Same Data, Different Results – MapViewer Lab 5

February 27, 2010

I chose to map deaths due to alcohol impaired vehicle accidents for 2008.  The data was divided by state across the US.  This information is extremely useful for the State Governments.  A state’s category as a high fatality state is brought before the local government and  the issue can be dealt with increased patrols and community awareness.   Also, knowing the changes in alcohol fatality numbers from year to year demonstrates the effectiveness of preventative measures or the need to improve them.

I found the data on a website for Alcohol Alert! , a product that can be installed in bars or restaurants.  For a price it measures one’s blood alcohol content and thus determines the individual’s ability (or inability) to legally drive.  Although an interesting product, this does not prevent an intoxicated driver from getting behind the wheel and seems ineffective.  As far as the accuracy of the site’s data, I doubt it is 100% accurate, as exaggerating the numbers would seemingly make the product a necessity.

The following map is made with equal number intervals and is in blue and red for easy contrast.  There are 5 classes, as more than 5 shades of color become difficult to differentiate among.

For the second map I kept the colors and numbers of classes the same for easy comparison to the first map.  However, I changed the way in which the data was sorted into class: this map is divided by a method called Jenkins’ Natural Breaks. 

The natural breaks method divides the data based on ‘clumps’ of information.  It provides a much more accurate selection of the high fatality states.  If you look at the classes, there is a large jump from the purple – red to the red.  While each class includes increases by 300 or 400 fatalities, the last red group jumps to thousands.  The natural breaks method separates the extreme states. 

The equal number method, on the other hand, splits the classes so that each contains an equal number of states.  This accurately disperses the low fatality states but crowds medium to high fatality states into one group.  Notice that the red group includes the entire uppermost half of the statistics.  Not only does this show an inaccurate view of the states in red, it also offers no clue as to where the states in red fall, as red includes such a large range.

For instance, observe Illinois.  In the first map the state is red, among the highest in alcohol fatality rates.  But when the red class from the first map is broken down, Illinois actually falls in as having an average rate as shown in the second map.  These dramatic differences are due to the way in which the data is classified.


One Response to “Same Data, Different Results – MapViewer Lab 5”

  1. ethingt1 said

    I would be interested to see this map as the number of fatalities per capita versus the number of overall fatalities. Obviously, states like California and Texas are going to have a high number of fatalities due to the large population. I think that knowing the number of fatalities per capita might be a better estimate to the alcohol problem within states. California and Texas may still have a high number of fatalities per capita, but it would also be interesting to see how states where drinking is common, such as Wisconsin, would fare in the number of fatalities per capita. Also, smaller states, such as Montana and Wyoming, have a small number of total fatalities due their small populations, but could potentially have drinking problems that could go unnoticed with this data.

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: