L11 Non-parametric

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.