Parametric and Non-Parametric Tests Flashcards
What is a parametric test?
They assume that the data follows a specific distribution, typically normal.
When should you use a non-parametric test?
When data do not meet the assumptions necessary for parametric tests, or when dealing with ordinal data or ranks.
What assumptions must be met for a t-test?
Normality of data, homogeneity of variances, and independence of observations.
Describe the ANOVA test and its purpose.
It compares means across multiple groups to determine if at least one differs significantly.
How do you interpret the results of a Chi-square test?
It assesses whether observed frequencies differ significantly from expected frequencies.
What is the Mann-Whitney U test used for?
It compares differences between two independent samples using ranks.
Explain the purpose of the Wilcoxon signed-rank test.
Used to compare two related samples where the data are not normally distributed.
What differences are there between the t-test and the Mann-Whitney U test?
The t-test assumes normality and equal variances; the Mann-Whitney does not.
What is the significance of homogeneity of variances in ANOVA?
It is crucial because unequal variances can lead to erroneous conclusions in ANOVA.
How can you test for normality before conducting a parametric test?
Using graphical plots like Q-Q plots, or tests like the Shapiro-Wilk test.
What are the key advantages of using non-parametric tests?
They are more flexible and can be used with ordinal data or non-normal distributions.
How does sample size affect the choice between parametric and non-parametric tests?
Non-parametric tests are more suitable for small sample sizes or when assumptions are not met.
What is the Kruskal-Wallis test?
It compares the medians of three or more independent groups.
How do you determine which type of ANOVA to use?
Based on the number of factors and the independence of samples.
What are the assumptions behind the Pearson correlation coefficient?
The data must be normally distributed and the relationship between variables linear.
Why is the Spearman’s rank correlation coefficient considered a non-parametric test?
It does not assume normality and works with rank-ordered data.
What statistical test would you use to compare the means of three or more groups?
ANOVA is typically used for this purpose.
What is the best test to use when comparing two related samples?
The paired t-test for parametric data, or the Wilcoxon signed-rank test for non-parametric data.
Why might parametric tests be more powerful than non-parametric tests?
Because they make more assumptions about the data, which can provide more precise results if those assumptions are met.
What are the limitations of non-parametric tests?
They often have less statistical power and may not provide as detailed information about the population.
How do you handle violations of assumptions in parametric tests?
Adjust the analysis method or transform the data to better meet the assumptions.
Why is it important to check for outliers before conducting a parametric test?
Outliers can significantly skew the results and violate the assumptions of normality and homogeneity.
What role does independence of observations play in statistical testing?
It ensures that the test results are valid and that the samples do not influence each other.
How do you interpret a significant result in a non-parametric test?
It indicates that there is a statistically significant difference between the groups or conditions tested.
What adjustments should be made for multiple comparisons in ANOVA?
Bonferroni correction or other methods to adjust for the increased risk of Type I errors.
How do you choose between a one-way and a two-way ANOVA?
Choose one-way for one independent variable and two-way for two independent variables.
What are the benefits of using the Wilcoxon signed-rank test over the paired t-test?
It does not assume normal distribution of differences and is less affected by outliers.
What methods can be used to assess the effect size in non-parametric tests?
Using rank-based methods or estimating the interquartile range difference.
How do you handle data that does not meet the assumptions of homoscedasticity?
Using robust statistical methods or non-parametric tests.
What are common errors in the application of Chi-square tests?
Not meeting the assumption of expected frequency counts being high enough in each category.
What should you consider when choosing between a parametric and a non-parametric correlation coefficient?
Consider the distribution of the data and whether the data meets the assumptions of the parametric test.
How do bootstrapping methods benefit non-parametric tests?
They allow for estimating distributions from the data without assuming a specific form.
What is the Friedman test and when should it be used?
Used for comparing more than two groups that are related, measured on at least an ordinal scale.
Can non-parametric tests be used on ordinal data?
Yes, they are especially suitable for ordinal data.
How do you ensure the reliability of test results in non-parametric tests?
By using appropriate sample sizes and ensuring the test assumptions are adequately met.
What are the consequences of using the wrong type of statistical test?
It can lead to incorrect conclusions, affecting the validity and reliability of the research findings.
Why is it necessary to use a control group in experimental designs using ANOVA?
To provide a baseline against which the treatment effects can be compared.
What are the implications of a low power in non-parametric tests?
It may fail to detect actual effects or differences when they exist.
How can transformations improve the applicability of parametric tests?
Making data more normal, thus meeting the assumptions of parametric tests more closely.
What should be considered when interpreting the results of statistical tests?
Ensure that all test assumptions have been met and consider the practical significance of the results.