistical significance of deviation from a thereotically expected distribution of two categories
Binomial Test
It is a parametrical test which is the alternative test to Wilcoxon test for dependent data
Sign Test
A data needs to be least interval scaled
Wilcoxon test
A data needs to be least ordina scaled
Sign Test
PARAMETRIC VS. NONPARAMETRIC
1. In terms of assumed distribution
PARAMETRIC - Normal
NONPARAMETRIC - Any
PARAMETRIC VS. NONPARAMETRIC
2. In terms of assumed variance
PARAMETRIC - Homogenous
NONPARAMETRIC - Any
PARAMETRIC VS. NONPARAMETRIC
3. Typical data
PARAMETRIC - Ratio or Interval
NONPARAMETRIC - Ordinal or Nominal
PARAMETRIC VS. NONPARAMETRIC
4. Data set relationship
PARAMETRIC - Independent
NONPARAMETRIC - Any
PARAMETRIC VS. NONPARAMETRIC
5. Usual central measure
PARAMETRIC - Mean
NONPARAMETRIC - Median
PARAMETRIC VS. NONPARAMETRIC
6. Benefits
PARAMETRIC - Can draw more conclusions
NONPARAMETRIC - Less affected by outliers
PARAMETRIC VS. NONPARAMETRIC
1. Choosing
PARAMETRIC - Choosing parametric test
NONPARAMETRIC -Choosing non-parametric test
PARAMETRIC VS. NONPARAMETRIC
2. Correlation Test
PARAMETRIC - Pearson
NONPARAMETRIC - Spearman
PARAMETRIC VS. NONPARAMETRIC
3. Independence measures, 2 groups
PARAMETRIC - Independence measures T Test
NONPARAMETRIC - Mann-Whitney Test
PARAMETRIC VS. NONPARAMETRIC
4. Independence measures, >2 groups
PARAMETRIC - One way ANOVA
NONPARAMETRIC - kruskal Wallis Test
PARAMETRIC VS. NONPARAMETRIC
5. Repeated measures, 2 conditions
PARAMETRIC - Matched paired
NONPARAMETRIC - Wilcoxon Test
PARAMETRIC VS. NONPARAMETRIC
5. Repeated measures, >2 conditions
PARAMETRIC - Oneway ANOVA
NONPARAMETRIC - Friedman’s Test
