1 / 12

Dueling Estimates: Closing Thoughts

Dueling Estimates: Closing Thoughts. Jennifer H. Madans. National Center for Health Statistics. Assumptions. Conflicting estimates are a problem for the Federal Statistical System They cause confusion among policy makers, the press and the public

hubert
Download Presentation

Dueling Estimates: Closing Thoughts

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Dueling Estimates:Closing Thoughts Jennifer H. Madans National Center for Health Statistics

  2. Assumptions • Conflicting estimates are a problem for the Federal Statistical System • They cause confusion among policy makers, the press and the public • They reduce confidence in the data produced by the system and each agency • The producers may understand and accept the causes but the users do not

  3. Assumptions • Producers and users would prefer if estimates of what appear to be the same phenomena were similar if not the same • Much attention has been directed toward the problem but results have been unsatisfying • We are still talking about it at this seminar

  4. The Challenge • Multiple estimates is a natural result of a decentralized federal statistical system • Focus on maximizing the utility of information for Departmental policy needs • Focus on maximizing quality of each data collection activity • Internal needs trump cross system needs

  5. Session Objective • Revisit possible solutions from a data collection life cycle point of view • Acknowledge the benefits and costs of different approaches • Start a discussion about what would be needed to have a more structured, proactive approach to reduce the impact of competing estimates and the associated costs

  6. Current Situation • Previous speakers outlined several different approaches • Front end – joint planning and testing with coordinated implementation • Back end -- Explanation of what is being measured; potential source of difference; joint analysis; joint publication

  7. Front End Solutions • Costly but hold promise • Increased cost of data collection • Reduced flexibility in collection cycle • Potential sacrifice of survey specific aims to achieve the system’s need for comparability

  8. Back End Solutions • Solutions of lowest data collection and staff cost but are probably least satisfying • Will always be needed even if other approaches are used

  9. System Solution • Statistical agencies produce multiple estimates of the same thing that generate little attention or confusion • Only a limited number are dueling • For these estimates, agree to designate which are ‘official’ • Official estimates can be a set of related estimates each with specific uses

  10. System Solution • Requires more of a change in how agencies relate to each • Limits statistical agency ability to communicate with users, the press and the public • Could have unintended impact of budgets

  11. Which Way Forward • System is too complex to embrace one solution • How do we determine what solution works best for different types of estimates? • Not all ‘dueling estimates’ are equally problematic • How do we identify the ones that need most attention?

  12. Which Way Forward • When are the costs of minimizing the existence of dueling estimates too high? • Are there other benefits of strengthening ‘System’ objectives in areas of data collection and dissemination of findings?

More Related