The Pennsylvania State University
The Graduate School

Department of Adult Education, Instructional Systems,
and Workforce Education and Development

THE RELATIONSHIP BETWEEN THE IMPLEMENTATION OF
HIGH PERFORMANCE WORK PRACTICES
AND NUCLEAR POWER PLANT PERFORMANCE

A Thesis in
Workforce Education and Development

by
William H. Lowthert III

Copyright 1996 BillisImproving@aol.com 

contact Bill Lowthert at BillisImproving@aol.com  or www.alwaysimproving.com/

Submitted in Partial Fulfillment
of the Requirements
for the Degree of
Doctor of Philosophy

May 1996

(Some formatting changed from Penn State standard to accommodate posting on the Internet website)

ABSTRACT

This research identified the implementation levels of ten recommended high performance work place practices in the American nuclear power industry. It also studied the relationship between the implementation level of each work place practice and measures of nuclear power plant performance.

The high performance work place practices studied were recommended by the United States Department of Labor's Office of the American Workplace in the guideline, Road to high performance, a guide to better jobs and better business results. The work place practices were taken from the participation, organization, and partnership category.

The three measures of nuclear plant performance were selected because they represent the performance areas designated as most important by nuclear industry leaders. These measures are nuclear safety, capacity factor, and cost of generation. This information is reported in electric industry publications.

The implementation levels of the recommended work place practices were obtained through a mail survey. Senior managers at all American nuclear power plants were surveyed. A 74% return rate was achieved. Descriptive statistics were used to study the industry wide implementation rates. The data were analyzed by applying the least-squares regression method to describe the relationship between each recommended work place practice and each measure of nuclear plant performance.

This research found that all ten work practices are implemented in the American nuclear power industry. However, there is opportunity for further implementation of the recommendations.

This research found a desirable relationship between the implementation level of five work practices and nuclear power plant safety performance as indicated by the NRC Systematic Assessment of Licensee Performance rating. It also found a desirable relationship between the implementation level of four work practices and plant capacity factor. There were no relationships identified between the implementation level of a work practice and cost of generation. There were no undesirable relationships identified where implementation of a work practice was related to decreased nuclear plant performance.

The data indicate that work place practices that involve teams have a moderate correlation with nuclear safety ratings. This relationship suggests that by organizing workers into teams, granting the teams substantial authority, and creating teams that cross functional and organizational boundaries, some nuclear power plants have been able to obtain better nuclear safety ratings.

Implementation levels of work place practices that provide workers some control over their work processes are related to higher capacity factors. The data indicate a relationship between plants with better electrical generation performance and workers being actively involved in continuously improving their work process, redefining their jobs, being involved in problem solving, selecting new technology, modifying their product or service, meeting with internal customers, and modifying their work processes.

ACKNOWLEDGMENTS

I wish to thank Dr. William J. Rothwell, thesis advisor and doctoral committee chair, for his advice and direction in this research and throughout my doctoral study. I also extend thanks to Dr. Kenneth C. Gray, Dr. Richard Walter, and Dr. Gordon E. Robinson for their assistance and efforts as members of my doctoral committee. Special thanks are given to Dr. Edward H. Klevans for his assistance in completing my defense.

Chapter 1

RESEARCH PROBLEM

The organizations that operate nuclear power plants to produce electricity in the United States have a particularly interesting management challenge. They apply a complex technology to control a potentially hazardous energy source to produce a commodity (Office of Technology Assessment, 1984; Wayland & Brusger, 1987). Their product, electricity, is in high demand but is rather easy to produce with less complex, nonnuclear technologies (Skinner, 1994). Other energy sources, including oil and gas, provide additional competition for energy consumers (Wayland & Brusger, 1987). Corporations that own nuclear power plants simply cannot shutdown the plants and acquire new sources of energy without major economic losses for their stockholders. The large capital investments made in most nuclear plants have not been recovered. Nuclear operators must succeed using the facilities available to them (Recovering Sunk Costs, 1995).

The electric power industry is facing a new challenge. Customers will soon be able to decide who supplies their electricity (Skinner, 1994). Customers will be allowed to buy from anyone who is willing to supply electricity to them. Electric utilities will soon face competition for markets that were reserved exclusively for them because of their franchise. Electric suppliers are doomed to exist in a world of regulated competition (Stelzer, 1994). The threat of deregulation is forcing electricity producers to reduce costs to be profitable (Warkentin, 1995).

Nuclear power plants must be economical to be successful during the shift from the noncompetitive, regulated environment to competition (Long, 1995). Economic success of a nuclear power plant is achieved through three strategies. The success strategies are adherence to the highest safety standards, production of competitively priced energy, and generation of sufficient revenue to provide a return on investment to the owners after operating costs and capital investments are covered (Byram, 1994). The economic threat is real. Even before the challenge of deregulation, three nuclear plants were shutdown permanently because of inadequate economic performance. The Trojan Nuclear Plant was shutdown in 1993. San Onofre Unit 1 and Yankee Rowe were shutdown in 1992 (Nuclear Energy Institute, 1994).

The industry recognizes that a major contributor to higher production costs at nuclear plants is the way the plants are managed and that improved management practices, not improved technology, will improve performance and competitiveness (Kuehn, 1994). Senior nuclear management is focusing on improving people and procedures rather than making technical design changes (Better Performers, 1995). At a recent industry wide strategic planning conference Gene Stanley, an industry vice president, stated "the focus needs to shift from technical performance to human performance" (Better Performers, 1995, p.10).

The nuclear power industry realizes that improvement in human performance is the best approach to meet the competitive economic challenges that it faces (Institute of Nuclear Power Operation, 1995; Better Performers, 1995). According to Dale Brethower (1995), a human performance technologist, high levels of human performance are usually accomplished through instructional technology and performance technology. Instructional technology attempts to improve organizational performance through improving the ability of individual to perform. Performance technology attempts to improve organizational performance through improving the work environment and the work practices of the organization.

Instructional technology is well established in the nuclear industry. A systematic approach to training design and delivery has been implemented (Smith, 1991). In 1982 industry standards for instructional technology were established. A training accreditation program was begun to enforce those instructional standards under the leadership of the Institute of Nuclear Power Operations (Coakley, 1991). In its 1994 annual report the Institute of Nuclear Power Operations (1995) reported that instructional technology continues to be applied effectively in the industry and that continual monitoring is performed through accreditation renewal activities of the National Nuclear Accrediting Board.

There is no evidence of nuclear industry research or broad application of performance technology that would lead to high performance organizations in nuclear power plants. In 1994 the United States Department of Labor's Office of the American Workplace published a guideline, Road to high performance, A guide to better jobs and better business results (1994). This guide combines common elements of high performance workplaces using empirical evidence from successful companies to develop workplace practices that lead to high performance. The Office of the American Workplace (OAW) groups these practices into three useful categories: skills and information; participation, organization, and partnership; and compensation, security and work environment. These categories cover workplace practices that could contribute to nuclear plant performance improvement.

The purpose of this study is to identify which of the characteristics of high performance workplaces recommended by the U.S. Department of Labor's Office of the American Workplace (1994) in the category participation, organization, and partnership are being practiced at American nuclear plants. It also studies the relationship between the implementation level of these recommendations and the performance level of individual nuclear plants. The results of this study will provide insight to nuclear managers concerning opportunities to improve nuclear plant performance and to achieve economic success through the implementation of high performance work practices involving participation, organization, and partnership.

The participation, organization, and partnership category is chosen for this study because its implementation appears to provide the greatest opportunity for organizational improvement and economic success for nuclear power plants. During a recent presentation to the American Nuclear Society Executive Conference Robert Byram (1994), a senior nuclear executive, stated that nuclear success will require that leaders involve the rank and file of their organizations in the development of strategy. The Electric Power Research Institute and the American Water Works Research Foundation believe that the single most important ingredient in the development of total quality programs is worker empowerment and involvement (Pittenger, 1991). According to the U.S. Department of Labor the implementation of the workplace practices concerning participation, organization, and partnership will contribute to a high performance workplace and generate better business results (1994).

This research identifies the high performance workplace practices in this category that are already in place at nuclear plants. If these high performance practices have not been fully implemented, performance improvements could be gained by their use.

This research study attempts to answer the following questions concerning the application of high performance workplace practices at American nuclear power plants.

Research Question 1

To what extent have the practices recommended by the U.S. Department of Labor's Office of the American Workplace (1994) in the category Participation, organization, and partnership been implemented at American nuclear plants?

Research Question 2

What is the relationship between the implementation level of the practices recommended by the U.S. Department of Labor's Office of the American Workplace (1994) in the category Participation, organization, and partnership and the performance level of nuclear power plants?

Nuclear power plant is an electrical generating station that uses nuclear fission to generate the energy necessary to produce electricity.

High performance workplace is an organization that effectively utilizes all of its people, leadership, and physical assets to achieve organizational and personal success.

High performance work environment is the application of physical, emotional, and leadership factors to support people in the achievement of organizational and personal success.

High performance work practices are organizational procedures and policies that allow individuals in an organization to perform effectively, such that both the organization and the individuals succeed.

According to the guideline Participation occurs when workers are involved in continuously improving their work processes, redefining their jobs, solving problems, selecting technologies, modifying the outcomes, and dealing with customers.

According to the guideline Organization occurs when people work in teams with substantial authority and reduced levels of management.

According to the guideline Partnership occurs when workers are involved in decision making on a range of issues traditionally decided by management alone.

Chapter 2

REVIEW OF THE LITERATURE

OF HIGH PERFORMANCE WORKPLACE PRACTICES

This study describes High Performance Workplace Practices as organizational procedures and policies that allow individuals in an organization to perform effectively, such that both the organization and the individuals succeed. Many phrases are being used to describe organizations that are achieving high performance. Terms such as high performance workplaces (U.S. Department of Labor, 1994; Becker & Steele 1995), high performance, high commitment organizations (Jewell & Jewell, 1992), high performance work organizations (National Center on Education and the Economy, 1990), high performing organizations (Thor, 1994), learning organizations (Senge, 1990; Watkins & Marsick, 1993) and high performance work practices (Huselid, 1993) are synonymous with high performance workplaces as used in this study.

During the last decade there has been substantial study and writing concerning high performance workplaces. Most of this writing has been broad in nature. It identifies characteristics of organizations that are successful and reports those characteristics. Readers, managers, and researchers are left to their own to apply the reported characteristics in specific situations. The advantage to this researcher in reviewing the available literature is to capture the identified characteristics of high performance and evaluate their usefulness to nuclear power plant organizations.

U.S. Department of Labor Guide

The federal government is attempting to improve work environments for all workers through the United States Department of Labor. A new division of the Department of Labor, the Office of the American Workplace, was created to promote the concept of high performance work organizations using progressive human resource management strategies and practices (Geber, 1995). According to Barbara Schrader (personnel communication, August 17, 1995), one of the first activities of the Office of the American Workplace was research and publication of a guideline for organizations to develop high performance workplaces, gain better business results, and create better jobs.

The U. S. Department of Labor's Office of the American Workplace published the guideline, Road to high performance, A guide to better jobs and better business results (1994) to lead organizations to high performance. This guide combines common elements of high performance workplaces using empirical evidence from successful companies to develop workplace practices that lead to high performance. The Office of the American Workplace grouped these practices into three useful categories: skills and information; participation, organization, and partnership; and compensation, security, and work environment. These categories cover organizational characteristics that could apply to most American industries. The purpose of the OAW study was to help create better jobs and better business results.

The skills and information category consists of continuous learning and information sharing. These two characteristics are often found together in high performance work environments. The OAW says that successful companies are changing from training for specific jobs to emphasizing broader skills such as problem solving and interpersonal interactions, team building, and basic business operation. Classroom, work based training, and rotations through a series of jobs are used to develop multiple skills. High performing organizations share financial, operational, technical, and strategic planning information concerning the company and competitors with employees. They apply new technologies to distribute information including e-mail and networks.

The participation, organization, and partnership category recognizes that employees are in a position to have a great impact upon company success. Employees are often closest to the customer and have the best knowledge about the product or service. This category attempts to provide greater worker autonomy and gain a greater sense of commitment from employees. Employee participation is enhanced through self-directed work teams, consultative committees of workers, and allowing workers authority previously reserved for managers. Organizational structures with few levels of hierarchy, streamlined processes, and cross functional teams build participation and empowerment of the workforce. The OAW admits that labor management conflicts and collective bargaining agreements may be impediments in this category. To achieve high performance, companies must work to end labor management confrontations and to develop higher levels of mutual trust and respect. Labor and management should jointly address issues that were previously decided by management alone. Internal governance systems should be based on trust rather than power or contractual rights.

In the compensation, security, and work environment category high performance organizations apply a variety of systems to reward attainment and performance, to establish an employee stake in the organization's success, and to accommodate the diverse needs of their workforce. Compensation is linked to corporate, team, and individual performance. High performance organizations consider employees as key investments. They use layoffs and downsizing as a last resort. Instead they provide an explicit commitment to employment security. The use of temporary and seasonal help meets business cycle needs without layoffs. When layoffs can not be avoided these organizations provide advanced notice, out placement service, and supplemental unemployment compensation. Supportive work environments are important to the productivity and commitment of the workforce. High performance organizations provide a balance between work and family life, initiate flexible work schedules, accommodate disabilities, provide child care, eliminate barriers that unfairly discriminate, and promote fitness, health and safety.

The Department of Labor's Office of the American Workplace states that the practices described in their guide must be implemented as part of a coordinated strategy, that is integrated into the business process. Companies with high performance employee practices perform well on critical corporate indicators such as return on investment and capital utilization (U.S. Department of Labor, 1994).

Barbara Schrader was the primary editor of the guideline. She explained that the research was performed by a team of Department of Labor employees and outside experts. The team assembled a data base of exemplary workplace practices. Existing Department of Labor data was reviewed. The Department of Labor Relations Cooperative Program data base was a beginning source of data. Agencies such as the Women's Bureau provided unique ideas. The team performed reviews of journals and magazines to identify potential high performance practices. They assembled the data into the Office of the American Workplace Best Practices Clearinghouse data base. As entries were made into the data base they were reviewed to ensure that the organization was actually successful and that existing labor laws and Department of Labor criteria were not being violated. As the data was assembled categories of practices began to emerge. The guideline groups high performance practices into categories (personnel communication, August 17, 1995).

Steve Marler was the lead writer for the guideline. Mr. Marler explained that the guideline was originally intended for the financial investment community. The Department of Labor was interested in creating a shift from the short term focus on quarterly financial results to a long term development centered strategy. The financial community reacted unfavorably. However the researchers identified a broader constituency. They believed that the high performance workplace practices would benefit operating organizations. The guideline was designed to persuade chief executive officers to include high performance workplace practices into their long term strategies. The workplace practices checklist portion of the guideline allows organizations to self check their practices against those in the guideline (personnel communication, August 17, 1995).

National Quality Criteria

The OAW intends their guide to become a nationwide guide to high performance. National guidelines and criteria concerning quality are already published and in use. Organizations may evaluate themselves against these criteria, use the criteria to focus improvement efforts, and compete for national quality awards. The Malcolm Baldrige National Quality Award Criteria provide core values and concepts that should direct organizations toward high performance (U.S. Department of Commerce, 1995). The Baldrige criteria core values and concepts are customer driven quality, leadership, continuous improvement and learning, employee participation and development, fast response, design quality and problem prevention, a long range view of the future, management by fact using measurement and analysis, internal and external partnerships, corporate responsibility and citizenship, and a orientation toward results.

The Malcolm Baldrige National Quality Award Criteria and the resulting core values and concepts were developed in response to the concern of business and government leaders that the nation was losing its ability to compete in the world market during the early 1980's. Forums, committees, and government conferences identified characteristics that would improve productivity and quality in America. The Malcolm Baldrige National Quality Improvement Act became law in 1987. It established the national quality award. The findings and purposes section of the law codifies expectations of Baldrige award winners (George, 1992). The award and criteria are managed by the United States Department of Commerce, National Institute for Standards and Technology (Brown, 1992). The Baldrige application form and its criteria have become one of the most widely used instruments for self assessment in America (Hodgetts, 1993).

There is overlap between these Baldrige core values and concepts and the OAW's categories of high performance. Areas concerning continuous learning, employee development, employee participation, and partnership development are very consistent. The Baldrige criteria provide more guidance on business issues such as results orientation, customer driven quality, and management by fact.

ISO 9000 Certification Standards

There are also generally accepted international quality standards that provide some support to the high performance categories of the United States Department of Labor's guideline, Road to high performance, A guide to better jobs and better business results. The International Organization for Standardization in Geneva, Switzerland creates and publishes ISO standards, including the ISO 9000 quality standards and their criteria for certification. ISO is a term derived from Greek that means equal (Bretz, 1994a). The intent of ISO standards is to ensure that suppliers are equal to specified quality requirements. It is not an award that only a few achieve, but a standard that all are expected to meet (Hutchins, 1993). More than 35 countries have adopted these standards. They are developed by 182 technical committees, including the American National Standards Institute. These committees draft standards, review proposals, resolve comments, and approved standards in their areas of expertise (Bretz, 1994a). The European Community has adopted ISO 9000 as part of their unification effort. American companies are beginning to seek ISO 9000 certification in response to foreign customer demands (Hockman, 1992).

The ISO 9000 standards provide a framework for verifying to customers how products are tested, employees are trained, records kept, and defects fixed (Henkoff, 1993). The high performance activities required for ISO 9000 certification include management responsibility, written total quality management policies, internal evaluation of organizational activities, document control, self auditing, corrective action programs, internal quality audits, training, and customer service (Berthelot, 1993).

Learning Organizations

"A learning company is an organization that facilitates the learning of all its members and continuously transforms itself" (Pedler, Burgoyne, & Boydell, 1991, p. 1). Chris Argyris discusses the concept of organizational learning (Argyris & Schon, 1978). He proposes that organizations often know less that the sum of the knowledge of individual members, even though organizations learn through the experiences and actions of their members. Organizational learning occurs because of responses to the internal and the external organizational environments, detecting and correcting errors, and embedding the lessons learned in personal memories and organizational memory maps.

Argyris and Schon (1978) categorize the ways organizations learn into groupings. Organizations as groups recognize the interaction and collective action of individuals. Organizations are agents or instruments for achieving social purposes. Organizations are structures that provide order, authority, and definition. Organizational change involves selecting new structures or modifying old ones to respond to changing conditions. Organizations as systems recognize self-regulating entities with cycles of action. This category considers conditions favorable to learning and self-regulation as factors important to high performance. Organizations as cultures are small societies with shared meanings and rituals that allow them to interact among themselves and their outside environment. The culture frames how an organization responds to stimuli. Organizations are political systems balancing interests and powers of their members to achieve results. In this environment of converting contention to cooperation, organizations learn by converting politics to inquiry (Argyris & Schon, 1978). The challenge to organizational intervention including development of high performance work environments is to meet the needs of all six categories of organizations.

Peter Senge (1990) describes organizations as complex interactions. He cautions against breaking the study of organizations into fragments. He encourages systems thinking. All individuals, groups, organizations, and their actions have an effect on all others. Senge presents high performing organizations as four core disciplines and the fifth discipline, systems thinking, that brings the other four together to cause success. Senge's core disciplines are personal mastery, mental models, shared vision and team learning. Personal mastery embodies clarifying what is important and continually learning to see current reality more clearly. Mental models are simple generalizations or complex theories that shape how we act. Shared vision is a force that creates commonality and a sense of purpose throughout an organization. It directs and gives coherence to diverse people and activities to meet common objectives. Team learning is the process through which the team develops and aligns to achieve the results the team desires.

Senge states that organizations must develop a learning cycle that includes new skills and capabilities, new awareness and sensibilities, and new attitudes and beliefs. This learning cycle must cause a fundamental shift in both individual minds and in the collective organizational mind. He emphasizes the need for organizations to change to orient themselves toward what they care about. Organizations should not change only because they need to change. They must employ the capacity to see the larger systems and broader forces that impact their existence (Senge, Roberts, Ross, Smith, & Kleiner, 1994).

Organizations realize that they must change to survive. This change must be directed toward continuous learning for continuous improvement. The line production organization must be the learning organization. Learning organizations cannot be parallel support organizations, as human resource and training organizations have become. The action imperatives of learning organizations must be implemented throughout the organization. These action imperatives are continuous learning opportunities, inquiry and dialogue, collaboration and team learning, systems to capture and share learning, people empowered toward a collective vision, and the organization connected to its environment (Watkins & Marsick, 1993).

Learning companies have distinguishing characteristics. They develop and revise strategy continuously. This allows learning to be included into company plans. They allow all members of the company to influence policy making. They use information technology to inform and empower employees and to provide accounting and budget information to assist employee learning. Internal departments and units see each other as both customers and suppliers with a goal of internal customer delight. Reward and recognition systems are flexible to reward performance in meaningful ways. Organization charts and roles are loosely structured to enable flexibility in assignments. Employees who interact with the outside environment provide information on the environment to the rest of the company. That information is collated and distributed. The key role of managers is to facilitate employee learning and continuous improvement. Joint learning ventures are undertaken with customers, suppliers, and companies in other industries. Benchmarking is common. People are encouraged to take responsibility for their own learning and development. Learning resources and facilities are available for their use (Pedler, etal., 1991).

Kline and Saunders (1993) propose ten steps necessary to become a learning organization. They are consistent with the descriptions previously provided. Wick and Leon (1993) have a slightly different view of learning organizations. They observe that learning organizations have a leader with a clear vision, a detailed action plan that is measurable, an ability to share information rapidly, inventiveness, and the ability to implement their plans.

The literature on learning organizations is consistent with the guidelines presented by the Office of the American Workplace (1994). Argyris and Schon's discussion of organizational learning (1978), Senge's disciplines (1990), the distinguishing characteristics proposed by Pedler, etal., (1991) and the other literature reviewed concerning learning organizations support and reinforce the guidance of the OAW. This research can rely upon the consistency of the OAW guidelines and the learning organization literature.

High Performance Work Environment Studies

Jeanne Meister (1994) provides lessons for building a world class work force. Her lessons are consistent with the guidelines of the OAW guide and the learning organization literature. Meister says that training must support the strategic organizational goals. All employees must have the opportunity for lifelong learning. Workers must be accountable for learning new skills. Organizations must encourage learning both inside and outside the classroom and teach employees a core curriculum that includes corporate citizenship, provides an organizational contextual framework for the employee, and develops needed competencies. Management must reinforce desired behavior through a program of rewards and recognition and ensure employees share their successes throughout the organization. Companies should hire people with the broad set of human traits needed, rather than attempting to train for judgment, initiative, and perseverance and should approach new employee orientation as a strategic process. Training should be provided to customers and suppliers.

Mills (1991) recommends a cluster organization with few managers, a weak chain of command and empowered workers who make decisions and take action with little management direction. As in other literature reviewed, Mills emphasizes the importance of communication technology. He acknowledges that cluster organizations will not succeed in all situations or with all people, however he proposes better results than are being obtained by current corporate structures. Unfortunately organizations are highly resistant to change. They are creatures of routine and repetition. Even though creativity is difficult for a large organization, it must be achieved for success (Albrecht & Albrecht 1987).

Hammer and Champy (1993) believe that a radical reinvention of corporations is required for organizational success. They propose throwing away old organizational systems and starting over by inventing a better way of doing work. They define reengineering as "the fundamental rethinking and radical redesign of business processes to achieve dramatic improvements in critical, contemporary measures of performance, such as cost, quality, service, and speed" (Hammer & Champy, 1993, p.32). Identifying obsolete assumptions is difficult for organizations because the most important ones are the ones less likely to be discussed. Organizational belief systems are at least as important as authority and financial systems. When an organization meets change, especially technological change, organizations are forced to reevaluate their belief systems. Innovation is the most important challenge facing American business, but innovation is not only products and technology. Innovation involves people and their work environment (Toffler, 1985). Kanter (1989, pp. 58-62) sees the value in restructuring or reengineering to achieve synergy. She recommends that changes made to organizations or processes must make the whole organization better. Synergy is achieved by implementing complementary organizational changes.

Creating and sustaining high performance in the face of relentless environmental change requires organizations to have the capability to redesign themselves. Organizations must sustain efficient, high performance over time while undergoing continuous change. Utilities are facing the new demands of a competitive, deregulated environment. These demands include achieving multiple goals simultaneously, relating and responding to multiple stakeholders, managing resources efficiently and adapting to constant change and unprecedented challenges in their environment (Mohrman & Cummings, 1989). Broad principles are recommended by Mohrman and Cummings (1989) to gain high performance. They recommend using multiple information processing systems to gather information, make decisions, and communicate. Organize into small self contained units to allow the freedom to respond quickly to changing conditions. Adapt flexible organizational structures that can be modified quickly to respond to uncertain environmental change. Emphasize the organization's human resources. Variety in assignments, constant training, feedback, reward systems, and information should be used to direct and motivate the workforce.

The OAW guideline and most organizational researchers ignore the impact of the physical work environment. They concentrate on the controlling and supporting aspects of the environment However, according to Smith and Kearny (1994) the physical work environment is important to performance. Physical workplace deficiencies often impact high performance and manifest themselves as workforce performance problems. In reviewing characteristics of high performance work environments the physical environment should also be considered (Smith & Kearny, 1994). High performing learning organizations occur when there is harmony between physical settings, technology, work processes, values, management style, and organizational philosophy. The physical work environment is an important part of the total work environment and a contributor to high performance workplaces (Becker & Steele, 1995).

