p ' 34:41! .v 09 . ‘ v i (.1 4.0%.... ,A. 1 .14.! I. A. .. vb s. .30 .I 103.: .. I .. I III III... 10" a... A l A I} 9a.: I . 1).. 12.7.2} . , a 4, '1“... x l. . . t v. . ta ‘ Iv. . x; a ‘ VJl.€-.ti . . . ‘IJ: .- . k}. l 5.1 [HKIIHNI‘ A.-fu.9|’.c {cl-O! I I .C)4 I 11 ‘4 V y Y . D. .A.{ ‘ 15.3 .l 0.! Vill.av. . ‘ if. ,7 .oxo. ~ . .viot..o .. 2's; Luann?“ , . c . :i i . évnllo s :rnu. .7! § .11 :.:-. i .:... 2mm . 1...; I... gins}... ‘Irh‘wxsmu .. Iv... -. n lfixsrplavn 55...! . :g. ... 4-t-“ou...,..... . . -11.-.. (mix. $5.. . .5. fififikmfil..tzf z . 11‘ Z... a 134.43...st $5.... an ABAI‘I‘J. .1 .gs Wfltim . . 9.1 I. . .lfifiaamn Mata; . .a . 5%.... 4.. H . ‘ .- . - “\ Allah"..- It!|' 5‘ 131.1211 4... 11.170... 1 in .. ‘74; z ... a . ‘31....»nfiuwfimuw ag%§.t ,- . .z. . . Mats IBRANES ll‘lllllll lllllllllillll W“ 93 (\mez 918 MICHIGAN STATE FIVE \ 3‘1 l This is to certify that the dissertation entitled THE EFFECT OF NETWORK AND PRODUCT CHAMPION TREATMENTS ON THE SOLICITATION OF MENTAL HEALTH TRAINING PARTICIPANTS presented by BRENDA DIANA BRYANT has been accepted towards fulfillment of the requirements for Ph.D. Psychology degree in ‘Tféégiflif(f:;L6éFA:%E%;:1¢DO Esther 0 .Ma (gig/$6?“ 0 Date July 2, 1992 MSUIJ an Affirmative Action/Equal Opportunity Institution 012771 I; l LIBRARY IM'Chlaan State L University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES mum on or before date duo. DATE DUE DATE DUE DATE DUE mm 1 51994 :li—jtl. MSU Is An Affirmdivo AdlonlEqual Opportunity lnditution amounts-m THE EFFECT OP NETIORE AND PRODUCT CHANPION TREATNENTS ON THE SOLICITATION OP NENTAL HEALTH TRAINING PARTICIPANTS 3? Branda Diana Bryant A DISSERTATION Submittod to niohigan stat. Univ-raity in partial fulfill-ant of tho raquiraaonta for tho degraa of DOCTOR OP PHILOSOPHY Dopartaant 0: Psychology 1992 pr na to 01‘ ad Che Sev Q ABSTRACT THE EPTECT OP NET'ORE AND PRODUCT CHAMPION TREATNENTS ON’THE SOLICITATION OP NENTAL HEALTH TRAINING PARTICIPANTS 3? Brenda Diana Bryant Research-based rehabilitation models and techniques for facilitating community employment of people labelled mentally ill have been developing over the past thirty years. Despite their availability, employment outcomes for this population are poor. A salient reason is the failure of the mental health system to wholeheartedly adopt these practices. This study attempted to facilitate the adoption of employment practices in mental health organizations by experimentally manipulating network and product champion treatments designed to solicit mental health organizations to send innovative training participants to an adoption workshop and to create an organizational environment that would be conducive to adoption. Results indicated that the network and product champion treatments were not significant predictors regarding sending behaviour; instead, budget emerged as significant. A second set of analyses focused solely on personnel identified by administrators in the network treatment - staff who might be helpful in identifying candidates for training. The product champion condition had a suppressing effect for these "network people" with regards to conduciveness of the environment to adopt. A third non-experimental set of anal cond It t trea‘ orgal the g treat is re them. as cu intent these analyses focused on "training participants." The network condition had a suppressing effect for training participants. It was suggested that the network and product champion treatments be either dropped and efforts focused on organizations with medium or larger budgets or, be modified if the goal was sending training participants. Else, these two treatments are appealing as strategies for change because it is relatively easy for an outside change agent to implement them, Unfortunately, the results of the study indicated that as currently conceived, these strategies did not have the intended effects. Implications for further efforts to employ these types of change strategies are discussed. ii IECtb mam ACKNOWLEDGEMENTS I am a lucky person. Yes, I worked diligently. But.... I also had friends. Thank you to Jennifer Stanley, Nancy McCrohan, Zora Ziazi, Anja Kusler and Mary Hulford, who not only talked me down off the "ledges" when things were hectic, but helped me visualize the ending. I also had a supportive family. Thanks to my mom, my two grandmothers and my sister who also believed. (Note that they were also relentless at asking when I was going to be done. well mom, I am done!) I had a fabulous committee. Thanks to Dr. Dozier Thornton, Dr. Larry Hesse and Dr. Steve Raudenbush who were not only competent and creative - but user friendly, and believed in the dissertation process as a learning experience. (And believe me I learned.) Finally, I had the best mentor and dissertation chairperson anyone could wish for. Thank you Dr. Esther Fergus. Your generous time, wisdom, confidence and inspiration lit.my way. I now endeavor to pass on the torch. To endings and new beginnings. iii TABLE 0! CONTENTS STATEMENT OP THE PROBLEM PAGE SOCIAL INNOVATION AND ORGANIZATIONAL CHANGE: A REVIE' OP THE LITERATURE RATIONALE OP THE STUDY EXPERIMENTAL HYPOTHESES METHOD PARTICIPANTS DESIGN MEASURES PROCEDURES RESULTS DISCUSSION SUMMARY REFERENCES APPENDICES BACKGROUND CHECK LIST MANIPULATION CHECK: MEASURES EMPLOYMENT INNOVATIONS: PERCEPTIONS UNCERTAINTY OP EMPLOYMENT INNOVATIONS EMPLOYAEILITY OP PERSONS LADELLED MENTALLY ILL PROFILE OP PRODUCT CHAMPION INTEREST AND MOTIVATION MEASURE ORGANIZATIONAL SUPPORT OP EMPLOYMENT INNOVATION APPROACH PROTOCOL: ADMINISTRATORS APPROACH PROTOCOL: NETIORK NAMES POLLOI-UP PROTOCOL CONSENT PORM MENTAL HEALTH ORGANIZATIONS: STRATIEICATION INTERCORRELATIONS OE DEPENDENT MEASURES iv 1 32 34 36 39 43 GS 69 SO 84 90 92 102 105 108 112 121 124 127 136 145 154 156 159 (Stroui the em past tr. Black, ' that pro ClubhOg transitio Stroul, 1! adequax STATEMENT OF THE PROBLEM As part of a directive from the National Institute of Mental Health (NIMH) (Stroul, 1986), research-based rehabilitation models and techniques that facilitate the employment of persons with mental disabilities have been developing over the past thirty years (Avison & Speechley, 1987; Black, 1986; Stein & Test, 1978; Black, 1970). A central focus of this effort has been on techniques and models that promote employment in the community. Such examples include the Lodge, Clubhouse and Assertive Community Training models, and supported and transitional employment techniques (Cnanna, Blankertz, Messinger, Gardner, 1989; Stroul, 1986). Experimental and quasi-experimental research has shown that given adequate support and training, the psychiatrically disabled are capable of working in competitive employment situations (Bond & Boyer, 1988; Black, 1986; Malamud, 1986; Fainlveather, 1980; Stein & Test, 1979). Despite the availability of tested employment innovations, employment rates for the population of persons with psychological disabilities are discouraging. Full- time competitive employment rates for discharged persons range from 10-30% (Anthony, Sharatt & Althoff, 1972; Anthony, Cohen & Vitalo, 1978), regardless of the follow-up period‘. For the severely psychiatrically disabled population, full time and part-time competitive employment rates drop below 16%, and these rates have changed little over the years (Anthony & Blanch, 1987). To further complicate the issue, even when competitive employment is secured, job tenure A word of caution. These base rate figures were acquired through investigation of several studies which were not necessarily comparable, i.e., heterogenous samples, different follow-up periods. Dec: 1975 rehal psyci 2 becomes a problem (Collignon, Noble Jr. & Toms-Barker, 1987; Stein & Test, 1978). Anthony & Blanch (1987) concluded that "the mental health and rehabilitation systems seem to have done a better job of teaching persons with psychiatric disabilities to be clients than teaching them to be workers." The literature (ie., Anthony & Blanch, 1987; Bond, 1987; Noble & Collignon, 1987; Black, 1986; Stein & Test, 1978; Fairweather, Sanders & Tomatzky, 1974) suggests that employment problems for the psychiatrically disabled occur primarily for two reasons: 1) mental health organizations (MHO) emphasize the medical model, viewing people as sick and dependentz; and 2) like institutions in general, MHO's are invested in maintaining control of their consumers. Typically, these two reasons are reflected in mental health policies and practices. Dependency attitudes are revealed in mental health workers’ low expectations and negative attitudes concerning the employability of their clients". Work and therapy are separated. Collaboration between vocational and MHO’s is minimal (Anthony & Blanch, 1987; Noble & Collignon, 1987; Backer, Liberman 8: Kuehnel, 1986; Black, 1986). There is little support for vocational skills Unlike, for example, vocational rehabilitation organizations which emphasize the rehabilitation model: viewing all persons as