Elicited Personal Information Retrieval

Elicited Personal Information Retrieval (EPIR) is a research technique to study Personal Information Management (PIM) retrieval. It combines the advantages of controlled laboratory experiment (control of some of the variables and measurement of others) with the advantage of a naturalistic observation (ecological validity).
The Problem with Measuring PIM Retrievals
Before the development of EPIR it was very difficult to measure retrieval of personal information as it occurs throughout the participant’s day. One solution was to give the participants 'artificial' information items and then observe their retrievals. While this method has the advantage of being able to measure and control related variables, it can be criticized for not representing real retrieval of personal information where the users are highly familiar with their own information and its organization.
Elicited Personal Information Retrieval
In EPIR to increase ecological validity, the tester asks participants to retrieve sample information items from their own personal information item collection, using their own computers. The retrieval task is elicited by the tester who asks the participant to retrieve one file at a time. This enables the tester to control some of the variables, such as the retrieval method (e.g. the tester can ask participants to navigate to their files for half of the files and search for them for the other half). The retrievals are video-recorded in order to measure the dependent variables which are typically: failure percentage, percentage of retrievals with missteps, and retrieval time. the effect of the operating system, presentation and depth on file navigation, retrieval of contacts on mobile phones, the amount of cognitive attention used in navigation compared to search, and sharing files using common repositories compared to sharing files via email. and logfiles. Unlike EPIR, both options record participants’ spontaneous retrievals. However, these options also have their limits: Diaries are typically used in small-scale qualitative studies with limited external validity. In addition participants are likely to omit important information (e.g. retrieval time). Logs are difficult to deploy (both technically and because of privacy issues) and are typically used only when a new prototype is tested. In addition there may be issues involved in interpreting user intentions from complex logfile data.<ref name="sharedfiles"/>
Notes and References

Main Menu

See Also

Main Menu