Data quality and positional accuracy. The following is adapted from OpenGIS® Spatial Referencing by Coordinates (Topic 2) specification.

The parameters that define a coordinate reference system are chosen rather than measured to satisfy the degrees-of-freedom problem in the changeover from observation to coordinate quantities. Coordinate reference systems are therefore by definition error-free (i.e., non-stochastic). A coordinate reference system is realised through a network of control points. The coordinates of those control points, derived from surface and/or from satellite observations, are stochastic. Their accuracy can be expressed in a covariance matrix, which, due to the degrees-of-freedom problem, will have a rank deficiency, described in geodetic literature.

Coordinate transformations between coordinate reference systems usually have parameter values derived from two sets of point coordinates, one set in system 1, the other set in system 2. As these coordinates are stochastic (i.e., have random-error characteristics) the derived transformation parameter values will also be stochastic. Their covariance matrix can be calculated.

Coordinates that have not been "naturally" determined in coordinate reference system 2, but have been determined in coordinate system 1 and then transformed to system 2, have the random error effects of the transformation superimposed on their original error characteristics. It may be possible in well-controlled cases to calculate the covariance matrices of the point coordinates before and after the transformation, and thus isolate the effect of the transformation, but in practice a user will only be interested in the accuracy of the final transformed coordinates.

Nevertheless the option is offered to specify the covariance matrix of point coordinates resulting exclusively from the transformation. It is outside the scope of this specification to describe how that covariance matrix should be used. Because a covariance matrix is symmetrical, only the upper or lower diagonal part (including the main diagonal) needs to be specified.

For some transformations, this accuracy information is compacted in some assessment of an average impact on horizontal position and vertical position, allowing specification of average absolute accuracy and, when relevant and available, average relative accuracy. Hence separate quality measures may be specified for horizontal and for vertical position in those objects.

Since:
2.0
• Class
Description
Closeness of reported coordinate values to values accepted as or being true.
Correctness of the temporal references of an item (reporting of error in time measurement).
Presence and absence of features, their attributes and their relationships.
Excess data present in the dataset, as described by the scope.
Data absent from the dataset, as described by the scope.
Adherence to rules of the conceptual schema.
Information about the outcome of evaluating the obtained value (or set of values) against a specified acceptable conformance quality level.
Result of a data quality measure organising the measured values as a coverage.
Quality information for the data specified by a data quality scope.
Adherence of values to the value domains.
Type of test applied to the data specified by a data quality scope.
Type of method for evaluating an identified data quality measure.
Degree to which data is stored in accordance with the physical structure of the dataset, as described by the scope.
Closeness of gridded data position values to values accepted as or being true.
Degree of adherence to logical rules of data structure, attribution and relationships (data structure can be conceptual, logical or physical).
Accuracy of non-quantitative attributes.
Accuracy of the position of features.
Accuracy of quantitative attributes.
Information about the value (or set of values) obtained from applying a data quality measure.
Closeness of the relative positions of features in the scope to their respective relative positions accepted as or being true.
Base interface of more specific result classes.
Description of the data specified by the scope.
Accuracy of the temporal attributes and temporal relationships of features.
Correctness of ordered events or sequences, if reported.
Validity of data specified by the scope with respect to time.
Accuracy of quantitative attributes and the correctness of non-quantitative attributes and of the classifications of features and their relationships.
Comparison of the classes assigned to features or their attributes to a universe of discourse.
Correctness of the explicitly encoded topological characteristics of the dataset as described by the scope.
Degree of adherence of a dataset to a specific set of user requirements.