The National Center on Education and the Economy commissioned a study of workforce skills in America to generate recommendations that would improve America's competitive position, increase our productivity, and ensure a more prosperous future. The study identified high performance work organizations as an alternative to the Tayloristic organizational policies in place at most American firms today. It identified the guiding principle of high performance work organizations as reducing the bureaucracy by giving workers direct authority over a greater variety of tasks. It recommended that organizations be given incentives and assistance from the government to pursue high productivity work organizations and to provide the necessary training for workers to succeed. Incentives are needed because workers are not prepared to meet the high technology, service oriented future that is rapidly approaching. High performance work environments are being created through the use of computers, high technology communications, and universal employee education (Commission on the Skills of the American Workforce, 1990).

Two nuclear industry wide efforts have been undertaken to identify characteristics of nuclear plant performance on a plant basis. Both studies were directed at error avoidance and reduction. They attempted to identify causes of specific errors and to identify the negative characteristics that caused the errors. The most useful effort to this research was the Nuclear Regulatory Commission (NRC) study that produced the NRC's human performance investigative process (Paradies, Unger, Haas, & Terranova, 1993c). The second effort was lead by the Institute for Nuclear Power Operations. It resulted in the development of human performance enhancement system programs at all American nuclear plants (Ryder, 1993). Additional nuclear industry research was conducted on specific tasks and job functions. Those of value to this high performance work environment research are also reported.

NRC's Human Performance Investigative Process

As the American nuclear power industry evolved and matured the role of the Nuclear Regulatory Commission changed from reviewing plant designs, overseeing construction, and issuing licenses to focusing upon plant operation and maintenance. This change in focus identified the need to evaluate human and organizational performance. NRC inspectors required a tool to apply during investigations at nuclear plants. The NRC's Human Performance Investigation Process (HPIP) is a systematic method of investigating nuclear operating experience and the human element of plant performance. The NRC applies the HPIP method to investigate events involving human performance problems.

The HPIP methodology includes HPIP modules to provide more information to the investigator concerning frequent causes of performance problems at nuclear plants (Paradies, Unger, Haas, & Terranova, 1993a). The HPIP modules direct NRC investigators to the six areas considered most likely to contain deficiencies that could be a root cause of the event. These areas important to human and organizational performance are procedures, training, verbal communications, organizational factors, human engineering, and supervision (Paradies, Unger, Haas, & Terranova, 1993b).

Procedures are written guidance provided to workers and supervisors. This guidance includes maintenance manuals, operating procedures, equipment or system testing procedures, and emergency procedures. HPIP is concerned with the technical accuracy, usability, and administration of the procedures. The three key areas of interest concerning procedures are workers not using a procedure when a procedure should have been used, workers following a procedure incorrectly, and the use of a wrong or incomplete procedure (Paradies et al., 1993b, pp. 15-21).

Training must support performance of tasks performed at the plant. A task analysis should be performed and learning objectives and lesson plans available to support task mastery. Continuing training programs should be provided to maintain job proficiency. Events could be caused by a lack of training on a particular subject or system and situations where workers do not have the necessary knowledge of a task, a system, a response, or other needed information because of faults in the training program (Paradies et al., 1993b, pp. 23-27).

Verbal communications are considered spoken communications between workers, supervisors, and work groups while preparing for or performing work. Verbal communication difficulties could occur because of technical inadequacies in the communication equipment or because of inadequacies in the communication itself. Verbal communications between workers, supervisors, and work groups are expected to be understood. Communications must be timely. Information provided during turnover or shift relief must be correct, complete, and adequate (Paradies et al., 1993b, pp. 29-33).

Organizational factors are aspects of the management systems that influence the reliability of human performance or enhance organizational effectiveness in preventing events. This area identifies problems that senior management has the authority and responsibility to control and correct. It addresses the management process, but it does not identify errors by individual managers. These aspects include administrative controls, organization structure, work processes, problem discovery system, problem correction system, and motivational programs. Standards, policies, and administrative controls must be straight forward, complete, clear, strictly enforced, and adequate enough to provide management direction. Standards, policies, and administrative controls must be used, adhered to, and followed correctly. Management should be involved in an adequate self evaluation program that provides timely correction of deficiencies. An adequate external operating experience review program should be used to identify potential weaknesses. Corrective actions for known deficiencies or recurring failures are implemented adequately through a timely corrective action program. An effective internal review program to detect trends in performance, identify potential problems and initiate corrective action should be in place. Management must hear about problems whenever their involvement is needed to ensure adequate corrective action. Management must communicate their beliefs, safety emphasis, and programmatic direction for the organization and the work process. Management must provide the appropriate work environment to prevent performance problems (Paradies et al., 1993b, pp. 35-44).

Human engineering aspects that support reliable, consistently correct human performance are considered. Factors such as arrangement of controls, lighting, noise, cleanliness, labeling, and complexity are included. Equipment, systems, facilities, instrumentation, and controls with which people work must be designed to allow proper human interaction. A satisfactory work environment that includes housekeeping, adequate lighting, temperature control, radiation and contamination control, and reasonable noise levels must be provided. Systems should not be overly complex or complicated. System design should minimize the number of knowledge based decisions routinely required. Monitoring of more than three simultaneous variables should be avoided since this could cause confusion or indecisiveness. Systems should allow the detection and correction of errors (Paradies et al., 1993b, pp. 45-52).

Supervision is important during preparation for the task and oversight during task performance. Supervisory responsibilities of interest to this area include preparation for work, scheduling, assignment of workers, pre-work briefings, safety, job plans, work area walk downs, on the job oversight during work, and crew teamwork. Workers are expected to question improper or unexpected indications in spite of supervisor or coworker forcefulness. Investigators determine if improper directions by a supervisor are carried out even though the workers know them to be improper (Paradies et al., 1993b, pp. 53-58).

Nuclear Organizational and Human Performance Research

The Human Performance Enhancement System is a program to improve human reliability through correction of conditions that cause human performance problems. It encourages reporting of near miss situations to allow correction of situations that could cause errors or performance problems. The Human Performance Enhancement System identifies general areas that may contribute to adverse human performance: verbal communications, written communications, equipment design, degraded equipment condition, environmental conditions, work schedule, work practices, work organization, planning, supervisory methods, supervisory oversight, training, personnel qualification, change management, resource management, managerial methods, and nonwork related distractions (Ryder, 1993). These general areas are consistent with the areas identified as important to organizational and human performance by the NRC's Human Performance Investigative Process.

An NRC sponsored study by the Strategic Management Research Center of the University of Minnesota attempted to develop a logical framework to link management and organizational factors to nuclear power plant safety. Researchers derived four outcomes that had immediate links to nuclear safety: efficiency, compliance, quality, and innovation. Efficiency measures included: plant availability for electric production, plant reliability, hours critical, and outage rates. Hours critical indicates the time the reactor was in operation. Outage rates indicate the plant shutdown lengths. Compliance refers to the extent to which rules are followed. Quality is an expression of excellence. Innovation is the development and application of new knowledge (Marcus etal., 1990).

The Nuclear Regulatory Commission's Office for Analysis and Evaluation of Operational Data collects and analyzes nuclear plant operational data and serves as a focal point for evaluation of nuclear plant safety performance. This office analyses human performance in nuclear plant operating events as part of its routine analysis. It publishes an annual report of its conclusions. In the annual report published in November 1994 the office identified human performance weaknesses in procedures, operator knowledge, command, control, communications, human machine interface, and impromptu operator actions (NRC, 1994).

The International Atomic Energy Agency (1989, pp. 13-14) states that accidents are not usually caused by a single failure or a single human error. Instead a combination of errors are the cause. It identifies the following on a very general level as characteristics involved in large accidents: design, procedures, training; numerous people involved, complicated chain of system interaction, judgment, decision making, conflicting process objectives, seriousness not understood, and lack of experience. The Electric Power Research Institute identifies nine factors that shape human performance: task complexity, personnel selection, facilities, work structure, training, system equipment, policies, documents, and support equipment (Kinkade, 1988, p.8). The American national standard on human performance measurements (American Institute of Aeronautics and Astronautics, 1993, p. 37) lists five representative performance shaping factors: operational factors, equipment factors, task factors, personnel factors and external environmental factors.

The NRC studies human performance in nuclear plant organizations. Three studies in specific areas of nuclear plant performance are valuable to this research. While studying the effects of local control station design upon operation the NRC identified several factors worthy of evaluation. These factors include physical environment, communication, procedures, necessary equipment, and training. These five factors are consistently identified by other researchers. Labeling, maintenance, component design, and accessibility were also mentioned (O'Hara, Ruger, Higgins, Luckas, & Crouch, 1990). The NRC has identified relevant human factors in maintenance: procedures, documentation, management, attitude, training, maintainability design, and personal accountability (Inaba, 1991). While studying approaches to team skills training for nuclear plant control room teams the NRC identified necessary operational team skills that support this research: information exchange, information evaluation, task assignment, performance direction, performance feedback, coordination, strategy development, problem solving, and decision making (Davis, Gaddy, & Turney, 1985).

