Abstract:
Lawyers are increasingly responsible for conducting research using legal databases and are looking to law librarians for training. As there is little information regarding law practitioner training, and even less which provides information about the actual search behaviour of the legal profession, much of this training has had to be based upon experience and best guesses of individual librarians.
This study was undertaken to investigate the actual search behaviour of practitioners using the Auckland District Law Society Library. Its purpose is to provide the training personnel in that library with information about the search habits of their potential trainees to improve current training initiatives.
It is based on data from transaction logs gathered from the public terminals in the Auckland District Law Society Library which are used by practitioners. An analysis of the logs collected revealed that: (1) the case summary databases, LINX and BRIEFCASE, were the databases most commonly used by practitioners; (2) the most common type of search conducted during the study was for commentary or case law on a particular subject; (3) the majority of search sessions comprised only a single query, but there were some instances where practitioner sessions would involve more than 10 queries; and (4) there was limited use of any of the advanced search features offered on FolioVIEWS.
Based upon these findings the following recommendations were made in relation to the existing training programme offered by the Library:
1. All training sessions include information regarding database concepts;
2. The library initiate additional lunch-time training sessions to inform practitioners of the databases currently available in the library and their content; 3. The library continue to teach advanced search techniques, particularly search construction, the use of synonyms and truncation, to help increase the levels of recall and therefore search success in practitioner searches;
4. The library continue to include information on Field and Phrase searching in both the beginners and advanced courses.
Although the purpose of the study was not to investigate the level of search 'failure' or 'success' attained by practitioners, this paper contains a discussion of the different measurement techniques which could be used to measure search effectiveness. It is argued that recall would be the most appropriate measure of search success and that, based upon a visual examination of the transaction logs, this is not being achieved in the majority of cases. Given this alarming observation it is argued that more attention should be paid to issues surrounding database and interface design and that the library become involved in a general education programme to help users recognise situations in which end-user searches may be inappropriate.
