College and Research Libraries Implementing Technological Change Peggy Johnson This article addresses some problems associated with technological change. The author exam- ines the nature of data and technology, organizations, and computer-mediated work. Concepts of sociotechnical systems design are related to library automation. Finally, a set of strategies for facilitating technological change is proposed. "It should be borne in mind that there is nothing more difficult to carry out, more doubtful of success, than to initiate a new order of things. "-Niccolo Machiavelli, The Prince nyone who has initiated changes in a library recognizes the truth of Machiavelli's state- ment. From moving the place- ment of a spine label, to implementing AACRII, to installing an automated sys- tem, innovations will likely meet some re- sistance. Automation is probably the most difficult change to implement, because it may alter virtually every library function, as well as the basic organizational struc- ture. In a study of 117 ARL (Association of Research Libraries) libraries, the introduc- tion of automated systems was the most frequently mentioned force contributing to organizational change. 1 More than a thousand libraries have installed auto- mated systems since 1973. 2 Even more have made automation a part of their op- erations through participation in national bibliographic utilities. The internal changes resulting from technological in- novation are legion, as are the number of staff who resist them. Introducing new technologies or re- vamping existing applications will always produce change. Change, however, does not necessarily lead to resistance, tension, and conflict. The intent of this paper is to propose a set of strategies that will aid li- brarians in the introduction of technologi- cal innovation. In order to provide a con- text for these strategies, data, technology, organizations, and computer-mediated work are examined. Concepts of so- ciotechnical design are then related to li- brary automation, and implementing strategies are discussed. SECTION I Politics of Data and the Power of Technology This is an age of extensive and increas- ing automation. While visible in banks, grocery stores, and libraries, it is often in- visible elsewhere. Much has been written about the speed of change and the amount of information being generated. Lucinda Conger calls the result ''cognitive whip- lash. " 3 Widespread automation has be- come possible because of a combination of explosive growth in computer power and advances in telecommunications, both at relatively low cost. Libraries generally turn to computers for relief from rising costs and increasing backlogs. They use the new technology to meet economic need and also to provide Peggy Johnson is Assistant Director, St. Paul Campus Libraries, University of Minnesota, St. Paul, MN 55108. 38 technological opportunities. The aim is to improve existing services, to introduce new services, and, ultimately, to apply computer technology to areas of manage- ment decision making. Libraries continue to expand the sophistication and number of computer applications. Change has be- come the constant. Computer technology was originally re- garded as a means for achieving savings at the technician and clerical levels. The · newer waves of library automation are af- fecting highly skilled personnel. While performing basic tasks such as circulation control, a system generates data of signifi- cant value to the library manager. A man- ager has the capability of forecasting de- mand flows, providing time horizons for various functions, and collecting and ana- lyzing a wealth of additional statistical data while the system engages in its pri- mary function. Early works on promoting innovation through the creative use of new technolo- gies were driven by a technological imper- ative. Current management literature sug- gests that technological determinism can be avoided. This does not mean that tech- nology is neutral. It requires a response from the user, but it does not prescribe the nature of that response. According to Shoshana Zuboff, the flexibility, memory, and remote access capabilities of informa- tion systems create new possibilities that provide choices in the design and intro- duction of applications. 4 The intrinsic power of information, or what Peter Keen calls the "politics of data,'' is a fundamental element in plan- ning for technological change. 5 Informa- tion as a resource symbolizes status, en- hances authority, and shapes relation- ships. Information may be public or pri- vate, formal or informal; regardless of its form, the owner of information owns power. For many years, catalogers held uniquely powerful positions because they alone could interpret the MARC format; reference librarians needed their help to decipher the message implicit in a serials tag such as "785 05 (Absorbed in part by:)." Computer systems frequently redistrib- ute information, breaking up monopolies. Implementing Technological Change 39 The display of a library's holdings in a na- tional database opens up its resources to a much larger user group. As libraries be- gan to participate in online, national, in- terlibrary loan networks, their control over collections and services was reduced. Little-known local collections became available for national use. The use of electronic data decreases in- dividual autonomy and increases the amount of control that can be exercised over library staff. Electronic data permit closer observation of subordinates' deci- sion making and help in efforts to develop additional output measures. Ruth Hafner has explored the nationwide process of evaluation encouraged by cataloging net- works. Peers, managers, critics, and con- sultants now evaluate catalogers' suc- cesses and failures. The power of individual catalogers ts diluted. 