One of the most important things in GIS is understanding the quality of the data used for any given project and one of the most important (and often ignored) tasks of any person who uses GIS software is to maintain the highest data quality. When data is created in-house,there is a very good chance it has been done with the goal of giving the data away, either to the customer who requested that data or uploaded to a data portal or clearinghouse where others will be able to download it for use in their projects. In turn, the data after delivery or upload doesn't have the name of the technician who created attached to it, but instead the name of the company or the agency which employes that technician. As companies and agencies never want their name muddied, a great way to not lose a job as a GIS technician is to create data of the highest quality utilizing the established standards. Data quality can vary based on several factors, such as the minimum and maximum map scales for which the data will be accurate, the established number of significant figures for the attributes, and the expected features which are represented. For example, if a United States vector layer which is intended for use at a scale of 1:500,000 and above where it's okay to have a rather generalized outline of the states since the nooks and crannies of each state cannot be resolved (recognized) at that zoom level. The data is a high quality choice for projects where the United States is zoomed out but of low quality for projects where the focus is a section of a single state's edge. In contrast, data which is intended to be used at a scale of 1:24,000 may have too many vertices to draw quickly or neatly.
To inform a user about where the data is of the best quality for a specific project, metadata, which is the data about the data, should be attached to each data layer. Metadata is a structured and established method of recording the who, what, why, where, and when of the data creation and intended uses. For convenience, ArcGIS utilizes a form that a technician can fill and the metadata is formatted correctly for several major metadata standards which is automatically attached to the data, however, far too few people actually complete this task.
In this chapter, data quality and metadata will be examined, including how to define data error, where that error comes from, and a bit more about metadata standards.