To Link or Synthesize? An Approach to Data Quality Comparison

Linking administrative data to produce more informative for subsequent analysis has become an increasingly common practice. However, there might be concomitant risks of disclosing sensitive information about individuals. One practice that reduces these is synthesis. In synthesis the are used fit a model from which synthetic then generated. The released end users. There some scenarios where user have option using linked or accepting synthesized data. linkage and susceptible errors could limit their usefulness. Here, we investigate problem comparing quality demonstrate through simulations how approached. These comparisons important when considering can supplied with highest-quality in situations one must consider risk/utility tradeoffs.