Administrative data may not be current. Often there is a delay between when the data are collected and when they are available for use by another party. The other issues in terms of timeliness of administrative data are the provision of the information by reporting units and the collection method. For example, the provision of administrative data may be subject to delays because of when reporting units of the target population submit or register the information. For example, tax data may require reporting to be undertaken quarterly. However, in some instances a taxpayer (i.e. reporting unit) may ask for an extension to this deadline. Hence, not all administrative data for the time period for which data are being received will be included in the data file acquired by the receiving agency. The extension data records may become available in the next quarter and will automatically be included in the data file at that time even though they correspond to an earlier time period.
Another issue that needs to be managed is that of duplicate records. As mentioned above, records may be included for a specific reference period at a later date. However, in some cases incomplete records are provided in one extraction of administrative data to a receiving agency. Hence, when the next extraction of administrative data occurs and those formerly incomplete records have been updated with the latest information the receiving agency ends up with two different records for the same reporting unit. This is how duplicate records can be created. These duplicate records need to be consolidated and the surplus record removed.
A decision needs to be made as to when the administrative data will be extracted for use by a receiving agency. As with any data there is a trade off between accuracy and timeliness. The longer a receiving agency waits to acquire the administrative data the more likely it is that the administrative data are more complete and accurate. However, this in turn delays the use of the administrative data for statistical purposes. As mentioned above, delays in the provision of complete data will always be apparent with administrative data. The important thing is to establish a set of guidelines for how to deal with the likely issue.
This page last updated 23 March 2011