The Electric Power Research Institute (1991) identified key organizational characteristics to make success happen in a study on cost and quality management. These characteristics are consistent with those identified: planning, communication, employee belief in the need to change, management committed to continuous improvement, authority and decision making shared with employees, information infrastructure, and cost and quality programs as part of the overall continuous improvement.

According to the Electric Power Research Institute, "the crucial problems facing utilities today ... often revolve around non-quantifiable issues that stubbornly resist the quantitative analyses upon which utilities have come to depend" (Shaw, Goldberg, & Goldberg, 1988). The research reviewed in this literature search relied heavily upon subjective judgment and expert opinion. All sources cited in this literature search relied upon a review of appropriate literature. Many used industry examples from successful organizations for characteristics that they recommended.

In the Road to High Performance, A guide to Better Jobs and Better Business Results the U.S. Department of Labor (1994) used studies by others as primary evidence and discussions of successful implementations as justifying examples for the characteristic. The Commission on the Skills of the American Workforce (1990) interviewed managers and workers, collected economic and labor market data, compared interview results to data, applied the judgment of the members of the commission and made recommendations. Alvin Toffler's research for The Adaptive Corporation (1985) uses historical trends, predictions of the future, and interviews with American Telephone and Telegraph personnel as input data. Mills (1991), Meister (1994), Hammer and Champy (1993), Watkins and Marsick (1993), Pedler et al. (1991), Becker and Steele (1995) and Kanter (1989) used examples of industry practices to justify the organizational performance characteristics that they recommend.

The NRC's human performance investigative process (HPIP) was developed using a search of applicable literature and interviews with NRC personnel, consultants, and industry personnel familiar with root cause analysis of human performance. This research allowed the development of an investigator's manual for use by NRC investigations of performance problems. The investigator's manual was pilot tested by NRC personnel prior to actual use in the field (Paradies et al., 1993c).

The research methodology for Organization and Safety in Nuclear Power Plants (Marcus et al., 1990) consists of a literature search to identify characteristics of possible interest. The characteristics were subjected to expert judgment and compared against Licensee Event Reports to test hypotheses. Licensee Event Reports are reports filed by companies operating nuclear plants if specific technical specifications or required practices are violated. They are a technical indication of nuclear safety performance. The International Atomic Energy Agency (1989) used an expert task force to review risk assessment techniques for application to human reliability analysis. The NRC used a literature search and workshop of experts to identify needed team skills for nuclear plant control room teams (Davis et al., 1985).

During the development of Putting Strategy to Work: Tools for Cost and Quality Management in the 1990's the Electric Power Research Institute (1991) performed a literature search and a survey to identify desired characteristics of cost and quality management. In a different study the Electric Power Research Institute used interviews to identify competitive production strategies for existing power plants (Wayland & Brusger, 1987).

Research into high performance work environments, high performance organizations, nuclear safety characteristics, and electric production improvement methods rely heavily upon reviews of literature and the opinion of experts. In 1988 the Electric Power Research Institute cautioned the electric power industry that it must rely upon qualitative measures to meet performance research needs (Shaw et al., 1988). In 1990 the NRC studied the use of expert judgment and reported results in Eliciting and Analyzing Expert Judgment (Meyer & Booker, 1990). They found expert judgment to be an informed opinion based upon the expert's training and experience. Expert opinion, subjective judgment, expert forecast, best estimate, and expert knowledge are synonymous with expert judgment. According to the NRC expert judgment is comparable to any other form of data. "Expert judgment data is no less representative of the underlying truth than data from instruments or any other source of data" (Meyer & Booker, 1990, p. 20).

Nuclear power plant performance can be measured from many approaches. The most comprehensive is advocated by Tim Martin, a nuclear industry organizational consultant. Martin (1994) ranks nuclear plants by three dependent variables: NRC Systematic Assessment of Licensee Performance scores, capacity factor, and operating and maintenance costs. He believes that for a nuclear plant to be considered high performing they must be in the top quartile in each area. According to Donald Hintz (1993), the president and chief executive officer of Entergy Operations, these variables are used in benchmarking activities for nuclear plants.

NRC Systematic Assessment of Licensee Performance

The NRC Systematic Assessment of Licensee Performance (SALP) process is an integrated assessment of nuclear power plant safety performance. It focuses on nuclear safety. It allows the NRC to make decisions regarding each nuclear organization's fitness to operate their plant and the need for modified NRC oversight. The SALP report communicates NRC observations and judgments concerning safe nuclear operation to both company management and the public (Nuclear Regulatory Commission, 1993a).

Four functional areas are assessed. The plant operations area consists of activities directly related to the operation of the reactor and the supporting equipment. The maintenance area covers all corrective, preventive, predictive, and diagnostic maintenance performed on the plant. It also includes inspections, equipment testing, and calibrations. Engineering assesses the adequacy of technical and engineering support to the plant and the company's design modification process. Plant support includes radiation protection, emergency preparedness, security, chemistry, and fire protection (Nuclear Regulatory Commission, 1993b).

SALP assessments normally cover an 18 month period. At the end of this period the SALP report is issued. Ratings are combined for multiple reactors on one plant site. Each functional area is rated into one of three performance categories. Category 1 is the best rating. It indicates that the organization was properly focused upon nuclear safety and attained a superior nuclear safety performance for the period reported. Category 2 indicates a good level of safety performance that is normally well focused upon nuclear safety. Category 3 indicates an acceptable level of safety performance with some performance dimensions not adequate (Nuclear Regulatory Commission, 1993b). Although the NRC assesses each functional area separately, consultants (Martin, 1994) and managers (Hintz, 1993) average the scores together to make comparisons between plants. SALP scores are collected, tabulated and reported by the Inside NRC newsletter. Appendix A contains SALP scores for the period 1992 to 1994 (NRC SALP Scores, 1995).

Capacity Factor

Capacity factor is an indicator of electrical power generation over a period of time. Capacity factor is calculated for a plant by dividing the amount of electricity actually produced over a period of time by the amount of electric power that would have been produced if the plant were operated at its design electrical rating for the entire period. It allows comparison of plant performance, because it is adjusted for the design electrical production rating for each plant. Nuclear News magazine publishes capacity factor ratings for three year periods to show sustained performance over a period longer than one year. This allows the data to account for different refueling cycles and varying outage lengths (Blake, 1994).

For purposes of comparison Nuclear News proposes that plants with capacity factors below 50% are poor performers. Plants with capacity factors above 70% are good performers (Blake, 1994). Nuclear News reports capacity factor ratings for nuclear plants annually. Appendix A provides the capacity factor data for each nuclear facility for the period 1992 to 1994 (Blake, 1995).

Cost of Generation

"The fate of the U.S. nuclear power industry rests on its ability to control operations and maintenance costs" (Kuehn, 1994, p. 23). Because of the heightened environmental awareness in America today, nuclear power could become a desired energy source. However this will happen only if the production costs can be controlled and reduced (Kadak, 1994). Since the beginning of the commercial use of nuclear energy to produce electricity, nuclear plants have held a cost advantage over coal fired electrical generating stations (Kuehn, 1994). In 1987 the unthinkable happened. The average cost of production per kilowatt of electricity for nuclear plants exceeded the costs of coal fired plants (Industry Unveils Economic, 1993).

There are wide variances in nuclear plant performance and cost of generation even though the technology and regulatory environments are the same (Kadak, 1994). The nuclear industry's attempts to reduce operating costs to remain a player in the increasingly competitive electrical industry are made more difficult by the need to balance safety and economics (Bretz, 1994b). Nuclear plant management practices are the cause of the poor cost trends (Kuehn, 1994). Economic reviews indicate that significant improvements in operation could be achieved. Many utilities operating nuclear units could improve economic performance (Industry Unveils Economic, 1993). Improvement is on the way. After operating and maintenance cost increases of about 8% per year through the decade of the 1980's nuclear plants have leveled their costs. Their challenge is to reduce the costs without compromising safety or reliability (Bayne, 1994).

Nuclear plants report cost information as part of their required utility filings to the Federal Energy Regulatory Commission. It becomes public information. These data are tabulated by McGraw-Hill's Utility Data Institute and reported in the Nucleonics Week newsletter. Total production costs, maintenance costs, net electrical generation, and other data are reported. The costs in mills per kilowatt of electricity produced are useful to this study because these data represent the costs for electricity actually produced. They allow comparison between plants of different sizes. The lowest cost per kilowatt produced is preferable (Braidwood, Prairie Island, and Oconee, 1994). Average operating and maintenance costs per kilowatt for the period 1992 to 1994 are reported in Appendix A (U.S. Utilities Tell FERC, 1995).

