Creating a relational MR database from an IBM SPSS Statistics.sav file

You can create a relational MR database from an IBM SPSS Statistics.sav file. The steps involved parallel the steps involved in creating a relational MR database from a Quanvert database.

1 Prepare the IBM SPSS Statistics.sav file. You might want to set up value labels for your categorical variables in IBM SPSS Statistics so that they are interpreted as categorical variables by the UNICOM Intelligence Data Model.

2 Create the metadata (.mdd) file: see Creating an MDM document from an SPSS Statistics.sav file.

3 3: Creating a new SQL Server database.

4 Exporting the .sav file data to the new database.

5 5: Changing the default data source.

Exporting the .sav file data to the new database

For information on working with the SQL Server Import and Export Wizard, see:

1 From the Windows Start menu, choose: Programs > Microsoft SQL Server > Import and Export Data

This opens the first page of the SQL Server Import/Export wizard.

3 From the Data Source list, select UNICOM Intelligence DM-2 OLE DB Provider, and then click Properties.

This opens the Connection tab in the Data Link Properties dialog.

4 From the Metadata Type list, select Intelligence Metadata Document.

5 Enter the Metadata Location, either by typing the name and location of the .mdd file, or by clicking Browse. This should be the .mdd file that you created from the .sav file.

6 From the Case Data Type list, select IBM SPSS Statistics File (SAV) (read-write).

7 Enter the Case Data Location of the .sav file. Either type the name and location of the file into the box or click Browse.

8 Click Test Connection.

▪Test connection succeeded means that you have entered all of the details correctly.

▪If Test connection failed is displayed, check that you have entered all of the details correctly.

9 Click OK to return to the Choose a Data Source page of the SQL Server Import/Export wizard, and then click Next.

This opens the Choose a Destination page of the SQL Server Import/Export wizard.

10 From the Destination list, select UNICOM Intelligence DM-2 OLE DB Provider, and then click Properties.

This opens the Connection tab in the Data Link Properties dialog.

11 From the Metadata Type list, select Intelligence Metadata Document.

12 Enter the Metadata Location, either by typing the name and location of the .mdd file, or by clicking Browse. This should also be the .mdd file that you created from the .sav file.

13 Select Open metadata read/write.

14 From the Case Data Type list, select UNICOM Intelligence Database (MS SQL Server) (read-write).

15 Enter the Case Data Location. This should be in the form of an OLE DB connection string, for example:

Provider=MSOLEDBSQL.1;Integrated Security=SSPI;Persist Security Info=False;Initial Catalog=MyDatabase;Data Source=LocalHost

16 If necessary enter your Login ID and Password and click OK.

17 Enter the project name in the Case Data Project box. Typically, this is the same as the database name.

18 Click Test Connection.

▪Test connection succeeded means that you have entered all of the details correctly and you have successfully connected to the database.

▪If Test connection failed is displayed, check that you have entered all of the details correctly and, if relevant, check your network connection to the server where the database is.

19 Click OK to return to the Choose a Destination page of the SQL Server Import/Export wizard.

This opens the Specify Table Copy or Query page of the SQL Server Import/Export wizard.

21 Select Copy table(s) and view(s) from the source database, and then click Next.

22 Select the VDATA table for both the Source and Destination, and then click Next.

23 Deselect Run immediately.

24 Select Save SSIS Package.

25 Select SQL Server, if you want to store the SSIS package onto a SQL Server installation, or Structured Storage File, if you want to store the SSIS package in a file.

This opens the Save SSIS Package page. The details on this page depend on whether you chose SQL Server or Structured Storage File on the previous page.

27 Enter a Name, Description, and the other details that are required for the SSIS package.