SW388R6

1
Kruskal-Wallis Test
2
Problem 1

• Based on the dataset GSS2000.SAV, is the
  following statement true, false, or an incorrect
  application of a statistic? Use 0.05 as the level
  of significance. Base your answer on the output
  for a Kruskal-Wallis test.
• For the population represented by this sample,
  there are differences in average rank for number
  of hours worked in the past week among groups
  defined by the variable "RS HIGHEST DEGREE".
• 1. True
• 2. True with caution
• 3. False
• 4. Incorrect application of a statistic

3
Request the Kruskal-Wallis Test
To compute a Kruskal-Wallis Test in SPSS, select
the Nonparametric Tests 2 Independent Samples
command from the Analyze menu.
4
Select the variables for the Kruskal-Wallis Test
First, move the variable hrs1" to the list of
test variables.
Second, move the variable degree" to the
grouping variable text box.
Third, click on the Define Groups button to
identify the groups of the degree variable to
include in the analysis.
5
Enter the values for the groups
First, enter the value 0 for the first group to
include.
Third, click on the Continue button to close the
dialog box.
Second, enter the value 4 for the second group to
include.
6
Complete the request
Second, click on the OK button to complete the
request.
First, make sure the Kruskal-Wallis H checkbox is
marked, and the checkboxes for the other
statistical tests are clear.
7
SPSS Output for the Kruskal-Wallis Test
The Kruskal-Wallis Test requires that the
variable be ordinal or interval. The variable
"NUMBER OF HOURS WORKED LAST WEEK" is interval,
satisfying this requirement. The kruskal-wallis
test requires that the independent variable
contain two or more groups. The variable "RS
HIGHEST DEGREE" contains 5 categories.
8
SPSS Output for the Kruskal-Wallis Test
The research hypothesis states that the average
rank for number of hours worked in the past week
is different for one or more categories of the
variable "RS HIGHEST DEGREE". The null
hypothesis states that the average rank for
number of hours worked in the past week is equal
for all categories of the variable "RS HIGHEST
DEGREE".
9
SPSS Output for the Kruskal-Wallis Test
The probability of the Kruskal-Wallis statistic
is 0.034, which is less than or equal to the
level of significance of 0.05. We reject the
null hypothesis and conclude that the analysis
supports the research hypothesis that the average
rank for number of hours worked in the past week
is different for one or more categories of the
variable "RS HIGHEST DEGREE". The answer to the
question is true.
10
Problem 2

• Based on the dataset GSS2000.SAV, is the
  following statement true, false, or an incorrect
  application of a statistic? Use 0.01 as the level
  of significance. Base your answer on the output
  for a Kruskal-Wallis test.
• For the population represented by this sample,
  there are differences in average rank for
  frequency of prayer among groups defined by the
  variable "RS HIGHEST DEGREE".
• 1. True
• 2. True with caution
• 3. False
• 4. Incorrect application of a statistic

11
Solution 2
The Kruskal-Wallis Test requires that the
variable be ordinal or interval. The variable
"HOW OFTEN DOES R PRAY" is ordinal, satisfying
this requirement. The kruskal-wallis test
requires that the independent variable contain
two or more groups. The variable "RS HIGHEST
DEGREE" contains 5 categories. The research
hypothesis states that the average rank for
frequency of prayer is different for one or more
categories of the variable "RS HIGHEST DEGREE".
The null hypothesis states that the average rank
for frequency of prayer is equal for all
categories of the variable "RS HIGHEST DEGREE".
12
Solution 2
The probability of the Kruskal-Wallis statistic
is 0.722, which is greater than the level of
significance of 0.01. We fail to reject the
null hypothesis and conclude that the analysis
does not support the research hypothesis that the
average rank for frequency of prayer is different
for one or more categories of the variable "RS
HIGHEST DEGREE". The answer to the question is
false.
13
Solving Kruskal-Wallis Test Problems
The following is a guide to the decision process
for answering Kruskal-Wallis Test questions in
SPSS.

• Is the level of measurement satisfied?
• Dependent variable ordinal or interval level
• Independent variable any level - two or more
  groups

No
Incorrect application of a statistic
Yes
True