resources, perhaps impaired, but capable of overcoming this impairment to achieve vocational goals. For example, Noble & Collignon (1987) relayed an experience where a mental health rehabilitation day program closed down because of lack of funds. Although several found and maintained employment in the community, when the doors of the program reopened, the vast majority left their jobs to go back to the program. The staff's response was simply that the clients could not handle the pressure of competitive work. Bl Ol Stc Fer exa (Fai only and is n dev Ave For 1 San 70? ’c A: 3 training, leaving mental health staff, who are required to develop and implement community vocational plans, with little or no technical training (MSU Regional Training Survey, 1989; Schalock, 1983; Bilvosky & Matson, 1977). Moreover, the mental health system refuses to wholeheartedly adopt tested community employment rehabilitation models and techniques which increase competitive work outcomes for the mentally disabled population - even despite concerns for new and effective methods by mental health consumers, their families and professionals in the field (Ml Alliance for the Mentally Ill Survey, 1989; Tashjian, Hayward, Stoddard & Kraus, 1989; Bond, 1987; Becker et al, 1986; Stroul, 1986; Tomatzky, Fergus, Avellar, Fairweather & Fleischer, 1980; Fainrveather et al, 1974). For example, results of a national diffusion experiment involving the Lodge model (Fairweather et al, 1974) indicated that of the 255 hospitals targeted nationwide, only 25 adopted the innovation - notwithstanding glowing results from experimental and quasi-experimental evaluation data over the last twenty years‘. But the mental health system’s resistance to these employment innovations is not new. Appareme it is common for practice to lag behind research and development (Rogers, 1983; Glaser, Abelson & Garrison, 1983; Tomatzky, Fergus, Avellar, Fairweather, Fleischer, 1980; Fairweather et al, 1974; La Piere, 1965). In ‘ For example, in a longitudinal experiment involving the Lodge model, Fairweather, Sanders, Maynard & Cressler (1969) found that after a 40 month follow-up period, Lodge members remained in the community significantly longer (80-10096 vs 20- 30% time), spent significantly more time in full-time competitive employment (40- 70% vs 0%), and incurred lower costs that those ex-patients who utilized other community treatment options. 4 a world faced with many crises and challenges to survival, practices linked to scientific knowledge represent a logical solution to the process of problem-solving. Yet, one of the most disconcerting phenomena in the study of adoption of social innovations is the wide gap that exists between the time at which an innovation is first developed and when it is finally adopted. Nonetheless, if we assume that employment has positive benefits for persons with mental disabilities‘, and we can identify employment innovations that do work, how do we close the gap? How can existing MHO’s be persuaded to incorporate new employment programs and techniques into their current practices? In an attempt to better introduce employment innovations into MHO’s, the Michigan Department of Mental Health provided funding to MSU for the development of a long-term vocational training program designed to increase knowledge and skills in both the understanding and adoption of employment innovations. In light of the documented resistance to expanded boundaries and roles, including programs that incorporate an employment context, it must be assumed that the doors will not be flooded with training participants eager to champion the adoption of these programs. Thus, there is a need to lnvestlgate efiacflvewaystoapproadtandpersuadeMHO'snotonlytosendpeopletomese talulngseesbm,bcflalsotosendpeoplewhowlllbelmerestedandmoflvamd For example, Fairweather (1980) would argue that work is beneficial because it is culturally rewarded not necessarily because it has inherent value; similar arguments about the inherent value of work could be made for leisure activities, for instance. attemp nanmé organiz Finally, recomrr 5 enoughtopushfortheadopflonandhnplementaflonoftheaeemploymem Innovations within their parent organizatlon. This dissertation reports the results of an experimental study which attempted to address the need to induce MHO's to send people to these MSU training sessions. It begins with a review of the literature pertaining to organizational change and innovation, which is followed by a rationale of the study. Finally, a description of the experiment, method, results, discussion and recommendations is presented. SOCIAL INNOVA110N AND ORGANIZATIONAL CHANGE: A REVIEW OF THE LITERATURE Although research concerning the adoption and diffusion of innovations has been conducted since the forties, beginning within disciplines that tried to diffuse their own technologies, interdisciplinary discussions of, and collaborations involving this issue did not begin until the sixties (Rogers, 1983). Relatively recent findings from experimental and retrospective research across many disciplines and varieties of social innovation and technologies (ie., Tomatzky & Fleischer 1990; Backer, Liberman & Kuehnel, 1986; Glaser, Abelson, & Garrison, 1983; Rogers, 1983; Tomatzky et al, 1980; Fairweather et al, 1974; La Pierre, 1965) suggest that there are predictable and consistent barriers to innovation adoption. PERSPECTNES OF ORGANIZATIONAL CHANGE Various models of organizational social change have been developed in an attempt to understand the process of social innovation (Rogers, 1983; Fairweather & Tor over 4 perspi theory, model. organiz mOSt efi grOUps became What wa Structure commUnr re'afiODSl‘ This perm agent, her Norman-0' the primap The Studies of: some, DSYC hYbrid COrr 6 & Tomatzky, 1976; Havelock, 1976;69; Rogers & Shoemaker, 1971). Based on over 4,000 studies of the social change literature, Havelock (1969) outlined three perspectives of dissemination and adoption. The Problem-Solver perspective was initially developed from organizational theory, particularly the writings of Max Weber (1958) concerning the bureaucratic model. Traditionally, the bureaucracy was seen as a rationalu designed organization: structures were geared towards accomplishing stated goals in the most efficient manner possible. However, influenced by research of informal work groups (ie., the classic Hawthome studies, Roethilisberger & Dickson, 1964), it became apparent that m factors, such as peer pressures, often dictated what was accomplished. Thus, the focus of change moved from organizational structures and tasks to organizational processes, and in particular, flows of communication. Ideally, this updated perspective assumes that if organizational relationships are positive, the organization will become more responsive to change. This perspective also introduced the role of the professional helper. This external agent, helped the client/organization identify and solve problems by providing valid information so that informed choices could be made - change was not necessarily the primary task of the agent. The Social lnteracfion perspective has its roots in the anthropological studies of the diffusion of cultural traits (Barnett, 1953), and later, was shaped by social psychology and sociology (ie., see the classic study of the diffusion of hybrid corn by Ryan & Gross, 1943). This perspective is concerned with the an the arg inte as; 7 diffusion (spread) of the innovation, and the match/mismatch between the innovation and old norms, values and roles. In particular, it addresses the movement of messages within the informal communication channels of the user or adopter, with emphasis on the user’s reference group. In a further description of this perspective, Rogers & Shoemaker (1971) delineated the diffusion of a innovation through stages (knowledge, persuasion, decision and confirmation) where certain kinds of communication (formal at the beginning, informal peer communication later) were of import. They also depicted differences between early and late adopters of social innovations. Similar to the problem-solver perspective, the social-interaction perspective underscores peer-to-peer communication and argues for a client orientation, but, prescribes a role for a change agent that is intent on creating change in a certain direction and/or promoting the adoption of a specific innovation. The Classical Research, Development, and Diffusion perspective (RD&D), the most widely accepted of the perspectives, particularly in industrial settings (T omatzky & Fleisher, 1990), dates back to early agricultural interventions, where change followed an orderly process, with separate, independent stages. Unlike the Problem-Solver and Social Interaction perspectives, this view of innovation diffusion emphasizes and begins with rigorous research concerning the need or problem. This beginning is followed by the development of the product, packaging and diffusion. The classical model assumed