Word Doc Week 6

Question 1

Parametric Tests

Answer 1

• Parametric tests assume a normal distribution of values, or a “bell-shaped curve
• assumes data from a population can be modeled by a probability distribution that has a fixed set of parameters

Question 2

Parametric Assumptions

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

Non-Parametric Tests AKA

Answer 3

• Rank Order tests
• Rank Sum Tests
• Assumption Free Tests

Question 4

How do Non-Parametric Tests work?

Answer 4

• By ranking data first
• Running analyses on the ranks next
• Does not analyse the actual scores themselves
• High scores have high ranks
• Low score have low ranks
• Removes problems with outliers and skew

Question 5

When do we use Non-Parametrics

Answer 5

• When assumptions are violated and can’t be fixed with transformations
• Particularly when sample size is small
• When the level of measurement is clearly nominal or ordinal

Question 6

Non-Parametric Correlation

Answer 6

1. Kendall’s Tau
2. Spearman’s rho

• Used when:
• one or both variables are clearly ordinal
• when assumptions underlying Pearsons r are not met
• can be used the relationship is not linear

Question 8

Non-Parametric Equivalent of Independent Samples t-test

Answer 8

• Mann-Whitney U test

Question 9

Wilcoxon Signed Rank Sum Test

Answer 9

Non-Parametric Equivalent of the Paired Samples t-test

Question 10

Chi-square

Answer 10

• Used to analysis the relationship between two categorical/nominal/grouping variables
• Comparable to Pearson’s r which measures strength, significance and direction of the relationship between two continuous/scale variables
• Chi-square is technically not a test of differences
• Actually a test of relationships
• Because variables are nominal/categorical the relationship might be a reflection of difference in pattern of responding