Defining the Concepts | Reasons for Doing an Assessment | History of Collection Analysis | How Collections are Measured -- Techniques | Organizing the Assessment | Presenting the Findings | Using the Results | Further Information | Self-Assessment #15
Defining the Concepts
Collection assessment is “an organized process for systematically analyzing and describing a library’s collection.” Assessments are conducted to provide several kinds of important information to libraries. They help clarify the library’s goals in the context of its mission and budget, supply data used to set funding priorities, and build a base for long-range planning and administration.
There are many ways to conduct assessments, some of which may be better suited to small libraries. However, the success of any method depends on how well it meets the goals of the evaluation, which in turn depends on the purpose and mission of the library. The process can be broken into two steps: assessment, in which the collection is described according to the different subjects and formats of the materials, and evaluation, in which the collection’s appropriateness for the community is judged. This training section discusses several methods and techniques for assessment. It draws upon Using the Conspectus Method: A Collection Assessment Handbook (Bushing, Davis, Powell, 1997) which is well suited to smaller libraries.
Reasons for Doing an Assessment
Collection assessment or collection mapping provides library administrators with a management tool for adapting the collection, an internal analysis tool for planning, a tool to respond systematically to budget changes, and a communication tool and data for resource sharing with other libraries. Library staff can also benefit by having a better understanding of the collection, a basis for more selective collection development, improved communication with similar libraries, and enhanced professional skills in collection development. For libraries involved in cooperative resource sharing, collection assessment is essential in determining how each library fits into the system and what should be expected for each library’s further growth in the context of the cooperative relationship.
When planning a collection assessment project, it is important to carefully define the goals for the program, choose the most appropriate method(s) to be used, and establish what information is needed. An assessment can be fully comprehensive or it can focus on specific areas, depending on the library’s needs (and the resources available to carry it out - evaluations can be expensive in terms of staff time!). It can be very tempting to gather all sorts of information because it seems interesting, but it may be useless if it does not fit into the parameters of the project. Be sure that everyone participating in the project understands what is expected and when tasks should be completed.
History of Collection Analysis
Collection assessment or collection mapping as it has more often been called in school libraries is not a new concept for libraries in the United States. Initially, item level records were the only means by which libraries were able to track the quality and character of their holdings. Pre-1956 imprints were recorded in the National Union Catalog, a huge set of books (754 volumes) containing both the bibliographic records and the holding symbols for libraries for titles published prior to 1956. These records primarily reflect the holdings of the Library of Congress and research libraries in the United States. The National Shelflist Count, another attempt by research libraries to track not just the number of items each held but also the rate of acquisition and the subject areas, provided yet another means of identifying collections of depth in a particular subject. However, it was not until the 1970s and the beginning of electronic library recordkeeping that libraries of all sizes began to develop practices and techniques to describe the nature and qualities of their collections in ways that provided a means by which to judge the depth and breadth of their collections.
A systematic approach to collection assessment was first developed by the Research Library Group (RLG) and was called “conspectus” which is a term simply meaning a survey or outline. The conspectus approach was organized using a Twentieth Century view of the disciplines in the modern university. This scheme included initially 25 major areas or divisions. Then the classification schemes (the in-house or home-grown organization originally developed for use just by the Library of Congress and the Dewey Decimal Classification with its mid-Nineteenth Century world view of knowledge) have been fitted into the this hierarchical structure. The conspectus organization is only necessary if one library wishes to compare its holdings with other libraries that use a different classification scheme. The conspectus organization allows us to define “geology” or “education” as the same body of knowledge regardless of how we organize it in our library. A library can identify the characteristics of a collection on or about these subjects and compare the quality and quantity of each subject, genre or format collection using a variety of measures.