6 Hafner describes the new technologies as a '' di- rect attack" on the autonomy of the cata- loger. 7 Networks maintain error tallies, hold training sessions, and, in several re- spects, have co-opted local decision- making responsibilities. Nature of Organizations Designing strategies for implementing technological change in libraries requires an understanding of organizational be- havior. It is useful to apply Harold Leavitt's model of organizations to li- braries. Leavitt identifies four increasingly complex, interrelated components: task, technology, structure, and people. 8 He notes that strong interdependence means a change in one component inevitably has planned or unplanned effects on the oth- ers. "Task" describes the day-to-day duties and the knowledge and operator skills necessary to perform them. Task is the easiest component to alter. ''Technology'' should be understood not as methodology but as the principles that govern how the tasks are accomplished. Learning MARC tags or the new principles in AACRII means changing the technology used to create catalog records. "Structure" refers to how responsibility is distributed, roles are coordinated, and information is dis- seminated. This is the organizational 40 College & Research Libraries structure. Automating interlibrary loan functions through national utilities fre- quently requires structural changes. The need to coordinate database protocols, to schedule a terminal, or to provide OCLC- trained backup may move ILL from a cir- culation to a reference responsibility. 11The organizational culture of a li- brary ... is nonverbal and intuitive, elusive, and resistant to efforts to control and change its nature.'' "People" is the most powerful and complex component in Leavitt's model. Warren Bennis and others define the peo- ple component as the culture of an organization-the system of norms, be- liefs and assumptions, and values that de- termine how people in the organization act, even when the action may be at odds with written policies and formal reporting relationships. 9 The organizational culture of a library is a potent force. It is nonverbal and intuitive, elusive, and resistant to ef- forts to control and change its nature. 10 A university administration may decide to restructure reporting lines so that the li- brary and the computing center· report to the same vice-president. This does not necessarily mean the library staff will per- ceive a common mission with the com- puter center or perceive the need to ex- plore common problems together. Changes in basic library premises are the most difficult to implement. With their multiple components, li- braries are complex organisms character- ized by homeostatic behavior. The inertia encountered in libraries is infamous. 11 In order to implement innovations success- fully, one must unfreeze the status quo. Peter Keen points out that dramatic chan9.e rarely occurs in complex social sys- tems. 2 Only small increments are possi- ble. Compromise, far from being bad, be- comes an essential aspect of the implementation process. A recognition of the importance of qual- ity of work life (QWL) to the individual is a significant factor in current management January 1988 and library literature. 13 High QWL implies the fulfillment of human needs that are the source of motivation and involve- ment. The objective is to provide those so- cial and psychological aspects of work that have been widely recognized, such as au- tonomy, discretion, variety, social con- nectedness (developing satisfying social relationships), feedback concerning per- formance, meaningful work, meaningful future (learning new skills and developing new talents), effective voice (having some say over how the work is performed), re- sponsibility, ability to see one's contribu- tion in the end product or service, and ad- equate and equitable reward. Robert D. Stueart writes that staff are libraries' ma- jor resource, and work satisfaction must be a prime consideration as preparations are made for change. 14 Any innovation that individuals or groups perceive as intruding on their terri- tory, limiting their autonomy, reducing their influence, or adding to their work- load will be resisted and will cause staff fears. 15 Resistance exists to the extent that innovation requires change in skills, be- havior, attitudes, social roles, and social context. Resistance and the difficulty of ef- fecting change are related to the type of component. It is usually easier to master a new task than a new technology. Adopt- ing a new technology is often easier than altering the organizational structure. Changes in the cultural fabric are the most threatening and are resisted most. The common thread in the resistance to change is the uncertainty and concern ex- perienced by the staff about how the change will affect their work. Computer-Mediated Libraries Interaction among th~ demands of new technologies, the nature of organizations and information, and the responses of the men and women who must work with the new technological systems are resulting in new conceptions ·of work organization and behavior. Zuboff' s comments on the nature of computer-mediated work are clearly descriptive of current trends in li- braries. 