CODI: Cornucopia of Disability Information
Chapter VI: Models Of Knowledge Utilization
Chapter VI: Models of Knowledge Utilization
I. INTRODUCTION TO MODELS
A. Criteria for models
Models discussed in this study provide frameworks for representing the factors or variables in the knowledge utilization process. Not only do the models identify the key variables and their relationship to one another but they also illustrate the sequence the variables follow for best results. While the diffusion research literature as meta-analyzed by Rogers (1983), has a system of propositions to support some of its models, the utilization-focused models seldom do. This study presents first the early models overviewed by other researchers and deemed widely accepted. Next it adds models from the literature based on the following broad criteria:
--includes the major steps in the knowledge utilization process rather than developing only one or two of those steps;
--the model links the utilization steps to the knowledge creation and diffusion components of the knowledge cycle;
--if proposed prior to the eighties, the model has been referenced or supported by other researchers.
Several authors have identified criteria for assessing models. Davis & Salasin (1979, as referenced in Glaser 1983:418-419) proposed 12 characteristics for a good model of change. Those characteristics are: practical, manipulable, economy of use, ease of communication, comprehensive, synergistic, phased, individual components workable, flexible and versatile, basis for subsequent evaluations, recognizes human characteristics, and calls attention to how the change process influences the rest of the system. Zaltman (1979:518-520), in discussing models of consumer behavior, defined the properties of a model as follows: (1) capable of explanation as well as prediction, (2) general, (3) high in heuristic power, (4) high in unifying power, (5) internally consistent, (6) original, (7) plausible (have face validity), (8) simple, (9) supported by facts, and (10) testable or verifiable (at least in principle). The purpose of the model dictates the degree to which each property is met within any given model.
While models help simplify the relationships between factors, they also can sometimes mislead. Zaltman warns us that since models simplify, it is easy to "draw artificial distinctions between concepts which in actuality are inseparable" (Ibid, 517).
Glaser, Abelson, and Garrison (1983:418) remind us that models are often hypothetical and heuristic and do not usually represent established laws or verities. Models, they say, may stem from theory, experience, or validated evidence.
B. Kinds of knowledge utilization models
There are many classification schemes for models in knowledge utilization. Weiss (1979), writing from a public policy perspective, refers to models driven by knowledge, problems, interactions, politics, tactics, or enlightenment. Dunn (1980:517), also writing from the public policymaking perspective, groups models as product-contingent, inquiry-contingent, problem-contingent, structure-contingent, or process-contingent. Glaser, Abelson, and Garrison (1983) refer to models as including the following perspectives: research, development, and diffusion; social interaction; problem-solving; planned change; and action research. Each of the three breakdowns of models includes those that begin with the research product and those that begin with the end user. Each breakdown includes an interaction-based model. All three include a problem-oriented one.
There are still other ways to group models such as by end users: researchers, public policymakers, administrators/ chief executive officers, practitioners, the general public, or targeted consumers. Models could be contrasted in terms of static vs. dynamic, systems-oriented vs. non-systems oriented, progressive and non-progressive, integrated and not integrated. They could be described contextually as related to individuals, organizations, and societies. They may be grouped based on the research traditions from which they are derived. They could be organized according to the phase of the process to which they relate. Lastly, they could be delineated by the time periods in which they first appeared in the literature.
The next section briefly highlights the more widely accepted models overviewed by Glaser, Abelson, and Garrison (1983; Human Interaction Research Institute, 1976) and by Weiss (1979). These models differ in their ability to handle complex vs. simple areas of knowledge utilization. Some are linear while others are not.
II. WIDELY ACCEPTED MODELS OF KNOWLEDGE UTILIZATION
Glaser, Abelson, and Garrison (1983:405-6) described five early and widely accepted models. The first three were originally set forth in 1969 by Havelock and the last two by Sashkin, Morris, and Horst (1973) as well as others. Each model represents a different perspective and area of emphasis in the knowledge utilization process.