The conspectus methodology includes a number of important elements in addition to taking a Twentieth Century view of the world of knowledge, an element of importance was the gradual development of a set of codes or shorthand to describe the depth and breadth of a collection segment so that the character or nature of the subject segment could be understood by others without making a bibliographic list of all the titles available. These codes have been adopted as an international standard. While it is not necessary for a library to use the codes, they can help a library identify its strengths and weaknesses and adjust their information resources as needed. Sometimes a library discovers inappropriate strengths and heretofore unknown gaps in the collection. Another element of importance of the conspectus approach was the ability of all types of libraries to compare their collections in a meaningful manor. WLN (a former regional bibliographic utility located in the Pacific Northwest) and RLG both developed software to make the whole process easier and to allow libraries to easily present their assessment findings to their funding bodies, granting agencies and others in data-rich graphic presentations as well as in narrative form. While this software is no longer being supported, much of what was done by the proprietary software can be duplicated in any library using standard spreadsheet software. This approach has been formalized in a book by Tony Greiner and Bob Cooper, Analyzing Library Collection Use with Excel, 2007. A library can adapt the ideas presented in this ALA publication to fit their own needs.
As a result of the many libraries and librarians who worked with the conspectus approach, we have a clearer understanding of what measures matter in our struggle to achieve excellence in our local settings. The collection information can be combined with statistics about use and the community of users in ways to provide insight about the ways in which a particular collection meets or fails to meet the needs of its community. By using a wide range of variables libraries can now more realistically consider their materials and how the segments of their collections fit the needs and interests of their user communities.
How Collections are Measured - Techniques
Evaluation and assessment techniques fall into two broad categories: collection-centered (counting holdings and checking lists to determine the collection’s scope and depth) and client-centered (conducting user surveys and gathering information on how clients use the collection). An effective assessment uses both types of techniques to gather two kinds of data: quantitative (including numbers, age, and/or use statistics) and qualitative (such as observations by informed staff and/or subject knowledgeable users). The type of data useful for your particular evaluation depends on the library’s purpose and mission. For example, a library that wants to provide many varied titles might compare its acquisition rate to annual publishing output, and might look at titles held per capita. If the library has very limited space and must keep growth to a minimum, data on turnover rates (how often items are circulated), acquisitions, and withdrawals (weeding) will be essential. A library that focuses on popular works would want information on circulation to use in turnover rates as well as statistics on in-house use per capita. Some examples of the types of quantitative and qualitative data you can collect include:
Examples of Quantitative Data
- Number of titles: A manual or automated shelflist count, a physical count or estimate of titles from the shelves, a count of acquisitions and expenditures, the percent of each subject’s yearly growth, the percent of acquisitions compared to published titles, and the percent of acquisitions compared to published titles.
- Age and timeliness of materials: The materials’ median or mean age, the range and distribution of publication dates, or the most common publication date range (i.e., 1972-1976 = 56% of the nonfiction).
- Use: Circulation statistics by type or subject and interlibrary loan borrowing statistics in each general subject area.
- Per capita measures: How many titles or items per community member.
Examples of Qualitative Data
- Percent of standard titles or items: Calculated by consulting “best” lists.
- Individual or group evaluation: Shelf scanning by the librarian, observation by an outside expert, or observation by a committee. (WLN, p. 19)
There are at least ten different techniques for collection assessment that have been more or less successfully used by libraries during the past twenty years. Of these, four measures are essential. They provide both quantitative and qualitative data, though they are not completely objective. By using a combination of techniques to complement and verify each other, the good judgment and experience of the librarian will compensate for much of the subjectivity. Regardless of the techniques you choose to use, it is very important to obtain data on the number of items by type and age. These two types of data are essential to the assessment process. There are additional techniques that can be used but as the practice of collection assessment has advanced it has become increasingly clear that the approaches listed below are the most essential. There is little need to go beyond these measures in gathering useful profile information about the collection and how it is or is not being used. These and other techniques are discussed in the sources cited at the conclusion of this section.
Determine the size of the collection/subject segment
This technique gathers quantitative data about the collection segment (subject, genre or media) including number of titles and the percent that this segment is of the whole collection. An automated system can often provide detailed reports of this information. If you don’t have an automated system, you may have to manually review the shelflist or the collection itself. For the most part, a multi-volume title is just one title for purposes of the assessment. Multiple copies of the same title sitting on the shelf do not add to the quality of the collection. If there is a need for multiple copies, that need should be evident in their use and circulation. If you have segments of your collection that are not in the catalog or cannot be counted by the computer for whatever reason, then you need to count them manually. While there are always things in circulation, your common sense and experience ought to enable you to estimate from the stacks how much might really exist. An exact number is not essential. A good estimate is usually enough for these purposes. You can easily calculate the percentage of the total collection that this segment represents. You are really acquiring two pieces of information in this process: (1) the number of titles/items in this segment of the fiction, nonfiction or media collection, and (2) the percentage this segment is of the total fiction, nonfiction or media collection. This is important because it allows you to easily compare the usage with the size by changing it to a percentage.
Determine the average age (mean or median) of the collection/subject segment
Run a system report to determine the mean (arithmetical average) or the median (the approximate year that divides the collection by age, often in the 1970s or ‘80s). Most automated catalogs will allow you to do this for the whole collection and for segments distinguished by media type and/or call number range. All of the research and our own experience show that most of what is used in a public library is the newest material. Material that has not been used is increasingly less likely to be used. The average age of any segment of the collection is a variable to be considered along with many other variables including the subject itself, the purpose of this segment, our users, and the nature of the discipline. For example, while we will want to have some new editions of classic fairy tales with award-winning illustrations perhaps, we may also want to keep some of the older editions and translations. For sports, we may want to only have available current information because the rules change, our knowledge of physiology changes along with the advice we give regarding conditioning and practice exercises, and the players and jargon change season to season. The median age indicates the point at which half the collection is older than the date indicated, revealing currency and/or retrospective strength. The mode shows what dates are most common, revealing clumps in the collection and often are the result of grants, special initiatives or the founding of the library. Again, although these numbers are relatively easy to obtain, they are not to be construed as meaning either good or bad things. They are numbers that need to be considered with the other variables such as the subject, the circulation or use, and the goals of the library. Assess the age of any reference or non-circulating collection separately from the rest of the collection on the same subject because any non-circulating collection (reference or research oriented) is likely to need to be either more current or possibly older than the regularly circulating collection on the same types of subject.
Determine the turnover rate for the collection/subject segment
Run a report that tells you the total circulation for this segment/subject/genre of the collection. Using the total number of items one can easily establish the turnover rate for a 12-month period. Thus, if the library has 400 books about investments and money management classed in the Dewey 300s and the circulation for that segment in the last 12-months was 200, the collection turned over only .5 times. We arrive at the turnover rate by dividing the number of circulations by the number of items/titles. Knowing that a number of titles will have gone out multiple times, we recognize that at least half of the collection, or 150 titles did not go out at all. Further, it is easy to understand that warehousing these items is expensive, their appearance is likely to be less than inviting, and the more there is, the more difficult it is for users to find what is relevant. Investment strategies need to be current. We see that this collection needs to be heavily weeded. We also might wish to consider if this is an area that should even contain many books now that there is so much of this type of information on the Web. Our system can also easily determine the percentage of the total circulation this segment of the collection had for the past twelve months. It might be something as small as .07% of the adult nonfiction but it represents as much as 1% of the total nonfiction collection. Does this make sense to us at this time?
If the segment of the collection we are assessing is our adult mysteries and we find that we have 1,000 titles, with a mean age of 1973 and a twelve month circulation of 3,000, then we might consider giving more resources to this collection to increase its size because it is being heavily used with a turnover rate of 3. This might be 35% of the total adult fiction circulation but it represents only 22% of the total adult fiction collection. We can understand that we might need to be buying more mysteries and fewer of some other area of adult fiction. We might also run a report regarding any of the titles older than 20 years to see what has not circulated more than once in the past two or three years and consider some of these for weeding. This information might prompt us to discuss the need for the purchase of more current mystery best sellers in both large print and audio book formats.
Objectively scan or examine the collection/subject segment
This technique is essential. It is well suited to libraries of all sizes. It has the advantage of providing relevant information quickly, but it can be highly subjective (especially if the person doing the assessment also does selection). Working in a team should be encouraged. Direct examination should not be used as the sole assessment technique. Shelf-scanning should be conducted after the statistical data have been collected; the data complements our observations in the stacks. You should make notes on what you see as you examine a segment of the collection. You do not need to take every item off the shelf. Begin by locating the beginning of this segment (i.e., the cookbooks) and the end of the segment. Put yourself in the shoes of the user. What do you see? What is the condition of the collection? Is it inviting? Ugly? Dirty? There are things one can see when examining the collection in subject segments that are not easily understood by browsing the bibliographic records. We see the collection of cookbooks, football books, travel CDs, or whatever, as the user sees them. We see the condition, the age, the extent (how many are actually available at any given time not just how many the library has), and we can observe things like the presence of too many textbooks, items with only black and white pictures, no indexes, etc. Deliberately and objectively doing shelf-scanning of collection segments in conjunction with some statistics generated by an automated catalog can provide the best type of information about the information resources.
Organizing the Assessment
The steps in organizing a collection assessment are much the same as those used in any project. They include: planning the project, doing the actual assessment work, recording the data and comments, analyzing the information, and then using the knowledge for a wide range of purposes. One way that libraries have used the information is to inform collection development and to change collections and resources to more accurately fit the needs of current users. The information should inform our policies and our budget allocations as well as being powerful in the political and funding arenas.
Planning the project
There are number of initial decisions to be made before beginning to assess the information resources of the library. Listed below are the questions you need to ask to plan the project. Some of these questions will lead you to answer other related questions and because of your unique situation, there may be some that do not have relevance for your library. Use these as a guide in planning your collection assessment.
- How will we use the information we gain from the process? (Policy? Budgeting? Advocacy? Operations? Selection? Weeding? Planning? Other?
- Who will be involved?
- What data can our system (if automated) provide?
- What data can we gather efficiently ourselves? What methods might we use? (Estimating? Sampling? Using others to record numbers that we can use to calculate mean ages or other information?
- What data are we going to gather? These are essential:
- Number of items in the segment
- Mean or medium age of the titles
- Percent this segment is of the total larger segment (i.e., all adult fiction)
- Circulation for the previous 12 months or most recent statistical year
- Percent this circulation represents of the total circulation for the larger segment (i.e., all adult fiction)
- Notes from doing shelf scanning (condition, general appearance, impression, types of materials, need for action, etc.)
- What additional information might be useful depending upon our purposes? Possibilities:
- Circulation figures for this same segment or the larger segment for the previous 12-month or year period and maybe the one before that for comparison and to identify trends
- Budget allocation for this segment or the larger segment
- The number of new items added to this segment during the previous fiscal year
- The possibilities are almost limitless but do not bog down the whole process by concentrating on “interesting” data that really will not influence your decisions
- How will we divide the collection into meaningful segments? (Fiction genres? Nonfiction meaningful segments – using Dewey schedules to identify meaningful pieces? Children’s materials? Young adult? Media collections? Special collections? Reference?)
- What will our data gathering and note taking form look like? (Spreadsheet with formulas built in? Chart? Other?)
- What will our timeline be? (Do we have a deadline? Remember that quick & dirty is always better than not at all.)
- What fiction and what nonfiction segments should we use for a trial run?
- How will our circulation practices influence the data we gather? (For example, if some collections circulate for a shorter period of time than others, how will that influence our understanding of how much use there is?)
- Once the questions have been answered and the form and format are ready for data gathering, it is time to gather the test data and to visit the shelves with the data in hand to take notes about the appearance and contents of the collection in a general way. Then analyze and discuss what you have found out about these segments. Is your process going to work? Do you need to revise your form or process? Make any changes, revise your timeline as needed, and begin your assessment.
The following is a sample form for gathering assessment data. It should spark ideas about how you wish to design your form. You can use a table in Microsoft Word or a spreadsheet program such as Excel. The advantage of a spreadsheet is that you can build in formulas to calculate percentages, averages, totals, etc., and you use the graph function to easily create graphical representations of your findings. The book by Greiner and Cooper gives some good examples.
Click here to see a sample Assessment Worksheet
Presenting the Findings
Once you have completed the assessment or that part that you are doing at this time, it is time to consider how you will present the information and to whom. This information will certainly be useful with staff and volunteers and the Board but how you present it will determine how well it is understood by others. Especially if you wish to use it in a budget presentation or in other advocacy activities, the information must be presented with graphics and in ways that convey your message. Books about the graphical presentation of such data are plentiful. A few are cited in the Further Information section below. Do not hesitate to present bare numbers in picture form. Calculate the size of the annual circulation figures into a stack of books. Have a local artist draw a picture of a tall local landmark (a tall building, a mountain, a hill) and a wobbly stack of books next to it that represents the annual circulation (change all circulations into a “book” for this purpose) so that the circulation numbers become a picture with the books reaching far beyond the local landmark, up into the sky and clouds. If you do not have a local tall landmark that will be useful, use another one - the Grand Canyon, the Statue of Liberty, be creative. Use pie charts, graphs and other visuals to convey the information you want people to remember. Do not give others useless information. Select the most important bits and pieces to make your points.
Using the Results
The results of the collection assessment should give you information that will help you adjust your CD policy to help you meet fulfill the collection aspects of the goals and objectives set for the library during the most recent planning process. These goals and objectives will relate directly to the service responses selected to meet the needs of the community. Summary information from your assessment can be placed in your policy to help staff and others understand the nature and direction of the library’s information resource efforts. A schedule for weeding, goal mean ages for various segments and other data can help conceptualize the nature and character of the collections for this library. You can be creative in your use of the knowledge you will have gained by doing the assessment. You should repeat the basic process every three to four years because technology and other societal changes including economics, politics, and demographics continue to change the ways in which libraries are used and the types and formats users expect to access through their libraries.
For Further Information
Baker, S. L. & Wallace, K. L. (2002). The responsive public library collection: How to develop and market it. 2nd. ed. Englewood, CO: Libraries Unlimited, Inc.
Bounford, T. (2000). Digital diagrams. New York: Watson-Guptill Publications.
Bushing, M., Davis, B., & Powell, N. (1997). Using the conspectus method: A collection assessment handbook. Lacey, WA: WLN.
Doll, C. A. & Barron, P. P. (2002). Managing and analyzing your collection: A practical guide for small libraries and school media centers. Chicago: American Library Association.
Greiner, T. & Cooper, B. (2007). Analyzing library collection use with Excel. Chicago: American Library Association.
Harris, R. L. (1999). Information graphics: A comprehensive illustrated reference. New York: Oxford University Press.
Kachel, D. E. (1997). Collection assessment and management for school libraries: Preparing for cooperative collection development. Westport, CT: Greenwood Press.
Robbins, N. B. (2005). Creating more effective graphs. Hoboken, N.J.: Wiley-Interscience.
Wallgren, A. (1996). Graphing statistics & data: Creating better charts. Thousand Oaks, CA: Sage Publications.
Zelazny, G. (2001). Say it with charts: The executive’s guide to visual communication. 4th ed. New York: McGraw-Hill.
Zelazny, G. (2006). Say it with presentations: How to design and deliver successful business presentations. New York: McGraw-Hill.