16 Much library work has become abstract. A person accomplishes a task through the medium of the information system. A reference librarian frequently responds to a reference question without ever touching traditional printed materi- als. An online database search may be conducted for an absent patron and there- sulting printout mailed directly to the re- questor. The library staff may be removed from the raw material, the product which results from the process, and perhaps even the user of that product. Work may no longer be tangible. Employees often go to extremes to re- tain a direct connection to their work. In the late 1970s, some catalogers found it impossible to create bibliographic records in MARC format. They continued to type 3-by-5-inch cards which were then coded with MARC tags. A 3-by-5-inch card pro- vided tangible proof that a title had been cataloged; online records may not provide the same sense of satisfaction. This is one of the reasons employees retain paper files. They may defend the practice as a necessary backup, but what they really want is something tangible. 17 The social fabric of a library is affected by automation. Once jobs are reorganized, new patterns of communication and inter- action become necessary. In time these patterns are likely to alter the organiza- tion's social structure. In automated cata- loging operations personal work stations can create an uncomfortable isolation. The local social network may become frag- mented and catalogers are distanced not only from public service operations but from other catalogers. Despite evidence of reduced face-to-face interaction, Zuboff indicates that technology can also make it easier to initiate dialogues and form coali- tions within the organization. 18 Dispersed cataloging activities are possible. Geo- graphic separation is not as significant as in the past.lt is important to acknowledge that information technology has power- ful, not necessarily negative, conse- quences for the structure and function of communication and social behavior in a li- brary. · Computers in libraries offer new possi- bilities for supervision and control. Tradi- tionally, supervision has depended on the quality of the relationship between super- visor and worker. By consulting printouts Implementing Technological Change 41 or visual displays, library supervisors can receive immediate information about how much employees are producing and in- crease observation without depending on face-to-face supervision. Remote supervi- sion and greater access to subordinates' production levels become possible with computer-mediated work. According to Zuboff, this may lead employees to limit their risk-taking and initiating behaviors. 19 The capacities of information systems can also alter relationships among library managers. Data can be available at all managerial levels. This new access raises a number of policy questions. What type of information is appropriate at each level of management? Who is responsible for eval- uating the data? What type of experience in statistical analysis is necessary? Library managers need the time and skill to un- derstand raw data. They must avoid the tendency to become bogged down in data, looking only at past and present exgeri- ences; they must plan for the future. Computer mediation of work has re- sulted in the birth of the information envi- ronment. 21 This environment is character- ized by speed of access, retrieval, and information processing. Increased access to data and, in particular, data relevant to specific decision making, results in greater ease and convenience for review and reor- ganization. Increases in speed and amount of access to data influence feelings about power and orderliness. The results of automation can be seen as neat and nice, as augmenting power, or as increas- ing regimentation and encroaching on in- dividual judgment and freedom. SECTION II Sociotechnical Systems Design Individuals who design automated sys- tems and plan implementation projects of- ten focus on the task and technology sub- systems and ignore or downplay the application's effect on people and the or- · ganizational structure. It is clear that new information systems inevitably affect how the staff relates to the tasks for which they are responsible, as well as to the organiza- tional structure and culture. Wilson Lu- quire's research into attitudes toward au- 42 College & Research Libraries 11The goal of STS design is to get the best match in the overall combina- tion of social and technical factors.'' tomation in academic libraries demonstrates the importance of consider- ing what he calls the psychological point of view when introducing innovation. 