Testing Road Network Completeness

This accuracy assessment was to understand what municipal road network dataset was "more complete" based on the lengths of line features and measure the difference using a grid.  For this module, we were provided with two datasets of the road network in Jackson County, Oregon; one presumably from Jackson County and the other TIGER 2000 data from the US Census Bureau.  A square grid with nearly 300 polygons was also provided.  After reprojecting one of the datasets to have both road networks in the same projected coordinate system, I used the Summarize Within tool to calculate the lengths of the road network segments within each polygon.  Using a Join, both attribute tables were combined and an additional column created. I then calculated the percentage difference between each road network in the column using a python expression in the provided pane.  A continuous color scheme and the Natural Breaks method was used to distinguish the variance between grids.

I used Excel to provide the numerical analysis for this lab, but had the intention of using PostgreSQL instead.  I wanted to use a series of joins and sub-queries but ran short on time.  I did however successfully install Microsoft SQL Server, PostgreSQL, and PostGIS as a result and have every intention on using SQL more for these types of analysis.