In follow up to yesterday‰??s overview of the hazmat headlines I provided, I would like to add some caveats to the process which have become increasingly important to me over the years. Specifically, there are many biases built into the system that have become increasingly important as the Internet has evolved over the last decade:
Biases related to the media source of the reports:
- The primary goal of the media is to draw attention, so if it bleeds, it leads. Many very interesting events get scant mention even though they are likely to be important learning opportunities.
- The role of the media is often to cooperate with emergency responders to manage evacuations and traffic patterns related to hazmat responses; sometimes that is the only information that is provided about the event.
- The media has a short attention span - there is little follow up to events without a news hook such as a related death or lawsuit. Work done to recover from an event or manage similar events in the future is seldom covered; I don‰??t think that a story is complete without that work being included.
- There is a chemical illiteracy in the media that often confuses the story. In the best case, a report will include the result of a google search of government resources with the worse case scenarios highlighted, whether they are relevant to the event or not. In the worst case, the chemical is incorrectly identified, furthering confusion in the virtual world.
Search engine biases
The search engines have allowed commercial considerations to increasingly impact the hits they deliver on a search. This has led to a noticeable impact on the results I see daily. Two ways this has played out are:
- Paywalls. About 20% of the reports I attempt to review are now blocked by the platforms. This was not a problem in 2011, and I expect it to get worse over the next few years, because
- The shrinkage of local media: local media sources are drying up quickly and the reliability of both coverage and accuracy is decreasing with time. This could be part of the explanation for the decreased number of reports in 2020-21, since Covid coverage usurped media space that hazmat issues took up.
There are also technical biases associated with the process:
- I suspect that there are many industrial services that use private hazmat responses that don‰??t share either events or Lessons Learned with the public unless evacuation or traffic issues arise.
- Most reports focus on identification of chemicals involved rather than a process description that would enable better hazard management. I understand this bias based on my work in investigating these events, but this limits the value of the media reports.
With these limitations in mind, I am thinking about reprioritizing the headlines I include in the summaries, and perhaps reducing their frequency to twice a week. Specific ideas are:
- Include safety culture news as ‰??education‰?? - I have been tracking ‰??safety culture‰?? in the news for two years and there is at least one story a day of interest on this topic. I could include these in the CHAS summaries under ‰??education‰??. Events in schools would move to the Public sector description.
- I propose to remove transportation events, as the information included in these tend to be sketchy and traffic-oriented. There are government database for these events that are not as specific as the information provided in many media reports, but which are probably more comprehensive in coverage.
- Include government actions in the headlines as opposed to as separate forwarded e-mails from the EPA and other government agencies.
To help me think about how to adjust these priorities, feel free to e-mail me directly, or better yet, take the CHAS Services 2022 survey at
Thanks for your interest in this work!
Ralph Stuart, CIH, CCHO
American Chemical Society Division of Chemical Health and Safety
For more information about the DCHAS-L e-mail list, contact the Divisional membership chair at membership**At_Symbol_Here**dchas.org
Follow us on Twitter **At_Symbol_Here**acsdchas
Previous post | Top of Page | Next post