a rational approach to change: tried and true innovations would naturally be accepted by passive users 1 9 Mc anc then deci: (Fain goab modifi Organi PfaCtiti. (Blaine) (1976) l model a deVEiOp The Usel SolufiOn Spanner. Mains it, 8 and organizations. Recently, however, researchers have questioned these assumptions (Blakely, Mayer, Gottschalk, Schmitt, Davidson, Roitman & Emshoff, 1987; Rogers, 1983; Blakely et al, 1984; Tomatzky et al, 1980; Barman & McLaughlin, 1978; Fairweather at al, 1974). Instead, it appears that consumers and organizations are quite active, often re-inventing innovations and adopting them to local sites (Blakely et al, 1987; Rogers, 1983), and/or making adoption decisions based on criteria that were often satisflcing rather than maximizing (Fairweather et al, 1974; Cyert & March, 1963). Likewise, frequently conflicting goals did not guarantee adoption throughout the organization. In response, a modified RD&D perspective has been developing which pays more attention to organizational dynamics, encourages research and development by the practitioner, and the use of full-time change agents to disseminate the innovation (Blakley et al, 1984). Based on the strengths of the perspectives he originally outlined, Havelock (1976) later developed a fourth perspective of social innovation. The Linkage model attempts to bridge the gap between the resource system (research and development) and the user, arguing for an interdependent, reciprocal relationship. The user is viewed as a problem-solver, who, because of a need, searches for a solution. The change agent’s role is characterized as a linkage or "boundary spanner" between the resource and the user subsystem. The linkage model retains the development and evaluation of innovations, but adds an interpersonal GUT (Hg (ESl (197i persr resea than . inhere eXDGrir approa tested i frElmer/i“ each pi diffusion Strategic the e’(Der Tor models' - c:ommelic as exDedie 038i} active 9 emphasis, with special attention paid to the adopter’s needs and local organizational processes. Finally, a fifth perspective, the Experimental Social Innovation and Diffusion (ESID) model, is proposed by Fairweather (1972) and Fairweather & Tomatzky (1976). Recognizing the importance of the variables outlined in the four perspectives summarized above, Fairweather noted that there was little evaluative research that determined what variables were more important in certain situations than others. In addition, Fairweather argued that these perspectives were inherently supportive of the status quo. Accordingly, Fairweather adopted a field experimental strategy to social change, but one that incorporated an action approach. This perspective emphasizes humanitarian values and experimentally tested innovations. The model assimilates important diffusion variables into the framework of the experimental model. Subsequently, it employs experiments at each phase of the innovation process (approach, persuasion, activation, and diffusion), and utilizes this feedback to change or enhance social change strategies urIIII the masses adopt. The change agent is seen as an advocate of the experimentally tested social innovation, one that is committed to adoption. Tomatzky & Fleischer (1990) contend that often these social innovation models, i.e., in particular the RD&D model, take on a centre-periphery communication bias. That is, they try to identify a central party and get information as expediently as possible to this centre, rather, than recognizing that users are often active participants in the change process and thus a reciprocal approach to innov orgar. declsi. Mintzl: much proces . . l sensmv 0f “chol ROgers, Should Charade dePloyer 3" Within FACTORE Of soda} ir 10 innovation adoption is required. Moreover, these authors suggest that many organizational change models assume a linear, step wise approach to adoption decisions, which is often not the case. Tornatzsky & Fleischer (1990) support Mintzberg, Raisinghami & Theoret’s (1976) perspective that adoption decisions, much like formal decision-making (House & Singh, 1987), entail a non-linear process involving movement back and forth, rich in feedback loops, and highly sensitive to new information. Such a viewpoint argues for careful understanding of "choices within contexts." Similar to others in the field (i.e., see Blakely, 1984; Rogers, 1983) Tomatzky & Fleischer (1990) suggest that "deployment" techniques should be chosen in accordance with the nature of the technology, the characteristics of the users and the deployers, the boundaries within and between deployers and users, and the characteristics of the communication mechanisms, all within a network approach. FACTORS ASSOCIATED WITH ORGANIZATTONAL CHANGE AND INNOVATION 9:ng m and Processes. Of import in the facilitation of social innovation is the context into which the innovation is introduced. Such a context involves a set of related goals, rules, assumptions, and expectations about behaviour and outcomes (T ornatzky & Fleischer, 1990; Downs & Mohr, 1979;76). Research in the field suggests that organizational processes and structures may override any effects of strategies taken to encourage adoption (Tomatzky & Fleischer, 1990; Tomatzky et al, 1980; Fairweather et al, 1974). ration comrr organi needs 1980; J structu channe reduce themse CIUO (Hc Zaitma n 11 Ideally, organizational missions outline goals within a context that maximizes rationality; organizations make use of expertise and specialization within a formal communication network, guided by rules and rewards. In reality, however, organizations have many goals, which may become blurred, generating multiple needs and priorities, and power struggles (House & Singh, 1987; Tomatzky et al, 1980; Zaltman et al, 1973; Burns.& Stalker, 1961). Consequently, organizational structures such as rules, rewards, formal communication and decision-making channels, originally designed to help facilitate operations and productivity, and to reduce uncertainty regarding tasks, often become paramount to the goals themselves, making the organization rigid, self-serving, and rewarding of the status quo (House & Singh, 1987; Fairweather & Tomatzky, 1976; Fairweather et al, 1974; Zaltman, Duncan, Holbek, 1973; La Pierre, 1965; Burns & Stalker, 1961). This emphasis on procedures and the maintenance of the status quo (ie., power structures) creates an environment which is impervious to social change, inhibiting flexibility, creativity and fluidity, necessary components to the process of the social innovation (T ornatzsky et al, 1980; Fleischer, 1979; Yin, 1977; Fairweather et al, 1974; Pincus, 1974). For example, correlational research (e.g., Rogers, 1983; Havelock, 1973; Haige & Aiken, 1970; Burns & Stalker, 1961) suggests that such aspects of organizations are negatively related to social change. Studies of organic organizations--i.e., those that are characterized by lateral communication, decentralized leadership and control, high level of networking-reveal that these types of organizations are more likely to adopt 12 innovations in comparison to mechanistic organizations where rules and procedures are emphasized (Aiken & Hage, 1970; Burns & Stalker, 1961 ). Theory and experimental research maintain that this difference in adoptability occurs because such rigid structures inhibit broad-based pm'llclpatlve, Interactive styles of decision-making, the very processes that are crucial for facilitating adoption, as people try to c0pe with change and/or tasks that are nonuniform (Rogers, 1983; Tomatzky et al, 1980; Fairweather et al, 1974; Havelock, 1973; Haige & Aiken 1970; Burns & Stalker, 1961; Schacter, 1959; Festinger, 1954). Public bureaucracies such as MHO’s tend to have formal, rigid, centralized structures (Backer, Liberman 8. Kuehnel, 1986; Fairweather et al, 1974). Barriers to social change within MHO’s include such factors as professional groupings (i.e., psychiatrists, psychologists) where power positions are guarded, and top heavy hierarchical structures yield rigid communication. Hence, inherently, MHO’s reduce opportunities for broad-based participative decision-making and ultimately, adoption (Backer et al, 1986; Tomatzky et al, 1980; Fairweather et al, 1974). Yet the phenomenon of social innovation within an organization is complex. Correlational research (i.e., see Tomatzky & Fleischer, 1990; Rogers, 1983; Zaltman, Duncan, Holbek, 1973; Wilson, 1966) suggests that the very aspects that are negatively associated with innovation adoption are actually positively related to the actual implementation of the innovation. That is, formalization and centralization, as organizational characteristics, impede the decision to adopt, but facilitate implementation. Experimental research (i.e., Fainrveather et al, 1974) sug inno oner adOp been aewl Showr isnot 13 suggests that this complex pattern exists because the implementation phase of innovation requires a more task-oriented approach in comparison to a