A. Research, Development, and Diffusion Model
The first model referred to as research, development, and diffusion (RDD) reflected the attitude that if research is made known and presented in the right way, targeted audiences would use it. That model involves a sequence of research, development, and packaging activities prior to dissemination, large-scale planning, division of labor, separation of roles and functions, and evaluation (Glaser, 1983:404). This is a linear model. It begins with the research product and its packaging rather than the ultimate user and their needs. The user has little, if any, input into the research design. The two-communities mindset is well entrenched in this model. If a feedback loop were added from the user to the researcher, that would be a beginning enhancement but still not sufficient for today's complexities. Weiss (1979:427) also referencing Havelock (1969) describes a related model, the knowledge-driven model. She points out that it derives from and is used mostly by the natural sciences. It usually includes the following sequence of events:
Basic Research -> Applied Research -> Development -> Application
"The assumption is that the sheer fact that knowledge exists presses it toward development and use" (Ibid). While this works well for the natural sciences, the social science knowledge--
...is not apt to be so compelling or authoritative as to drive inevitably toward implementation....Social science knowledge does not readily lend itself to conversion into replicable technologies, either material or social. Perhaps most important, unless a social condition has been consensually defined as a pressing social problem, and unless the condition has become fully politicized and debated, and the parameters of potential action agreed upon, there is little likelihood that policy-making bodies will be receptive to the results of social science research (Ibid).
B. Social Interaction Model
The second model, the social interaction model, focuses on human relationships and influencing strategies at each stage of the dissemination and adoption processes (Glaser et al, 1983; Zaltman et al, 1977). The importance of face-to-face contacts and of understanding communication networking and psychological reference groups enter into this model. Note that in this model the interaction begins from the researcher's perspective only at the dissemination stage not at the research design stage.
The interactive model, according to Weiss (1979:428-9) and Zaltman (1977:159), is not linear. Instead policymakers seek information, knowledge, and opinions from a variety of sources including social scientists but not limited to their research findings. Administrators, practitioners, politicians, planners, journalists, clients, interest groups, aides, friends, and social scientists may contribute information useful to policymaking.
Seldom though do they have conclusions specific to the policy issue nor do they have convergent evidence. "In this model, the use of research is only one part of a complicated process that also uses experience, political insight, pressure, social technologies, and judgment" (Weiss, 1979:429).
In a study of innovation adoption Bingham, Freeman, and Felbinger (1984) reported support for the interactive, nonlinear model. They said: "the interactive terms made an independent contribution to the explained variance in every case" (Ibid, 333). They added that in one instance the interactive terms explained 60% of all of the variance. They recommended greater attention be paid to developing and refining interactive models. These authors also report that Downs and Mohr (1976:712-713) proposed that interactive models would be best for developing an integrated theory of knowledge utilization (Bingham et al, 1984:311).
C. Problem-Solving Model
Generally, in the problem solving model a problem exists, a decision has to be made, and a gap in information, understanding, or knowledge exists for making the decision. The pending decision drives the search for knowledge and subsequent application of that knowledge. Goals are generally already agreed upon and the research is to help identify and select appropriate means to reach the goal. While in many instances knowledge is sought among existing research findings, Weiss (1979:427) says, occasionally policymakers will commission social science research to fill knowledge gaps. Sometimes those gaps remain.
The model includes the following sequence of events.
Define pending decision --> Identify missing knowledge --> Acquire social science research --> Interpret research for the decision context --> Choose a policy
Lack of information can thwart the effects of this model; bypassing this step also can thwart its effects. Weiss points out that occasional studies have affected decision-making but usually on "relatively low-level narrow-gauge decisions." Most studies seldom leave any discernible mark on the direction or substance of policy (Ibid, 428). The extraordinary circumstances that might influence policy decisions directly would include:
a well-defined decision situation, a set of policy actors who have responsibility and jurisdiction for making the decision, an issue whose resolution depends at least to some extent on information, identification of the requisite informational need, research that provides the information in terms that match the circumstances within which choices will be made, research findings that are clear-cut, unambiguous, firmly supported, and powerful, that reach decision-makers at the time they are wrestling with the issues, that are comprehensible and understood, and that do not run counter to strong political interests (Ibid, 428).
The prevalence of this model of research utilization may account for some of the disillusionment regarding the applicability of social science research to social policy. "Because people expect research use to occur through the sequence of stages posited by this model, they become discouraged when events do not take the expected course" (Ibid).
Boyd and Menlo (1984:59-74) expanded the problem solving model. Their prescriptive model for knowledge utilization in problem solving has three realms: cognitive, instrumental, and the psychosocial. The instrumental realm has seven stages: (l) situational problem (with tactical seeking questions and choices), (2) problem situation (with strategy seeking questions and choices), (3) general inquiry statement, (4) knowledge-seeking question (that if not answered can lead to new areas of scholarly inquiry), (5) scientific generalization (with consideration of ontological and epist-emological issues), (6) practice generalization (emphasizing ontological issues), and (7) design generalization (emphasizing valuation issues). The cognitive realm relates to each stage with its thinking styles, thinking formats, process knowledge, repertoire use, verbal language usage, etc. The psychosocial realm adds the feelings, needs, personality characteristics, dynamics, norms, pressures, and so forth to round out the model.
D. Planned Change Model
The fourth model, the planned change model, assumes that "change occurs through a consciously controlled, sequential, and continuous process of data generation, planning, and implementation" and that stabilization and support of the changes are important (Glaser, 1983:405). Because this model stems from research on organizational change, it is not unusual that Weiss does not mention this model in her discussion of models applicable to policymakers.
E. Action Research Model
The fifth model, the action research model (as defined by Patton), resembles the planned change and problem-solving models while emphasizing research creation by, and research use within, the same organization. Critics of this model complain that efforts are not always made to transfer knowledge from action research projects into the knowledge pool for all organizations to readily access. On the other hand, this model has the advantage of increasing use among potential users housed in the same organization as the researchers. "Due to an emphasis on trust and control, individual managers in the public sector will tend to use predominantly that research which is produced within their own organizations. Thus, it is natural to expect that R&D managers will also be R&D users" (Rich & Goldsmith, 1982: 432).
F. Widely accepted models and the prosumer approach
The prosumer approach captures the best elements of each of the widely accepted models. The prosumer approach is flexible, beginning with either the problem or the consumer-need-based product. It is interactive and interdependent not just during dissemination and use but throughout every phase of the knowledge cycle. It includes planned change but is not restricted to planned change as its only form of change. The prosumer approach encourages research within organizations as well as between and among organizations. Unlike action research based on tall hierarchical organizational structures, the prosumer approach emphasizes flat structures. Communication is open in all directions. Responsibility, power, and control is shared. The prosumer approach suggests ways to help evidence from many sources converge into a common body of knowledge on a subject.
III. OTHER MODELS IN THE LITERATURE
A. Systems-oriented models
Several models have a heavy systems orientation. The characteristics of systems models according to Glaser, Abelson, and Garrison (1983) include: (1) interactive components such as inputs, transformations, and outputs; (2) "a dynamic flow of 'signals' (i.e., messages) or 'effects' (i.e., influences), (3) a feedback system for modifying activities, (4) boundaries among the supersystem, system, and subsystem, and (5) a controlling mechanism."
Glaser et al (1983:406-407) reviewed the following:
--Closed vs. bipolar loop problem-solving model (Havelock, 1977)
--Framework on innovating in organizations (Munson and Pelz, 1980)
--Innovation factor system (Souder & Rubenstein, 1976)
--Cybernetic societal guidance model (Etzioni, 1965)
--Systematic assessment of new technology model (Dobrov, 1978)
--Social network technical/political/cultural cycles model (Tichy, 1980)
--Technology delivery crisis model (Wenk, 1979)
--A formula-type model, e.g., A VICTORY/Decision determinants analysis (Davis, 1971/1978; Davis and Salasin, 1975/1979)
The A Victory model has been tested, refined, and renamed the decision determinants analysis (DDA). It originally consisted of nine factors that created the anachronym, A VICTORY: Ability, Values, Idea/Information, Circumstances, Timing, Obligation, Resistance, and Yield. The model was envisioned as useful for accounting for organizational behavior related to the use of either promising new knowledge or validated innovative procedures/practices/products. Forty questions concerning the potential change situation and a basic working list of about fifty prescriptive suggestions accompany the model (Glaser, et al, 1983:30; Davis and Salasin, 1975).
Within an action research context, A. W. Clark (1975) described the client and practitioner relationship at the system rather than individual level. Each is in a separate and closed system that can become open and interdependent. The bases of interdependence is the task system, external system, value system, reward system, and power system. The technical systems surrounding the client, practitioner, and action researcher also influence interdependency. The study emphasizes the need for exchange with adequate rewards for each party in the relationship.
