Related product Link Resolver (SFX)

SFX Navigation

Usage Stats Introduction

Overview

SFX provides 21 statistical queries that provide some useful insights about how your patrons are using your institutions SFX interface. These queries are available by clicking the “Queries” link under the Statistics heading on the main SFX Admin page. Queries can be run as needed or can be scheduled to run at regular intervals with the results delivered via email.  A thorough description of SFX usage statistics is available in the SFX General User’s Guide Version 4 beginning on page 247.

Options

Most queries are customizable with the following options:

Query Parameters – This gives the choice of how the data is displayed. Depending on the query, this can be a choice of dates, what targets and objects to display or the number of items, “All”, “Top 10”, etc.

Dates to Query – This gives the choice of what dates queries gather the statistics from. An administrator can either specify a specific date range or choose dates available from a drop down menu such as “Yesterday” or “Last year”

Filters- This is a specialized part of the query that allows the administrator to include or exclude Institute, Formats, Usergroup, Faculty or IP Address.

Output – An administrator can choose the method of how the data retrieved is delivered and in what format. The formats available are HTML and Plain Text. Plain Text is best for creating an excel spreadsheet of the data. HTML is best for viewing as a web site.  The delivery options are to your screen immediately or via email. If “Plain Text” is chosen as the format and “Screen” as a delivery method, a text document will be downloaded to your device.

A Selection of Useful Queries

To make this queries more useful to new SFX administrators, this document will highlight some of the queries and discuss how other institutions use this data to better understand their patron’s behavior.

 1. Number of requests and clickthroughs per day

This query provides a basic overview of how SFX is used. If you set the query parameter to total by day, you can see how your SFX usage varies by day of the week.

 2. Number of requests and clickthroughs per source

This query provides an overview of SFX usage based on the source of the SFX request. In this context, “source” refers to the starting point of the user’s action. Source names are generally ugly but readable by humans. For example, “info:sid/sfxit.com:azlist” refers to the SFX A to Z list. This query produces two columns, “Requests” and “Clickthroughs”. A “request” is any time a user clicks the SFX link button to get to a specific article or item. A “clickthrough” indicates that a user actually got to the link and viewed it or downloaded it. This query is valuable to discover where users are beginning their searches. For example, if a large number of users are starting with the A-Z list this could indicate that the user had a specific journal in mind. This query is also very valuable for evaluating the usage of non full text resources such as abstract and indexing databases such as BioAbs.  Finally, if the administrator has activated Google Scholar, this query is a method of measuring how users might be using the web as a way of getting to resources.

 4. Number of requests and clickthroughs per service type

This query shows SFX by service type. That is, how many request did you get for “getFullTxt” services and how often did your users “Clickthrough” into that target from the SFX menu. This query also can help evaluate how often your online catalog is being used as it counts the number of times someone as used the “get Holding” service which is the one service that is used when you activate links to your local catalog.

7.  Number of clickthroughs per target.

This query shows the number of times specific targets were selected in the SFX menu. This query is an excellent method of determine what targets you have activated are being reached via an SFX request, i.e. from the screen that pops up when a user clicks the SFX button on an article. To help illustrate this query, here is a list produced by this query:

Query 7 : Number of clickthroughs per target.
Target Clickthroughs
EBSCOHOST_ACADEMIC_SEARCH_COMPLETE 4812
LOCAL_CATALOG_EXLIBRIS_VOYAGER 2360
SPRINGER_LINK_BOOKS_COMPLETE 2302
GOOGLE_SCHOLAR_FREE 2178
WWW_SEARCH_ENGINES_FREE 2029
MISCELLANEOUS_FREE_EJOURNALS 1805
SAGE_PREMIER_2007 1404
GALEGROUP_IT_EXPANDED_ACADEMIC_ASAP 1301
WILEY_ONLINE_LIBRARY_2010_FULL_COLLECTION 1235
SPRINGER_LINK_JOURNALS_STANDARD 1146
DOAJ_DIRECTORY_OPEN_ACCESS_JOURNALS_FREE 972

What these numbers tell the administrator is that users are finding full text in places one would expect such as Academic Search Complete but also from places that you would not expect to be so heavily used such the sources that come from free ejournal services on the web.

10. Most popular serials selected by target, 11. Most popular journals selected by source, 19. Most popular journals

These three queries all count the number of times specific journals were requested and clicked through to. Each displays the data in a different way. By target organizes the data by the source of the request such as a database or journal collection. Therefore individual titles will show up under different sources as many titles are available from different resources. Most popular journals by source gives both the title and the issn, the most popular journals just by the issn. This is not an indication of the overall usage of a journal but only its popularity within SFX. The data is by default presented from the journal with the highest number of requests to the journal with the lowest. Journals with a high number of clickthroughs indicate users want to use the journal and also that the institution is providing adequate access through SFX. However, if a journal is being requested a large number of times and is receiving very “0” or very few clickthroughs, this could be an indication the institution might consider purchasing the item. Or it could mean that SFX is not adequately indicating the institution’s holdings for the item.