• Single swaps, using a graphical user interface to perform data swapping on categorical databases of essentially unlimited size, with user-specified swap rates, swap attributes and (optionally) equality or inequality constraints on unswapped attributes. The swapping is performed on the user's local machine, allowing the DSTK to be used on confidential databases.
  
  
• Batch swaps: in-depth investigations of multiple swap rates and multiple choices of swap attributes–all one-attribute swaps, all two-attribute swaps, or both.
  
  
• Risk-utility calculations: for both single and batch swaps, the DSTK calculates disclosure risk and data utility, the latter as absence of data distortion.
  
  
• Record weights, which are employed in data distortion calculations. This allows DSTK users to quantify distortion caused by data swapping in population-level estimates derived from survey data.
  
  
• Visualization and manipulation of results of batch swaps, including visualization of (distortion, risk) frontiers.

The DSTK provides a Java class library for performing customized data swapping tasks.

The DSTK was produced by the Digital Government Research Program at the NISS, with support from the National Science Foundation and the National Center for Education Statistics. It was written by Ashish Sanil, William "Jimmy" Fulp, Shanti Gomatam, Chunhua "Charlie" Liu and Alan Karr.