22 However, the technical application itself cannot be ignored. One solution is to ap- proach systems design and implementa- tion in a way that combines both social and technical sensitivity. Several researchers, primarily Enid Mumford and her colleagues, have devel- oped a sociotechnical approach to systems design. Sociotechnical systems (STS) de- sign acknowledges the interaction of tech- nology and people; its goal is to produce systems that not only are technically effi- cient but that also lead to high job satisfac- tion. 23 Calvin Pava uses STS design in his strategies for automating offices. 24 Ac- cording to Pava, planners who use STS design recognize that each application has a technical subsystem (the tools and tech- niques that transform the input into a product) and a social subsystem that joins the disconnected jobs together and coor- dinates them. The goal of STS design is to get the best match in the overall combina- tion of social and technical factors. 25 STS interventions typically involve the restructure of work method, the rear- rangement of technology, or the redesign of the social structure. STS design aims to unfreeze the existing organization in or- der to facilitate change. The techniques proposed by Mumford parallel the model for change developed by Kurt Lewin. 26 Le- win's model states that successful change must move through three steps: (1) un- freeze: create a climate for change and get contracts/commitments with users; (2) · change: analyze, design, develop, and in- stall the new system; (3) refreeze: institu- tionalize the new system. This model em- phasizes that stable organizational systems with supporting political coali- January 1988 tions need to be disturbed before change begins. The first step in STS intervention is designed to begin the unfreezing pro- cess and involves assessing the existing social system, primarily by measuring the job satisfaction of the work group. The in- tent is to create an environment open to exploring alternatives. Lewin's change step begins when the job satisfaction data are used to set human (social) objectives directed at increasing job satisfaction. This can be the responsi- bility of the work group. Alternatives are developed that deal with work group structures, task allocation, and individual job designs. The technical experts exam- ine alternatives for meeting the technical objectives of the system but usually do so prior to initiating the design effort. Once technical and social alternatives have been identified, the entire design team selects those alternatives that meet both objec- tives. If successful, the result should be improved task efficiency and job satisfac- tion. After installation of a new system, the concluding step is refreezing. In this stage, the organization accepts the changes and becomes committed to them. If the organization has successfully worked through the preceding steps, re- freezing should occur easily. Although STS design is time consuming and can require special training, it can be used to introduce technological change with a minimum of stress and upheaval. As a result, the financial cost of implemen- tation may be reduced. Serious staff/man- agement problems can be avoided. Ac- . cording to Mumford, the successful implementation of a new computer sys- tem can provide the opportunity for ac- quiring human as well as technical gains. 27 Organizations characterized by inertia may be reluctant to take a risk. 28 Fear of the unknown is the critical element. Change becomes more difficult to im- plement as the level of change increases. This can be linked to the four interrelated components of an organization: task, technology, structure, and people. The more sweeping and sophisticated the change, the more difficult it is to imple- ment. Librarians who wish to manage change successfully must recognize the human factor at all levels. SECTION III Strategies for Change How should a librarian plan for change? Management is traditionally defined as the act of controlling; strategy is the act of planning and directing. Considering the nature of organizations, technology, and the power of information, planning for change is a more appropriate concept than controlling change. Bennis defines planned change as the conscious, deliber- ate, and collaborative effort to improve the operations of a human system. 29 The most successful implementation projects tend to be those that result from a systematic ef- fort to consider all components of the or- ganization, particularly the social fabric. This paper concludes with a set of strate- gies for facilitating technological change based on STS design. 1. Communication. Theroleofcommuni- cation is paramount because most resis- tance stems from communication failure. Appropriate information will assist the li- brary organization to unfreeze and change. Staff need to know what change is going to mean to them in terms of their jobs, salaries, personal satisfaction, sta- tus, etc. Library managers should strive for open communication both horizon- tally and vertically, formally and infor- mally. They should be prepared to convey information repeatedly. They must also be available and open to questions. They must be prepared to deal with resistance. Resistance demonstrates the employee's need to have some power over the situa- tion. Managers should recognize that re- sistance may be appropriate; some inno- vations are poor. The employee usually has something of value to communicate about the nature of the system. The library manager should view the implementation process as joint problem solving, not as a win/lose conflict. 