More recently, several authors have pointed to the systems approach as the best way to organize and account for the social impacts of science. Those impacts may be behavioral, structural, or intensity of penetration of science on social action (Cozzens, 1987:315). Holzner, Dunn, and Shahidullah (1987) describe the social system of knowledge as "the complex of institutions, organizations, occupations, and their norms, social roles, and resources that constitute the social arrangements within which knowledge-related activities are conducted" (Ibid, 182-183). While their view of the social system of knowledge institutionalizes the validation of information and the quality of knowledge, they do not restrict the system to only scientific knowledge.
They say: "it makes little sense to call all codified professional expertise scientific, yet this type of information is of vital importance to the functioning of the knowledge system" (Ibid, 183).
What constitutes appropriate knowledge is determined in various domains of the knowledge system. Those domains include industrial engineering, mass communications, medicine, education, agriculture, and so forth. Some domains have clearly "specialized and differentiated structures, while others have more diffuse structures" (Ibid, 187). The domains identify the standards and the "degree of reliance on scientific knowledge" (Ibid, 187).
Institutions and organizations in the knowledge system, Holzner and others say, facilitate or limit the impact science has on society in six areas: "(1) the structure of science specialties, (2) scientific conclusions and evidence, (3) scientific methodologies, (4) science-based technologies, (5) science-based policies and organizational designs, and (6) structure of scientific institutions" (Ibid, 188).
Augsburg workshop participants who reviewed this proposed knowledge system said: "we must distinguish among science as a mode of thought, science as an archive of knowledge, science as new research results, and science embodied in trained manpower" (Cozzens, 1987:314).
The functions of the knowledge system that occur across and within different domains, according to Holzner, Dunn, and Shahidullah (1987), include: knowledge production, knowledge structuring, knowledge storage, knowledge distribution, knowledge utilization, and knowledge mandating. Those functions occur in different domains such as industry, agriculture, education.
They believe the advantages of adopting a knowledge system perspective for designing impact indicators include: (l) enables "measuring of causal interconnectedness"; (2) enables the "transition to a multidimensional indicator system"; (3) "can identify the pathways of impacts and can assist in the identification of areas of policy intervention... or points of leverage"; (4) "can yield indicators relevant to the multiple and at times conflicting goals of actors at all levels and regions of the system" (Ibid, 180).
B. Policy-oriented models
Weiss (1979) described four additional models, all policy-oriented: political, tactical, enlightenment, and intellectual enterprise. Under the political model, policy-makers use research to support their predetermined position. "When research is available to all participants in the policy process, research as political ammunition can be a worthy model of utilization" (Weiss, 1979:429). Under the tactical model, policymakers use the fact that they are sponsoring the research to demonstrate responsiveness and excuse their delayed actions in terms of waiting for the research results.
They may also use research to deflect criticism for unpopular policy outcomes. Support of research may also lead to building a constituency of academic supporters who can assist under congressional reviews. In the enlightenment model users of research cannot pinpoint a specific study or group of studies on which their policies or decisions are based. "The imagery is that of social science generalizations and orientations percolating through informed publics and coming to shape the way in which people think about social issues....Research sensitizes decision makers to new issues and helps turn what were nonproblems into policy problems" (Weiss, 1979:429).
Deficiencies associated with this model include the promulgation of invalid as well as valid generalizations. Incomplete, inadequate, oversimplified, and wrong findings, Weiss says, often seep into the ground waters of knowledge along with the quality research studies. The percolating process sometimes is so long that the research results may be out of date by the time they arrive or they may never reach the policy makers. Then too, "as more studies are done, they often elaborate rather than simplify. They generate complex, varied, and even contradictory views of the social phenomena under study, rather than cumulating into sharper and more coherent explanation" (Ibid, 430).
Social science research is used also as one of the intellectual pursuits of society (Ibid, 430-431). "Social science and policy interact, influencing each other and being influenced by the larger fashions of social thought" (Ibid, 430). Emerging policy interest in a social issue often leads to appropriations of research funds that subsequently attract social scientists to certain areas of study or confinements within those areas of study. "Later, as social science research widens its horizons, it may contribute to reconceptualization of the issue by policy makers. Meanwhile, both the policy and research colloquies may respond, consciously or unconsciously, to concerns sweeping through intellectual and popular thought"(Ibid).