Other Measures of Nuclear Performance

The Institute of Nuclear Power Operations (INPO) coordinates the World Association of Nuclear Operators Performance Indicator program for nuclear plants in the United States. These are technical indicators useful to managing a nuclear plant and comparing performance in specific technical areas to national targets. The indicators are industrial safety accident rate, volume of low level radioactive waste generated, personnel radiation exposure, thermal performance (how efficiently a plant converts thermal energy to electricity), nuclear fuel reliability, nuclear safety system performance, unplanned automatic scrams (reactor shutdowns caused automatically by safety systems), unplanned capability loss factor (energy not available to be supplied to customers because of plant problems), and unit capability factor (energy available to supply to customers and desired to be sold by management). INPO also performs evaluations at nuclear plants, provides assistance to plants, analyzes events, and facilitates information exchange (Institute of Nuclear Power Operation, 1995). Performance indicator data is reported for the industry at large. Performance indicators for individual plants and the results of evaluations and assistance visits are maintained as proprietary information between INPO and the plant. This data is not available for use in public research.

Cantor, Walker, and Fuller (1992) describe a discrepancy based methodology for evaluating organizational and human performance concerning safety in nuclear power plants. This methodology uses indicators of performance to identify deviations from expected organizational and human outcomes. It compares and correlates the deviations with other indicators to confirm, clarify, and explain the deviations. Each performance indicator has a predetermined standard that indicates safe operation. This system identifies and allows correction of performance deficiencies.

According to Roger Kaufman (Kaufman & Kaufman, 1992), a performance technologist from Florida State University, nuclear power plants are chartered to support the economy by providing safe affordable energy, without any penalty to the safety, health, or well being of employees or the public. When evaluating a high risk operation, an evaluator should compare the results expected to the actual deliverables. Criteria for evaluation of high risk operations such nuclear power operations should include considerations for the safety of the general population, the safety of employees, and the cost effective delivery of products and services. For nuclear power plants this would include delivery of electrical energy. Generation of electricity using nuclear power must not only be safe, it must also make positive economic and societal contributions. Evaluators should consider with three levels of results. The mega-level is concerned with the usefulness that the organization delivers to society. The macro-level considers the quality, price, and timeliness of delivery of the product to consumers. The micro-level studies the contributions that individuals or small groups make to internal organizational clients. Kaufman suggests that mission objectives for the organization be understood. Criteria, consequences, and inputs of conditions need to be assessed. Data that is independently verifiable should be collected and used. Also data based on judgments and attitudes collected from surveys or interviews should be analyzed. (Kaufman & Kaufman, 1992).

The review of the literature identifies the guideline for creating a high performance work environment published by the U.S. Department of Labor (1994). The Road to High Performance, A guide to Better Jobs and Better Business Results will provide a good basis for specific research on characteristics of high performance work environments. Other literature on high performance supports the characteristics proposed in the Department of Labor guide.

Tim Martin's (1994) three parameter approach to judging nuclear plant performance provides an accepted measure of selecting high performing and low performing plants. The parameters used by Martin are NRC Systematic Assessment of Licensee Performance ratings, cost of production, and capacity factor.

Chapter 3

RESEARCH METHODS

This study identifies the extent to which the practices recommended by the U.S. Department of Labor's Office of the American Workplace (1994) in the category Participation, organization, and partnership have been implemented at American nuclear plants. It also studies the relationship between the implementation level of those practices and the performance level of nuclear power plants.

Struening and Guttentag (1975, p. 3) suggest steps for a research study to ensure successful completion, acceptance, and applicability. They report the crucial steps as conceptualizing the problem, reviewing relevant literature, developing a research strategy, determining a research design, selecting a sample, choosing measures, selecting personnel to conduct the study, maintaining data collection standards, analyzing the data, and communicating the results. These steps are applied during this study.

The problem is fully addressed in Chapter One of this thesis. The answers to the two research questions stated in Chapter One are pursued. Research questions one is, to what extent have the practices recommended by the U.S. Department of Labor's Office of the American Workplace (1994) in the category Participation, organization, and partnership been implemented at American nuclear plants? Research question two is "what is the relationship between the implementation level of the practices recommended by the U.S. Department of Labor's Office of the American Workplace (1994) in the category Participation, organization, and partnership and the performance level of nuclear power plants?" The relevant literature has been reviewed and reported in Chapter Two of this thesis.

Research Strategy and Design

This study is a status study in which data concerning variables were collected but not manipulated. This strategy allowed the researcher to report status, association, and relationships among data but does not support inferences regarding causal relationships between variables (Baker & Schutz, 1972). This study strategy was directed toward answering the research questions.

This research is composed of four design components: acceptance of the recommendations of the U.S. Department of Labor concerning workplace practices, identification of the status of implementation of those recommendations in nuclear plants, determination of performance levels of individual nuclear plants, and comparison of actual nuclear plant performance levels to the implementation status of the recommendations. All four components combine to answer the research questions.

A key to this research was the acceptance of the recommendations identified in the Road to High Performance Workplaces, a Guide to Better Jobs and Better Business Results by the U.S. Department of Labor's Office of the American Workplace (1994). If the researcher cannot depend upon the work that led to the development of the recommended practices, then the results and conclusions of the study have little value. A review of the literature on high performance workplaces indicated that this document will become a primary guide to high performance work practices in the United States. Related literature discussed in Chapter Two supports the recommendations of the Department of Labor guide. Interviews with participants in the Department of Labor study suggest that the recommendations are valid. These interviews are discussed in Chapter Two. This study accepted the recommendations as valid for the improvement of organizational performance.

The Participation, organization, and partnership category of the guide was identified by the researcher as the area of interest to be studied, because the researcher believes this is an area of opportunity for improvement by the nuclear industry. This belief is supported by the literature reviewed and reported in Chapter Two.

The implementation level of the U.S. Department of Labor recommendations at nuclear plants was identified using mail surveys following Dillman's (1978) Total Design Method. The method is described in the paragraph on maintaining data collection standards in this chapter.

There are three valid measures of success for nuclear power plants in the United States. They are easily observed and available to researchers. The measures are nuclear safety performance as measured by the Nuclear Regulatory Commission's Systematic Assessment of Licensee Performance ratings, operating performance as measured by production capacity factors, and cost performance as measured by cost of generation per kilowatt hour of electricity produced. These measures define success for a nuclear power plant (Hintz, 1993).

Comparison of actual nuclear plant performance levels to the level of implementation of the recommendations at individual nuclear power plants is described in the section concerning data analysis in this chapter.

Because of the nature of this study and the research questions posed, a census that surveys all operating commercial nuclear power plants in the United States was performed. One senior manager from each of the 70 nuclear power plants in the United States of America was surveyed regarding the extent of implementation of the high performance workplace practices at the plant. Typically the senior manager surveyed is the vice president for operations or the nuclear plant manager. The nuclear plants are listed in Appendix A. This study deals with the population rather than a sample.

Data concerning the status of implementation of the Workplace practices recommended by the Department of Labor's guide (1994) was collected through a mailed questionnaire. According to Udinsky, Osterlind, and Lynch (1981) questionnaires are used most often to gather opinions and factual information in educational research and evaluation. Questionnaires are appropriate when the information can be requested in written form, when a large geographical area is to be surveyed, and when efficiency is important (Udinsky et al., 1981). Data was collected following Dillman's Total Design Method (1978). Dillman's method includes determining the kind of information sought, deciding on a question structure, wording the question precisely, formatting the questionnaire, formatting the pages, pretesting the questionnaire, writing the cover letter, preparing the return envelopes, assembling the mail out package, and sending follow-up mailings. Dillman's method was followed, except for his recommendation that questionnaires be bound into a booklet with front and back cover sheets. Instead the advice offered by Udinsky et al. (1981) to keep mail questionnaires short enough to be completed within ten to fifteen minutes and to give the appearance to the responder that the questionnaire will not require excessive time to complete was utilized.

By following Dillman's Total Design Method the survey solicited nuclear manager's beliefs concerning the implementation level of workplace practices at their nuclear power plants. A statement of the practice as suggested in the Department of Labor guide was provided. A scale was constructed for each workplace practice to allow selection from a range of no implementation to fully implemented. Responding managers were asked to select the position on a scale that identifies the extent to which each workplace practice was implemented in their plant on January 1, 1995. According to Dillman (1978) the respondent's task is to select the most appropriate response on the continuum. Dillman's (1978) advice to keep questions consistent in structure and ordered response to avoid confusing the responder was followed.

