Preparing the Data for Independent Samples Analysis

In order to ensure consistent dataset names to refer to during the following instructions, give the dataset a name.

1. From the menus choose:

   File > Rename Dataset...

2. Type originalData as the new name of the dataset.
3. Click OK.

   You will now be able to more easily refer back to this dataset later.

4. To aggregate the data, from the menus choose:

   Data > Aggregate...

   

5. Select Customer ID [custid] as a break variable.
6. Select Primary credit card [card], Number of items [items], and Amount spent [spent] as variables to aggregate.
7. Select card_mean in the Summaries of Variables list and click Function.
8. In the Aggregate Functions dialog, in the Summary Statistics group, select First.
9. Click Continue.
10. In the Aggregate Data dialog in the Summaries of Variables list, select items_mean and spent_mean and click Function.
11. Select First in the Summary Statistics group as the function.

    The dictionary information for Primary credit card [card] will be passed to the new variable card_first. Note that there is no difference between choosing First, Last, Minimum, or Maximum because the values for Primary card [card] are the same for each customer.

12. Click Continue.
13. In the Aggregate Data dialog, select items_mean and spent_mean in the Summaries of Variables list and click Function.
14. Select Sum in the Summary Statistics group as the function.
15. Click Continue.
16. In the Save group, select Create new dataset containing only the aggregated variables and type independentSamples as the new dataset name.
17. Click OK.

    The aggregated dataset is almost ready for analysis.

18. In the Variable View In the Data Editor window for the new independentSamples dataset, set the role of items_sum and spent_sum to Target.

These selections will allow you to run the independent samples and one-sample nonparametric tests on "automatic."

Next