2. Purpose. Clearly stating goals is an es- sential element in the process of bringing about change. Management should state Implementing Technological Change 43 the reason for initiating a new system. Clear understanding by staff regarding the necessity for change is critical. They can then claim partnership in the change process. 3. Leadership. Success of a project de- pends on the support of middle and top management. Change is less avoidable if supervisors support fully the innovation. Large-scale change is a process of coalition building on the part of management. The presence of a clearly identified individual or group who champions and guides tech- nological activity is essential. This may be the director of automation or an automa- tion steering committee. This person or group must have the prestige, visibility, and legitimacy to move the project for- ward. 4. Incentives. Incentives may be formal or informal. Quality of work life studies have shown that employees are motivated by a mixture of incentives. The need for variety and opportunities to learn new skills and develop new talents is fre- quently a high priority on the part of staff and can be designed into the system. Job rotation, job enlargement, and job enrich- ment are more feasible in an automated environment. Further, it is more effective and efficient to have a staff person per- form several functions. The use of work groups can provide an incentive by satisfying the need for social connectedness. Small groups provide a link between the individual and the larger organization. They carry the added ad- vantages of functioning both as a medium and as a target of change. Small groups within the library offer security in the midst of change. Access to information conveys power, and an increase in social power may be perceived as an incentive. Perceiving a proposed change as congruent with self- interest complements incentive. 5. Consequences. Managers should make consequences clear. Organizations with highly visible consequences are quicker to suggest alternatives when faced with or- ganizational changes and to make the nec- essary adjustments to react to change. When the effects of actions are clear, adap- 44 College & Research Libraries tion is easier. Procedures that enhance vis- ibility of consequences will enhance effi- ciency. 6. Time. Adequate time is needed for implementation. A staff needs time to ex- plore options, compare alternatives, and consider consequences. A gradual ap- proach toward automation alleviates so- cial inertia and helps reduce stress and tension. 7. Incremental Implementation. Introduc- ing a project in several overlapping phases provides several benefits. Phasing makes it easier to see and define what benefits or problems are associated with each func- tion; the impact on users is less drastic. Local priorities can be used to determine the sequence in which functions are im- plemented. Gantt charts are frequently used to present visually the idea that com- plex systems can evolve out of phased components. Using prototypes in one or two sites on a full-production basis is another way to im- plement incrementally. This method of- fers a way to assess system impact on hardware and on users and to test train- ing. Prototypes provide time to make needed changes before the system is ex- panded to include all users or all sites. 11Collaboration is an essential ingre- dient in effective planned change. People want a say in how their work is performed." 8. Collaboration. Collaboration is an es- sential ingredient in effective planned change. People want a say in how their work is performed. Participative involve- ment increases commitment and coopera- tion. It offers wider resources for problem solving. The degree of collaboration achieved by staff members has a direct, positive impact. However, extensive par- ticipation of users at the design stage may .. increase the planning time and costs of the system. Libraries must recognize the trade-offs. Systems designed without the active involvement of users may initially January 1988 appear cost-effective on the basis of tech- nical criteria, but, in fact, higher costs are often incurred in the long run. These higher costs result from resistance to change, poor system utilization, high turnover, and absenteeism. 9. Design. The design process should be open ended. Design is a dynamic, reitera- tive process; a certain degree of incom- pleteness allows for creativity. The library should expect the innovations to change over time. Variations can be legitimate forms of adaptation without implying fail- ure of implementation. The design pro- cess should be initiated with few details. The details of a work system should be limited to what is absolutely essential be- cause overspecification narrows options. Developing alternatives is useful. It will allow exploration of more options and in- crease the chances for the acceptance of one. Practical design considerations should evolve out of an awareness of ergonomics. Plans for input and display variables, work station layout, and data manipula- tion and transfer should consider the li- brary staff who will use