\[ \newcommand\si[1]{\mathrm{#1}} \newcommand\SI[2]{#1\,\si{#2}} \newcommand\matr[1]{\mathbf{#1}} \DeclareMathOperator*{\argmax}{arg\,max} \DeclareMathOperator*{\argmin}{arg\,min} \]

GIS data quality

Dr. Huidae Cho
Institute for Environmental and Spatial Analysis...University of North Georgia

1   Data quality

Indication of how good data are.

Need to be complete, compatible, consistent, and applicable for the analysis being performed.

Problems may include errors, accuracy, precision, bias, resolution, and generalization.

2   Accuracy vs. precision

accuracy-vs-precision.png

3   Resolution and generalization of raster datasets

resolution-and-generalization-of-raster-datasets.png

5   Completeness

5.1   Data completeness

Covers the entire study area and period.

Attributes exist for each feature.

Only one set of attributes are assigned to each feature.

5.2   Quality completeness

Compatibility: Spatial scale and measurement scale (e.g., ratio vs. ordinal)

Consistency: Data sources, calibration, boundary changes, different data encoders

Applicability: Using appropriate data layers for functions

6   Sources of error

Conceptual errors: The way in which we perceive, study, and model reality

Errors in source data: Human error, equipment problems, conversion errors

Data encoding errors: Methods and conditions under which it is carried out are important

Data editing and conversion errors

Data processing and analysis errors

Data output errors

7   Checking for errors

Visual inspection

Double digitizing

Examination of error signatures (e.g., feature ID, vertex ID, field name)

Statistical analysis: Checking outliers and extreme values