C. Integrated organization-oriented model
Michael Huberman (1987) developed an integrated model of research utilization that is based on a confirmatory study conducted in a Swiss setting. His beginning steps to the development of a superordinate conceptual model integrates the instrumentalist and transactional paradigms that lie on either extreme of a conceptual continuum in research utilization. The instrumentalist paradigm assumes a passive, reproductive use of research information "housed within a Weberian organization" (Ibid, 590). The transactional or conflict-theoretic perspective, on the other hand, assumes an active strategic use of research, one that transforms information "to preserve cognitive consonance or to legitimate interventions within a 'bargained' social environment" (Ibid). He illustrates how each paradigm is partially valid and are "empirically interwoven in ways that are also conceptually plausible from a cognitive, social-psychological, and intraorganizational perspective" (Ibid).
Huberman's integrated model has four parts: (1) General Model, (2) Organizational Model: Researchers, (3) Organizational Model: Users, and (4) Dissemination Effort Model. Each breakdown represents a set of major variables and their hypothesized interrelationships. The General Model outlines the major variables. The organizational factors within the researcher and user contexts drive the linkage/coupling variable set. The degree of linkage between researchers and users directs the intensity and quality of the dissemination effort. That effort produces costs to researchers and a positive or negative impact for users. Secondary effects may occur from the dissemination effort or from the impact/use. In a separate more classic-based chain of actions, use is not directly dependent on what disseminators do. That chain flows from the users and predictors of local use to the impact/use variable set without direct linkages or dissemination efforts.
When looking at the model from the researchers' perspective, the organizational factors include the researchers' experience level with diffusion and utilization as well as the status and associated rewards or disincentives for promoting diffusion and utilization. The researchers' nesting in their home organization, Huberman says, relates to hierarchical level, stability, proximity to influential units or persons, relative centrality or marginality of the researchers' conceptual and methodological stance. From the users' perspective, the organizational factors relate to experience with research, the status of research in the organization, familiarity with or mastery of the scientific research, and characteristics of organizations.
Linkage and coupling from the researchers' perspective and the users' perspective are interdependent. Details about the intermediaries are given on the researchers model. Those include credibility, communication skills, mastery of substance of study, degree of infrastructure, and level of involvement.
Impact on users may be positive or negative, durable or not. Potentially positive uses include conceptual, instrumental, and strategic use. Negative ones include confusion, uncertainty, delayed action, intra-organizational tensions or conflicts. Other user-related factors under impact/use include the extent of use, scope of impact, relative influence, and degree of transformation vs. fidelity. Impact on researchers may also be positive or negative; though in Huberman's model, impacts are described in terms of conceptual, attitudinal effects, instrumental effects, and costs. The costs relate to quality of the study or other tasks, conflicts, prestige, credibility, and negative effects on career.
The last of the four breakdowns of the integrated model centers on the dissemination effort. This breakdown elaborates on the researcher and user contexts, the linkage mechanisms, predictors of local use, and the actual dissemination effort. Among the researcher contextual factors are the user-centeredness of the study and the orientation towards diffusion and utilization in the study. Among the user contextual factors are perceptions about the study's worth, influence, realism, and linkage to needs or priorities. The credibility or reputation of the research team, commitment of key leaders, and the presence or absence of an institutional mandate factor into the user contextual variables set.
Huberman separates from the user contextual variable set the following predictors of local use: users' understanding of the main findings, time and resources devoted to their use, compatibility with organizational objectives or users' opinions, and their perceived quality or validity. These predictors cluster differently for different user publics.
Of the variables listed under the dissemination effort Huberman says "the strongest, most systematic correlations are found for the list under D & U savviness: user-specific products, multiple channels, redundancy and reinforcement, use of personal contacts, and different types of follow-through. Research on communication and on attitude change heightens the face validity of these factors, many of which have found their way into advertising and political consultation" (Ibid, 604-5).
