Module 1.3 Data Quality Assessment

Here we are in Lab 3 Module 1.3 continuing to dive deeper into Data Quality, this week's being Assessment. Our objective is to learn how to determine the quality of road networks, by determining the completeness of a road network with comparing it to another road network. Some learning outcomes are to be able to determine the completeness of the road network and to be able to summarize analysis in textual, visual and numerical terms.

Our task of this analysis was to evaluate the relative completeness of two road datasets provided to us TIGER_Roads and County Street_Centerlines across a gridded study area. By comparing total road lengths within each grid polygon, the assessment aimed to identify which dataset offers broader spatial coverage and where discrepancies may indicate outdated or missing data.

This kind of accuracy assessment is essential for ensuring that spatial analyses, routing models, and infrastructure planning are built on reliable basemaps. It also helps highlight areas where local datasets may outperform national ones or vice versa.

To assess the relative completeness of the TIGER_Roads and County Street_Centerlines datasets, I created a uniform grid across the study area to serve as spatial units for comparison. Using the Spatial Join tool, I intersected road segments from both datasets with these grid polygons and calculated the total road length within each cell separately for each source. From these values, I derived two metrics: the absolute difference in length (Difference_km) and the percent difference (Percent_Diff), using the County Centerlines as the base. To ensure clarity in the visual representation, I filtered out grid cells with extreme percent differences beyond ±100%. The resulting data was then symbolized using a choropleth map, highlighting areas where one dataset significantly outperformed the other and revealing spatial patterns in road network completeness across the county.

The TIGER_Roads dataset demonstrates a broader coverage across the county, outperforming Street_Centerlines in 61% of grid cells. However, the Street_Centerlines dataset still provides more detailed coverage in 39% of the grid, that may reflect in more recent or higher-resolution data collection.

Here below is a map showing the results of the analysis of the relative completeness of the two road datasets.