Voter density question

4891
10
Jump to solution
08-13-2014 07:04 AM
ELTerrell
New Contributor III

So I'm trying to map voter density by party.  I have uploaded the xy coordinates of addresses of everyone voting in an election including their party.  I'd like to make a density map by political party.  There are 1,000 voters or rows/addresses.  However, when I create a density map using the input as the voter addresses, it comes up with a scale of up to 88,000.

I'm not sure why it isn't just bringing in the rows, maybe thats the total number of excel squares?  Any ideas?  Is it something in environments that I need to change?  Thanks in advance for your help.

0 Kudos
1 Solution

Accepted Solutions
IanMurray
Frequent Contributor

You should be able to use either and get the same results, point density can only use point data, and can search by a circle(radius), torus, rectangle or wedge, whereas kernal density can take point or polyline inputs and can only search by radius for density.  In your case since you are using points and only need to search by a radius either one will work.

What Dan was trying to explain is that the output of this tool gives its output in units per area unit, so in your case voters per whatever are unit you selected on the tool.  So if you used square map units as your area unit scale factor and your map unit is decimal degree, then it calculates the density of voters per square decimal degree( or a 1 degree latitude by 1 degree longitude square box.)  So even though you only have 1000 voters, if your area is significantly smaller than 1 degree squared(approximately 4761 square miles), then your density per degree squared is much larger, since it scales it up for the larger area.

Example:

If you have one person per square foot(1ft by 1ft), then if you scale that to people per square miles, it would be 5280x5280 people per square mile(5280ft by 5280ft).  It doesn't care how many people are in your input, just the people per area unit you selected when you ran the tool.

Check your dataframe to see what your data frame unit is in(You can see it on the bottom right hand corner of your map).  If it is in decimal degree, try re-running the tool after you change the are unit scale factor of your tool to a more useful unit, or change the projection of your data frame to one that uses feet, yards, or miles as a map unit if you want people per square mile, or to one that uses meters or kilometers for people per square kilometer.

If you are still having trouble, post up your data and maybe someone can help you out.

Also, Dan posted up the tool help, it should be able to help you figure out how to use it properly.

Regards

View solution in original post

0 Kudos
10 Replies
DanPatterson_Retired
MVP Emeritus

if you are using the Point Density tool or any other density tool, the units are in number of observations per unit area

ELTerrell
New Contributor III

Where do I find the number of observations per unit area?  Is that when creating the density map or somewhere after it's created?

0 Kudos
DanPatterson_Retired
MVP Emeritus

That is the number... ie 88,000 would be the number of observation per unit area..if your areal units are in square kilometers, then it would be 88,000 people per sq km (or square miles, or square whatevers)

0 Kudos
ELTerrell
New Contributor III

So, the points I have mapped are 1,000 in total (1,000 people).  How is it picking up the 88,000?  What does the 88,000 mean (if there's only 1,000 points/people)?  How can I get it to switch to the 1,000 people? 

0 Kudos
DanPatterson_Retired
MVP Emeritus

It has nothing to do with the actual number of people...but the number of people per unit area

If you have 1000 people in 1 square km, then the density would be 1000 per sq km BUT if your area were

0.1 sq km, then your density would be 10,000 per sq km so if you want to convert from density to the number of observations, you multiply by the area where the people are found in...I hope you are making too much of this...

ELTerrell
New Contributor III

Then in a county with 1,000 voters plotted out, how can I indicate with the range/scale the numbers of people voting so it's understandable?  So far, I've got  a really dark color indicating 88,000 in one area though there's only 1,000 total.  Is there a way I can maybe do the multiplication so it's reflected in the range/scale?  My fear is people will look at that and either say it's wrong of have no idea what it's showing.

0 Kudos
ELTerrell
New Contributor III

One other thing, can you explain to me the difference between kernal and point density?  Not sure which I should be using so I just picked one.

0 Kudos
IanMurray
Frequent Contributor

You should be able to use either and get the same results, point density can only use point data, and can search by a circle(radius), torus, rectangle or wedge, whereas kernal density can take point or polyline inputs and can only search by radius for density.  In your case since you are using points and only need to search by a radius either one will work.

What Dan was trying to explain is that the output of this tool gives its output in units per area unit, so in your case voters per whatever are unit you selected on the tool.  So if you used square map units as your area unit scale factor and your map unit is decimal degree, then it calculates the density of voters per square decimal degree( or a 1 degree latitude by 1 degree longitude square box.)  So even though you only have 1000 voters, if your area is significantly smaller than 1 degree squared(approximately 4761 square miles), then your density per degree squared is much larger, since it scales it up for the larger area.

Example:

If you have one person per square foot(1ft by 1ft), then if you scale that to people per square miles, it would be 5280x5280 people per square mile(5280ft by 5280ft).  It doesn't care how many people are in your input, just the people per area unit you selected when you ran the tool.

Check your dataframe to see what your data frame unit is in(You can see it on the bottom right hand corner of your map).  If it is in decimal degree, try re-running the tool after you change the are unit scale factor of your tool to a more useful unit, or change the projection of your data frame to one that uses feet, yards, or miles as a map unit if you want people per square mile, or to one that uses meters or kilometers for people per square kilometer.

If you are still having trouble, post up your data and maybe someone can help you out.

Also, Dan posted up the tool help, it should be able to help you figure out how to use it properly.

Regards

0 Kudos
ELTerrell
New Contributor III

I see, that makes a lot of sense.  Thank you both for your help!

0 Kudos