What is the main difference between parametric and non-parametric tests?
Parametric tests assume data follows a normal distribution; non-parametric tests do not assume normality.
Understanding the distinction is crucial for selecting the appropriate statistical method.
What type of data is used in parametric tests?
Interval or ratio scale data (continuous, normally distributed).
These data types allow for the application of parametric statistical methods.
What type of data is used in non-parametric tests?
Nominal or ordinal data (non-normal or ranked).
Non-parametric tests are suitable for data that do not meet the assumptions of parametric tests.
What are the assumptions for parametric tests?
- Normal distribution
- Homogeneity of variance
- Independent observations
These assumptions must be met for the results of parametric tests to be valid.
Give examples of parametric tests.
- Student’s t-test
- ANOVA
- Paired t-test
- Pearson correlation
These tests are commonly used when data meet the parametric assumptions.
Give examples of non-parametric tests.
- Chi-square test
- Mann-Whitney U test
- Wilcoxon signed-rank test
- Kruskal-Wallis test
- Spearman rank correlation
Non-parametric tests are often used when data do not meet the assumptions required for parametric tests.
What is the parametric counterpart of the Mann-Whitney U test?
Unpaired t-test
This relationship helps in choosing the appropriate test based on data characteristics.
What is the parametric counterpart of the Wilcoxon signed-rank test?
Paired t-test
Knowing counterparts aids in understanding the relationship between different statistical tests.
What is the parametric counterpart of the Kruskal-Wallis test?
One-way ANOVA
This comparison is useful for analyzing data across multiple groups.
What is the parametric counterpart of the Spearman rank correlation?
Pearson correlation coefficient
Both tests assess relationships between variables but under different assumptions.
Which test is used to compare means between two independent groups (parametric)?
Unpaired t-test
This test is fundamental in hypothesis testing for comparing two groups.
Which test is used to compare medians between two independent groups (non-parametric)?
Mann-Whitney U test
This test is appropriate when the data do not meet parametric assumptions.
Which test compares proportions in categorical data?
Chi-square test
This test is widely used in categorical data analysis.
Which test is used for correlation in ranked (ordinal) data?
Spearman rank correlation
This test is suitable for assessing relationships in non-parametric data.
What happens to the power of test when using non-parametric instead of parametric?
Non-parametric tests are less powerful (require larger sample to detect difference).
Understanding power is essential for effective study design.
Which test is used when sample size < 30 and data non-normal?
Non-parametric test (e.g. Mann-Whitney or Wilcoxon).
This guideline helps in selecting the right statistical approach for small samples.
Which test is used to compare more than 2 groups when data are non-normal?
Kruskal-Wallis test
This test extends the Mann-Whitney U test to multiple groups.
Which test checks paired differences in non-normal data?
Wilcoxon signed-rank test
This test is crucial for analyzing repeated measures or matched samples.
What does ANOVA test compare?
Means of 3 or more independent groups (parametric).
ANOVA is a fundamental technique in statistical analysis for comparing multiple groups.
Which test checks association between two categorical variables?
Chi-square test
This test is essential for understanding relationships in categorical data.
