Data Quality Data quality a measure of how
- Slides: 29
Data Quality • Data quality – a measure of how well the GIS data represents the target domain • Related terms: – Data uncertainty – Data error – Data accuracy
Data Quality • Micro level components – factors that pertain to the individual data elements • Macro level components – factors that pertain to the data set as a whole
Data Quality • Micro level components – positional accuracy – attribute accuracy – logical consistency – resolution
Data Quality • Positional (spatial) accuracy – difference between location of an object as it is described in the data and its actual location – bias: systematic error – precision: standard deviation of error
Data Quality High bias: a systematic error Low bias: a random error ‘truth’ data
Data Quality High precision: all errors about the same distance ‘truth’ data Low precision: errors vary greatly in distance
Data Quality • Attribute accuracy – are geographic objects identified correctly – invalid attribute values – missing attribute values – ‘mixed-up’ attribute values
Data Quality • Logical consistency – logically consistent relationships between geographic objects – e. g. if a lake edge forms the boundary of a state, the lake boundary line should be identical to the state boundary line
Data Quality Logical inconsistency among different data, due to generalization Water Not PA Land PA
Data Quality • Resolution – raster: length of a side of a grid cell in real world units – vector: size of the smallest geographic object represented (minimum mapping unit)
Data Quality • Macro level components – completeness – time – lineage
Data Quality • Completeness – coverage • proportion of data available for the area of interest – classification • how well the classification is able to represent the data – verification • amount and distribution of field measurements or other independent sources of information that were used to develop the data
Data Quality • Completeness – coverage • proportion of data available for the area of interest Area of data availability
Data Quality • Completeness – classification • how well the classification is able to represent the data Agriculture Grains Orchards Forest Urban Water Deciduous Coniferous
Data Quality • Time – commonly, the date of the source material used to create the data – some data do not change significantly over the time data usage (elevation data) – other data can change rapidly (demographic data and land use)
Data Quality • Lineage – the history of a data set: the source data and processing steps used to produce the data – each data source and processing step introduces a level of error into the final data product – lineage should be encoded in documentation detailing how the data was produced and who did it
Data Quality • Sources of error – Error in spatial data cannot be completely eliminated, but it can be managed – trade-off between cost of creating and maintaining data and level of error
Data Quality • Sources of error – data collection – data input – data storage – data manipulation – data output – use of results
Data Quality • Sources of error – data collection • error in field data collection • errors in existing maps used for digital data creation
Data Quality • Sources of error – data input • inaccuracies in digitizing (operator and equipment) • discretization of geographic entities (e. g. vector digitizing of forest ‘edge’ • error in attribute entry
Data Quality • Sources of error – data storage • numerical precision
Data Quality • Sources of error – data manipulation • error propagation in multiple overlay operations • ‘sliver’ polygons
Data Quality Data manipulation: sliver polygons Water Not PA Land PA
Data Quality Vector to Raster Conversion
Data Quality Raster to Vector Conversion Original
Data Quality • Sources of error – data output • scaling inaccuracies (printer dpi) • instability of the medium
Data Quality • Sources of error – use of results • misinterpretation of data • no acknowledgement of data uncertainty
Data Quality • Metadata – data about data – data quality is described in the metadata – standards for metadata and data sharing developed by National Committee for Cartographic Data Standards (NCDC) and, currently, Federal Geographic Data Committee (FGDC) – www. fgdc. gov – PASDA example
Data Quality • Metadata – digital spatial data that is derived from USGS paper maps conform to National Map Accuracy Standards (NMAP) http: //rockyweb. cr. usgs. gov/nmpstds/nmas. html – in place since 1940 s – as set of accuracy ‘requirements’ that all published maps conform to
- Data quality dimensions
- Gibbons jacobean city comedy download
- Measures air temperature
- Measure page quality
- Data quality and data cleaning an overview
- Data quality and data cleaning an overview
- Secondary content analysis
- Data quality and data cleaning an overview
- Data collection planning in the measure phase starts with
- Mode grouped data
- Mode of grouped data formula
- Numerical measure of how alike two data objects are
- Measures of central tendency for grouped data
- Quality assurance vs quality control
- Pmp quality management
- Pmp gold plating
- Quality improvement vs quality assurance
- Compliance vs quality
- Concept of quality assurance
- Management gurus and their contributions
- Quality is free: the art of making quality certain
- What is tqm
- Uji validitas dan reliabilitas
- Iso 8000
- Spatial data quality
- Quality assurance definition
- Data quality standards iso
- Total data quality management
- Data quality management best practices
- What is data quality in gis