Lecture 6

Question 1

When to use non-parametrics?

Answer 1

• When assumptions are violated and can’t be fixed with transformations, especially when sample size is small (for example less then 20)
• when the level of measurement is Cleary nominal or ordinal

Question 2

What are the assumptions that are made about the non-parametric data?

Answer 2

• homogeneity of variance
• normality
• measured using at least an interval level of measurement (i.e., interval or ratio, not nominal or ordinal)

Question 3

What can be done if those assumptions are not met?

Answer 3

• use non-parametric equivalents

Question 4

When to use Kendall’s non-parametric correlation?

Answer 4

This is reccommended for smaller data sets.

Question 5

What is the Mann-Whitney U test?

Answer 5

Comparing independent samples t-test result with non-parametric equivalent.

Question 6

What does a chi-square test?

Answer 6

It a test of relationships, but because of the variables are nominal/categorical the relationship might actually be a reflection of a difference in the pattern of responding.

Question 7

Chi-square significance test establishes whether there is a ____ _____ between two ___ _____.

Answer 7

Significant relationship; categorical variables.

Question 8

What is a standardised residual?

Answer 8

Difference between what you actually observed and what you would have expected under the null hypothesis, expressed as a standardised number.