The ODURT (http://www.oplin.org/odurt) reports on queries and articles for each specific database delivered by OPLIN to Ohio's public libraries. Counts of database sessions are also available through the zip code stats reporter. These numbers are most useful in spotting trends in usage over time.

The counts of queries, documents, and sessions in an individual month mean very little. What is important is how these numbers change over time, and the "trend" is the direction of activity. It may be increasing by a certain percentage, for example, or decreasing by a certain percentage over a given span of time. Data that are measured and reported consistently over time can be useful for observing trends since the manner of reporting remains the same. Change may be assumed to be due to some other variable, such as increased usage.

The database vendors are not required to have their stats to OPLIN until the end of the first week of each month.

A single query is any event in which a patron or staff member types a keyword or other search term in a query box and clicks on 'search' or 'find' or 'go,' causing the database to conduct a search and issue a results list.

For our purposes, an article is a full text document. It might be an actual magazine or newspaper article from other sources (as is the case in most of the EBSCO databases). It might also be an informational document created for the database itself (as is the case in some of the encyclopedias, almanacs or NoveList for example). It specifically is not an abstract, or a citation. Our count is the count of every full text document patrons or staff members view, download, email or print. It is the count of documents "used."

Because the databases delivered by OPLIN come from a variety of database vendors, the technology that allows for browsing varies. Some of them use menus and subject trees in their overall design. We do not track the hits or click-throughs in a central fashion because we do not believe they can be reported consistently across products. Some of the vendors report this activity, however, in their own reporting tools.

In addition to the stats provided in ODURT, there are other stats available directly from most database vendors. A list of the vendor stat sites and how to log into them is available at http://oplin.org/content/statistics-0

It varies from product to product. Vendors may give some indication of browsing, number and length of sessions, and particular sections or component titles that were viewed.

EBSCOadmin is EBSCO's own tool for checking EBSCO database usage statistics, creating reports, and customizing certain interface options. Since it is maintained by EBSCO and not OPLIN, a separate ID and password are required. If you do not know your ID and password, please contact the OPLIN Support Center, support@oplin.org for assitance. For further details on how the tool itself works, click on the "Help" link at the top of the screen after logging in.

This page lets you view how many times databases were accessed from within particular zip codes during a selected month. The OPLIN authentication mechanism stores the domain name of the resource (e.g. ebscohost.com, heritagequestonline.com), the month and year, whether access is in-library or remote, and the zip code of the user. We began collecting these statistics in August 2006.

Sometimes multiple resources have the same domain. For example, all of the separate EBSCOhost resources, including NoveList, are recorded as "ebscohost.com." The vendors have web-based tools that you can use for more detailed statistics. Visit http://oplin.org/content/statistics-0 for more information.

Our system records all authentication requests, even unsuccessful ones. So if someone tried to access a subscription resource that your library doesn't subscribe to (either by being in your library or using one of your library cards), then our system records the attempt. This can be useful information; perhaps there is local demand for a resource that your library currently does not have.

If the session is from within a library, the zip code we store is that of the main library. This is true even of branch libraries; the recorded zip code is the main library zip code. For remote sessions, we either ask the user to enter her zip code or her library card number. We record the zip code as entered, or for library card validation we record the zip code of the main library in that system.

Most OPLIN databases are licensed for all of Ohio, so we have tried to remove barriers and make it easier for remote users to get to subscription content. Using geographic data on IP addresses, we are able to tell when users are in Ohio, and we wish to let them into the databases as quickly and easily as possible. Usage in these sessions is not credited to particular libraries, but to a generic statewide account; therefore we ask those users for their zip codes so that we can get some picture of the distribution of usage locally. Previously, OPLIN has only tracked the number of searches and the number of documents viewed in subscription databases. A count of the number of sessions is a third statistic that is recommended by various groups (ICOLC, Project COUNTER, NISO). We had no measurements of how much of our database usage originated from within libraries and how much was remote; our objective was to start collecting some numbers to help us track this.

Except for in-library usage, you can't. We know that library service areas overlap, and that libraries can have patrons registered all over the state. It doesn't matter if several libraries track stats in the same zip codes. These numbers are not reported to the State Library, and they are not used to determine funding levels between library systems. What is ultimately important about database usage statistics is not the actual numbers, but how they change over time. What we want to see is a constant, steady rise in usage. Removing access barriers is the first stage of raising usage (and entering a library card which can be up to 14 digits long is definitely a barrier).