Online Program Home
  My Program

Abstract Details

Activity Number: 375 - Negative Results: They're Essential!
Type: Invited
Date/Time: Tuesday, August 1, 2017 : 2:00 PM to 3:50 PM
Sponsor: Committee on Professional Ethics
Abstract #322212 View Presentation
Title: Negative Findings in Environmental Epidemiology
Author(s): Sidney Stanley Young*
Companies: CGStat
Keywords: environmental epidemiology ; publication bias ; negative results

Hundreds of environmental epidemiology papers show associations between air quality and ozone, and human mortality. Many consider these associations to prove causation, but increasingly, that assumption of causation is being questioned. In 2015, a journal rejected a submitted paper reporting no association without review- saying "The issue...was laid to rest in the mid 1990s... The EPA and other regulatory bodies have long since concluded these associations are causal." The editor cited a 1993 paper that had ~10k observations (unavailable for reanalysis); the submitted paper had ~2M observations (available for reanalysis). The journal advertises that it will publish any science so long as the data and methods are sound, but the rejection without review was upheld by the section editor. The negative result of this rigorous analysis would be an important contribution to environmental epidemiology- if not for publication bias. The ASA Ethical Guidelines for Statistical Practice assert that statisticians need to be aware of those of other disciplines, and that reviews should be based on the science, without bias. These Guidelines will be discussed from the author's perspective.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

Copyright © American Statistical Association