In reporting preliminary results after 12 months of data collection, Huberman said that "the models have performed well in the field" (Ibid). Some unbundling and refinement of variables has been needed. He indicates that initial confirmation exists for the "nonpertinence of factors not included on the model, which is itself another validation index" (Ibid, 606). The 1987 initial data analysis also reinforced the credibility of the linkage theorem of research dissemination.
D. Practitioner-oriented models
Two models in particular stand out among practitioner-oriented models: (1) the concerns-based adoption model (CBAM) and the progressive or scientist-practitioner model. While the CBAM does not link its utilization component readily with the research creation and dissemination components of the knowledge cycle, it has features worthy of consideration for subsequent model development. This author chose to place it in the practitioner framework, even though it could have just as easily been placed under the organizational framework, because of its teacher orientation.
1. Concerns-based adoption model (CBAM) The concerns-based adoption model (CBAM) addresses the organizational and individual perspectives of innovation adoption (Hall, Wallace, & Dossett, 1973 in Loucks & Hall, 1977). This model stems from the educational system and has been used primarily with teachers. Two important dimensions pervade the model: (1) the affective dimension, i.e., concerns about the innovation; and
(2) the behavioral dimension, i.e., skill and sophistication levels in the use of the innovation. The model juxtaposes a broad sequence of self-related, task-related, and impact-related concerns about the innovation with the sequential skill development in the use of the innovation: orienting, managing, and integrating.
In a more detailed form the model consists of seven stages of concern about the innovation and eight levels of use of the innovation. The stages of concerns begin with lack of awareness and then move into the information, personal, management, consequence, collaboration, and refocusing. The informational stage of concern relates to the general nature of the innovation and its present uses. The personal concern involves reflection on what in the present role needs to change to use the innovation. The management concern stage encompasses logistical issues related to the pending change. The consequence stage is concerned with the impact on the practitioner's audience, i.e., students, clients, patients, etc.. In the next stage the potential user is concerned about collaboration needs and benefits surrounding the innovation. Then the last concern relates to broadening one's perspective regarding the innovations' relationship to other similar ones. A key question that may arise at this stage is: Would another innovation achieve the same goals more effectively?
The levels of use of the innovation associated with the CBAM are: nonuse, orientation, preparation, mechanical use, routine, refinement, integration, and renewal. From nonuse the potential user orients himself or herself to the innovation and decides whether to adopt it (unless it is mandated already). Next the potential user prepares for adoption whether through training, resource development, procedural modification, and so forth. Mechanical use precedes routine use and "reflects lack of effective management and lack of anticipation of more than day-to-day needs, problems and events" (Loucks & Hall, 1977). From an established routine for problem handling surrounding the innovation, the teacher moves on to refinement of usage, including eliciting feedback from recipients of the changes, e.g., students.
In the last two levels of innovation use, the user integrates usage with other practitioners and looks for other innovations to supplement or perhaps even replace the innovation.
2. Progressive model The roots for the progressive or scientist-practitioner model are embedded in the 1949 Boulder Conference sponsored by the National Institute of Mental Health and the American Psychological Association (under David Shakow's leadership). That conference focused on curriculum development for educating students in clinical psychology and drew representatives from 71 universities, mental health agencies, and allied professions. Barlow, Hayes, and Nelson (1986:4-10) state that this consensus-building event put the stamp of approval on the logic and integrity of the scientist-practitioners' model.
Their book describes the attempts to implement the model and the factors in the growing field of psychotherapy that thwarted full implementation at that time. Obstacles included: (1) disillusionment with traditional research methods for the clinical setting, (2) little control over the training of psychotherapists, and (3) limited use of traditional research methods by practitioners since the conference.
Prior to that conference Frederick C. Thorne published in American Psychologist (1947) his views on how the practice of clinical psychology would develop. Although it didn't develop that way, his ideas are relevant to envisioning new directions in knowledge utilization for the rehabilitation field.
