Statistics for the Behavioral Sciences by Frederick J. Gravetter, Larry B. Wallnau ISBN 10: 1305504917 ISBN 13: 9781305504912

FOCUS ON PROBLEM SOLVING 327
Group Statistics
VAR00002
N
Mean
Std. Deviation
Std. Error
Mean
VAR00001
1.00
8
8.0000
2.92770
1.03510
2.00
8
12.0000
3.07060
1.08562
Independent Samples Test
Levene’s Test for Equality of
Variances
F
Sig.
t-test for Equality of Means
t
df
VAR00001
Equal variances assumed
.000
1.000
22.667
14
Equal variances not
assumed
22.667
13.968
Independent Samples Test
t-test for Equality of Means
Sig. (2-tailed)
Mean
Difference
Std. Error
Difference
Lower
95%
Confidence
Interval of the
Difference
Upper
VAR00001
Equal variances assumed
.018
24.00000
1.50000
27.21718
2.78282
Equal variances not
assumed
.018
24.00000
1.50000
27.21786
2.78214
FIGURE 10.8
The SPSS output for the independent-measures hypothesis test in Example 10.2.
FOCUS ON PROBLEM SOLVING
1. As you learn more about different statistical methods, one basic problem will be deciding
which method is appropriate for a particular set of data. Fortunately, it is easy to identify
situations in which the independent-measures t statistic is used. First, the data will always
consist of two separate samples (two ns, two Ms, two SSs, and so on). Second, this
t statistic is always used to answer questions about a mean difference: On the average, is
one group different (better, faster, smarter) than the other group? If you examine the data
and identify the type of question that a researcher is asking, you should be able to decide
whether an independent-measures t is appropriate.
2. When computing an independent-measures t statistic from sample data, we suggest that
you routinely divide the formula into separate stages rather than trying to do all the calculations
at once. First, find the pooled variance. Second, compute the standard error. Third,
compute the t statistic.
3. One of the most common errors for students involves confusing the formulas for pooled variance
and standard error. When computing pooled variance, you are “pooling” the two samples
together into a single variance. This variance is computed as a single fraction, with two SS
values in the numerator and two df values in the denominator. When computing the standard
error, you are adding the error from the first sample and the error from the second sample.
These two separate errors are added as two separate fractions under the square root symbol.