Nonparametric

1
Nonparametric

• Three or more groups

2
Kruskal-Wallis Analysis of Variance of Ranks Test

• Kruskal-Wallis Analysis of Variance of Ranks Test

3
Example

• A police psychologist wants to determine whether
  relaxation or guided imagery enhance the
  rank-orderings of the crime-solving abilities of
  police officers. He/she compares them to a
  control group.
• 6 in the control group, 6 in relaxation, 6 in
  imagery

4
Example -- continued

• Number of Samples
• Nature of Samples
• s Known
• Independent Variable

5
Example -- continued

• Dependent Variable and its Level of Measurement
• Target Population
• Inferential Statistical Technique
• Kruskal-Wallis Analysis of Variance of Ranks Test

6
Example -- continued

• H0
• Type of training will have no effect on
  rank-ordered crime-solving ability
• H1
• Type of training will have an effect on
  rank-ordered crime-solving ability
• Decision Rule
• If the p-value of the obtained test statistic is
  less than .05, reject the null hypothesis

7
Example -- continued

• Obtained Test Statistic
• ?2 (2) 5.96, p .16
• Decision

8
Results Section

• The results of the Kruskal-Wallis Analysis of
  Variance of Ranks Test involving type of training
  as the independent variable and rank-ordered
  crime-solving ability as the dependent variable
  were not statistically significant, ?2 (2) 5.96.

9
Discussion Section

• It appears that the type of training police
  officers receive has no effect on rank-ordered
  crime-solving ability

10
Kruskal-Wallis Test and SPSS for Windows

• Statistics, Nonparametric Tests, k Independent
  Samples
• Move DV to Test Variable list
• Move IV to Grouping Variable
• Define Groups
• Make sure K-W is checked

11
Interpreting the Printout

• Mean ranks
• Chi-Square (obtained test statistic)
• Significance (p-value)

12
Sample Printout