When collecting data there are several issues that you will have to consider in order to ensure that the information that you are collecting is appropriate for the purposes that you are collecting it for. These include what data you should be collecting and its quality. For instance, the data may be coming from a variety of different sources in which case you may need to match users up between systems. This may be easy as there is a common user id in the systems, or require some mapping exercise if not. You will also need to check that the systems are actually logging the types of event that you are interested in. When the logs were turned on they may not have been set to collect data that you need. You will also need to consider the format that you collect the data in, which will depend on how you want to process the data. Each of these is discussed:
- The data to collect.
- Data quality and completeness.
- Formats for collection.
- Data collections tools that can sometimes be useful.
We have also produced a guide and a number of recipes on collecting and extracting data.
- Extraction of tracking data from DSpace
- Extract authentication counts from the EZproxy Authentication and Access Software
- Extract circulation data from the Horizon Library Management System
- Extract entry statistics from the Sentry library gate entry system
- Extract trace data from log files
- Extract user ID and course code from CIRCE MI
- Extract anonymised loan data from Talis Alto