BIBCO Core Record Study: Final Report

Prepared for the PCC Policy Committee

By David Banush,

Cornell University Library
June 11, 2001

I. Executive summary

• Findings

  The BIBCO core record, while intended to provide cataloging at an acceptable level of quality and detail, with the intention of increasing the pool of usable copy, has been less widely implemented than initially hoped. The vast majority of BIBCO participants continue to contribute mostly or exclusively full-level records. David Banush of Cornell University undertook a qualitative study of cataloger and manager attitudes toward the core record, with the stated goal of assisting the PCC Policy Committee in its planning for and marketing of the BIBCO program.

  The study consisted of in-depth telephone interviews with 20 catalogers and 20 managers from BIBCO institutions. Diane Cellentani of Marketing Backup conducted the interviews, which took place between March 19 and March 29, 2001. Marketing Backup tabulated the data and prepared a preliminary report on the findings. David Banush further analyzed the data for this report. The results of the analysis suggest the following:

  • Managers are happier with core than catalogers, and overall there is less reported dissatisfaction with core than the BIBCO statistics and anecdotal evidence would suggest.
  • Despite the reported levels of satisfaction, both catalogers and managers are divided about the benefits and problems of core. For example, respondents are almost equally divided on the time savings resulting from core in original cataloging. Nearly as many respondents were satisfied with the quality of core records as not, and many feel that core provides inadequate access to materials.
  • The primary elements of a quality bibliographic record, as identified by the respondents, are all present in the core record. Both catalogers and managers define quality bibliographic records primarily in terms of adherence to rules and standards. A widely shared definition of quality from a user's perspective does not emerge in the survey results. Further, there is evidence that the more expansive view of quality promoted in BIBCO training--which includes the notions of timeliness and cost-effectiveness--has not been accepted by cataloging professionals.
  • An interesting dichotomy exists between the use of core in copy cataloging and in original cataloging. While most catalogers and managers are happy to find and use core records, far fewer are interested in creating them. Even among those who claim to be satisfied in general, as well as among those who are satisfied with particular aspects of the core concept, there exists a palpable unease with broader use of core records in original cataloging.
  • Both catalogers and managers have a number of concerns about the program's implementation, marketing, and management. Participants voice dissatisfaction with the utilities, particularly OCLC, and their handling of BIBCO records. Many believe training and documentation are inadequate, and a number of those surveyed question the current recruiting and publicity strategies of the program.
  • Misunderstandings and false perceptions about what the core guidelines actually require continue to linger. A number of those interviewed, for example, believe that use of core prohibits more than one added entry or subject heading. Such perceptions may be related to the dissatisfaction with training and documentation.
• Recommendations.

  A full report, including recommendations, has been forwarded to the PCC Steering Committee for consideration as they prepare the agenda for this November's PCC Policy Committee meeting at the Library of Congress. In this report, the recommendations are arranged into three distinct groups. Each group of recommendations offers different degrees of possible change to the BIBCO program. The recommended changes are intended to assist in making the program more attractive to both present and potential members, as well as more relevant to the changing needs of the library community in the medium and long terms.

II. Background, rationale, and significance of the study

The BIBCO core record was intended to provide cataloging at an acceptable level of quality and detail while also increasing both the speed of cataloging and the pool of usable copy. Few libraries, however, have used the core record in their original cataloging. The vast majority of BIBCO participants continue to contribute mostly or exclusively full-level records. In LC's fiscal year 2000, only about one-third of BIBCO contributions were core records. In the current fiscal year (as of May 2001), the level of core contributions has actually dropped to about 27% of total output.

No quantitative or qualitative research has yet investigated the possible reasons underlying the cataloging community's response to the concept of the core record. To explore some of the reasons behind the low use of core, David Banush of C ornell University proposed a study on catalogers' attitudes toward the BIBCO core record. He developed a research proposal and submitted it to the PCC Policy Committee in November 2000. The Committee subsequently endorsed the proposal, with revisions. (The text of the final proposal is available online at the BIBCO home page. The study moved from proposal to reality in January 2001, after the project was generously funded by Sarah Thomas, University Librarian at Cornell.

The study had the following goals:

• To explore working catalogers' attitudes toward the core record
• To explore cataloging managers' attitudes toward the core record
• To examine both groups' understanding of the issue of cataloging quality
• To help identify strengths and weaknesses in the current PCC strategies for promoting BIBCO membership and core record creation.
• To provide the PCC Policy Committee a list of specific recommendations to address perceived weaknesses in the marketing of, and planning and training for, the BIBCO program.

To achieve these ends, the study outlined the following objectives:

• Identify an appropriate pool of candidates for interviews.
• Engage the services of a consultant to assist in the design of the interview scripts.
• Conduct in-depth telephone interviews with approximately 20 catalogers and 20 cataloging managers from BIBCO institutions.
• Collect and analyze the resulting data and write a report of the findings, with the consultant's assistance.
• Present the findings and a list of specific action items to the PCC Policy Committee
• Disseminate the findings through publication in the professional literature.
• Prepare the groundwork for further qualitative or quantitative investigation of cataloger attitudes and core records.

The consultant employed for this project was Diane Cellentani of Marketing Backup, a research-consulting firm based in Columbus, Ohio. A professional researcher with over 20 years of experience in feedback research, Ms. Cellentani has conducted over 250 focus group and interview research projects on library issues. She holds an MBA in marketing. Her previous clients include Cornell University Library and OCLC.

III. Methodology

The methodology used was a series of in-depth telephone interviews with 20 working catalogers and 20 cataloging managers from BIBCO institutions. For the purposes of this study, managers were defined as those having primary responsibility for decisions relating to BIBCO in their institutions. Those interviewed were recruited via invitation on PCC listservs, with an additional follow-up invitation going out to BIBCO liaisons. All potential participants subsequently received a short questionnaire for screening purposes. The screening questionnaire was developed in consultation with members of the Policy Committee.

Recruiting and screening began February 14, 2001; the process was complete by March 15. In total, there were 47 volunteers. In some cases, there were multiple volunteers from several institutions; in those instances, the first person in each category (cataloger or manager) reached by telephone was included in the survey. The final group of 40 was drawn from 29 institutions, representing 70% of BIBCO libraries. Interviews began on March 19 and were complete March 29.

Each interviewee was asked a series of prepared questions designed to gather the required feedback. The questionnaire was developed in consultation with members of the Policy Committee, and was pre-tested with several volunteers before being finalized. Interview sessions were flexible, allowing further probing of responses where appropriate. The length of the interviews ranged from 15 to 70 minutes. Interviewees' identities and responses have been kept strictly confidential.

Diane Cellentani, the consultant, assisted in the developing the structure of the questionnaire. Her firm, Marketing Backup, conducted the interviews. Marketing Backup also prepared an initial analysis of the data and submitted a preliminary report on the results. That report has been further analyzed; it serves as the basis for this document.

Telephone interviews were chosen because they are one of several demonstrated methods to collect reflective, in-depth information about a specific topic. While other methods of gathering this type of data (in-person interviews, focus groups) offered advantages for research of this nature, the logistics and expense of planning and scheduling focus groups rendered those methods impractical for this study. Phone interviews provided many of the advantages of other methods while also permitting a more flexible timetable and a practical means of reaching a widely dispersed population.

This was a qualitative study. Statistically representative, quantitative data could have been gathered through other techniques, most notably survey research. However, the limited range of possible responses in that kind of survey would not have provided the sort of open-ended feedback that could most optimally assist the PCC Policy Committee in assessing the current marketing of the BIBCO program and planning for its future. Moreover, the length of time required to plan, execute and analyze the results of a mailed survey would have been considerably greater than telephone interviews, and the cost of such a survey significantly higher.

The choice of front-line catalogers and cataloging managers enabled a clear focus on those library staff most closely involved in the process of setting cataloging policies and creating records. Comparing and contrasting the attitudes of these two key stakeholders provided insights into the differences between working catalogers and their supervisors. These comparisons should assist the Policy Committee in its planning and decision-making.

The concentration on current BIBCO participants permitted a more narrowly focused exploration of why institutions already committed to the program have failed to utilize the core guidelines more frequently. The insights gathered from this group will be used by the PCC Policy Committee to assess current approaches to BIBCO participation and encourage greater use of the core standard among these members, as well as address concerns of members regarding the perceived shortcomings of core records. The data gathered may also be valuable in planning efforts for the recruitment of additional participants to the program.

IV. Survey findings

Participants' profiles

As noted above, participants came from 29 of the 42 current BIBCO institutions. The institutions represented were:

Arizona State University, Brigham Young University, Cleveland Public Library, Columbia University, Cornell University, Harvard University, Joint Forces Staff College, National Agricultural Library, National Library of Medicine, Northwestern University, New York University, Oberlin College, Princeton University, Queens Borough Public Library, St. Louis University Law Library, Stanford University, UC Berkeley, UCLA, University of North Carolina-Chapel Hill, University of Chicago, University of Colorado, University of Dayton, University of Florida-Gainesville, University of Maryland, University of New Mexico, University of Texas-Austin, University of Wisconsin-Madison, Vanderbilt University, and Yale University

21 (72.5%) of these institutions contribute at least some core records; 8 (27.5%) do not. These figures closely parallel the overall BIBCO production profile for the current fiscal year, in which 30 of 42 institutions (71%) have contributed at least some core records. Of those that do contribute, several (e.g., Yale, Brigham Young, Colorado) produce very small numbers of core records.

The length of time these institutions have been BIBCO members ranged from 1 year to the program's onset (6 years). The average length of participation was 3.4 years.

Catalogers' seniority ranged from 3 to 25 years; the average was 13.5 years. Their length of time as BIBCO contributors ranged from just over 5 years (recorded as 5 years) to 1 year. The average length of time as contributor was just over 3 years.

There appears to be no clear correlation between catalogers' seniority and their attitudes toward the core record. Neither were there any manifest differences between the attitudes of catalogers or managers working in longer-term BIBCO member institutions and those who have joined more recently.

General comments

In general, the results suggest the following:

• Managers are happier with core than catalogers, and overall there is less reported dissatisfaction with core than the BIBCO statistics and anecdotal evidence would suggest.
• Despite the reported levels of satisfaction, both catalogers and managers are divided about the benefits and problems of core. For example, respondents are almost equally divided on the time savings resulting from core in original cataloging. Nearly as many respondents were satisfied with the quality of core records as not, and many feel that core provides inadequate access to materials.
• The primary elements of a quality bibliographic record, as identified by the respondents, are all present in the core record. Both catalogers and managers define quality bibliographic records primarily in terms of adherence to rules and standards. A widely shared definition of quality from a user's perspective does not emerge in the survey results. Further, there is evidence that the more expansive view of quality promoted in BIBCO training--which includes the notions of timeliness and cost-effectiveness--has not been accepted by cataloging professionals.
• An interesting dichotomy exists between the use of core in copy cataloging and in original cataloging. While most catalogers and managers are happy to find and use core records, far fewer are interested in creating them. Even among those who claim to be satisfied in general, as well as among those who are satisfied with particular aspects of the core concept, there exists a palpable unease with broader use of core records in original cataloging.
• Both catalogers and managers have a number of concerns about the program's implementation, marketing, and management. Participants voice dissatisfaction with the utilities, particularly OCLC, and their handling of BIBCO records. Many believe training and documentation are inadequate, and a number of those surveyed question the current recruiting and publicity strategies of the program.
• Misunderstandings and false perceptions about what the core guidelines actually require continue to linger. A number of those interviewed, for example, believe that use of core prohibits more than one added entry or subject heading. Such perceptions may be related to the dissatisfaction with training and documentation.

Satisfaction with Core Records

Overall satisfaction

Respondents were asked to rate their overall satisfaction with core records on a 5-point Likert scale, ranging from very satisfied to very dissatisfied. Managers reported a much higher degree of satisfaction than did catalogers. 18 of the 20 managers (90%), and 12 of the 20 catalogers (60%), report being very satisfied or satisfied with core records. 2 managers (10%) and 5 catalogers (25%) reported being neither satisfied nor dissatisfied, while 3 catalogers (15%) indicated they were either dissatisfied or very dissatisfied with core records overall.

Yet the complete picture is more complicated than the figures alone would suggest. Respondents were asked an open-ended follow-up question in which they elaborated on the reasons for their response. In the open-ended responses, 30% felt that core records did not provide sufficient access, 22.5% felt that core did not yield any time savings, and 10% felt that core was simply not as good as full. Indeed, even among catalogers and managers who reported being at least satisfied overall, there was skepticism or reluctance expressed about core records.

Selected comments:

• "My problem is, I don't want to change because I feel more is better."-- Manager, very satisfied overall
• "The records have all the information that is needed. I appreciate the authority work that is done."--Manager, very satisfied overall
• "I don't find time savings."--Cataloger, very satisfied overall
• "Catalogers say creating core doesn't save significant time."--Manager, very satisfied overall
• "I'm not certain that [core records] are allowing appropriate access."--Manager, satisfied overall
• "We do not create core records. We use core records. The quality is generally good [with core] ... main problem is that records don't include as much subject analysis as we would like."--Manager, satisfied overall
• "The [core records] I have used did not have to be modified ... We have not begun creating core records ourselves ... It is only just a little more work to do full instead of core."--Cataloger, satisfied overall

Satisfaction with Specific Features of Core Records

Participants were asked to rate their satisfaction with the following features of core records:

• Quality of core records created by other libraries
• Timeliness of new core records in bibliographic utilities
• Time savings for performing copy cataloging
• Time savings for performing original cataloging
• Usefulness of core records to end-users
• Adequacy of training and documentation for core record creation

70% of respondents indicated that they were satisfied or very satisfied with the quality of records created by others. 60% stated that they were at least satisfied with the time savings for copy cataloging, and nearly as many (57%) were satisfied or very satisfied with the time savings for original cataloging.