the system. In ad- dition, planners should consider work flow. The system should be designed with elements that facilitate its use. 10. Training. Ease of learning the new technology is a key factor in acceptance. Training is most successful if it is linked to the library's specific organizational con- text and tied to day-to-day operational needs. Local development (or at least modification) of training materials is desir- able because it clearly provides this link. Online training manuals and simulations are useful, as is online documentation. Training sessions should be presented in a comfortable setting similar to the work en- vironment. SUMMARY AND IMPLICATIONS It is clear that automation is becoming an integral part of internal operations as li- braries implement new systems and re- vise existing applications. Plans to imple- ment library automation projects must recognize the power of technology and the politics of data. Introducing automa- tion in a library setting requires an under- standing of the nature of organizations and the new environment of computer- mediated work. This paper has presented a set of strategies for the effective intro- duction of planned technological change. These strategies reflect a STS design ap- proach. The social subsystem is as impor- tant as the technical subsystem. When the technical subsystem is based on informa- tion technology, the implementation is made more complex by the power and politics of information. The strate_gies proposed in this paper Implementing Technological Change 45 can be time consuming and expensive. However, the beneficial results will ulti- mately outweigh any liabilities. Imple- menting technological innovation using STS design productively involves library staff in the change process. Stress andre- sistance are reduced and changes are more smoothly assimilated. Recognizing the contributions that library staff can make is central. Clearly, the introduction of a new computer system provides the opportunity for realizing human as well as technical benefits. REFERENCES AND NOTES 1. Automation and Reorganization of Technical and Public Services, SPEC Kit 112. Washington, D.C.: ARL, Office of Management Studies, . Mar. 1985. 2. Richard Boss, "The Automated Library Systems Industry: A Time of Change," Library Systems Newsletter 6:89-93 (Dec. 1986). 3. Lucinda D. Conger, "Predictions," Online 11:44-45 (Jan. 1987). 4. Shoshana Zubo££, "New Worlds of Computer-Mediated Work," Haroard Business Review 60:148 (Sept./Oct. 1983). 5. Peter G. W . Keen, ''Information Systems and Organizational Change,'' Communications of the As- sociation for Computing Machinery 24:20 (1981). 6. Ruth Hafner, Academic Librarians and Cataloging Networks (Westport, Conn.: Greenwood, 1986), p.3. 7. Ibid., p.76. 8. Harold J. Leavitt, ''Applying Organization Change in Industry: Structural, Technological and Hu- manistic Approaches," in Handbook of Organizations, ed. J.G. March (Chicago: Rand McNally, 1965). 9. Warren G. Bennis, Kenneth D. Benne, and Robert Chin, The Planning of Change, 4th ed. (New York: Holt, 1985), p.154. 10. Ibid., p.166. 11. David W. Lewis, "An Organizational Paradigm for Effective Academic Libraries," College & Re- search Libraries 47:337 (July 1986). 12. Keen, "Information Systems," p.25. 13. The reader is referred to a number of ·columns and case studies, edited by Charles Martell and focusing on QWL in libraries, which appeared in Journal of Academic Librarianship during 1983, 1984, and 1985; and to JeanS. Decker's article "QWL in Academic/Research Libraries," Technical Services Quarterly 3:51-58 (Fall1985/Winter 1985-86). 14. Robert D. Stueart, "Preparing Libraries for Change," Library Journal109:1724-26 (Sept. 15, 1984). 15. Ibid., p.27-28. 16. Zubo££, "New Worlds of Computer-Mediated Work," p.142-52. 17. Ibid., p.144. 18. Ibid., p.146. 19. Ibid., p.147. 20. Ibid., p .148. 21. Ibid., p.149. 22. Wilson Luquire, "Attitudes Toward Automation/Innovation in Academic Libraries," Journal of Academic Librarianship 8:344-51 (Jan. 1983). 23. Enid Mumford and Mary Weir, Computer Systems in Work Design: The ETHICS Method (New York: Wiley, 1979). 24. Calvin H. P. Pava, Managing New Office Technology: An Organizational Strategy (New York: Free Pr ., 1983). 46 College & Research Libraries January 1988 25. Mumford, Computer Systems in Work Design, p.3. 26. Kurt Lewin, "Frontiers in Group Dynamics/' Human Relations 1:5-41 (1947). 27. Mumford, Computer Systems in Work Design, p.290. 28. Maggie Moore and Paul Gergen, ''Risk Taking and Organizational Change,'' Training and Develop- ment Journal39:72-76 Oune 1985). 29. Bennis, The Planning of Change, p.4. IN FORTHCOMING ISSUES OF COLLEGE & RESEARCH LIBRARIES The Emergence of the Book by Glenn Anderson Time Management in Academic Libraries by Helen M. Gothberg and Donald E. Riggs The Academic Librarian and Information Technology: Ethical Issues by Lawson Crowe and Susan H. Anthes Library Materials: Paper, Microform, Database by Michael K. Buckland Characteristics of the Monographic Scholarship of Foreign Literary Studies by Native- Speakers of English by John Cullars Users' Persistence in Scanning Lists of References by Stephen E. Wiberley, Jr., and Robert Allen Daugherty Turnover of Professional Staff in Academic Libraries by Dee Ann Allison and Eva Sartori