Geographic Information Systems Stack Exchange is a question and answer site for cartographers, geographers and GIS professionals. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have just lost the past 8 hours on this, and literally getting nowhere - hoping someone can help before I go mad!

I am trying to determine the correlation between sinkholes and a number of different variables. My sourced data is as follows:

Sinkhole Database: 6460 points
Fault Lines: Line files
Surface Streams: Line files
Karst Landscape: Polygon file
Depth of soil to bedrock: DEM

Using ArcMap 10.1

Can anyone point me in the right direction as to how to analyse the correlations between each variable and the points? I have set up buffers around the river and fault lines, and have extremely simplistic counts at each of the boundaries, but would prefer something slightly more data rich!

Any "how to" videos would be greatly appreciated!


share|improve this question

closed as unclear what you're asking by Devdatta Tengshe, whuber Aug 29 '13 at 15:00

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question.If this question can be reworded to fit the rules in the help center, please edit the question.

Many people would start by using a GIS to create a table of tuples, one per point, containing attributes describing the geographic relationships between the sinkholes and the other features. Such relationships can include distance, bearing, presence/absence (as in karst landscape), and numerical value (as in depth of bedrock). That's merely the point of departure: now you abandon the GIS--it has done its job and typically is not the best tool for the next step--and move into exploratory data analysis and formal statistical analysis. But precisely how you do that depends on your objectives. – whuber May 16 '13 at 17:15
+1 Similar exercises would have the user list as many relationships as possible and perform a regression or principal component analysis. This may not be appropriate in cases where you are missing data for specific important variables, but it's a start and it's simple. – Radar May 16 '13 at 23:22
Many thanks to you both – Christography May 17 '13 at 12:42
@whuber It looks like your (and Radar's) comment resolved this one for the asker so I think you should convert it to an answer we can upvote to get off Unanswered list or get refined with more detailed objectives. – PolyGeo Jul 18 '13 at 1:17
@PolyGeo It is unclear whether this question was resolved. I'm still not sure what it was specifically asking. – whuber Jul 18 '13 at 13:06