Formerly, clinical technology was mainly empirical, i.e., based on study and 'experience' rather than experiment....Ideally, diagnosis (description) and treatment of each individual case may be regarded as a single and well-controlled experiment. The treatment may be carefully controlled by utilizing single therapeutic factors, observing and recording results systematically, and checking through the use of appropriate quantitative laboratory studies. In addition to the general scientific orientation to the individual case there are frequent opportunities in clinical practice to conduct actual experiments to determine the validity of diagnosis or the efficacy of treatment....Individual clinicians are encouraged to apply experimental and statistical methods in the analysis of case results and larger scale analyses are made of the experience of the whole clinic over a period of years. Thus, the clinician comes to regard each individual case as part of a larger sample (Ibid, 159-160, quoted in Barlow, Hayes, and Nelson, 1986:5).
Agras and Berkowitz (1980) subsequently developed a model around the scientist-practitioner which Raymond Kent (1985) later adapted. These model builders refer to the model as a progressive model of clinical research. The model begins with an assessment of the status of the given intervention field and of sources from whence new techniques may emerge, such as clinical innovation, basic research, or new theoretical developments. Then practitioners research via clinical observations and uncontrolled clinical tests while scientists begin with basic theoretical and laboratory research. Each may generate new interventions or measurement that may be subjected to short-term outcome studies that compare the new procedures to no treatment.
For component analysis either single subject studies or group and laboratory studies may be conducted by the practitioner and the scientists respectively. Then the effectiveness of the intervention is compared with alternative interventions in short-term comparative outcome studies. Once passed those tests, the new intervention may be considered from an efficiency or long-term perspective. The final steps in the model involve evaluation of training methods or methods of dissemination and subsequent evaluation of the field efficacy of the new interventions.
The progressive model, also referred to as the integrated applied research model, while comprehensively and systematically going far beyond the current reality of most intervention research, is not unrealistic (Kent, 1985). "Each of the stages of research is attainable and indeed should eventually be accomplished for treatments that become clinical standards. Above all, the model shows the interactions among several sources of information and evaluation" (Ibid, 3).
Kent's version of the model intertwines research and practice closely in the short-and long-term perspectives as well as at several levels: the conceptual, the observational, the experimental, and the interpretive. His article promotes practitioner involvement in advancing the state of the art as a research consumer and a practical scientist. He provides questions to pose to research studies and describes varied approaches to single-subject designs. He ends his article with a description of the body of knowledge a practitioner needs to be an effective change agent. Barlow, Hayes, and Nelson (1986) also identify the research strategies that can be more readily conducted by practitioners in contrast to the more traditional modalities of research scientists.
Halgin (1986) identified ways training clinics could restructure their educational approaches to assure graudates recognize the value of scientist-practitioners. "Course work, clinical and research supervision, and conferences should teach students that being clinical practitioners, consumers of research, and research innovators are not separate and independent roles" (Goldried, 1984:479 cited in Ibid). Phillips (1989) acknowledged six major obstacles to the scientist-practitioner model along with the importance of collaboration among researchers and practitioners. He identified the knowledge, skills, and attitudes needed by researchers and practitioners to successfully collaborate.
Those include researchers recognizing that the world of practice is the scientists only true laboratory and that its boundaries do not respect those of their disciplines. For practitioners it includes recognizing the gambling status of practice and the need to maximize gains and minimize losses from applications of science to practice.
Persons (1991) identified the benefits of the scientist-practitioner model and ways researchers could better align their psychotherapy outcome studies to more accurately represent current practices of psychotherapy. She offered a case formulation approach to psychotherapy research that would take into account the individualization and idiographic practices in psychotherapy rather than relying on only standardized and atheoretical approaches.
E. Consumer-oriented models
While the marketing model is applicable to organizations, it is geared more to individuals in a given community or society. Often it is linked to tangible products. Fine (1981), however, suggested it was equally applicable to services, ideas, and social issues or causes for both the nonprofit and profit-making sectors of society. Consumers and their needs are the focal point of the marketing model and each of its components. Identifying and satisfying the consumer's needs is interwoven into every feature of the marketing mix. Marketing strategists such as McCarthy and Perreault (198 ) have identified the four p's--product, place, promotion, and price--as the key controllable variables of a marketing mix that is in turn tailored to specific customers [people]. McCarthy and Perreault define product as the "need-satisfying offering of a firm"(Ibid, 288). In the marketing mix, product takes on an expanded meaning to encompass the concept of the right product for the target market. It includes the product's features, accessories, installation, instructions, service, warranty, product lines, packaging, and branding.
