NEW
Font size
WorksheetsL11 Non-parametric
Total questions: 20
Worksheet time: 10mins
What is the main difference between parametric and non-parametric tests?
Parametric tests are used for larger sample sizes, while non-parametric tests are for smaller sample sizes.
Parametric tests use raw data, while non-parametric tests use adjusted data.
Parametric tests are only used for categorical data, while non-parametric tests are for numerical data.
Parametric tests make assumptions about the population distribution, whereas non-parametric tests do not.
What is a key assumption of parametric tests?
That the sample size is at least 30.
That the independent variable is categorical.
That the population distribution is normal.
That the data have no outliers.
What type of data do parametric tests require for the dependent or criterion variable (DV/CV)?
Nominal.
Ordinal.
Binary.
Continuous (interval/ratio).
When should non-parametric tests be used?
When the data are skewed but have equal variances.
When the assumptions of parametric tests are met.
When data violate the assumptions of parametric tests or when the variable of interest is categorical.
When sample sizes are extremely large.
What are some situations where non-parametric tests are necessary?
When analyzing symmetrical distributions.
When dealing with categorical data, small sample sizes, or non-normally distributed variables.
When the effect size is very large.
When testing for differences between means of normally distributed variables.
What is the main drawback of non-parametric tests compared to parametric tests?
They cannot be applied to continuous data.
They make more assumptions than parametric tests.
They are less powerful and sensitive.
They only work with symmetrical distributions.
How do non-parametric tests handle data?
By normalizing the data.
By calculating confidence intervals.
By ranking the data instead of using raw scores.
By averaging outliers across distributions.
What happens to information about variance when continuous data are converted to ranks?
Variance increases.
Variance is unaffected.
Some information about variance is lost, reducing sensitivity.
Variance becomes proportional to the ranks.
What is the general recommendation when unsure about using parametric or non-parametric tests?
Use parametric tests by default.
Conduct both tests and compare the results.
Always prioritize non-parametric tests.
Increase the sample size and choose parametric tests.
What is a chi-square test used for?
To compare the means of two independent groups.
To assess the strength of a linear relationship.
To measure the effect size of categorical data.
To assess the relationship between two categorical variables.
What is a common non-parametric alternative to the independent-samples t-test?
Kruskal-Wallis test.
Mann-Whitney U test.
Spearman’s rho.
Wilcoxon signed-rank test.
What is the Spearman’s rho test used for?
To test the equality of variances.
To calculate means of categorical variables.
To assess linear relationships between variables.
To measure the strength and direction of the association between two ranked variables.
Why might researchers prefer parametric tests when assumptions are met?
They are easier to conduct.
They are better at detecting differences or relationships.
They are less affected by outliers.
They provide results in ranks rather than raw values.
What happens to power when non-parametric tests are used?
Power increases.
Power becomes independent of the sample size.
Power decreases due to the loss of information about variance.
Power remains unaffected by the choice of test.
What is an example of a situation where non-parametric tests must be used?
When analyzing the mean of two normally distributed variables.
When testing the correlation between two interval-level variables.
When analyzing the relationship between two binary categorical variables, such as yes/no responses.
When testing for differences in normally distributed sample means.
What is an example of a decision point for choosing between parametric and non-parametric tests?
When variables are not normally distributed but parametric tests are still applied.
When categorical variables are ranked.
When the variances of the groups are equal.
When both variables are continuous and have no outliers.
What assumption is violated if a Pearson correlation is calculated with non-normally distributed variables?
The assumption of independence.
The assumption of linearity.
The assumption of normality.
The assumption of homogeneity of variance.
What is the main strength of non-parametric tests?
They are better at detecting relationships in small samples.
They work with a wider range of data types and make few assumptions.
They eliminate outliers from the dataset.
They are more sensitive to variance in the data.
What does the Mann-Whitney U test compare?
The mean ranks of paired groups.
The ranks of two independent groups.
The means of normally distributed variables.
The proportions of variance between two groups.
What is the key benefit of conducting both parametric and non-parametric tests?
It reduces the likelihood of Type 1 errors.
It allows researchers to confirm conclusions when assumptions are violated.
It provides a greater level of sensitivity in all cases.
It helps rank categorical data for comparison.