process oriented one needed earlier on. Other organizational characteristics that have received attention in the adoption literature are slack and size. The unavailability of slack resources has been suggested as a reason for the lack of implementation of complex innovations (i.e., March & Simon, 1958). However, experimental and survey research has shown that while slack resources may be a necessary condition for innovation, it is not a sufficient one (T ornatzky & Fleischer, 1990; Tomatzky et al, 1983; Roger, 1983; Fairweather et al, 1974; Cyert & March, 1963). Size of the organization has been shown to be one of the most powerful predictors of organizational adoption (Kelly & Brook, 1988; Aiken & Hage, 1970; Cyert & March, 1963). However, critics propose that it may actually be an indicator of other organizational structures, i.e., formalization, hierarchy, resources and so on (T ornatzky & Fleischer, 1990) It has been suggested (e.g., Tomatzky & Fleischer, 1990; Tushman & Nadler, 1986; Fairweather et al, 1974; Gailbraith, 1973; La Pierre, 1965) that organizational processes, like organization structures, are also important within the context of adoption behaviour“. Do internal organizational processes nurture an environment that is open to change? Does the organization foster informal linkages and communication within and beyond its boundaries so that new Note that such processes are indices of behaviour rather than attitudes, which have shown to be poor indicators of implementation behaviour (Rogers, 1983; Tomatzky et al, 1980; Fairweather et al, 1974). infon orga span innox mane proce to a c denot 80 th. Shoer 14 information, practices, innovations can be brought in and/or shared? Does the organization reward innovative behaviour or have a role defined for "boundary spanners," "product champions"? Does top management support social innovation in general and/or the identified innovation? For example, top management support has been recognized as a salient variable in the adoption process: although broad-based participative decision-making has been shown to a crucial factor in promoting an environment that is open to adoption, research denotes that sanction by top management is needed for actual implementation, so that necessary resources are released (Fainlveather et al, 1974; Rogers & Shoemaker, 1971). mm, Research on the utilization of the RD&D model and studies of innovation fidelity (i.e., Blakely et al, 1984) have been instrumental in pointing out that adopters are not passive consumers of innovations. Extensive literature reviews (i.e., Rogers, 1983; Glaser et al, 1983; Tomatzky & Klein, 1981) yield five basic attributes of innovations on which users focus when making a decision to adopt. These include: relative advantage, compatibility, complexity, trialibility, and observability. The relative advantage of the innovation (the best predictor of adoption) is defined by issues of costs, convenience, prestige, comparison with present techniques. Compatibility of the innovation addresses the issue of its congruence with existing values, norms, past experiences and 15 needs within the organization’. Complexity is the degree to which an innovation is perceived as difficult to understand and use; trialibility concerns the degree to which an innovation can be experimented with on a trial basis; observability is associated with the visibility of the innovation’s results. MEI—MEN.- Social innovation is inherently an uncertainty- arousing phenomenon. There may be tentative concerns about the effectiveness of the innovation and/or its unforseen consequences. Adoption may require role and procedural changes, the learning of new skills. It has been proposed that people are motivated to reduce uncertainty (Louis, 1980; Schacter, 1959; Festinger, 1954). Thus, a large degree of uncertainty may result in organizational resistance. Conversely, a high degree of certainty may lead to adoption (Tomatzky & Fleischer, 1990; Backer et al, 1986; Fernandez-Sandin, 1986; Rogers, 1983; Tomatzky et al, 1980; Becker, 1970). Schacter (1959) argues that increased interpersonal contact and interaction to facilitate coping with potentially negative future events are necessary to the extent that novel stimuli enhances people’s need for uncertainty reduction. Broad-based participation in decision-making has been cited as an effective tool for reducing uncertainty and facilitating an atmosphere conducive to adoption (i.e., 7 To those who doubt the importance of this factor consider this example. A public health campaign was launched by the Peruvian government to introduce innovations that would improve citizens’ health. A peasant village of 200 families, who had little understanding of the relationship between sanitation and health, was approached to teach them to boil water before using it. Only eleven housewives were persuaded to do this. In their culture, boiled water was linked to sickness: only the ill used hot water (Wellin, 1955 as cited in Rogers, 1983). Tom. phas level 16 Tomatzky et al, 1980; Fairweather et al, 1974). For example, in the approach phase of a longitudinal adoption experiment that manipulated (among other things) level of staff and group enhancement (directed to talk with others about the innovation or not), Tomatzky et al (1980) found a significant interaction between level of staff and group enhancement that affected perceptions of the Lodge innovation: groups composed of line staff only, and/or line staff and administrators in the high group enhancement condition perceived the Lodge innovation as significantly less different in comparison to the administrator-only groups and all those similar groups in the low groupoenhancement condition. Furthermore, results revealed that line-staff involvement in the approach phase significantly increased the probability of accepting a persuasion workshop. The process of social innovation, particularly regarding complex innovations, is often difficult in organizations that foster predictability and rigid structures. Such an environment inhibits broad-based involvement in the decision-making process- a means by which uncertainty can be dealt with. Thus, attempts at reducing this uncertainty surrounding change should be incorporated into any social innovation strategy. Rogers (1983) posits that people primarily seek information about two concerns: (1) the cause-effect relationship of the innovation in achieving desired outcomes-i.e., what is it?, does it work?; and, (2) the consequences of the innovation--i.e., why does it work, what will it accomplish in my situation? Mare cf Mmgnlcetlon Ohmic Because organizations require information about the innovation to make informed decisions regarding adoption, it 1 Re Fai Ber mec to la are r decis GXDer servic. safisfie among exDarin innovau approac Works;S Up a de adding”. reCerVed implemer Re 1956; Ry 17 it seems logical to understand the best way to communicate this information. Research (e.g., Fernandez-Sandin, 1986; Rogers, 1983; Beyer & Trice, 1982; Fairweather et al, 1974; Chappanis, 1971; Coleman, Katz & Mentzel, 1966; Berrelson & Freedman, 1964; Ryan & Gross, 1943) indicates that although mass media are effective methods for creating knowledge awareness of the innovation to large numbers, intensive efforts to communicate at a more interpersonal level are more effective in conveying complex information, and in affecting the actual decision to adopt and implement. For example, Chappanis (1971), in an experiment designed to provide researchers with different kinds of information services (phone, literature search, written), found that users were much more satisfied with interpersonal service for complex problems, but no differences among services emerged when the problem was simple. In an longitudinal experimental study designed to persuade state hospitals to adopt a complex social innovation program, Fairweather et al (1974) found that although less active approaches (brochures and workshops) were effective in "getting a foot in the door," significantly more hospitals that received an more intense approach (setting up a demonstration ward) actually agreed to adopt the Lodge program. In addition, of those hospitals that agreed to adopt, significantly more hospitals who received a personal consultant, in comparison to a manual only, actually implemented the program. Retrospective diffusion studies (i.e., Rogers, 1983; Coleman, Katz & Menzel, 1966; Ryan & Gross, 1943) indicate that evaluations of innovations are very SUSCE opri a diff. 0966 inforrr domo adOpt excha dmusi Sockfi Uncen in the 1959; indnnd (Lazar Deer c 0°”WCH meanir i Milena: rm. CU” 18 susceptible to subjective peer evaluations. That is, peer to peer discussions and opinions often dictate decisions concerning innovation adoption. For example, in a diffusion study regarding adoption of a new drug "gammanym," Coleman et al (1966) found that although doctors were approached by a drug company and informed of the tested drug, subjective