PARAT maintains its own workspace. It can only analyze and de-identify data in this workspace.
In version 1.12 of PARAT, there are two ways to import data into the PARAT workspace:
- Any csv file can be imported into PARAT
- Data in an Access database can be imported into PARAT via an ODBC connection
If you have Excel data files then you can export these as csv files quite easily in the "Save As" menu option. Most other tools that manipulate data (e.g., statistical analysis tools) have the option to export files as csv. To import a csv file, please follow these steps:
1) Open PARAT
2) From “Data” menu select “Load Data”
3) Click the folder icon to browse for a file
4) Select an “.csv” file to import from
5) If the csv file has a header row, select the "skip" checkbox
6) Create a name for the table
7) Modify the variable names as necessary
8) Set the variable types as necessary
9) Click “Upload”
and for an Access table:
1) Open PARAT
2) From “Data” menu select “Load Data”
3) Click the folder icon to browse for a file
4) Select an “.mdb” file to import from
5) Select the Access table from the drop-down menu that you want to import from
6) Create a name for the table
7) Click “Upload”
Data in other databases can be exported to the instance of SQL Server that PARAT is using to maintain its workspace. For example, if you have data in another SQL Server or Oracle database, you can export that table to the SQL Server instance that PARAT is using and it will appear in list of data sets within PARAT.
The PARAT tool can only analyze single tables. Therefore, if you have a relational data set, you will need to "flatten" it to a single table before PARAT can analyze it.
As of v1.13 of PARAT, extensive effort has been spent optimizing the data import functions to be able to handle very large data sets. Therefore, now a few million record data set would be readily importable by PARAT in a couple of minutes or less. Also, the import functions have been improved to handle Unicode csv files to accommodate languages other than English.
When importing mega data sets, you should be cognizant of the database size limits that are imposed by the version of SQL Server that you are using. For example, SQL Server Express 2005 has a 4GB database size limit (see the feature comparison here: http://www.microsoft.com/Sqlserver/2005/en/us/compare-features.aspx).
The author(s) retain all copyright to this knowledgebase article. Please include a citation to the web page if you reuse this material. More information is available at our lab web site: http://www.ehealthinformation.ca/.
Categories 