The questionnaire was constructed of the survey items of interest. Questions were ordered according to Dillman (1978). Easiest questions are first. Potentially objectionable questions are last. In this survey questions concerning the relationship between management and unions could be sensitive. Union relations questions were placed last. Questions were grouped according to content. Groupings of questions supported cognitive ties between questions. Vertical flow of answers was maintained. Answers on the questionnaire were consistently formatted from no implementation to fully implemented to avoid confusing the responder.

As suggested by Dillman (1978) the questionnaire was pre-tested by two types of experts. Five nuclear executives, representing the potential user group, were asked to review the questionnaire for clarity, terminology, appropriateness, and expected response. Five colleagues reviewed the questionnaire for clarity, terminology, bias, and consistency. Both groups of pretest responders were asked for suggestions for improvement. Time to complete the survey was estimated by colleagues who timed their completion of the questionnaire.

Upon approval of the questionnaire by the professor in charge of the program the questionnaire was submitted to the Pennsylvania State University, Office for Regulatory Compliance for protection of human subjects review and approval. After protection of human subjects approval, ten members of the population were sent questionnaires by following all of the procedures to be used in the survey. This test was designed to identify defects in the questionnaire or the survey procedure. After the first three questionnaires were satisfactorily completed and returned the questionnaire and survey procedure were considered satisfactory.

When the questionnaire was mailed, it was attached to a cover letter, an instructions and consent form, and a return envelope. The cover letter, questionnaire, and instructions and consent form are in Appendix B. A current list of nuclear plant managers was used for the mailing. One questionnaire was sent to the vice president or plant manager of each American nuclear plant. If a response was not received the name and address were verified and corrected as necessary. A follow up mailing was sent to non responding plants.

Research Question 1

Data analysis results provide information to answer the two research questions stated in Chapter One. Research question one examines the extent the practices recommended by the U.S. Department of Labor's Office of the American Workplace (1994) in the category Participation, organization, and partnership been implemented at American nuclear plants. The Department of Labor recommends ten workplace practices in this category. The practices studied are listed in Table 1.

Each workplace practice is studied independently. A frequency distribution is provided to indicate the extent of implementation for each workplace practice. The mean, standard deviation, and median of the implementation level of each practice is reported.

Table 1

The null hypothesis must be clearly stated in measurable terms. It should be directly related to the research being conducted (Baker & Schutz, 1972). The null hypothesis related to research question one indicated the practices recommended by the U.S. Department of Labor's Office of the American Workplace (1994) in the category Participation, organization, and partnership have not been implemented at American nuclear plants. Since this research deals with census data, the test of the null hypothesis was straight forward. Descriptive, rather than inferential statistics, were used to analyze this data for question one. The null hypothesis was tested for each workplace practice. If any implementation was reported for a workplace practice, then the null hypothesis for that workplace practice was rejected and the results discussed. If no implementation was reported, then the null hypothesis was accepted for that practice.

Research Question 2

Research question two examines the relationship between the implementation level of the practices recommended by the U.S. Department of Labor's Office of the American Workplace (1994) in the category Participation, organization, and partnership and the performance level of nuclear power plants. As in question one, each workplace practice is studied independently. The mail survey provides data measured on the interval scale which describes the extent of implementation of each workplace practice at each nuclear plant. Martin's (1994) method of ranking nuclear plants using three variables, NRC Systematic Assessment of Licensee Performance scores, capacity factor, and operating and maintenance costs provides measures for nuclear plant performance. The data for these variables are reported in Appendix A.

The data were analyzed by applying the least-squares regression method to describe the relationship between each recommended workplace practice and each measure of nuclear plant performance. The null hypothesis related to research question two is stated as there is no relationship between the level of implementation of the practices recommended by the U.S. Department of Labor's Office of the American Workplace (1994) in the category Participation, organization, and partnership and the performance level of nuclear power plants at the 95% significance level

(H₀:B=0; p< 0.05). If the slope of the regression line indicated a relationship exists (H₁:B¹ 0; p< 0.05) the null hypothesis was rejected and the relationship was described.

A Pearson product moment correlation coefficient was calculated to describe the relationship between the variables of interest. The coefficients were reported according to Fitz-Gibbons and Morris (1978). Coefficients from - 0.2 to + 0.2 are reported as no correlation. Weak correlation occurs from - 0.2 to - 0.4 and + 0.2 to + 0.4. Moderate correlation occurs form - 0.4 to - 0.6 and + 0.4 to + 0.6. Strong correlation occurs from - 0.6 to - 1.0 and + 0.6 to + 1.0.

Assumptions

The three measures of nuclear plant success proposed by Martin (1994) and Hintz (1993) are accurate measures of nuclear plant performance. This is a reasonable assumption because these measures are already accepted throughout the nuclear power industry as indicators of performance. Although advocates of high performance work practices suggest that there are other measures with which to judge performance, those measures are generally untested and not accepted in the electrical generation and nuclear power industry.

Senior managers at operating nuclear power plants know the implementation level of the recommended high performance workplace practices at their plants. This is a reasonable assumption, because each manager surveyed has operating responsibility for the plant.

The implementation status of the workplace practices on January 1, 1995 at each nuclear plant is related to the performance of the nuclear plant during the time frame 1992 to 1994. This assumes the implementation of the practice would have sufficient time to impact operation and the plant performance. The January 1, 1995 data point was chosen because it provided the best point in time to assess the implementation of the work practices. The three year period to measure performance is necessary to remove inaccuracies caused by variations in annual operating cycles among nuclear plants.

This research strategy allows reporting status and relationships among data but does not support inferences regarding causal relationships between variables (Baker & Schutz, 1972). According to Couch (1987), conclusions concerning causal relationships are possible only when the experimental method has been applied by manipulating independent variables and observing the impact upon dependent variables.

The responses consists of opinions of the responding managers rather than data retrieved from corporate files. Only one senior manager from each plant was surveyed. No attempt was made to ensure that other managers from that plant would respond in the same manner as the respondent. The opinions could be challenged concerning the level of implementation of the workplace practice. In the study published by the Nuclear Regulatory Commission on expert judgment Meyer and Booker (1990) report that expert opinion is comparable to any other data and is as representative of truth as any other source. Since the responders are responsible for management of the nuclear plants being surveyed their opinions are considered expert. This limitation is minimized.

Chapter 4

ANALYSIS OF FINDINGS

This study identifies the implementation level of selected high performance workplace practices at American nuclear power plants. The workplace practices studied were recommended by the U.S. Department of Labor's Office of the American Workplace in the guideline, Road to high performance, A guide to better jobs and better business results (1994). The practices are from the category, Participation, Organization, and Partnership.

This study also examines the relationship between the implementation level of the Department of Labor recommendations and the performance level of individual nuclear power plants. Performance levels are reported in Appendix 1. Implementation levels of the recommended workplace practices were obtained by a mail survey of senior managers at nuclear power plants.

These findings answer the two research questions. The data analysis and findings are reported in this chapter.

Research Question 1

To what extent have the practices recommended by the U.S. Department of Labor's Office of the American Workplace (1994) in the category Participation, organization, and partnership been implemented at American nuclear plants?

Table 2 is the frequency distribution of the implementation level of work practice number 1 as a percentage of fully implemented as reported by nuclear power plant senior managers for their nuclear power plant. Zero percent indicates "not at all" implemented. One hundred percent indicates "fully" implemented.

Table 2

Table 3 provides descriptive statistics to describe the implementation level of work practices at nuclear power plants.

Table 3

The null hypothesis related to research question one is stated as the recommended work practices have not been implemented at American nuclear plants. In order to accept the null hypothesis for an individual workplace practice, the mean of the implementation level for that work practice must equal zero percent implemented. Since all the means of the implementation level are greater than zero percent, the null hypothesis is rejected. This research indicates that all ten workplace practices recommended by the U.S. Department of Labor guideline in the category Participation, Organization and Partnership have been implemented in the American nuclear power industry to the extent presented in table 3.

What is the relationship between the implementation level of the practices recommended by the U.S. Department of Labor's Office of the American Workplace (1994) in the category Participation, organization, and partnership and the performance level of nuclear power plants?

High Performance Work Practice #1

The first work practice studied is workers are actively involved in continuously improving their work process and redefining their jobs.

Table 4 describes the relationship between the implementation level of work practice number 1 at each nuclear power plant and the three measures of performance for each nuclear power plant being studied. The performance measures are nuclear safety performance as measured by the Nuclear Regulatory Commission's Systematic Assessment of Licensee Performance ratings, operating performance as measured by production capacity factors, and cost performance as measured by cost of generation per kilowatt hour of electricity produced (Hintz, 1993).

Table 4

Since the t-ratio for capacity factor is greater than the critical t value for a confidence level of 95% (± 1.96) we reject the null hypothesis and apply the regression equation. The study indicates that there is a relationship between capacity factor and the implementation level of work practice number 1, workers are actively involved in continuously improving their work process and redefining their jobs.

