Other Computer Solutions for Chapter 14 SPSS. Two

advertisement
Other Computer Solutions for Chapter 14
SPSS.
Two-independent-samples t-test. Follow these steps:
Analyze
Compare Means
Independent Samples T-Test…
Move dv to Test Variable(s): box
Move iv to Grouping Variable: box
Click Define groups button
Supply integer values for groups to be compared
Continue
OK
Mann_Whitney U-test. Follow these steps:
Analyze
Nonparametric Tests
2 Independent Samples…
Move dv to Test Variable List: box
Move iv to Grouping Variable: box
Click Define Groups… button
Supply integer values for groups to be compared
Continue
OK
Dependent-samples t-test. Follow these steps:
Analyze
Compare Means
Paired Samples T-Test…
Select first variable in the pair
Select second variable in the pair
Click on the “arrow” button in the middle of the box to move to Paired
Variables: box
OK
Wilcoxon Matched-Pairs Signed-Ranks test. Follow these steps:
Analyze
Nonparametric Tests
2 Related Samples…
Click on first variable in the pair
Click on second variable in the pair
Click on the “arrow” button in the middle of the box to move to Paired
Variables: box
OK
SAS. Many of the analyses illustrated in this chapter can be accomplished using Procedures in
SAS. Proc Ttest provides the independent samples t-test, Proc Npar1way provides the Wilcoxon Rank
Sum Test and the Median Test, and Proc Univariate provides the dependent samples t-test. The
application of these tests within SAS is illustrated using examples from earlier in the chapter. More
specifically, the data example used for the independent samples t-test (see Equation 14.12) and later for
the Mann–Whitney U-Test is used to illustrate the use of Proc Ttest and Proc Npar1way, while the
dataset from the dependent samples t-test example (see Equation 14.22) is used to illustrate the use of
Proc Univariate for conducting a dependent samples t-test.
data twogroup;
input SES score;
datalines;
0 13
0 13
0 21
0 16
0 22
0 15
0 18
0 20
0 22
0 14
1 40
1 28
1 18
1 32
1 17
1 27
1 19
1 21
1 16
1 30
;
proc ttest;
var score;
class ses;
proc npar1way;
var score;
class ses;
run;
data dependent;
input reading math;
difference=reading-math;
datalines;
58 43
47 49
38 40
35 45
43 38
47 49
34 38
56 64
53 52
44 49
;
proc univariate;
var difference;
run;
Download