|
Help (wait for the button to appear above) |
|
||||
The Two sample item on the Z Statistics submenu on the Stat menu provides confidence intervals and/or hypothesis tests for the difference in two means from independent samples (first selected minus second selected). The results are displayed in the Results panel.
In order to use the Z statistics procedures, the population standard deviation must be known for each population. If it is not, it can be estimated from the data, and the T statistics option may be appropriate.
In the dialog box, select the variables representing the two samples, then supply (in the order selected) the standard deviation values for each population The two sample Z procedure is illustrated in an example using the reading data. In this example, the control group is chosen as the first variable, and the treatment group as the second, and the standard deviations are assumed to be 12 for each population.
Click the Next -> button to select either a confidence interval (specifying confidence level) or a hypothesis test (specifying null mean difference and alternative hypothesis). In the reading example, the comparison of interest is the difference between the two groups, and the analysis is done using a 90\% confidence interval.
Click on the Calculate button and the output is displayed in the Results panel. In both the confidence interval and the hypothesis testing case, the output contains a table of the user-supplied values for the two population standard deviations. In the confidence interval case, as in this example, the output also includes of the estimated difference of means, standard error, number of degrees of freedom, and the upper and lower confidence limits. In the hypothesis testing case, the output includes the null value for the difference ("Delta0"), the estimate, the standard error of the estimate, the Z test statistic, and the p value.
