The Association of Research Libraries (ARL) has conducted three national surveys of interlibrary loan (ILL) service since the early 1990's and the Iowa State University (ISU) Library has been involved in all three. The first study was based upon fiscal year (FY) 1991 data, the second was based upon FY 1996 data, and the third utilized FY 2002 data. The study methodology, sampling approach and statistical measures have evolved over the years, but there are some basic measures of service at Iowa State that can be traced across all three studies. Comparisons can be also be made between Iowa State and the other surveyed libraries. Table 1 highlights the results of these three studies.

The ISU Library offers three basic services - ILL borrowing, ILL lending, and document delivery - and all three services have been evaluated by ARL. Interlibrary Loan Borrowing and Lending services have been evaluated in all three of the ARL studies, but the document delivery service has only been evaluated in the 2002 study.

Interlibrary Loan-Borrowing Service

ILL borrowing is often considered the most important service offered by an interlibrary loan operation, simply because it involves service to your local clientele. For the borrowing service, four evaluation criteria are usually measured: cost to the library (average cost per transaction, including staff salaries and wages, benefits, equipment depreciation, fees to paid to suppliers, etc) fill rate (percentage of total requests that are successfully completed); turnaround time (the number of days it takes from the time ordered by the patron until it arrives in ILL); patron satisfaction, as measured by a questionnaire.

Of these four criteria, only cost was measured in all 3 ARL studies. The first ARL study, conducted with FY 91 data, showed that Iowa State had a 25% higher cost than the mean for its borrowing service. In the FY 96 study ISU's cost dropped to $19.63 and was just 7% higher than the mean. In the FY 02 ARL study, the cost dropped again to $18.89, but was still 8% higher than the mean. Not factoring for inflation, ISU borrowing costs had dropped 19% from FY 91 to FY 02, but consistently exceeded the average of the libraries surveyed.

In terms of fill rate, the ISU Library has consistently performed better than the libraries surveyed by ARL. The measure was first sampled in the FY 96 study: ISU had an 88% fill rate compared to the study mean of 85%. In FY 02, Iowa State's fill rate improved to 95% compared to the study mean of 86%. From 1996 to 2002 Iowa State's performance improved at a significantly higher rate than the rest of the libraries surveyed.

The third evaluation criteria of turnaround time, or delivery speed, basically showed a better performance by the ISU library than the others surveyed. In FY 96 the average turnaround time for borrowing requests at ISU was 14.70 days compared to the mean of 15.60. A further break down of the data, however, showed ISU to be slower (16.90 days) than the mean (14.90 days) for non-returnables, or copies of documents. For returnables, ISU was faster than the mean: 14.30 days compared to 16.90 days. There were significant improvements made in the turnaround time for all surveyed libraries 6 years later. ISU's turnaround time dropped to 5.60 days for non-returnables and 8.30 days for returnables. There was no overall average calculated in FY 02, but both of the ISU figures bettered the study means. Turnaround time is also tracked internally using an ILL data software called Clio. This figure has continued to drop (see table 2). The average turnaround time for non-returnables (copies) was 7.4 days in calendar year 2001 and dropped to 2.8 days in calendar year 2004. For returnables (loans) it was 10.0 days in 2001 and 7.3 days in 2004.

Patron satisfaction, as determined by a survey of users, was only conducted in ARL's FY 96 study. 100% of the ISU patrons surveyed were satisfied with the time it took to obtain the materials requested; 98% were satisfied with the quality of the copy or loan; 100% were satisfied with the ILL staff.

