I am by trade a GIS Specialist and I am responsible for the maintenance of a great deal of spatial data. The data I deal with on a daily basis can be related to health, conservation, and real estate to name a few. I have recently dealt with statistics for hospital asthma admittances and saw an opportune time to do my first Instructable. I am going to do a general summarization of my process to get hospital asthma admittances counts and rates into a format to graph.
The data I started out with was a polygon feature class for all of the US Census tracts within my county. These polygons contained a table that contained the Tract number, total population, and the count of asthma admittances per Census tract for 2010, 2011, & 2012. I exported the table as a csv so I could import it into Excel or even Plotly. In this instance I decided to use Excel.
After I imported the data into Excel I had the three columns of data as previously stated. This is all fine and dandy but if I was to graph this data it would look like certain tracts had a higher instance than others. In order to correct this so that it will not be as misleading I had to do what is called normalizing data. To normalize the data in Excel I used the following formula.
=round((sum(Asthma Count / Total Population) * 1000),0)
I decided to round the count to the nearest whole number and for the rate per capita to be per 1000 people.
If you look at the graph created before normalizing titled “Asthma Admittances by Census Tract” you will see that for the year 2012 Census Tract 30.01 has the highest count of asthma admittances with over 140 admittances and Tract 12 had only around 25 admittances. Now, if you look at the graph titled “Asthma Admittances Per 1000 People” you can clearly see that Census Tract 12 is the true problem area with around 24 occurrences for every 1000 people and Tract 30.01 is only around 16 admittances per 1000 people.
This is only a very generalized method of how I came up with the data to use in Plotly. I believe I has able to create a very good graph that is able to get it's point across without detracting from the main focus that the data is supposed to convey. If I was to present this to someone I would also create multiple maps the the viewer could get a better idea of the location of the Census Tracts. These graphs would be used to aid in the process for identifying areas that need some kind of aid, assistance, or even lead to a more focused study to find the source of the problem. I also attached a simple map for the 2012 Asthma Data to visualize the locations and the severity.