Place involves getting the right product to the target market when and where it is wanted. Place includes the channels and locations for distribution of the product as well as the use of middlemen/women, and the levels of service. Promotion involves personal and mass selling, advertising, publicity, and other forms of promotion that stimulate interest, trial, or purchase/use by final customers or others in the channel. In the area of price, strategic decision areas relate to objectives, the level over the life cycle of the product, flexibility, geographic terms, discounts, and allowances. The main objective is to make the price right for the customer and for the company's survival and growth.
The four controllable variables--product, place, promotion, and place--lay within the circle of uncontrollable variables. The uncontrollable variables include the environment--cultural, social, political, legal, economic, and technological, competitive--and the resources and objectives of the company or organization. The marketing model encompasses the assessment of need, market strategy (target market and market mix), market plan, plan implementation, and evaluation. More and more the consumer has become an integral part in the selection of products, their design, their promotion, and marketing.
F. Other models and the prosumer approach
The prosumer approach, like the systems models, includes (1) interactivity, (2) dynamic flow of influences, (3) feedback, and (4) controlling mechanisms for debating and developing consensus about knowledge on a given subject. Boundaries associated with closed systems, however, are deemphasized in the open-system oriented prosumer approach. The social system of knowledge as envisioned by Holzner, Dunn, and Shahidullah are compatible with the prosumer approach as long as the domains of consumer knowledge are a prominent consideration along with the other domains. The prosumer approach, from the policymaker's viewpoint, makes research results available to all interested individuals. It uses public forums and media announcements regarding consensus panel results to help minimize misinformation that comes through the percolating process. The integrative organization-oriented model of Huberman in which he trys to bridge the gaps between the extremes of the research utilization conceptual continuum is laudable. However, it needs the benefits of the scientist-practitioner conceptualization, and the prosumer approach to maximize the bridging and truly become superordinate.
The prosumer approach supports the mindset and structural changes recommended under the scientist practitioner model. It also supports the consumer-driven marketing strategies now in vogue.
Chapter 6 has overviewed a variety of knowledge utilization models. Some of those models begin with the research product while others begin with the potential users. They range from simple to complex and from consumer-oriented to policymaker-oriented. Many reflect the two communities theory of knowledge utilization that views users and researchers as residing in two differing and alien cultures requiring linkers or integrators. They also reflect elements, such as "interaction," "problem-oriented," "planning," and "interdependency" of a prosumer approach. The practitioner-oriented models pointed to the importance of professionals contributing to knowledge production as well as use. The marketing model pointed to the importance of consumers in the entire process. The integrative organizational model linked the instrumentalist and transactional extremes of research utilization models. While Huberman viewed his model as potentially superordinate, its omission of the scientist-practioner conceptualization and the prosumer mindset leaves it less likely to reach superordination. The knowledge system model also broadened the scope of past models; however, it too falls short by not looking at new ways of relating in the process of knowledge production and use. The prosumer approach captures the best of all the models. It is flexible for changing times and changing conditions. Its uniqueness lies in the ways roles and relationships interact to maximize the production and use of knowledge and technology to improve the quality of life of people with disabilities.
The prosumer approach is interactive and interdepend-ent throughout creation, diffusion, and use of knowledge and not just dissemination and use. The prosumer approach suggests ways to help evidence from many sources converge into a common body of knowledge on a subject. The structures associated with the prosumer approach are relatively flat rather than hierarchically tall so that influencing is not just one-way but two-way and geometrically beyond.
The prosumer approach looks at the problem solving approach to knowledge utilization as valuable when repre-sentatives of all groups affected participate, when literature bases and expertise are equally accessible to all affected parties, and when consumers understand the knowledge cycle and how they can best participate in it. The prosumer approach does not restrict itself to a sequence of stages but rather multiple entry points and exit points. The planned change model does not serve the knowledge utiliza-tion field well when survival depends on grasping opportuni-ties and adapting rapidly to changing environmental condi-tions. Flexibility is not its strength. The prosumer model, on the other hand, is flexible to changing times and changing conditions. Because it is characterized by a way of relating with respect to knowledge production and use, it adapts readily to internal and external complexities.
A LOOK AHEAD:
Chapter 7 identifies problems with current models and identifies trends that affect the future state of the art in knowledge utilization.