What a week in the news…Paris goes back to jail. It’s certainly a story that has people riled up, engaged and talking. Should Paris go back to jail? Is 45 days an unreasonable sentence? Should she have dressed up for her court appearance? How is she spending her time in jail? What is that medical condition that allegedly got her out of jail the first time around? Is she getting preferential treatment because she is a celebrity?
I thought I would take a moment to refocus the discussion and talk about what I believe is the most important aspect of the story and what people should be talking about. Drinking and driving in the U.S. is very much a problem and every year it results in senseless deaths. I came across some statistics on the Mothers Against Drunk Driving (MADD) website that emphasize the various dimensions and magnitude of the problem.
Many traffic fatalities involve alcohol. In 2005, over 30% of all traffic fatalities involved a driver who had a blood alcohol level of 0.08 or more.
The problem touches many. Nearly three out of ten people in the U.S. will be involved in some way in a drinking related accident and numerous others will suffer from the loss of a friend or family member.
Drinking and driving kills innocent children. Motor vehicle accidents are the primary cause of death for children (under the age of 14), and 21% of these deaths involve alchol.
The problem is getting worse. Between 2005 and 2006, alcohol related fatalities increased by 2.4%, despite the fact that traffic fatalities as a whole actually declined over this time period.
I was curious about the geographic dimension of the problem. A heat map of drinking related fatalities for 2005 reveals something very interesting. While the problem of drinking and driving is certainly geographically pervasive, reaching nearly every portion of the United States, southern California stands out as a real trouble spot. In fact, in Los Angeles County, the fatality rate for 2005 was more than 55 times the national average!
//
So, circling back to Hollywood. I believe the real question is: What can Paris Hilton and other high profile individuals do to help this national problem and to put an end of all of the senseless deaths that occur as a result of drinking and driving? As the national president of MADD, Glynn Birch, recently noted: “Celebrities like Hilton need to clean up their act and need to be spokespersons for the cause.” I agree wholeheartedly with this – It’s a simple solution that could have a big impact.
7 Responses to Drunk Driving: "That's Hot". Paris Hilton as a Role Model for MADD
Leave a Reply
New GeoCommons Maps- Ancient Near East bradconner
- CO BLM Oil & Gas Leases 01/30/2012 ConnorBailey
- BVTRIERG azmisy
- geotaps mariosadio
- U.S. Bank Exposure to Europe thefactfile
- 2011_Mulch_Order dcacner
If this really is a rate heat map then I am even more concerned about the Appalachian inferno. Is there something about the distribution of the data producing that cluster or is that real. If it is real that seems to be a pretty big cultural problem. I wonder if you first regressed drunk driving fatalities versus “poverty” and then looked at the residuals what pattern would remain.
Where did you get the data?
Either way I totally agree about drunk driving. I had a niece put into a coma and know at least 2 friends of the family killed by drunk drivers. I am all for breathalyzers in all cars. I know it is pretty invasive but I am not sure how you stop people who drink and drive without a license. Well I guess we can have that debate another time…
Hi Steven,
I am assuming that you zoomed out one notch so that you could see the whole United States at once. When you zoom out that far with county level data the heat map algorithm averages the data to a geographic mean. The farther out you zoom the wider the search radius of the heat mapping algorithm looks for geographic entities with high values (in this case drunk driving fatalities). As a result you get a hot spot aggregating to the middle of the high values (very similar to a kernel density analysis in ESRI’s spatial analyst). As you zoom in the the search radius decreases and you get more detail on the exact geographic units with high values providing more detail. The downside is when you use this approach extreme zooms out become a bit meaningless with small geographic data units. In which case you want to use a thematic map where the coloring is consistent and there is not search radius or averaging/blurring function. I am getting into the geographic weeds on this, but always happy to explain how the technology works.
Geowanking aside breathalyzers required to turn on the ignition of a car is a great idea.
Hi Steven,
I am assuming that you zoomed out one notch so that you could see the whole United States at once. When you zoom out that far with county level data the heat map algorithm averages the data to a geographic mean. The farther out you zoom the wider the search radius of the heat mapping algorithm looks for geographic entities with high values (in this case drunk driving fatalities). As a result you get a hot spot aggregating to the middle of the high values (very similar to a kernel density analysis in ESRI’s spatial analyst). As you zoom in the the search radius decreases and you get more detail on the exact geographic units with high values providing more detail. The downside is when you use this approach extreme zooms out become a bit meaningless with small geographic data units. In which case you want to use a thematic map where the coloring is consistent and there is not search radius or averaging/blurring function. I am getting into the geographic weeds on this, but always happy to explain how the technology works.
Geowanking aside breathalyzers required to turn on the ignition of a car is a great idea. Appreciate the feedback.
I saw yesterday that MADD is saying that Paris Hilton is one of 500,000 people who drive drunk, have their license suspensed, and drive anyway each year. They say they are trying to raise $500,000 by the end of June to combat these 500,000 unsafe drivers; I’ve pasted part of their email below:
“MADD’s Campaign to Eliminate Drunk Driving would help stop these 500,000 Paris Hiltons by 1) replacing long license suspensions with restrictions on driving that require offenders to use ignition interlock devices and 2) increasing penalties for driving while suspended to help deter people from committing this crime.
There are three things you can do to help stop these 500,000 offenders:
1. Donate to MADD online and help us meet their $500,000 goal at http://support.madd.org/combat500
2. Forward this email to on to your friends and family
3. Email your legislators to let them know you would like better ignition interlock laws at https://secure2.convio.net/madd/site/Advocacy?id=477”
It seems to make sense — why not try to solve the larger problem?
Hey Seagor:
Thanks for the explanation and I totally understand what you mean. What is the size of the geographic unit you are aggregating?
Are you really sure this is a rate and not just a total # of fatalities that you are aggregating. The reason I ask is because when I zoom in to regions of the us all the hotspots occur over major metropolitan areas. The strength of the hot spot seems to be highly correlated with population size. This would tend to happen with counts rather than rates.
Thanks
Steve
Hi Steve,
The data is at the county level and you can get all the details on it here – http://www.geocommons.com/data_set/show/2343 or go back to the original source here – FTP Source: ftp://ftp.nhtsa.dot.gov/fars/ . Although the work to aggregate the data and geo-reference it is a real pain in the butt. There is also a data set in GeoCommons that has the same data at the zip code centroid level with a lot more statistics – http://www.geocommons.com/data_set/show/232.
The data is the total # of fatalities, so yes there would be a high correlation with population. You could download a county population data set from GeoCommons and the drunk driving data set and create a normalized rate. Maybe one of the data folks here will give it a shot.
best,
sean
Drunk driving isn’t the problem. People who get into accidents or cause accidents while drunk driving are.
Penalties should be drastically increased for causing accidents while drunk. Meanwhile, penalties should be lowered for merely having a certain blood alcohol level. I know many people who drive much more carefully when they have had a few drinks. And I know at least a few people who can operate an automobile at a .08 bac quite successfully. Better than most 80 year olds can operate an automobile sober.
I’m not encouraging people to drive when they are drinking, but we should punish people for specific actions…not bad behavior. You get behind the wheel while intoxicated, then that is your responsibility…not the government’s. I would have no problem instituting the dealth penalty for drunk drivers that cause fatal accidents. Give people freedom, but also make them take responsibility when they screw up.
One thing to keep in mind: 70% of fatal accidents involve people who were sober. People make stupid decisions all the time while driving.