The statistical beta weight of 18.4 for capacity factor indicates a positive relationship with work practice number 1. The data indicate that plants with a higher level of implementation of work practice number 1 have a larger capacity factor, which is an indicator of increased electrical production. The R-sq. value indicates that 8.4% of the increase in capacity factor can be explained by the relationship with work practice 1. The correlation coefficient for capacity factor (.289) is indicative of a weak relationship.

Since the t-ratios for nuclear safety rating and cost of generation are less than the critical t value for a confidence level of 95% (± 1.96) we fail to reject the null hypothesis. The study does not prove that there is a relationship between nuclear safety rating or cost of generation and the implementation level of work practice number 1, workers are actively involved in continuously improving their work process and redefining their jobs.

High Performance Work Practice #2

The second work practice studied is workers modify their work processes to correct quality, production, safety, or other procedures.

Table 5 describes the relationship between the implementation level of work practice number 2 at each nuclear power plant and the three measures of performance for each nuclear power plant being studied. The performance measures are nuclear safety performance as measured by the Nuclear Regulatory Commission's Systematic Assessment of Licensee Performance ratings, operating performance as measured by production capacity factors, and cost performance as measured by cost of generation per kilowatt hour of electricity produced (Hintz, 1993).

Table 5

Since the t-ratio for capacity factor is greater than the critical t value for a confidence level of 95% (± 1.96) we reject the null hypothesis and apply the regression equation. The study indicates that there is a relationship between capacity factor and the implementation level of work practice number 2, workers modify their work processes to correct quality, production, safety, or other procedures.

The statistical beta weight of 19.9 for capacity factor indicates a positive relationship with work practice number 2. The data indicate that plants with a higher level of implementation of work practice number 2 have a larger capacity factor, which is an indicator of increased electrical production. The R-sq. value indicates that 10.5% of the increase in capacity factor can be explained by the relationship with work practice 1. The correlation coefficient for capacity factor (.323) is indicative of a weak relationship.

Since the t-ratios for nuclear safety rating and cost of generation are less than the critical t value for a confidence level of 95% (± 1.96) we fail to reject the null hypothesis. The study does not prove that there is a relationship between nuclear safety rating or cost of generation and the implementation level of work practice number 2, workers modify their work processes to correct quality, production, safety, or other procedures.

High Performance Work Practice #3

The third work practice studied is workers are actively involved in problem solving, selecting new technology, modifying their product or service, and meeting with internal customers.

Table 6 describes the relationship between the implementation level of work practice number 3 at each nuclear power plant and the three measures of performance for that nuclear power plant. The performance measures are nuclear safety performance as measured by the Nuclear Regulatory Commission's Systematic Assessment of Licensee Performance ratings, operating performance as measured by production capacity factors, and cost performance as measured by cost of generation per kilowatt hour of electricity produced (Hintz, 1993).

Table 6

Since the t-ratios for nuclear safety rating and capacity factor are greater than the critical t value for a confidence level of 95% (± 1.96) we reject the null hypothesis and apply the regression equation. The study indicates that there is a desired relationship between nuclear safety rating and the implementation level of work practice number 3, workers are actively involved in problem solving, selecting new technology, modifying their product or service, and meeting with internal customers. There is also a desired relationship between capacity factor and work practice number 3.

The statistical beta weight of - 0.49 for nuclear safety rating indicates a desired negative relationship with work practice number 3. The lower the Nuclear Regulatory Commission's Systematic Assessment of Licensee Performance score, the better the nuclear safety performance for the plant. The data indicate that plants with a higher implementation level of work practice number 3 have better nuclear safety performance. The R-sq. value indicates that 11.8% of the decrease in NRC scores can be explained by the relationship with work practice 3. The correlation coefficient for nuclear safety rating (- .344) is indicative of a weak relationship.

The statistical beta weight of 23.2 for capacity factor indicates a positive relationship with work practice number 3. The data indicate that plants with a higher level of implementation of work practice number 3 have a larger capacity factor, which is an indicator of increased electrical production. The R-sq. value indicates that 16.7% of the increase in capacity factor can be explained by the relationship with work practice 3. The correlation coefficient for capacity factor (.408) is indicative of a moderate relationship.

Since the t-ratio for cost of generation is less than the critical t value for a confidence level of 95% (± 1.96) we fail to reject the null hypothesis. The study does not prove that there is a relationship between cost of generation and the implementation level of work practice number 3, workers are actively involved in problem solving, selecting new technology, modifying their product or service, and meeting with internal customers.

High Performance Work Practice #4

The fourth work practice studied is when individuals or teams make suggestions, they receive feedback about their suggestions.

Table 7 describes the relationship between the implementation level of work practice number 4 at each nuclear power plant and the three measures of performance for that nuclear power plant. The performance measures are nuclear safety performance as measured by the Nuclear Regulatory Commission's Systematic Assessment of Licensee Performance ratings, operating performance as measured by production capacity factors, and cost performance as measured by cost of generation per kilowatt hour of electricity produced (Hintz, 1993).

Table 7

Since the t-ratios for all three performance measures are less than the critical t value for a confidence level of 95% (± 1.96) we fail to reject the null hypothesis. The study does not show a relationship between the measures of nuclear power plant performance and the implementation level of work practice number 4, when individuals or teams make suggestions, they receive feedback about their suggestions.

High Performance Work Practice #5

The fifth work practice studied is workers are organized into teams with substantial team authority.

Table 8 describes the relationship between the implementation level of work practice number 5 at each nuclear power plant and the three measures of performance for that nuclear power plant. The performance measures are nuclear safety performance as measured by the Nuclear Regulatory Commission's Systematic Assessment of Licensee Performance ratings, operating performance as measured by production capacity factors, and cost performance as measured by cost of generation per kilowatt hour of electricity produced (Hintz, 1993).

Table 8

Since the t-ratio for nuclear safety rating is greater than the critical t value for a confidence level of 95% (± 1.96) we reject the null hypothesis and apply the regression equation. The study indicates that there is a desired relationship between nuclear safety rating and the implementation level of work practice number 5, workers are organized into teams with substantial team authority.

The statistical beta weight of - 0.54 for nuclear safety rating indicates a desired negative relationship with work practice number 5. The lower the Nuclear Regulatory Commission's Systematic Assessment of Licensee Performance score, the better the nuclear safety performance for the plant. The data indicate that plants with a higher implementation level of work practice number 5 have better nuclear safety performance. The R-sq. value indicates that 21.1% of the decrease in NRC scores can be explained by the relationship with work practice 5. The correlation coefficient for nuclear safety rating (- .459) is indicative of a moderate relationship.

Since the t-ratios for capacity factor and cost of generation are less than the critical t value for a confidence level of 95% (± 1.96) we fail to reject the null hypothesis. The study does not prove that there is a relationship between capacity factor or cost of generation and the implementation level of work practice number 5, workers are organized into teams with substantial team authority.

High Performance Work Practice #6

The sixth work practice studied is cross functional teams and other mechanisms are used to increase innovation across organizational boundaries.

Table 9 describes the relationship between the implementation level of work practice number 6 at each nuclear power plant and the three measures of performance for that nuclear power plant. The performance measures are nuclear safety performance as measured by the Nuclear Regulatory Commission's Systematic Assessment of Licensee Performance ratings, operating performance as measured by production capacity factors, and cost performance as measured by cost of generation per kilowatt hour of electricity produced (Hintz, 1993).

Table 9

Since the t-ratio for nuclear safety rating is greater than the critical t value for a confidence level of 95% (± 1.96) we reject the null hypothesis and apply the regression equation. The study indicates that there is a desired relationship between nuclear safety rating and the implementation level of work practice number 6, cross functional teams and other mechanisms are used to increase innovation across organizational boundaries.

The statistical beta weight of - 0.47 for nuclear safety rating indicates a desired negative relationship with work practice number 6. The lower the Nuclear Regulatory Commission's Systematic Assessment of Licensee Performance score, the better the nuclear safety performance for the plant. The data indicate that plants with a higher implementation level of work practice number 6 have better nuclear safety performance. The R-sq. value indicates that 13.8% of the decrease in NRC scores can be explained by the relationship with work practice 6. The correlation coefficient for nuclear safety rating (- .371) is indicative of a moderate relationship.

Since the t-ratios for capacity factor and cost of generation are less than the critical t value for a confidence level of 95% (± 1.96) we fail to reject the null hypothesis. The study does not prove that there is a relationship between capacity factor or cost of generation and the implementation level of work practice number 6, cross functional teams and other mechanisms are used to increase innovation across organizational boundaries.

High Performance Work Practice #7

The sev