Managers were more likely to be very satisfied than catalogers with quality of records produced by other libraries (35% versus 10%), time savings for copy cataloging (45% versus 20%), and usefulness of records to end users (35% versus 25%). On the other hand, catalogers were more likely than managers to be very satisfied with the time savings for original cataloging (25% versus 10%). However, as many catalogers were either dissatisfied or very dissatisfied with the time savings involved for original cataloging as were very satisfied. Only 10% of managers were dissatisfied with the time savings, and none were very dissatisfied.

Nearly a third of catalogers were dissatisfied with the adequacy of training and documentation, over twice the number of managers giving that response. 55% of both catalogers and managers felt they could not report their levels of satisfaction with timeliness of records in the utilities, suggesting either that they have little opportunity to assess the timeliness of the records, or that there are insufficient numbers of records on which to base an assessment.

Those who were dissatisfied or very dissatisfied with any feature of core records were asked why. The largest number of dissatisfied responses had to do with training and documentation. Most respondents felt that the training emphasized the selling points for using core and spent too little time on what a core record is and how it differs from a full record. Some respondents felt the approval process was too long, while others mentioned that the documentation needed to be more complete.

Defining quality bibliographic records

One of the principal issues surrounding the use of core records has been the issue of quality. Among the goals of BIBCO has been to encourage cataloging professionals to consider quality from a user's perspective. The program has sought to balance timeliness and cost-effectiveness with reasonable access, and to move away from a view of quality that adheres narrowly to the rules of bibliographic description. The more expansive view of quality has not, however, been widely accepted, and there have often been questions raised about the quality of core records.

To get a better sense of what managers and catalogers in BIBCO libraries view as a quality record, participants were asked to give their definition of a quality bibliographic record. The interviewer provided no prompts on this question; respondents were free to name any elements they found necessary for a quality record.

The two major points that emerged were accurate description and sufficiency of access points. Managers were slightly more concerned with the appropriateness of subject analysis than other elements of the description. 85% of managers and 75% of catalogers felt that succinct, pertinent subject headings were an essential trait of a quality record -- the highest number for any given characteristic. Overall, catalogers were more concerned with the fullness of records. 60% of catalogers (but 45% of managers) felt that a quality record must contain all appropriate access points, including subjects, series, and added entries.

The results suggest that managers were more concerned than catalogers with general accuracy of description. 50% of catalogers and 70% of managers noted that the record must match the item in hand or otherwise accurately describe it. However, absolute accuracy of description was more important to catalogers than to managers. 7 catalogers (35%) noted that a quality record must contain no typographical or ISBD punctuation errors; no managers mentioned that characteristic. A majority of both catalogers and managers (50% and 60%, respectively) noted that a quality record must follow national descriptive standards (i.e., AACR2R, LCRIs).

Authority work was the third most frequently mentioned element of a quality record; 50% of managers and 40% of catalogers noted that completed authority work was an essential trait. Interestingly, however, only 20% of managers and 25% of catalogers cited having completed authority work as a benefit of core records.

No managers or catalogers in this survey mentioned either timeliness or cost-effectiveness as elements of a quality record. Indeed, the responses suggest that adherence to the rules remains the central concern for both managers and working catalogers in BIBCO libraries. The more expansive view of quality promoted in BIBCO training has yet to gain significant support in the professional cataloging community.

Use of core records in cataloging

Original cataloging

Despite noting a high overall satisfaction rate with core records, only about 22% of respondents almost always perform original cataloging at the core level. Nearly twice that number (40%) almost never catalog at the core level. Most respondents only cataloged monographs at the core level, though a few did handle other formats, notably videotapes.

