Serious users could be affected by the way we edited and need to understand our processing of the data records. Furthermore, users can be affected by the fact that we continue this process and not all records have been fully normalized and georeferenced. See our complete methods description for details of our data processing. In all cases specimen pages in the website provide the verbatim donor data along with our edits of that data. (more detail below and in this section).
We chose to work, at least to start, only with data in which every occurrence record is based on the existence of a preserved specimen in a museum. If there are ever any questions about the identification of the specimens, we can simply examine them and verify the data. Museums also typically record detailed spatial and temporal data for every specimen collection, while most every other data source summarizes these data in one way or another.
Over 40 museums contributed their data, however, since data quality, basic data definitions, and formats vary greatly among donor institutions we dedicated large amounts of time to basic data compilation and management, and standardization of data formats and definitions among all contributing institutions. Basic steps in our data manipulation are briefly described in the pages in this section (often with links to more detail).