What is the difference between the Attribute Accuracy Report and the Quantitative Attribute Accuracy Assessment?

The Attribute Accuracy Report is a complete description of the accuracy of the attribute information in the data set, include the identification of tests used, testing methodology, results obtained, etc.

In developing the metadata standard, reviewers asked that a summary method of providing results also be permitted. While this summary method would not relieve producers of the obligation to provide a thorough report, it would allow users, especially those using a data catalog, to quickly exclude those data sets that obviously would not serve a user's needs. The Quantitative Attribute Accuracy Assessment (and its horizontal and vertical accuracy counterparts) were provided in response to this request. A producer identifies the test used by name and the value obtained from the test. The derivation of these values should be described in the accuracy reports.

What does SDTS say about attribute accuracy?

Attribute Accuracy:

Accuracy assessment for measures on a continuous scale shall be performed using procedures similar to those used for positional accuracy (providing a numerical estimate of expected discrepancies). The report of a test of attribute accuracy shall include the date of the test and the dates of the materials used. In the case of different dates, the report shall describe the rates of change in the phenomena classified. Spatial variations in attribute accuracy may be reported in a quality overlay.

Accuracy tests for categorical attributes may be performed by one of the following methods. All methods shall make reference to map scale in interpreting classifications.

Deductive Estimate

Any estimate, even a guess based on experience, is permitted. The basis for the deduction shall be explained. Statements such as "good" or "poor" should be explained in as quantitative a manner as possible.

Tests Based on Independent Samples

A misclassification matrix shall be reported as counts of sample units crosstabulated by the categories of the sample and of the tested material. The sampling procedure and the location of sample units shall be described.

Tests Based on Polygon Overlay

A misclassification matrix shall be reported as areas. The relationship between the two maps shall be explained; as far as possible, the two sources should be independent and one should have higher accuracy.

Return to: Attribute Accuracy