Interlibrary Loan-Lending Service

The evaluation criteria for interlibrary loan-lending service are very similar to the borrowing service. Over the course of ARL's three cost and performance studies, the following have been measured: 1) Lending cost 2) Lending fill rate 3) Lending turnaround time. In the FY 91 ARL survey only cost was analyzed. Table 1 shows the average lending cost at Iowa State to be $7.61, which was 30% less than the study mean of $10.93. In the FY 96 study, the average lending cost for Iowa State was determined to be $7.51. This was 1% less than the FY 91 ISU cost (not adjusted for inflation) and 20% less than the FY 96 average of $9.48. The third ARL study showed ISU's lending cost rising to $9.50, a 20% increase over the FY 96 study and 2% higher than the study mean of $9.27.

The fill rate for ILL lending has been studied in both the FY 96 and FY 02 ARL studies. In FY 96 the fill rate was determined to be 61% for Iowa State and 58% for the average library surveyed. In FY 02 ISU's fill rate dropped to 51% while the study mean stayed at 58%.

Turnaround time for ILL-lending is determined by when the request is received and the material is sent to the borrowing library. The FY 02 ARL report was the only one to address this performance measure. In terms of non-returnables (i.e. copies of documents) ISU took an average 2.1 days to send out the material compared to the study mean of 1.35 days. For returnables (e.g. books, microfilm, videos, etc.) the ISU Turnaround time was 2.80 days and the study mean was 1.69 days.

Document Delivery Service

The ISU Library's document delivery service is the newest service to be offered by the ILL/DD unit. It is a fee-based service that is offered to any individual (affiliated or unaffiliated with Iowa State), organization without a formal library, or other fee-based document delivery service. The 2002 ARL cost and performance study was the only one to examine this service (Table 1). The average cost of an ISU document delivery request was $10.23, while the study mean was $10.76; a difference of -5%. The ISU fill rate was 90% compared to the study mean of 86%. The ISU turnaround for non-returnables was 2.90 days while the study mean was 1.94 days. For returnables the ISU figure was 3.50 days and the study mean was 2.32 days.

Conclusions

In terms of ILL-borrowing, the ARL ILL studies show Iowa State's service to be performing better than the average library in the key areas of fill rate and turnaround time. Furthermore, internal data show turnaround time to continue to drop since the last study in FY 2002. Much of this improvement is no doubt due to joining the Rapid ILL service in October 2003. There is no cost to the patron to borrow materials on a regular basis, and the only patron satisfaction survey conducted by ARL showed ISU patrons to be very happy with the service. The one area of concern for borrowing is the cost to the library. ISU's borrowing cost dropped from FY 1991 to FY 2002, but it has also been consistently higher than the average library surveyed. During the last few years, two ILL staff positions have been eliminated due to attrition. This should have the effect of lowering borrowing costs, but this could be offset by a continued drop in the volume of borrowing requests.

The three ARL studies clearly showed that more attention needed to be paid to the Lending service. In the FY 02 study, lending costs were higher, the fill rate was lower, and the turnaround time was slower. Because of this, a number of steps have been taken to improve in this area: a high level staff member (LAIV) was named to coordinate the lending service; ILL staff took on the task of union listing serials in OCLC to improve fill rate; the ISU Library joined the Rapid Consortium where requests are routed according to actual journal holdings; and steps were taken to shorten the time from when records were updated to when materials were delivered. Already some improvement in the fill rate is being recognized. In FY 05, internal statistics showed a fill rate of 59% and in the first two months of FY 06 the fill rate has gone up to 64%. The reduction in staffing mentioned above should have a salutary effect upon lending costs, as should the rapid growth in lending volume: new records for filled lending requests have been set in each of the last three fiscal years. The turnaround time figure may have improved because of the demanding delivery standards of the Rapid project and the Greater Western Library Alliance (GWLA), but this is not a statistic that is routinely tracked.

As for the document delivery service, ISU appears to offer a cost efficient service with a good fill rate, but with a delivery speed that could use improvement. The base charge to non-ISU clients for document delivery service is currently $12.00 and $5.00 for ISU-affiliated clients. At $12.00 the ISU Library is fully recovering the costs of the service and is subsidizing fully half the cost of sending materials to it clients that have ISU Library cards.