Thinking about Data Quality When Organizing Agencies

I find this fascinating: a view inside an agency’s thought processes about organizational measures that will (or could) affect how the public, politicians, or pundits perceive the usefulness and quality of agency-generated data. From the Bureau of Justice Statistics:

However, the methodological challenges involved in filling these major gaps preclude doing so under BJS’s current funding; it would require increased and sustained support in terms of staff and fiscal resources. Two strong organizational measures are suggested to reduce the likelihood that BJS and its officials are inappropriately treated in the future: 1) BJS’s current administrative position within the Office of Justice Programs (OJP) is detrimental to the agency’s function, it is recommended that BJS be moved out of OJP; and 2) that the position of BJS director be made a fixed-term presidential appointment with Senate confirmation.

Note the two parts of the recommendation:

  • Moving the agency.
  • Fixing the term and changing the appointment/removal process.

The former is pretty run-of-the-mill, but it says a lot about how the designers see OJP inside Main Justice.

The more interesting recommendation is to go to a fixed term. The question is always “how long?”, with the natural answer being “something longer than 4 years”.

It’s like an IRC (without party-balancing rules) for criminal justice data collection efforts. Another interesting example of attempts to politicize and then de-politicize agencies (note the date: 2009).