6 Survey Results: Executive Summary
Interestingly, 18 respondents indicated that they have used a “system developed locally/in-
house.” Data storage systems were the most common solutions listed by these respondents, but the
underlying foundation of these varied from spreadsheets to relational databases to more sophisticated
data warehouses. They reported storing a variety of data, including e-resource usage, expenditures, and
journal holdings and/or overlap. The emphasis of these local data systems appears to be on integrating
data from multiple sources. Web or proxy server logs were also mentioned.
Most respondents indicated that data is collected annually, with only about a third reporting that
data is collected monthly or quarterly. The most commonly reported other frequency was “as needed” or
“ad hoc,” usually reported in addition to the other frequencies. The data that librarians need for collection
evaluation is generally directly accessible. Just over one-third reported that most of the data, and another
third reported that some of the data, is accessible. Another 11 respondents reported that the data is
accessible upon request, and only four reported that some data is not accessible at all. Among the other
responses, a distinction was made with ease of access.
Data Collection Dream Tools
The purpose of this question was to stimulate the development of tools not already available or created.
In retrospect, this question could have been phrased more clearly, as some of the responses (8 of 42) were
for tools that already exist, e.g., any ERM, the WorldShare CAS, Greenglass, and Tableau. Of the expected
responses, 17 were for improvements to existing systems, primarily the integrated library systems and the
resource usage tools. The requested improvements centered on ease of use and integration with other
data, notably cost and print usage. Also requested were improvements in generating reports and the
ability to analyze data at levels that are higher (e.g., consortial) and lower (e.g., patron groups) than the
individual library. Of the responses that could be considered “dream tools,” the key concern was for data
aggregation and integration, between and within systems. Some responses were very general:
“It would blend financial and usage data in an accurate, useful, actionable way and would be open
source and scalable to consortial/shared activities.”
“Internal database to allow all collected data to be in one place and have the ability to run reports
and combo reports to have a better ‘big picture’ of what data is collected, allow efficiency, and
help expedite the annual reporting.”
“Allow data aggregation and analysis from disparate data collection systems.”
Others requested specific combinations, notably for the aggregation of e-resource usage data,
print circulation, and expenditures. It is clear that the respondents were requesting data management
and analysis tools that brought together data to answer questions related to collection coverage, usage,
and efficiencies. This requires bringing data out of the silos and integrating the counts of titles and/
or volumes, records of usage, and costs. There were two suggestions that were quite different from
the others:
“Would be great if our automated monitoring systems (gate counters, environmental monitors)
would auto-report to a server.”
“A tool that scrapes bibliographic information from grant proposals, faculty annual reports,
materials in the institutional repository, course management sites, etc. but that allows
for anonymity.”
Data Analysis Methods and Frequency
Our survey asked respondents to indicate the use or interest in using a myriad of measures and methods
for collection analysis. These were organized into the four categories in the key textbook, Fundamentals
of Collection Development and Management (2nd ed.) (Johnson, 2009), and adapted here:
Interestingly, 18 respondents indicated that they have used a “system developed locally/in-
house.” Data storage systems were the most common solutions listed by these respondents, but the
underlying foundation of these varied from spreadsheets to relational databases to more sophisticated
data warehouses. They reported storing a variety of data, including e-resource usage, expenditures, and
journal holdings and/or overlap. The emphasis of these local data systems appears to be on integrating
data from multiple sources. Web or proxy server logs were also mentioned.
Most respondents indicated that data is collected annually, with only about a third reporting that
data is collected monthly or quarterly. The most commonly reported other frequency was “as needed” or
“ad hoc,” usually reported in addition to the other frequencies. The data that librarians need for collection
evaluation is generally directly accessible. Just over one-third reported that most of the data, and another
third reported that some of the data, is accessible. Another 11 respondents reported that the data is
accessible upon request, and only four reported that some data is not accessible at all. Among the other
responses, a distinction was made with ease of access.
Data Collection Dream Tools
The purpose of this question was to stimulate the development of tools not already available or created.
In retrospect, this question could have been phrased more clearly, as some of the responses (8 of 42) were
for tools that already exist, e.g., any ERM, the WorldShare CAS, Greenglass, and Tableau. Of the expected
responses, 17 were for improvements to existing systems, primarily the integrated library systems and the
resource usage tools. The requested improvements centered on ease of use and integration with other
data, notably cost and print usage. Also requested were improvements in generating reports and the
ability to analyze data at levels that are higher (e.g., consortial) and lower (e.g., patron groups) than the
individual library. Of the responses that could be considered “dream tools,” the key concern was for data
aggregation and integration, between and within systems. Some responses were very general:
“It would blend financial and usage data in an accurate, useful, actionable way and would be open
source and scalable to consortial/shared activities.”
“Internal database to allow all collected data to be in one place and have the ability to run reports
and combo reports to have a better ‘big picture’ of what data is collected, allow efficiency, and
help expedite the annual reporting.”
“Allow data aggregation and analysis from disparate data collection systems.”
Others requested specific combinations, notably for the aggregation of e-resource usage data,
print circulation, and expenditures. It is clear that the respondents were requesting data management
and analysis tools that brought together data to answer questions related to collection coverage, usage,
and efficiencies. This requires bringing data out of the silos and integrating the counts of titles and/
or volumes, records of usage, and costs. There were two suggestions that were quite different from
the others:
“Would be great if our automated monitoring systems (gate counters, environmental monitors)
would auto-report to a server.”
“A tool that scrapes bibliographic information from grant proposals, faculty annual reports,
materials in the institutional repository, course management sites, etc. but that allows
for anonymity.”
Data Analysis Methods and Frequency
Our survey asked respondents to indicate the use or interest in using a myriad of measures and methods
for collection analysis. These were organized into the four categories in the key textbook, Fundamentals
of Collection Development and Management (2nd ed.) (Johnson, 2009), and adapted here:
