# IBM SPSS Modelar Experimental

This article explains how to connect IBM SPSS Modeler with Treasure Data using **Datatank v2**, Treasure Data's recommended method for exporting data to external tools, replacing the deprecated Presto ODBC Driver.

By combining Treasure Data with IBM SPSS Modeler, you can prepare and model your data using powerful statistical tools after retrieving it from Treasure Data’s platform.

## Step 1: Prepare IBM SPSS Modeler

Download and install IBM SPSS Modeler from IBM's official site. This article assumes you’re using **IBM SPSS Modeler**.

Once installed, you should see the standard SPSS Modeler interface.

## Step 2: Use Datatank 2.0 as a datamart

Using [Datatank 2.0](https://docs.treasuredata.com/smart/project-product-documentation/about-data-tank-2-0) or other datamarts, that provides PostgreSQL interface compatibility, SPSS's various functionalities and features can assist in efficiently visualizing this data and creating Dashboards.

While you can export data to Datatank 2.0 via Activation to PostgreSQL, you will need to integrate Datatank 2.0 with SPSS as well. You must first install PostgreSQL database drivers into your SPSS environment.

- [Datatank 2.0](https://docs.treasuredata.com/articles/pd/about-data-tank-2-0)


## Step 3: Access Datatank v2 from SPSS Modeler

To complete this step, please refer to IBM's official documentation for instructions on importing external data sources:

- [IBM SPSS Modeler: Amazon RDS PostgreSQL Connection](https://www.ibm.com/docs/en/watsonx/saas?topic=connectors-amazon-rds-postgresql-connection)


## Deprecation Note

2025-05: Treasure Data packaged Presto ODBC driver is deprecated and no longer maintained. We recommend all users to migrate to Datatank-based workflows for SPSS and other intermidate storages that SPSS supports.