18 Survey Results: Survey Questions and Responses Retention varies by specific type of data Retention varies from 2 years (EZProxy logs) to 10 years (circulation records). The retention varies by system. Varies—Data with IDs: less than two years without IDs: indefinite. Does not have a retention schedule/policy N=13 Common practice is 3 years retention there is no formalized retention policy in place. Every several years, or so Indefinite (8 responses) Indefinitely Varies greatly depending on the type of data and why it was collected We automatically anonymize circulation data 30 days after an item is returned as long as there are no outstanding fines/fees. We clear from the Libraries’ ILS all patrons who have been gone at least one year on an annual basis. We do not get rid of ILL patron data. There is no way to anonymize ILL data, and we’d lose all related transactions if we removed patron data. 12. Please enter any comments you have regarding library data collection and retention. N=22 All historical data not purged from systems. While data may contain personal identifiers, it is used in aggregate to count users and usage patterns for administrative and reporting purposes. Although we do not currently have a data retention schedule or policy document, we are in the process of developing one and have reviewed our existing policies in order to be explicit about how data is used. As noted earlier, the library doesn’t engage in “learning analytics” per se, but it definitely engages in collecting usage statistics. With the exception of on-site professional training and ILL requests, identifiers are not used. Circulation records are purged weekly. Currently, we do not have a retention guideline that is specific to library records. However, we do have two general retention guidelines, the Administrative and Management Records guideline and Fiscal Records guidelines that apply to all university offices. These would apply to library administrative and financial records in lieu of a more specific guideline. Data collection and retention is driven by external reporting needs as well as for specific decision- making related functions within individual units or departments. Data collection varies across units, departments, and even librarian/instructor. We have not developed specific retention policies to govern learning analytics data collection because the data we are using for learning analytics is covered by existing categories in the data retention schedule. I’m not certain it is accurate to describe use of the data as learning analytics. Various departments and libraries collect data. There is no comprehensive program to use the data in the way I think you are hoping to understand. Identifiers are not included with on-campus usage of databases, journals, and e-books. It is generally best practice to retain all information until it is no longer perceived to be needed.