Metadata Education Project

Metadata education suggestions and materials for:

Spatial Analysis

Learning Material | Preparatory topics | Complementary topics | Vocabulary

Learning Outcomes

Conviction

Motivation

Skills

Knowledge

Preparatory topics:

Complementary topics:

Vocabulary

Vocabulary definitions

Material for this topic

Increased confidence in analysis results

Analysis is the process of identifying a research question, modeling that question, then investigating and interpreting the results.

Spatial analysis is the same process, but using spatial data to perform analyses such as:

To answer a research or management question, one or more of these and other types of analysis steps may be taken.

Where does metadata fit into spatial analysis? Traditionally, metadata has been used to document data, but documenting process is also important. For instance, formal metadata has a section devoted to describing "process steps", the steps taken to create, update, transform, and manage a dataset, so that the users have a descriptive history of the dataset. This history of the dataset's sources and processing steps (formally refered to as Lineage), is very helpful in determining the dataset's fitness-for-use for particular applications or analyses.

Metadata or documentation of the process steps that contribute to a spatial analysis procedure is also important. Valid scientific technique requires the ability to replicate the result. Similarly, valid spatial analysis should also be replicable.

Time and money is invested in a GIS project to create data or format existing data, and then perform analysis using that data, to provide an answer pertaining to a research question or application situation. Eventually the GIS project or a particular aspect of the project is completed, and the GIS analyst(s) responsible for the work disperse to other duties or to other jobs. Later on... months or years later, a question is raised about the results of the GIS project. Even if the same analyst(s) are still around, they probably won't remember specific details of procedure unless it has been documented.

Some potential situations they may be faced with:

So, how was the original analysis accomplished, anyway? Some GIS systems automatically log commands that were issued, but these logs can be difficult to interpret due to obscure file-naming conventions and trial and error attempts. If errors were discovered in a dataset, then corrected, how is it possible to judge between the original results and the results of the second analysis using the corrected data, if the procedure cannot be exactly replicated? Slight variations in coordinate precision, area calculation precision (number of significant digits), tolerances used, classifcations, and even different order of analysis steps can contribute to significant changes in results. Without good process documentation, or metadata, it is difficult to replicate analysis results or compare different analysis procedures with a high measure of confidance.

Skills: creating "Process metadata"

Documenting process steps in the Content Standard for Digital Geospatial Metadata (CSDGM):

Example exercises to demonstrate the importance of metadata

Back to Course Topics