Respondents were asked why they did or did not use core in their original cataloging. Among those who reported almost always using core, several mentioned that library policy guided their decisions. In those institutions where the choice between core and full was left completely to the cataloger, use of core was far less common. Many participants noted that certain categories of materials or formats were excluded by institutional policy from core-level cataloging, thus hindering their use of it. Other barriers to further use of core included reaction of public services colleagues and the perception that core hinders access to end-users (despite the finding that 70% of interviewees also said they were satisfied or very satisfied with the usefulness of core records to end users.) The perceived lack of time savings was also cited by a number of participants. Some selected comments:

• "Our directive from above was to use core. My manager told us to catalog at core level."--Cataloger, almost always creating core.
• "We catalog less than a quarter of our work at core level. We don't catalog at core level if we do not feel comfortable or unsure when we need to add the authority record for a series, when the material is ephemeral [or] when we catalog non-book formats . . . "-- Manager, sometimes creating core.
• "I perform original cataloging for 2 major subject areas: music and Latin American studies . . . we need the best access to this material." -- Cataloger, almost never creating core
• "At this particular library, we have not gotten complete buy-in from public services . . . we need to convince public services."--Manager, almost never creating core.
• "After training, people couldn't see that much difference between core and full . . . we were not convinced to use core cataloging."--Manager, almost never creating core.

Copy cataloging

Most of the respondents accept core records as they find them: only a handful (10%) report making changes to them. 95% of managers and 80% of catalogers reported that they accept core records without change. 60% of respondents reported being very satisfied or satisfied with the time savings of core for copy cataloging, and 70% said they were very satisfied or satisfied with the quality of core records created by others. These findings stand in stark contrast to the use of core records for original cataloging.

When asked what changes they made to core records they encountered, most respondents noted that they added notes (504s and 505s, as well as 500s), additional 7XXs, and additional 6XXs. Others added or modified series statements to reflect local practices.

Future directions, enhancements, and changes to BIBCO

In an open-ended question, respondents were asked what they would recommend for PCC and BIBCO to encourage broader acceptance of core level bibliographic records and to increase BIBCO contributions and membership. There was no clear consensus on specific changes or strategies, with the exception of training and documentation. 60% of respondents felt that training and supporting materials for the program were unsatisfactory. Others suggested educating librarians about core and BIBCO through better training and outreach, especially at the state and local levels. 30% noted that support from the utilities, especially OCLC, as well as from LC, was lacking. 35% cited the need for further studies on time savings and quality of BIBCO records, while 10% noted that outreach to public services is insufficient. Some selected comments:

• "De-emphasize core . . . Good leadership hasn't existed for BIBCO. There needs to be more money . . . to further strategic goals of documentation, training, meetings, and the website . . . BIBCO should have a presence at any professional group." -- Cataloger.
• "To improve BIBCO, we need better documentation and a forum for ongoing Qs & As for participants. I want an A-Z manual on everything for BIBCO."--Cataloger.
• "Give higher OCLC cataloging credit to those who create PCC core or full records." -- Manager.
• "We should get paid for OCLC credit when we catalog at PCC [level]. It is frustrating when a library creates an acquisition record and they get credit . . . [Libraries] need more training at local areas . . ." -- Cataloger.
• "Create records only once on our local systems and load them nationally. Why do we have to re-input into OCLC? Also, we need to make it easier to change records on OCLC." -- Cataloger.
• "A lot more libraries would accept core if pricing breaks were available for creating core records." -- Cataloger.
• "Promote at the local level . . . Perhaps have more meetings at [state] conventions . . . There is a perception that BIBCO is an elitist group."--Manager.
• "Major failure is not sharing among OCLC and RLIN . . . PCC should work with [library] directors. They need top-down support for core." -- Manager
• "PCC could also get visibility from LC. Anytime LC can mention core level it would be beneficial. If core is associated with LC, it would enhance the image of core." -- Manager.

V. Conclusions

The responses suggest decidedly ambivalent attitudes toward core records. While many respondents indicated general satisfaction with core, and were mostly satisfied with specific features of the records, the open-ended comments revealed considerable uneasiness about the creation of core records and the support mechanisms (training, documentation, utilities' cooperation) underlying the BIBCO program. Clearly, the program's emphasis on cataloger judgment remains problematic for many participants. The "more, better, faster" message has not been well-received, particularly by front-line catalogers, and even many managers remain skeptical that the cost-benefit analysis comes out in favor of core records.

It is hoped that the study's results and the recommendations that follow will provide the PCC Policy Committee with useful information with which to proceed as its members plan for the future of the BIBCO program

VI. Recommendations.

A full report, including recommendations, has been forwarded to the PCC Steering Committee for consideration as they prepare the agenda for this November's PCC Policy Committee meeting at the Library of Congress. In this report, the recommendations are arranged into three distinct groups. Each group of recommendations offers different degrees of possible change to the BIBCO program. The recommended changes are intended to assist in making the program more attractive to both present and potential members, as well as more relevant to the changing needs of the library community in the medium and long terms.