Data Quality: Assessment

This week in GIS 5935, I concluded the last of a three-part module on data quality in GIS. This module focused on data assessment where we were tasked with determining completeness of two different road networks by summarizing our analysis in textual, visual, and numerical terms. Completeness is one of the five components of data quality where it essentially refers to "how well the data set captures all the features" (Boldstad and Mason 2022:611).  Further, assessment of completeness can help inform the analyst on important information missing in a data set that might be critical for a spatial analysis project.

For this assignment, I assessed completeness by examining differences in road length for a section of Washington County, Oregon in two different data sets:  TIGER Roads (U.S. Census Bureau) and Street Centerlines (Washington County). In addition, I used a grid consisting of 297 1-km by 1-km polygons to serve as base units of analysis. For the analysis I used various tools and functions in ArcGIS including Clip, Summarize Within, Spatial Join, Calculate Field, and Calculate Geometry. 

The total length of roads (km) was higher for the TIGER data.  In the grid polygon, this consisted of 11,253 km of roads in the TIGER layer with less for the County layer (10,671.2). Of the 297 grid polygons, TIGER data was more complete in 162 with the County data more complete in 132.  In three grid sections, the lengths were identical. 

In addition, I calculated a percentage that reflected the differences in length between the two data sets across the grid. This involved subtracting the total length of TIGER roads from the total length of the County centerlines, dividing by the length of the County centerlines (as the base), the multiplying by 100%.  The map choropleth map below shows how the differences are symbolized where I establish three distinct categories:  -197.5-10, -9.9-10, and -10.1 to 80.9.  Here I wanted to show at opposite ends, greater than 10% differences as well as those with less than 10%:

I found this assignment interesting and helpful in allowing me to better understand aspects of data quality, especially the component of completeness, along with further enhancing my cartographic design skills.

 
2022  Boldstad, Paul and Steven Manson
      GIS Fundamentals:  A First Text on Geographic Information Systems, 7th edition. Minnesota:  Eider
      Press.

Comments

Post a Comment

Popular posts from this blog

My GeoJourney begins today!

Image
                                             In Spring 2023 I began the graduate certificate in GIS at the University of West Florida.  The first and current class I am taking in the program is GIS 5050, Geographic Information Systems.  One of the first course assignments requires students to create a blog that they will use not only for this class, but future ones in the certificate program.  While I have yet to start work on my own GIS projects for class, I thought I would share one from my course project in the last GIS class that I took (GIS 5107) at the University of Florida in Spring of 2010. This map shows the distribution of graves and artifacts at the Old Oakland Cemetery in Oakland, Orange County, Florida and the site's proximity to Highway 50 and a sinkhole immediately to the northwest.  The map also shows contou...
Read more

Computer Cartography!

Image
This week I started Computer Cartography (GIS 5007), which is my second course in the graduate certificate in GIS at UWF. I hope to enhance the knowledge I learned in the previous class, Introduction to GIS (GIS 5050), while expanding my knowledge and skill base in GIS and in cartographic design principles. Recently I resumed archaeological fieldwork with my Gulf Coast State College students at an important historic/archaeological site in Panama City.  We are currently exploring sites associated with the "lost" town of St. Andrews (1827-1863). For more information on the project, please check out My Story Map . Below is a cut and zoom of the 1855 U.S. Coastal Survey Map of St. Andrews City.  I hope to eventually work with this map/image formally in GIS, including overlaying it with modern aerial photographs and topographic maps, georefencing archaeological field data to it, and incorporating all the proper map elements in the final layout.
Read more

Surfaces: TINs and DEMs

Image
In this week’s lab, we learned about the use of surfaces in GIS, specifically creating, editing, and analyzing TINs and DEMs. Through the assignment, I was able to create 3D visualizations of elevation models, use various input data sets to create and modify TINs, and compare the properties and derivatives of TIN and DEM elevation models.  TIN is abbreviated for triangulated irregular network; it is one model that is used to represent surfaces including elevations, through a combination of point, line, and area features (Boldstad and Manson 2022:40). DEM refers to digital elevation model or digital terrain model (DTM) and it is commonly used for terrain analysis (2022:66). TINs are essentially a DEM but in the form of vector-based data whereas other DEMs can be derived through a raster grid.  One of the drawbacks of TINs, however, is that they are less widely available than raster surface models, and tend to be more expensive to build and process because of the complex data st...
Read more