In effect, the researchers took one number – 15 deaths identified from a survey of 3,299 households – and extrapolated that to come up with 4,645 deaths across the island. That number came with a very large caveat, clearly identified in the report, but few news media accounts bothered to explain the nuances.The problem is that real journalism requires bothering to explain nuances.
Sunday, June 03, 2018
In case you wondered …
… Did exactly 4,645 people die in Hurricane Maria? Nope. - The Washington Post.
Subscribe to:
Post Comments (Atom)
Isn't this report actually appearing in the WSJ? Why is that not real journalism?
ReplyDeleteIt says the report appeared in the New England Journal of Medicine. I see no mention of the WSJ. But even if it had appeared there, it isn't the brand that counts. It's the product. Good journalists go out of their way to make things clear. They don't run with a sensational number and leave out the cautionary details.
ReplyDeleteI stand corrected. I meant the Washington Post. But see both Jeff's and Rus's comments.
DeleteFor what it's worth, all the hubbub appears to trace back to a May 29 story in the Washington Post that was clear about the methodology of the study. The problem here appears to be other people taking the estimate and running with it on social media for their own purposes, using the term "Harvard study" for credibility that increases the possibility that almost nobody will bother to read the full Washington Post article or doubt the methodology if it serves the political needs of the moment.
ReplyDeleteThat's not a defense of the obviously farcical official estimate of 64 dead, just a comment on how quick everyone is now to use whatever factoid makes them feel superior to their political antagonists.
Doing an albeit quick search for the news on the study, I found only 1 that asserted the 4,645 number without explaining that it was the middle number in a range that a small sample would indicate. Here is that report: Cable News Largely Ignores Deaths of 4,645 in Puerto Rico to Focus on Roseanne. While calling the main stream media out on what it was giving headlines and attention to, Roseanne instead of probably thousand(s) of death, the news source seems not to have realized that it was not doing its due diligence.
ReplyDeleteWhat some did, was similar to this Reuter's report: Study hikes Puerto Rico's Hurricane Maria death toll to 4,645. If you never read beyond the 1st couple of paragraphs, you would be left with the idea that 4,645 is an assertion, not the mid number between 793 and 8,498, which is the range the study firmly indicates ~~ which agrees with those other estimates of the toll being around 1,000. Deviations from the mean that are possible but highly improbable are death counts below 793 or over 8,498. Don't be surprised, however, if we find the numbers to be 800 or 8,000, if we can ever find the exact count. We'll need to find a way to replicate this study with a higher sample, however, if we are to close in on a tighter range.
The headlines also touted the 4,645 number. It's not a terrible number, but a far reach from both 800 or 8,000.
Since the political talk about "fake news" from the administration, there has been the phenomenon of Fox News touting how everyone else has fake news but them. Here, we see that the Washington Post reports that other news sources have been getting the report wrong, but not saying which ones they are talking about. Did they mean only the Democracy Now 1-paragraph report? Or did they not name names so that we would suspect everyone but them. And yet, here is their own misleading headline from 4 days earlier: Report: Hurricane Maria left 4,645 dead in Puerto Rico
The current estimate — from the Puerto Rican government — is 1,431. The point remains that if you are going to cite something, cite it accurately and precisely. It's called reporting.
ReplyDeleteHre's some interesting wording from a Huffington Post article from just this morning:
ReplyDelete[San Juan Mayor Carmen Yulín Cruz's] latest attack on Trump was in response to a new Harvard study published last week in the New England Journal of Medicine that that death toll from the hurricane is now conservatively estimated to be 4,645. The official death toll remains at 64."
Yes, it's so true, the big news to report is that the official death toll of 64 is grossly ridiculous, outrageously inaccurate and should never have been spoken or considered.
That noted, if Mary Papenfuss had simply said "estimated" instead of "conservatively estimated", her wording would have been okay, and if she could then afford to squeeze a couple more words in, to note the 1,431 count of the Puerto Rican government, say, more elucidating.
This thread has me interested in the reporting. Here's another, which notes that the research authors hike their estimate to be 5,740 from 4,645, thus 4,645 would indeed be conservative. The issue is with what is called "survivor bias", that when there was only one in the household, there was no one left behind to report his or her death in the sampling: Hurricane Maria Death Count over 5,000–not 64, New Study Finds.
ReplyDeleteWhat I love about "Left Voice" and other leftist zines, is the political relativity that left means left, not something like Obama and other Democrats espouse, which are on the right, way right, indeed, as opposed to way way right, or even way way way right :)