Some brief observations on Nate Silver’s analysis of hurricane coverage

August 29, 2011

On his FiveThirtyEight blog, Nate Silver has a post entitled “How Irene Lived Up to the Hype“.   In it, he makes an attempt to determine which hurricanes of the past 20 years received the most media coverage.  He describes his methodology thusly:

The Web site is a searchable database of millions of news accounts — mostly newspaper and magazine articles, but also some sources like television transcripts. While it lacks representation of things like blogs and social media, it contains a highly comprehensive sample of what we might think of as the traditional media.

It’s easy enough to conduct a series of searches on in order to determine how much press coverage past Atlantic hurricanes have received. The only tricky part is that the further you go back in time, the fewer sources the database has available, so we’ll have to adjust for this.

We’ll accomplish this by creating a statistic which I’ll call the News Unit (or NU). This is defined by taking the total number of stories that mentioned the storm by name (for instance, “Hurricane Hugo” or “Tropical Storm Hugo”; either one is considered acceptable) and dividing by the average number of stories per day that were available in the database during that period. I then multiply the result by 10 just to make things a little bit more legible — so essentially, a News Unit consists of one-tenth of all the stories published on a given day.

For instance, there were 13,326 stories published that used the term “Hurricane Gustav” or “Tropical Storm Gustav” during the period when that storm was active, from Aug. 25 through Sept. 4, 2008. And on average, there were 56,200 stories published per day during that period in the database. Dividing 13,326 by 56,200, and then multiplying by 10, gives a result of 2.37. So Gustav produced 2.37 News Units worth of coverage while the storm was active.

The outcome is surprising. By his reckoning, Hurricane Ivan of 2004 had the greatest proportion of coverage, followed by 1992’s Andrew.  Katrina is number 14 on the list, with Wilma and Rita of that year coming in 12th and 4th, respectively.

Hurricane Ivan at landfall

Must have had a good press agent

–  As Katrina’s last advisory was written on the morning of Tuesday August 30, the bulk of Katrina coverage had yet to be published, as the storms ultimate impact had yet to be realized.  I haven’t studied any accounting for the deaths attributed to Katrina,  but I would think it reasonable to estimate that more than 50% of the people who comprised the storm’s death toll were still alive at that point. If Silver’s ranking of the storm is accurate, it shows that Katrina was phenomenally under-covered in the run-up to landfall.

– I’m curious to what the cut-off date was for Ivan coverage.  Recall that advisories on Ivan ended within 24 hours of landfall (September 16) . However, a circulation was clearly evident afterwards and that circulation wound its way back to the Gulf of Mexico (track), causing the National Hurricane Center resumed advisories six days later. Ivan ultimately dissipated on September 24.  If Silver used the latter date, then Ivan’s numbers are certainly inflated.  as “Ivan” would of been mentioned in numerous stories about Hurricane Jeanne, which was bearing down on Florida during the tail end of Ivan’s “second” life.

– Similarly, Andrew’s numbers are helped by the storm continuing from south Florida to a second landfall in Louisiana, two days later (and ultimate disspation two days after that), thereby gaining more post-storm coverage.  Given this boost, it is probably fair to say that Andrew was also under-covered pre-landfall, given its ultimate impact.

– A commenter expressed skepticism that Rita had more coverage than Katrina. That, given the methodology that restricts the most memorable parts of Katrina coverage, is actually unsurprising to me.  Rita was the first great storm after Katrina, and accordingly got more coverage than it would have pre-Katrina.  The massive evacuation was a big story pre-landfall, with the deaths in the bus explosion being a heavily covered story.


One Response to “Some brief observations on Nate Silver’s analysis of hurricane coverage”

  1. Dominic Caraccilo Says:

    The quality of life and security for the citizens has been largely restored and we are a large part of why that has happened.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: