---

The Value of Information in Library Catalogs
by Joe Matthews

"The new power is not money in the hands of the few, but information in the hands of the many." -John Naisbitt, Megatrends

The purpose of this article is an attempt to discover the value of the information contained within a library's catalog. It specifically assesses the value of MARC (Machine Readable Cataloging) records and location/status information. Almost thirty years ago, Slamecka, (1970) suggested that;

The development of a pragmatic measure of information utility must rank as the most important task of information science research in the next decade.

Machlup (1979) stated that "it seems more reasonable to keep use and effect of use separate." Thus, the focus of this paper is on the content of the library catalog rather than on the value or effect of using the library catalog and, by extension, the library itself.

Information
Information is data (numeric and text) that is organized and imbued with purpose or intelligence resulting from the assembly, analysis or summary of data into a meaningful form (McGee and Prusak, 1993; Walker, 1993). Or, when data is given context it becomes information. The intent of information is to "inform."

The value of a library is found in its collection of information resources. A library creates a catalog of bibliographic records as a finding aid to assist users in discovering what information resources are located within the library or elsewhere.

It is quite clear that valuable information is contained in the books and other materials of a library's collection. However, the information is difficult to retrieve without the added value brought to the library's collection by the process of organizing and describing it (cataloging) so that the user is able to locate the desired item(s) or source of electronic information. Once a library reaches a certain size, it becomes almost impossible for locate materials without some sort of organizing principles, classification, cataloging, shelving and storage guidelines.

A library's catalog, that meets the objectives first articulated by Charles Cutter (1904), have stood the test the time. They make as much sense in the context of Internet searching as they did in the days of the card catalog.

The bibliographic records of a library's catalog are the surrogates of the "real thing" - the information package. To create a catalog that can meet Cutter's objectives, the cataloger creates a bibliographic record that describes in broad terms the contents of the information package. This descriptive bibliographic record has recently been described as metadata--or data about the information found in the information package.

The creation of a library's catalog is a process that adds value. Taylor (1986) has suggested that there are three major processes that add value to information: organizing, analyzing, and judgmental. Within each of these three major processes there are additional specific activities as shown in Figure 1.

 

Typically, catalogers perform a number of these "value-adding" processes when they are creating or editing bibliographic records for the library's catalog. Figure 2 shows how a MARC cataloging record can add different kinds of value to an information package (Taylor, 1986).

 

The quality of the MARC record is a topic that has received scant attention in the library literature. Yet poor quality (incomplete, inconsistent, inaccurate) records would have a significant impact on the ability of the library to deliver quality services.

MARC rules proscribe the content designation--the machine readable tagging rules (codes and conventions that identify specific data elements) within a record. Anglo-American Cataloging Rules, 2nd Edition or AACRII and other similar cataloging rules used in other countries describe the format and location of descriptive data within a record. Subject heading terms (and formatting) fall outside of descriptive cataloging rules and under subject heading systems/thesaurus conventions. Authoritative agencies such as Library of Congress, etc. determine these conventions. Konovalov (1999) suggests that "subject headings need to be knowledgeably designed and carefully applied by professional catalogers in order to help our customers improve both precision and recall of their searches."

The MARC standard has added value to library service in other ways. It has enabled libraries to share bibliographic records for cooperative projects, e.g., production of union catalogs, union lists of serials, etc. More importantly it has allowed vendors to develop MARC-based software to streamline and support a wide range of library functions. This has meant savings to libraries as they purchase new software or upgrade existing automated library systems.

Given the assumption that MARC records are beneficial to the library, the obvious question arises "What is the value of a MARC record?" To begin addressing this question it is important to understand the nature of value itself.

Value

A number of approaches have been taken in an attempt to value information as Badenoch et. al. (1994) have illustrated in Figure 3:

 

The pragmatic approach taken by economists, who hold that value is the worth of something that contributes to wealth, has yielded the best results. Adam Smith's distinction between "value-in-exchange" and "value-in-use" is one of the foundational principles of economics.

The value-in-exchange theory is most easily understood in terms of prices. People exchange money for products and the price paid is the accepted indicator of the products' value. Accountants and systems analysts have also applied these value-in-exchange theories to the arena of "cost-benefit analysis." King and Schrems (1978) define a benefit as "the consequence of an action that protects, aids, improves, or promotes the well-being of an individual or organization. Benefits take the form of cost savings, cost avoidance, improved operational performance, better allocation of resources, and ‘intangibles,' e.g., better understanding of a particular situation." Benefits are centered on the customers--in libraries the users or the consumers of information system outputs.

When preparing a cost/benefit analysis, dollar values are generally known or can be accurately estimated for the cost side of the analysis. Difficulty typically arises on the benefit side where dollar values are more elusive. Rarely do we hear of "runaway benefits or benefit overruns" (King and Kraemer, 1981). Benefits will vary entirely upon the perceptions of the party seeking to acquire them.

Extending the concept of cost-benefit analysis, Glazer (1993) has suggested a method to calculate the value of a firm's information assets. He demonstrates that the value of information can be aggregated from: the information exchanged as a result of transaction between the firm and its suppliers; the information exchanged as part of transactions within the firm; and the information exchanged between the firm and its customers.

The concept of "exchange value" not only includes the agreed upon price between two parties, but also the time and effort an individual is willing to invest in order to receive the perceived benefits. The "value of information, then, has meaning only in the context of its usefulness to users" (Taylor, 1986).

Cost/benefit is not the only method for quantifying value. Managers are often asked to justify an investment by analyzing the "return on investment" or ROI. The problem with determining a ROI for an information service or for information itself are the unique properties of information that may make monetary measures misleading or inappropriate. For example, the value of information is not diminished when it is used repeatedly or consumed by different individuals.

To overcome the limitations of exchange or price theories of value, economists have developed a second set of theories called value-in-use. The value-in-use or "utility theory" approach allows economists to focus in a more satisfactory manner on wants, usefulness, satisfaction, demands, etc. This may be a better means of addressing the value of libraries and information services.

The fundamental building block of information for any library is the bibliographic record that describes an item in the library's collection. From the user's perspective, however, the value is the information that can be found in the item itself and not the descriptive bibliographic or MARC record (unless you are student looking for a few bibliographic citations to add to a term paper!). Nevertheless, a MARC record does have value and that value will vary, often significantly, depending upon the use, user or "owner" of the MARC record.

Originally, the Library of Congress was interested in distributing MARC bibliographic records because it was felt that if bibliographic records could be shared the costs associated with cataloging could be reduced (Avram, 1975). MARC has become a worldwide "lingua franca" for libraries that want to exchange cataloging data. Shapiro and Varian (1999) have pointed out that people will often assign a greater value (and pay a premium) for timely information, e.g., stock market data, etc. The same also holds true for MARC bibliographic records--online access to the MARC database is generally more expensive than records on CD-ROM or microfilm. The exception is the access to MARC records provided by the Library of Congress, which requires a Z39.50 interface.

Among the earliest providers of Library of Congress MARC records was an organization that today is known as OCLC. Through its international communications network, OCLC now provides access to its database of bibliographic records to more than 34,000 member libraries. Membership entitles a library to search, retrieve and pay for a MARC bibliographic and/or authority record that has been contributed by the Library of Congress or other OCLC member libraries.

If a library fails to find the desired record, OCLC expects the library to produce an original cataloging record and add it to the OCLC database. OCLC recognizes the value of this record by waiving any charge for adding an original cataloging record to the OCLC database. The library does incur direct labor and overhead costs for the production of the original cataloging record.

As of March 1, 1999, OCLC had a bibliographic database that contained 40,871,887 bibliographic records. Of these, fourteen percent were originally created by and distributed by the Library of Congress while seven percent were records created by the Library of Congress and input by participating member libraries. The remaining seventy-nine percent are original cataloging records contributed by the various OCLC member libraries (OCLC Newsletter, 1999). Bibliographic records are being added to the OCLC database at the rate of one every fifteen seconds. The amount of original cataloging that needs to be done by a member library has declined over time as more and more libraries have joined OCLC.

The value of the MARC records found in OCLC's WorldCat database is quite dramatic. Over the ten years between 1989-1998, OCLC has derived some $486,700,200 of income simply by re-selling MARC records to member libraries assuming an average of forty-two percent of revenue was generated by Current Cataloging and RetroCon services (OCLC Annual Report, 1998).

Of course, the staff of OCLC has definitely contributed to the value equation by developing the necessary software and maintaining a computer system to provide access to these MARC records, as well as creating and updating the necessary infrastructure to link a large number of libraries to the OCLC network. Thus it is not reasonable to suggest that the value of the MARC records for OCLC is the total component of the WorldCat cataloging revenue, but it is clear that some proportion of the revenue must be allocated to the OCLC bibliographic database--the MARC records.

Another obvious questions then arises: What proportion of financial revenues should be allocated to the database of MARC bibliographic records and what proportion should be allocated to the system that manages those records? Most libraries decide to join OCLC because of its large and diverse database of bibliographic records rather than the user-friendly software or its communications network. While arbitrary, it would seem safe to assume that seventy-five percent of the value of the OCLC cataloging system should be attributed to the database of MARC bibliographic records. Thus, the value of a MARC record for OCLC might be $8.93 [($486,700,200 x 75%)/40,871,887(number of holdings symbols)]. Within the OCLC system, every library has its holding symbol (a code to indicate who owns a particular item) linked to a bibliographic record.

Another perspective would also suggest that some MARC bibliographic records might be more "valuable" than others. In an attempt to answer this question, the author examined a sample of 1,000 OCLC MARC bibliographic records selected at random across the entire WorldCat database. As shown in the Figure 3, on average the LC contributed records have more holding symbols attached than do the original cataloging records submitted by member libraries. This would suggest that the LC contributed records are, in fact, more valuable for OCLC - they generate more revenue.

 

Does the value of the MARC bibliographic record for OCLC diminish over time? That is, is the record used less and less or do the number of holding symbols added to a MARC record decline over time? Unfortunately, data to answer these questions are not available. Even if the value of the MARC record should decline with time, the availability of other services, e.g., interlibrary loan, may in fact revive or maintain the value of older MARC bibliographic records. This situation may be somewhat analogous to rare/out of print books. For some books, value declines with age; for others the value declines for a while, then goes back up - sometimes surpassing the original value.

Enhanced MARC
Records Some researchers have suggested that enhancing MARC bibliographic records by adding content would improve the success that an OPAC searcher experiences. The main advantage of enhanced bibliographic records, as first articulated by Pauline Atherton (1978) in the study which resulted in a report titled "Books Are For Use," is that the success of the user is increased as the number of keywords is increased. As content is added to the MARC record, the user is more likely to find records that match a search request. Peis and Fernandez-Molina (1998) found that enhanced MARC bibliographic records, using Table of Contents information, increased both precision and recall when searching.

Consider a hypothetical library with a collection of 50,000 titles. On average, it cost the library about $50.00 to purchase and catalog each title in the library's collection [a recent study suggests that the average cost of copy cataloging a title is $16.25 (Morris and Wool, 1999)]. For the purposes of this analysis, costs of the building, shelving, and annual operating costs are ignored. Thus, the library's collection can be valued at $2,500,000.

Assuming a library with 50,000 bibliographic records were to find 20,000 enhanced MARC records when its bibliographic records were matched against the Blackwell enhanced database, this would cost the library $20,000. If success in searching is improved by twenty percent, as the result of the user having access to enhanced MARC records, then use of the collection will likely increase by a corresponding 10 % or more -- the actual percent increase will vary by size and type of library. A Return On Investment or ROI analysis would suggest that the organization has improved its library collection ROI by $250,000 ($2,500,000 multiplied by ten percent). Thus, in this case it is fair to suggest that the value of the enhanced MARC records correspond to the improved library collection ROI of $250,000.

Dividing the improved library collection ROI of $250,000 by the 20,000 enhanced MARC records would suggest that the value of these records is $12.50 per record.

It is also important to note that this kind of analysis is making an important assumption. Namely, that each item borrowed by a library user has value for the user to meet an information need. And personal experience would suggest that this is not always the case.

Value of Holdings Symbols
OCLC reports that, as of March 1, 1999, there are 692,821,411 holding symbols in the OCLC database and there are 34,307 members (OCLC Newsletter, 1999). This means that, on average, there are 18.5 holding symbols per MARC record and an average of 20,377 MARC records linked to each library. Unfortunately, data about the median or maximum number of holding symbols linked to bibliographic records are not available from OCLC.

Using this holdings information, OCLC has been providing an optional Interlibrary Loan service since April 1,1979. Approximately 92 million ILL transactions have been completed using the OCLC Interlibrary Loan system--the Lubbock City-County Library entered the ninety-second millionth transaction on June 30, 1999, while the University of Oregon Library entered the ninety-fisrt millionth transaction on May 10, 1999. The interval between the ninety-first millionth and the ninety-second millionth was only fifty-one days. On average a library is charged eighty cents per ILL transaction--covering the display of holdings and request for service.

What then is the value of the holdings symbol information? Approximately ten percent of the $1.158 billions of revenue generated by OCLC over the last ten years has come from InterLibrary Loan services (OCLC Annual Report, 1998). Dividing this revenue by the total number of holding locations (692,821,411) provides a potential value of 16.7 cents per holding location.

So what then is the value of the "holdings symbol" information? Again, acknowledging the arbitrary nature of the allocation, it might reasonably be suggested that fifty percent of the revenue associated with the Interlibrary Loan system should be attributed to the "holdings symbol" information while fifty percent should be attributed to the computer hardware and software supporting this activity. Thus, the value of the holding symbol location information to OCLC is about eight cents per holding symbol.

Impact on a Library
The value of a MARC record for an individual library would appear to be a fairly straight-forward calculation. Imagine a typical library that has been automated for some time. All of its retrospective conversion has been completed and the library must now obtain MARC records for all of the additions made to the collection each year. The library is able to obtain a machine-readable copy of a MARC record from a commercial or other source for "x" %, e.g., 97%, of the items, which are added each year to the library's collection. Via copy cataloging the library will add its holdings, call number, price and other library-specific information to the MARC record. An individual library receives value when using one of these widely available MARC records since copy cataloging involves less staff time and thus costs less than original cataloging.

For the remaining portion of the materials being added to the collection, the library will need to do original cataloging. The cost of original cataloging varies considerably, but for the purposes of this article an estimate of $50.00 per record is used.

Total cost of cataloging = Cost of shared cataloging [($/record x % of collection additions/year) + staff costs to select and use the shared cataloging records] + Cost of original cataloging [% of collection additions/year x staff costs to create the original cataloging records].

For purposes of illustration, consider a library that adds 11,000 titles to its collection each year. Assuming that the library finds shared cataloging records for 95.5% or 10,500 titles, the costs would be $1.75 (cost of the MARC record) x 10,500 titles + staff costs ($20 per title x 10,500 titles) or $228,375. The costs for all of the needed original cataloging would be equal to $50 per title x 500 titles or $25,000. Thus, the total costs for cataloging this library's annual additions to its collection would be $253,375 or $23.03 per bibliographic record.

To identify the potential savings by using a shared cataloging service, it is necessary to calculate the costs of original cataloging for all 11,000 titles. If this were done then the costs would be $500,000 (11,000 titles x $50 per title). Thus, the savings would amount to $296,625. It is important to note that these are not true savings but cost avoidance. The library is not going to have an additional $300,000 to spend for additional materials or new library services.

Yet, the value of these MARC records remains. In the case of our illustration library, the value would be the $296,625 divided by the 11,000 titles or $26.97 per record!

But the value of the MARC record must be more than the cost savings that accrue to a specific library during the cataloging process. This must be so since the library, as an institution, and the individual user, is not concerned with the savings but rather with the level of access that the MARC bibliographic and other types of records afford when the user (library customer or staff member) is searching the library's catalog.

The vast majority of automated library systems installed in the majority of North American libraries today provide both phrase or browse searching, e.g., author, title, subject, call number, etc. as well as keyword/Boolean searching, e.g., limited to a specific field or across the complete record.

Our illustrative library has a collection valued at $2,500,000 as previously noted. The use of the collection ROI would be based on the number of annual circulation x $26.97 per volume per year. In this case, the use of the collection ROI would be $449.50 [($2,500,000 divided by an annual circulation of 150,000) times $26.97]. However, it should be noted that a use of the catalog ROI is not exactly equivalent to a library's collection ROI for the following two reasons: a) not all materials checked out are found by consulting the library's catalog (patron's browse the shelves directly), and b) not all use of the library's materials is reflected in the annual circulation statistics (in library usage).

MARC Authority Records
Research has shown that there is a direct correlation between the number of cross-references and the success a patron experience's while searching of the library's catalog (Mandel and Herschman, 1983). Thus, an authority control component of an automated library system, with its associated cross references, is an important tool that a library can use to improve the success rate experienced by its users while searching the OPAC.

Until a library's database reaches more than 150,000 bibliographic records, one rule of thumb suggests that there is a 1 to 1 correlation between the number of bibliographic records and the corresponding number of authority records. Then, as the size of the database grows, the number of authority records will decline. A library with one million bibliographic records should expect to load approximately 650,000 related authority records. The number of authority records will also vary by type of library--public libraries will have fewer authority records due to a larger proportion of its collection being fiction.

Currently, the Library of Congress Name Authority file contains about 4,560,000 records and there are about 245,000 Subject Heading records (LCSH). About sixty percent of the Subject Heading authority records contain see and see also cross-references.

If success in searching is improved, e.g. 20%, as the result of the user having access to cross-references, then use of the collection will likely increase by a corresponding 10 % -- the actual percent increase will vary by size and type of library. Then a Return On Investment or ROI analysis would suggest that the organization has improved its library collection ROI by $500,000 (collection valued at $5,000,000 divided by 10%). Thus, it might be suggested that the value of the MARC authority records correspond to the improved library collection ROI of $500,000? Assuming the library has 100,000 authority control records and its collection ROI has increased $500,000, then the value of these MARC authority records would be $5.00 per record.

Value of an Index
Improving the library OPAC users success--by identifying more of the library's collection that would be relevant to an information need--improves the rate of return on a library's collection. One way to improve user success is to provide new or improved indexes. If such advanced searching tools improve success by twenty percent, using our hypothetical library, the value of these advanced searching capabilities would be equal to $1,000,000 (assuming a ten percent increase in the use of the collection and the base value of the collection being $10,000,000).

Location & Status Information
One piece of information that is likely to have a high value for library users is location/status information. If the library patron is further able to ascertain the location and status of an item before actually visiting the library, then it might also be reasonable to assign value based on the time saving or value of time for the library patron.

In the corporate library environment, consider an automated library system with an OPAC accessible via a local area network (LAN) or an Intranet. The location and status information is a by-product of an automated circulation control system. The corporate library's users have access to the library's catalog and hence the library's MARC bibliographic and/or authority records. The library user is able to conduct a search and identify a set of records of potential value. Using the OPAC, the library user is further able to identify the location and current status of the desired item(s). Discovering this information online, the user typically spends five minutes conducting the search and noting the location and status information. The required "information package" might even be located down the hall in a colleague's office rather than in the library.

Were the library's catalog along with its associated location and status information not accessible online, then the user would need to visit the library in order to determine this information. The user would, on average, spend twenty minutes on this process.

For professionals, a conservative average cost of their time (with benefits) might be $60 per hour (or $1 per minute). Given an online system, the organization would save $15 per library visit. If the library received 100 library online visits per day, then the organization would save $1,500 per day or $7,500 per week or $390,000 per year. Thus, it would appear that the location and status information, in conjunction with a MARC record, is very valuable to some organizations. Assuming our sample library with 100,000 volumes, then the value of the location and status information would be $3.90 per volume.

Another approach to establishing the value of location and status information is to identify the increased control over the library's collection afforded by the automated system. Assuming a hypothetical library with an annual circulation of 50,000 items and costs of $50 to purchase and catalog a title. A library with a manual circulation control system will typically experience losses of three percent or losses of 1,500 titles per year (Baker & Lancaster, 1991). With an automated system the amount of circulation loss can be reduced to one-half of one percent or 250 items per year. Thus the savings associated with the automated circulation control system with its location and status information amount to $62,500 (1,250 titles not "lost" in circulation x $100). $62,500 divided by the 50,000 annual circulation in our hypothetical library provides a value of $1.25 per item.

Value for an Individual
Typically the library patron does not think about nor appreciate the data and information contained within a MARC record and thus, by implication, the value of the records found in the library's catalog. The patron is using the library's catalog to get pointed in the right direction to find a specific item or to browse for items that might be of value.

If the library patron is a student, professor or professional, then the bibliographic citation may be important and thus the patron will find portions of the MARC record itself to have some value. This value is reflected in the tens of thousands of individuals who have personal bibliographic software, which allows the user to download MARC records in order to maintain personal libraries and the production of bibliographies. These MARC records then assist the individual as they write papers, articles and books. It is nonetheless very difficult to assign a value, even a time value, of this improved productivity for any specific individual.

Should the library's collection not contain an item being sought be the patron, then the patron may request that the library borrow the desired item from another library. The accuracy of the bibliographic citation then becomes very important; thus the value of the MARC record itself may be higher in an interlibrary loan (ILL) transaction.

Conclusion
A library creates and maintains a catalog as a tool to assist users in finding information resources. The value that has been added indirectly to the information package or directly to the library's catalog as a result of the cataloging process is difficult to determine. It has been demonstrated that the value of information itself, particularly the value of a MARC record will vary significantly--depending upon the provider, library or user of the record.

Additional value also results since bibliographic record providers usually generate a variety of print and other products for specific member customer libraries, which result in additional revenue. As libraries increasingly provide access to electronic materials, whether text, images, audio or video files, the role of the library's catalog to provide accurate and robust access to this information will become increasingly important. Thus, while a library's physical collection may decline in coming years, the value of the library catalog to provide access to both the library's physical collection and to electronic resources will only increase. Understanding the actual value of the information components of a library's catalog will allow librarians to be better informed about the costs and benefits of maintaining or enhancing a library's catalog.

Joe Matthews is Vice President of Business Development at EOS International. He may be reached at jmatthews@eosintl.com.

 

References
Atherton, P. (1978). Books are for use: Final report of the Subject Access Project to the Council on Library Resources. Washington, DC: Council on Library Resources.

Avram, H. D. (1975). MARC: Its history and implications. Washington, DC: Library of Congress.

Badenoch, D., C. Reid, P. Burton, F. Gibb and C. Oppenheim. (1994). The value of information in The Value and Impact of Information, Feeney, M. and M. Graves (Eds.). London: Bowker Saur, 9-77.

Baker, S. L. and F. W. Lancaster, (1991). The measurement and evaluation of library services. 2nd edition. Washington, DC: Information Resources Press.

Bates, M. (1989). Rethinking subject cataloging in the online environment. LRTS, 33(4), 400-412.

Borgman, C. L. (1996). Why are online catalogs still hard to use? Journal of the American Society for Information Science, 47(7), 493-503.

Broadbent, M. (1992). Demonstrating information service value to your organization. Proceedings of the IOLIM Conference, 16, 65-83.

Burk, C. F. and F. W. Horton (1988). InfoMap: A complete guide to discovering information resources. New Jersey: Prentice Hall.

Chan, L. M. and D. Vizine-Goetz, (1995). Feasibility of a computer-generated subject validation file based on frequency of occurrence of assigned LC Subject Headings. Annual Review of OCLC Research. www.oclc.org/oclc/research/publications/review95/part1/chan.htm

Cummings, M. (1986). The economics of research libraries. Washington, DC: Council on Library Resources.

Cutter, C. (1904). Rules for a Dictionary Catalog. 4th Edition. Washington, DC: Government Printing Office.

Dillon, M. and P. Wenzel (1990). Retrieval effectiveness of enhanced bibliographic records. Library HiTech, 8(3), 45-46.

Drabenstott, K. M. and D. Vizine-Goetz (1994). Using subject headings for online retrieval: Theory, practice, and potential. San Diego: Academic Press.

Feeney, M. & Grieves, M. (Eds.) (1994). The value and impact of information. London: Bowker, Saur.

Furrie, B. (1998). Understanding MARC - Bibliographic. Washington, DC: Library of Congress.

Glazer, R. (1993). Measuring the value of information: The information-intensive organization. IBM Systems Journal, 32(1), 99-110.

Griffiths, J. and D. King, (1993). Special Libraries: Increasing the information edge. Washington, D.C.: Special Libraries Association. International Federation of Library Associations (1961). International Conference on Cataloguing Principles. London: IFLA.

Kantor, P. (1995). Savings to the nation resulting from the existence of LC cataloging records. A consulting report. Washington, DC: Library of Congress.

Keyes, A. (1995). The value of the special library: Review and analysis. Special Libraries, 86, 172-187.

King, D. W., N. K. Roderer, & H. A Olsen. (1983). Key papers in the economics of information. White Plains, NY: Knowledge Industry.

King, J. L. & K. L. Kraemer (1981). Cost as a social impact of information technology. In N. L. Moss (Ed.), Telecommunication and Productivity. Reading, MA: Addison-Wesley.

King, J. L. & Schrems, E. L. (1978). Cost-benefit analysis in information systems development and operation. Computing Surveys, 10, 19-34.

Koenig, M. E. D. (1990). Information services and downstream productivity. Annual Review of Information Science and Technology, 25, 55-86.

Koenig, M. E. D. (1992). The importance of information services for productivity. Special Libraries, 83, 199-210.

Konovalov, Yuri (1999). Cataloging as a customer service: Applying knowledge to technology tools. Information Outlook, 3(9), 25-27.

Machlup, F. (1979). Uses, value, and benefits of information. Knowledge: Creation, Diffusion, Utilization, 1, 62-81.

Marshall, J. (1993). The impact of the special library on corporate decision-making. Washington, D.C.: Special Libraries Association.

McGee, J. and L. Prusak (1993). Managing information strategically. The Ernst & Young Information Management Series. New York: John Wiley & Sons, Inc.

Morris, D. E. and G. Wool (1999). Cataloging: Librarianship's best bargain. Library Journal, 124(11), 44-46.

OCLC Annual Report (1998). Dublin, OH: Online Computer Library Center. OCLC Newsletter (1999). Dublin, OH: Online Computer Library Center, March/April.

Peis, E. and J. C. Fernandez-Molina (1998). Enrichment of bibliographic records of online catalogs through OCR and SGML technology. Information Technology and Libraries, 17(3), 161-172.

Saracevic, T. & P. B. Kantor (1997a). Studying the value of library and information services. Part I. Establishing a theoretical framework. Journal of the American Society for Information Science, 48 (6), 527-542.

Saracevic, T. & P. B. Kantor (1997b). Studying the value of library and information services. Part II. Methodology and taxonomy. Journal of the American Society for Information Science, 48(6), 543-563.

Slamecka, V. (1970). Methods and research for design of information networks. Library Trends, 18, 551-568.

Smith, K. W. (1998) OCLC: Yesterday, today and tomorrow. Journal of Library Administration, 25 (2/3 and 4), 251-270. Statistical Report '98 (1998).

Statistical Report '98: Public Library Data Service. Prepared by the Public Library Association, a division of the American Library Association. Chicago: American Library Association.

Taylor, A. (1999). The organization of information. Englewood, CO: Libraries Unlimited.

Taylor, R. S. (1986). Value-added processes in information systems. Norwood, NJ: Ablex. Walker, R. Ed. (1993).

AGI Standards Committee GIS Dictionary. Association for Geographical Information.

 

Information Outlook Main Page | This Issue's Table of Contents | Back Issues of Information Outlook
SLA Home Page | Join SLA Now | Feedback | Search

• Executive Connections

  Guest Blog: SLA President Gloria Zamora

• SLA Blog

  David Shumaker Honored as Exceptional Educator with 2009 Rose Vormelker Award

• Public Policy Connections

  Justice Department Submits Views on Amended Google Book Search Settlement