evaluations of the use of the drug by doctors in their own personal network was the most significant reason for drug adoption. Similarly, Ryan & Gross (1943) concluded that farmer-to-farmer exchange, regarding usage of a new seed, was the determining factor in the diffusion of the innovation. Why this reliance on peer networks? Recall that theory (i.e., Festinger’s Social Comparison Theory, 1954) purports that people are motivated to reduce uncertainty. Accordingly, people talk to each other to gain clarity and feedback in their environment, particularly in uncertain situations (Louis, 1980; Schacter, 1959; Festinger, 1954). In addition, homophin--the degree to which pairs of individuals who interact are similar in certain characteristics, beliefs, attributes (Lazarsfeld & Merton, 1964)-- has been cited as another reason for this peer to peer communication flow. Communication is said to be more effective and comfortable among similar individuals because they share common languages and meanings (Rogers, 1983; Lazarsfeld & Merton, 1964). Of import to understanding this peer-to-peer evaluation is the impact by influential near peers in the network. Reviews of studies depicting diffusion curves (the cumulative number of adopters plotted over time) indicate that adoption takes bc "bl cor per asl ofn inth afler ado; innov inierc adopt Exam; the (in 19 off when opinion leaders adopt (i.e., see Rogers, 1983). This trust of opinion leaders is said to be highest when networks are heterophilious (non-similar), and boundaries must be crossed (Rogers, 1983). Apparently, opinion leaders or "boundary spanners," play a role in reducing uncertainty by bridging links, conveying and/or interpreting new information. Opinion leaders are often perceived as having greater competence-«but not too much, else they are seen as too different from their followers. For example, Lee (1977) found that modeling of innovative behaviour occurred in Korean women when the model adopter was in the same network and/or one intermediary step beyond the network; otherwise attempts to induce modeling outside the network yielded no effect. In essence, the diffusion effect is the increasing influence on individuals to adopt or not adopt resulting from the "activation" of peer networks concerning the innovation (Rogers, 1983). Retrospective studies of diffusion purport that the interconnectedness of the individual to the social System is an indicator of adoption potential (Rogers & lfincaid, 1981; Lee, 1977; Coleman et al, 1966) For example, Coleman et al (1966) found that isolated doctors were the last to adopt the drug gammanyn. WM Diffusion research has shown that all things being equal, not every target adopts at the same time (Rogers, 1983). Based on diffusion curves, researchers have developed adopter categories according to "innovativeness," or the degree to which an individual adopts new ideas relatively earlier than other members of a system (Rogers & Shoemaker, 1971; Rogers, 1961 majt idea: inforr onen innov by im t0 ab: compl uncen cDDIRIOr adODte DUI. Tc 20 1962). The five adopter categories include: innovators, early adopters, the early majority, the later majority and laggards. Typically, Innovators are portrayed as venturesome and eager to try new ideas (Rogers, 1983). They display cosmopolitan behaviour; that is, they seek out information, and have social relationships and communication patterns that are oriented outside of their social system/community (Merton, 1957)”. Although innovators may be marginal members within the system, they launch new ideas by importing them from the outside system. Innovators usually have the money to absorb losses on risky endeavours, and the ability to understand and apply complex technology, as well as the ability to cope with a high degree of uncertainty. f Early adopters are depicted as respectable, credible, with a high degree of opinion leadership (Rogers, 1983). Other potential adopters look to the early adopter for evaluations of the innovation; change agents often seek these people out. To maintain this credibility, it is posited that the early adopters are judicious when considering adoption: they are bound by the conservativeness of their organization or community (Becker, 1970). Their role is to reduce uncertainty by conveying subjective evaluation of the innovation to near peers viz a viz interpersonal networks. For example, studies of earlier adopters (Rogers & Sovenning, 1969; Coleman et al, 1966; Ryan & Gross, 1943) found that earlier adopters travelled to out of town meetings or to urban centres significantly more often than later adopters. BVBEB interc not at DBCGE WES C agent 21 The early mdoflty are portrayed as deliberators, adopting just before the average number of people do (Rogers, 1983). Seldom leaders, they provide interconnections in the system. The late mdorlty are characterized as skeptical (Rogers, 1983). They do not adopt until after the average members do, generally because of economic necessity and/or peer pressure. Laggards, the last to adopt, are depicted as traditional, and look to what was done in the past (Rogers, 1983). They are generally suspicious of change agents and innovations. In a study of personality variables and innovativeness, Loy (1969) successfully distinguished among four of the adopter categories-the exception was innovators. Personality characteristics associated with the early adopters included the following: adventuresome, imaginative, dominant, unsociable, self- sufficient. Based on 900 empirical studies available since the late sixties, Rogers (1983) delineated generalizations of characteristics, attitudes and behaviours related to earlier adopters in comparison to later adopters”. Such generalizations include: empathy, or the ability to project self into a role of another such as innovators, change agents, R & D workers who are outside of his/her system; more favourable attitudes towards change; greater ability to cope with uncertainty and risk; greater influence or degree of opinion leadership (although they may not ° Rogers notes that although these generalizations have come from studies ending in the late sixties, everything he has read up until the time of his book parallels these findings. J 22 have formal status in the organization); greater communication behaviour i.e., greater social participation, more highly interconnected in the system; and greater cosmopolitan behaviour. WM Change agents or product champions are individuals who influence clients concerning innovation decisions (Rogers, 1983). They may be outside individuals solicited by the organization or outside individuals "championing" an innovation not solicited by the organization. They may also be individuals inside the organization, who introduce and advocate for innovation in general or for a specific one (T ornatzky et al, 1990). Such "insiders" have been referred to as boundary spanners, bureaucratic entrepreneurs, adoption agents, product champions (Hill, 1982). Research has demonstrated the efficacy of change agents and product champions in the process of social innovation (i.e., Gerwin, 1988; Parkinson & Avlonitis, 1986; Rogers, 1983; Tomatzky et al.1980; Chakrabarti, 1974; Fairweather et al, 1974). For example, in a dissemination experiment, Fairweather et al (1974) found that the actual implementation of the Lodge program dependended upon the emergence of a "product champion" that could organize a cohesive problem-solving group and emphasize tasks, while keeping morale high in the face of adversity. Apparently, these insiders advocate for an innovation and/or keep change on the organizational agenda, provide communication links between internal boundaries related to aspects of the adoption process, and take responsibility for coordinating its actual implementation. per beh 198. StUC proc dorr statL Simil Signj relate 23 Although there is little research that examines these “product champions" per se, there is some evidence that suggests that they mirror characteristics and behaviours of the early adopter (T ornatzky & Fleischer, 1990; Fernandez-Sandin, 1986; Hill, 1982; Chakrabarti, 1974) For example, Hill (1982), in an experimental study that varied dissemination techniques and examined characteristics of internal product champions, concluded that the personal characteristics of adventureness, dominance, communication potential, formal decision-making power, and social status had value in predicting collective adoption of a university program”. Similarly, Fernandez-Sandin (1986) found that being "venturesome" was significantly correlated with implementation behaviour. Success of the change agent/product champion has been shown to be related to such aspects as assuming an active role (Rogers, 1983; Fairweather et al, 1974; Fleigel et al, 1967) and a client orientation or priority to clients’ needs (Rogers, 1983). Also, homophily to the client or organization has been shown to be related to successful adoption. For example, Placek (1975) concluded that the 1% of homophily was a determining factor in the failure of a family planning intervention where welfare workers were to work with mothers. Although most of these women did not want more children, little adoption of birth control techniques took place. Typically, welfare workers were white, middle class, college educated ‘° This finding re formal decision-making power may be more a characteristic of the university system. The literature suggests that product champions may not necessarily have formal power (i.e., see Tomatzky & Fleischer, 1990; Rogers, 1983; Hill, 1982). pers high cllen (Torr 1969, in Inc OCCUF 0i soc Fleiscr avaiiab Change Organjz serviceE and so bUSlnes was Sta probes and the. & Fleisc acCordjp °filsb 24 persons working with mothers, of whom 80% were African Americans with little high school education. Credibility of the change agent/product champion in the client’s eyes has also been shown to be of import to adoption behaviour (T omatzky & Fleischer, 1990; Rogers, 1983; Lionberger & Chang, 1970; Repetto, 1969). For example, Repetto (1969) found that credibility of change agent aides in India was a crucial component in the decision to have a vasectomy; credibility occurred when the aides showed their operation scars to their potential adopters. WM To some extent, organizational adoption of social innovation is depends upon its external environment (T ornatzky & Fleischer, 1990). Among others, environmental considerations might include the availability of a labour market to match innovation technology; the expertise of change agents; the growth demands of clients/markets; the extent of alternative organizations and/or competition; access to resources and technically related services; government regulations; zeitgeist of the times; cultural norms and values; and so on. For example, Benvignatti (1982) and Mansfield (1968;1977) found that business firms were most likely to invest in innovations when the business cycle was stable, rather than when it was recessive or at the height of activity. Mansfield proposed that firms may perceive these high and low points as uncertain periods, and therefore are inclinded to wait for more stable, predictable times. Tomatzky & Fleischer (1990) also point out that the importance of such factors may vary according to the degree to which the organization is responsible to others outside of its boundaries. In the case of MHO’s, they are accountable not only to gm LEE org; 25 governmental bodies for monies and resources, but also the general public, (Backer et al, 1986) and in theory, consumers. LESSONS FROM THE LITERATURE. The following is a summary of the literature concerning social innovation and organizational change. 1. Organizational innovation appears to depend on a number of factors: the context in which it occurs; the organization’s environmental context; attributes of the innovation itself; the uncertainty aroused by the innovation; the nature of communication; and characteristics of potential adopters (or units) and product champions/change agents. 2. Organizational change does not seem to emanate from unilateral decisions. Diffusion research suggests that adoption decisions are activated by peer networks. Thus, there is a need for a network approach that incorporates broad-based participative decision- making. 3. Organizations that are rigid tend to impede the facilitation of the decision to adopt by limiting broad-based participative decision- making. 4. A certain amount of support from top administration is necessary, particularly to sanction behaviours and commit resources associated with the actual implementation of the innovation. 5. It seems that little spontaneous adoption occurs in MHO’s. They tend to be rigid organizations that need external stimulation to induce adoption. 6. The more complex the innovation, the greater the need to communicate information at an interpersonal level. 7. Innovations are inherently uncertainty-arousing. This uncertainty may impede adoption. Interpersonal strategies designed to increase broad-based participative decision-making concerning the innovation have been shown to be effective in reducing uncertainty. 26 8. Intense, interpersonal strategies that incorporate broad-based participative decision-making not only facilitate the decision to adopt, but also the actual implementation. 9. There seems to be a gap between positive attitudes towards the innovation and actual change. Innovation adoption and implementation have been shown to be more certain when a “product champion" is identified or emerges from within the organization to support the adoption process. 10. It is posited that product champions can be identified by their risk-taking characteristics, communication potential, influence, network expansion and cosmopolitan behaviour. MTIONALE OF THE STUDY As discussed earlier, this study emanated from the need to procure mental health organizations to adopt and implement tested employment innovations. In consideration of the bureaucratic nature of MHO’s «that is their rigid nature; past dissemination efforts-~that is their documented resistance to employment innovations; the environmental context of budget cuts, lays offs and closures --a situation of uncertainty; and the complex nature of employment innovations themselves --also a situation predisposed to uncertainty, the problem suggested an action oriented perspective, one that nudged MHO’s to adopt and implement employment innovations. Ergo, this research incorporated an Experimental Social Innovative Diffusion (ESID) perspective. Recall that the ESID perspective of social and organizational change is comprised of four phases of action research: 11 27 approach, persuasion, activation, diffusion. The present research focuses on the approach phase." Given the arduous task of introducing new employment innovations into a system that does not necessarily support and have knowledge of these practices, what is the most effective strategy to approach and persuade MHO’s to incorporate new employment programs and techniques into their current practices? This research attempted to answer this question. In particular, it addressed three aspects of this question: (1) How can we solicit MHO’s to send a training participant to the MSU trainings to learn knowledge and skills required toedoptandlmplementtheseemploymentlnncvatlons'f; (2) Howcanweeollclt hemosthterested,moflvated,hnovafivepersons(persommorelkelyto championtheselnnovatlons)toattend MSUtralnings? And (3) Howcenwelnltlate anemrlromientwlthln MHO'sthatIsconducivetoadoptlonand Implementation cftl'leseemploymentlnnovatlons (sothatthesetralning par’dclpantsgobacktoan environment that Is open to these employment Innovations)? Consistent with an action orientation, research has demonstrated the value of a "product champion." Although the literature has attempted to vary such characteristics as social status, it is void of attempts at intentionally soliciting persons who would best characterize, and/or take on, this innovative role. Accordingly, one intention of this research was to experimentally test two strategies The MSU research team plans to investigate aspects of the other three phases at a later date. 28 aimed at soliciting the most Interested, motivated and Innovatlve mental health workers to attend an MSU training aimed at skill and knowledge development required for adoption and implementation of employment innovations for persons labelled mentally ill. Hence, one strategy was to experimentally vary the selection criteria. MHO’s (who were intially randomly assigned to each treatment condition) were asked to select (or not) a training participant who most resembled qualities associated with a "product champion," l.e., risk-taking behaviour, communication potential, influence, network expansion and cosmopolitan behaviour. Ideally, a proactive search for "product champions" (as opposed to others) should lead to more appropriate training participants; persons who will nudge the organization towards change and increase the likelihood of adoption and implementation of employment innovations within their parent organization. Now, this component of the study raised two concerns: (1) what is the best way to solicit organizations to actually send a body; and (2) what would be the most effective method to select persons with product champion qualities? A fiyer could announce the training, describe the positive attributes of employment innovations and note what type of person should come. As such it would address some of the factors associated with adoption. However, use of a flyer represents a mass media approach. Thus, this procedure is contrary to the literature, which suggests that an intense interpersonal approach is necessary when conveying complex information. Hence, all contact was aimed at a more personal level. For 29 example, all MHO administrators were personally contacted by phone”, informed of the training, and asked to select people whose