0

I have a series of feature classes that represent trail systems with hiking times. They are straight lines between two known points. I generated these classes by interviewing various people on their hiking times between the two points. The end result is a series of classes with overlapping lines between points that are represented in another class separate from the lines.

I.e.: One feature class for known locations - point file. One feature class for Hiker A - polyline file. One feature class for Hiker B - polyline file. etc...

They are all contained within a single feature dataset.

I need to run various statistical analyses on the times that people reported - i.e. I need to find the mean and median of the times people reported for that distance.

So my question, is what is the best way to get this information in a format that can be analyzed with Excel or R?

I've tried working with the Network Analyst and Topology, but I can't seem to export data from either of them. Any ideas?

2 Answers 2

0

I assume that the time is stored in each line, so you could do :

merge all the lines in a single feature class.

then you can use summary statistics to get the mean and median of your times (no need to do it in excel or R).

Note: if you need more complex stats , you can export the table using "table to table" (either in txt for R or directly in xls for Excel)

2
  • Is there any way to include the origin/destination point for each line? There isn't any other identifying information in the line classes.
    – Josh
    Commented Jun 2, 2014 at 9:47
  • if the direction does not matter, you can use "spatial join", once with the FIRST statement, and then second time with the "LAST" statement to aggregate the ID of your origi/destination points. If you care about the direction, then you can use feature vertices to points to get the from and to points of your segment, then you join the origin/destination to those points by location, then you join the points back to the lines by attributes.
    – radouxju
    Commented Jun 2, 2014 at 13:53
0

A more formal approach with just as many steps as that which radouxju proposes would be to use a route system. It may be intuitive that you are indeed dealing with a system (or set) of routes, but assembling the route system, then exporting the route system is what you want to do. (Explore the Linear Referencing tutorial in the ArcGIS for Desktop help, watching for these steps.)

  • Create Routes by adding a Name/Description column to your line features and giving all the lines with the same from-to combination the same name such as "Hike from A to B" versus "Hike from B to A" as obviously they are different hikes. Once you have those named, then load them with the Create Routes tool.

  • Create a Route Event table -- These are waypoints or POIs along the route including the start and end points for the routes! Most people initially think there is little need for these points then later decide to add them because most travelers along routes want knowledge such as "How long (or how far) until the shady tree everyone stops at for lunch?"

  • Optionally, Calibrate Routes using DISTANCE such as marker posts (or DURATION in time) as a measurement.

  • Finally, as suggested above, you can perform "Table-to-Table" exports to get the route system out into Excel where worksheets inside the workbook contain the summary, details, and statistics for the route system.

If there are only a handful of hiking trail, then this more formal approach is perhaps overkill. If you are responsibly managing a large trail system and/or want to make this information published in a web-services environment, then the ESRI linear reference tools and route structure will pay off.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.