Purpose -The paper aims to provide details of a study conducted at Hunter College Libraries in fall 2005, the focus of which was how presentation of initial digital resource pages (or gateway pages) on the library's web site impacted students' subsequent steps in the research process. Design/methodology/approach -A group of 16 students from English and History classes at Hunter College were recruited to participate after having had basic library instruction. They were given computer-based key tasks to perform in a proctored classroom setting, using the library's homepage. A second group of students was recruited to participate in two small focus groups. The methodology and exercises were developed in part using guidelines from a taxonomy of user behavior developed by librarians at Hunter College, and recommendations from usability literature by Krug, Neilsen and Rubin. Findings -Results from the computer-based key tasks exercises were bifurcated. Completion rates for computer-based key tasks using the in-house developed Hunter College Library database grid, with less than 80 percent (37 percent-73 percent) students successfully completing all the tasks, was inferior compared to performance using the Serial Solutions access page and the Academic Search Premier database, both commercially-developed products, with most of the tasks successfully completed by at least 80 percent of the students. Originality/value -This study is unique in that the focus is not on the usability of an entire library web site, rather, on the presentation of select, highly visible gateway pages that get a lot of use.

Purpose -This paper seeks to address the administration of Standardized Assessment of Information Literacy Skills (SAILS), the use of SAILS report data, and respondent perceptions of the utility of SAILS at institutions that comprise the "All Institutions Benchmark" -the institutions which participated in the SAILS testing through the Spring 2007 testing period. Design/methodology/approach -An online survey was used to collect qualitative and quantitative data for the study. Overall response rate for the survey was 63.9 percent. Disregarding unusable responses, the response rate for the data analysis was 57.8 percent. Findings -The large majority of institutions used convenience sampling in administering SAILS. With regard to the SAILS report data, there are indications that those institutions that received support for data analysis were more likely to find the results useful; the utility of the SAILS report data can also be correlated with institutional type. Practical implications -Results from this study suggest that SAILS can be most effective if there is statistical/institutional research support for data analysis, if the sampling method for selecting test takers is more rigorous, if SAILS is used in conjunction with other instruments, and if the SAILS data is correlated with other institutional data. Originality/value -There have been few published studies on the large-scale administration of standardized information literacy assessments. Moreover, there has been no other published research study assessing multiple institutions' experience of administering SAILS and using SAILS results. Institutions interested in the use of SAILS and other well-known standardized information literacy tests will find this paper especially relevant.