characteristics mirror those of product champions (or not ‘3. In this sense, personal contact, combined with information regarding the actual innovation, was posited to reduce uncertainty. In addition, because administrators were asked to actually select a person with product champion qualities themselves, they may feel more invested in the entire selection and adoption process. But there is a flaw in this approach. Research suggests that administrators do not always optimize their flow of information by selecting sources of information who are relatively experienced or knowledgeable (Kearns, 1989). That is, they are often bound by their own peer network. Accordingly, they may not be aware of others, outside of their network, who might be more appropriate. Therefore, a second strategy experimentally varied the actual initial approach when contacting administrators, whereby a broad-based participative network approach (or not) was also instituted. For example, in the network condition of the approach treatment administrators were personally contacted by phone and asked to consider appropriate persons who mirror product champion qualities (or not). Next, they were asked to give the researcher names of two staff members within ‘2 Telephone advocacy has been experimentally shown to be as effective as face-to- face advocacy regarding the adoption and implementation of a complex innovation (i.e., see Fernandez-Sandin, 1986). ‘3 Information concerning the attributes of employment innovations and rewards for attending the training session was sent to all research participants after initial contact. 30 the organization who might be helpful in selecting appropriate people, and in turn these two people were contacted and asked to give names of two additional staff members. Jointly, all these people were asked to choose someone to come to the training, one who possesses (or not) product champion qualities. It was posited that not only would the activation of a broader participative network within the organization result in greater "sending" behaviour «because it was assumed that the participative network would reduce uncertainty and peak interest, but also a more effective selection of appropriate training participants --by virtue of a instigating a larger peer network who could provide more information about appropriate direct service training participants within the organization. Another part of the study concerned the question how could we initiate an environmentwlthln MHO'sIhatwould beconduclvetotheultlmateadoptlon and implementation of these employment innovations? Recall that broad-based participative decision-making has been shown to be instrumental in reducing uncertainty, and hence, to facilitate decisions to adopt and implement complex innovations. Accordingly, it was postulated that the network approach would also yield a reduction of uncertainty within the participating organizations because it would give people the opportunity to talk with each other about employment innovations. It was also postulated that the network approach would positively affect other factors associated with adoption and implementation such as personal attitudes concerning employment innovations, personal attitudes concerning the employability of persons with mental illness, organizational support for employment 31 innovations, and knowledge seeking behaviour, In comparison to other MHO’s where administrators, only, were approached. In summary, selection criteria (product champion vs not) was instigated to select appropriate innovative training participants; approach strategy (network vs not) was varied to create a broad-based participative network to reduce uncertainty, peak interest (thereby increasing sending behaviour), and to increase the effectiveness of the selection process regarding training participants. Ultimately, it is hoped that this networking approach, combined with a proactive search for product champions will lead to MHO’s adoption and implementation of employment innovations. In summary, this study aspired to experimentally test two strategies (network/no network and product championlno product champion) aimed at (1) solldtingpctentialadoptlonunltstomgtralnlng partlclpantstothedlrectservlce MSUtralning; (2) solicitingthen'lostlnterested, motivatedandmmu’alnlng participants; and (3) lnIIiatlnganva_ron_Mwlthln parentMHO’sthatfacllItates the ultimate adoption and Implementation process regarding employment Innovations. 32 EXPERIMENTAL HYPOTHESES -.-_-:' ntal 0 come .. : : ° NI L aini . P.1i in_-—».~ “ 1. The network condition will increase the possibility of a training participant from a potential adoption unit attending the direct service MSU trainings in comparison to the no network condition. 2. The product champion condition will increase the possibility of a training participant from a potential adoption unit attending the direct service MSU trainings in comparison to the no product champion condition. 3. The network condition will increase the possibility of training participants from a potential adoption unit to attend both the direct service and the administrative MSU trainings in comparison to the no network condition. 4. The product champion condition will increase the possibility of a training participants from a potential adoption unit to attend both the “ There were two kinds of training sessions that MSU offered, one for administrators, and one for direct service personnel. Although this experiment was directed at soliciting direct service training participants, attendance at the administrative training sessions by personnel from potential adoption units in the experiment was also recorded. It was thought that this outcome would be of import since involvement of both top management and front line people has been documented as a salient variable in the adoption process. The two kinds of training sessions were also open to Michigan Rehabilitation Service staff, and the public associated with mental health including consumers, providers, AMI members etc. The direct service training session included six days of workshops covering topics such as values associated with employment and persons labelled mentally ill, employment models and how to implement them, research associated with employment and persons labelled mentally ill. The administrative training sessions covered similar topics, but were only two days in length. Please note that the dependent variables associated with sending training participants reflect attendance for direct service and administrative training sessions offered by MSU during the late spring and summer of 1991. Identical workshops were offered approximately 6 to 9 months later and held if attendance warranted it. This data do not take into account training participants sent beyond the first two training sessions, that is beyond the summer of 1991. Presumably these potential adoption units would represent the "late majority" and/or "laggards". 33 direct service and administrative MSU trainings in comparison to the no product champion condition. e .n.. o-ri entaloutcomeH "theses: , DN INN! TV Tainin- P ic‘ a 5. Potential adoption units in the product champion condition will to a greater degree identify training participants with product champion qualities than those associated with the no product champion condition. 6. Potential adoption units in the network condition will to a greater degree identify training participants who are more interested and motivated concerning employment innovations than those associated with the no network condition. a nda erimen al come H otheses: FA l ITATIN 3 ND IV Enm'ronment for Adoption 7. The network condition will increase the likelihood that administrative and direct service training participants identified show significame more behaviours, perceptions and attitudes that are characteristic of an environment that is conducive to adoption and implementation of employment innovations in comparison to those in the no networking condition, including: a) more knowledge seeking behaviour, b) more positive attitudes towards employment innovations, c) more positive attitudes concerning the employability of persons with mental illness, d) less uncertainty regarding employment innovations, and 9) more positive perceptions concerning organizational support for employment innovations. 8. The product champion condition will increase the likelihood that administrative, network persons, and direct service training participants identified show significantly more behaviours, perceptions and attitudes that are characteristic of an environment that is conducive to adoption and implementation of employment innovations in comparison to those in the no product champion condition, including: a) more knowledge seeking behaviour, v! 34 b) more positive attitudes towards employment innovations, c) more positive attitudes concerning the employability of persons with mental illness, d) less uncertainty regarding employment innovations, and s) more positive perceptions concerning organizational support for employment innovations. WW. 9. Is budget of the potential adoption units and/or experience with employment innovations related to the outcomes associated with sending training participants? 