Randomly chosen MBA students were asked their opinions about the ideal number of children for a married couple. The sample data were entered into MegaStat, and the following results were produced. Hypothesis Test: Independent Groups (t-test, unequal variance)

$\begin{array}{rc}
\text { Men } & \text { Women } \
2.812 & 2.1538 \text { mean } \
1.2505 & 0.5547 \text { std. dev. } \
11 & 13 \text { sample size } \
& 13 \text { pooled df }
\end{array}$
$ 0.6582 $ $ \quad $$ \quad $difference (Men - Women)
$ 0.40722 $ $ \quad $$ \quad $standard error of difference
0$ \quad $$ \quad $$ \quad $$ \quad $$ \quad $ hypothesized difference
$ 1.62 $$ \quad $$ \quad $$ \quad $$ \quad $t
$ 0.065 $$ \quad $$ \quad $$ \quad $$ \quad $P-value (one-tailed, upper)

F-test for equality of variance
1.56375 variance: Men
$ 0.307692 $ $ \quad $variance: Women
$ 5.08$$ \quad $F
$ 0.01 $$ \quad $P-value To compare the means, would it be appropriate to use a test that assumes equal variances?

Question

Randomly chosen MBA students were asked their opinions about the ideal number of children for a married couple. The sample data were entered into MegaStat, and the following results were produced. Hypothesis Test: Independent Groups (t-test, unequal variance)

$\begin{array}{rc}
\text { Men } & \text { Women } \
2.812 & 2.1538 \text { mean } \
1.2505 & 0.5547 \text { std. dev. } \
11 & 13 \text { sample size } \
& 13 \text { pooled df }
\end{array}$
$ 0.6582 $ $ \quad $$ \quad $difference (Men - Women)
$ 0.40722 $ $ \quad $$ \quad $standard error of difference
0$ \quad $$ \quad $$ \quad $$ \quad $$ \quad $ hypothesized difference
$ 1.62 $$ \quad $$ \quad $$ \quad $$ \quad $t
$ 0.065 $$ \quad $$ \quad $$ \quad $$ \quad $P-value (one-tailed, upper)

F-test for equality of variance
1.56375 variance: Men
$ 0.307692 $ $ \quad $variance: Women
$ 5.08$$ \quad $F
$ 0.01 $$ \quad $P-value To compare the means, would it be appropriate to use a test that assumes equal variances?

Quizplus · Accepted Answer

The F-test for equality of variance resulted in a p-value of 0.01, which is less than the significance level of 0.05. Therefore, we reject the null hypothesis of equal variances and conclude that the variances are unequal. When the variances are unequal, it is more appropriate to use a test that assumes unequal variances, such as the t-test with unequal variance, which was used in this case. Choosing a test assuming equal variances would not be appropriate in this case.

Randomly Chosen MBA Students Were Asked Their Opinions About the Ideal