10. Are there other salient factors related to the outcomes associated with sending training participants? METHOD PARTICIPANTS Seventy-six potential adoption units were defined by the researcher (see below). Potential participants consisted of 1) 76 mental health administrators who were in charge of potential adoption units, 2) all persons that were selected by potential adoption units to participate in the MSU direct service training concerning employment innovations for persons who have been labelled mentally ill, and 3) all network persons associated with potential adoption units, that is, those potential adoption unit staff contacted in the network condition. All participation was subject to voluntary consent (see participant recruitment below). Area community mental health boards (CMH), among other things, are in charge of support services for those persons labelled mentally ill, and represent potential adoption units by virtue of their role as service providers. Currently there 35 are 55 CMH boards operating in Michigan. The boards are roughly equivalent in size with the exception of the Detroit-Wayne CMH. This board is significantly larger than other CMH’s, i.e., operating budget is $249,000,000 versus $32,000,000 the next largest CMH operating budget. As such, the Detroit-Wayne CMH board was sub-divided into smaller "adoption units". Smaller adoption units were defined as those programs/services who currently: a) serve persons labelled mentally ill, and b) have programs that are amenable for employment innovation adoption, i.e., day treatment, drop in centres case management, residential programs, and/or c) have programs that are work related, i.e., Lodge, ACT, Clubhouse, supported employment, work related activities. Such sub-divisions yielded an additional 22 potential adoptions units, resulting in a population of 76 adoption units. Because observed differences in adoption might occur because there are structural and resource differences between the boards (Szilvagyi, 1990), potential adoption units were blocked on three salient variables and then randomly assigned within blocks to one of the four experimental conditions. (The use of blocking variables had the added possible benefit of increasing the power of the design.) 36 Mug, Potential adoption units were blocked according to (see Appendix M for blocking details): 1) Experience with Employment Innovations, i.e., potential adoption units scored 2 points for every employment innovation already operating in their district including the Lodge, ACT, Clubhouse, supported employment programs, and work related activities. 2) Budget of the Potenlial Adoption Unit, i.e., low (less than 3 million), medium (greater than 3 million, less than 8 million), and high (greater than 8 million). 3) Wayne Community Mental Health Board Subdivisions, i.e., included 22 sub-divisions. DESIGN The experiment was a 2 x 2 randomized block design (see figure 1). The two independent variables were approach strategy (network vs no network) and seleclion criteria (product champion vs no product champion), with budget and experience with employment programs used as blocking variables for the analyses.“5 ‘5 Location, i.e., Wayne County vs not, was utilized as an independent variable in the analyses associated with sending behaviour, but not for secondary outcomes. 37 FIGURE 1: EXPERIMENTAL DESIGN NETWORK SELECTION CRITERIA I __ _ ii ‘ NO NETWORK NETWORK l I ! NO PRODUCT l CHAMPION ' l I PRODUCT ! CHAMPION The outcomes included: Dependent Variables Associated with Sending Training Participants 1) the number of potential adoption units Who sent/did not send someone to attend the direct service MSU training sessions concerning employment innovations for persons with mental illness; 2) the number of potential adoption units who sent/did not send people to both the direct service and administrative trainings; DependentVarlablesAssociatedwithSending Innovative DIrectServiceTrainlng Participants 3) mean scale scores depicting qualities of the selected direct service training participants, including, product champion qualities, and interest and motivation concerning adoption and implementation of employment innovations; as: uni h‘ 38 Dependent Variables Modated with Facilitating an Environment Conduclve to Adoption 4) the number who engaged/did not engage in knowledge seeking behaviour; 5) mean scores associated with a measure of personal attitudes towards employment innovations; 6) mean scores associated with a measure of personal attitudes concerning employability of persons labelled mentally ill; 7) means scores associated with a measure of certainty regarding employment innovations, and; 8) mean scores associated with a measure of perceived organizational support for employment innovations. Within each of the three blocks, potential adoption units were randomly assigned to one of four conditions. Administrators within these potential adoptions units were than identified. The two experimental treatments consisted of: Network Treatment 1) mm, Contacting administrators by phone and asking them to personally select an interested and motivated person(s) with product champion qualities (or not) to attend the direct service MSU training sessions (see Appendix I for approach protocol for administrators); 2) NM Contacting administrators by phone and asking them for names of two persons who could help select an interested and motivated person(s) with product champion qualities (or not) to attend the direct service MSU training sessions, followed by phone contact to these two people (see Appendix J for approach protocol for network persons) for names of two more persons to help select, asking that all parties jointly attempt to select an interested and motivated person(s) with product champion qualities (or not) to attend; 39 Product Champion Treatment 3) NO figuct mm. Contacting administrators by phone and asking them to personally select (or with network staff) an interested and motivated person(s) to attend the direct service MSU training sessions; 4) Product chamM. Contacting administrators by phone and asking them to personally select (or with network staff) an interested and motivated person(s) with product champion qualities to attend the direct service MSU training sessions. After the initial contact, all voluntary participants (administrators and network people) were mailed a standardized information package and a questionnaire. This package included an outline of the training program, the positive attributes of employment innovations, and the reward associated with completion of the MSU training sessions: a MSU training certificate.“I Recall that positive attributes and rewards have been shown to be related to the adoption process. MEASURES 1mm A yes/no categorical variable was used to record 1) whether or not potential adoption units sent a direct service training participant(s), and 2) whether or not they sent at least one person to the direct service training and at least one person to the administrator training. ‘° Past experience has shown that giving out a MSU training certificate is considered prestigious and rewarding among persons in the mental health field. 40 mm This check list consisted of short answer questions aimed at gleaning historical information, i.e., how long have you worked in your current job?; and multiple choice questions aimed at gleaning demographic information i.e., age, gender, race, education (see Appendix A). W This measure was designed to help assess the fidelity and contamination of the treatments associated with each of the four conditions. It contained short answer questions and multiple choice questions concerning administrators’ and (network) staffs’ actual participation in the selection process, use of the selection criteria in the selection process, and their communication with other potential adoption units (see Appendix B). A Likert type item which measured the administrators’ and network people’s feelings about being part of the selection process was also included. W This measure consisted of two yes/no variables regarding knowledge seeking behaviours involving employment innovations, alter MHO’s were approached. Specifically, I) did administrators and network people seek out MSU for information regarding the training?; 2) did they seek out information from others regarding employment innovations? This measure also consisted of a Likert type item designed to assess administrators’ and network persons’ interest in attending the MSU administrative training session. These items were incorporated into the Manipulation Check Measure (see Appendix B). 41 P of Innovations Persons Labelled M11111, This measure consisted of seven Likert type items designed to assess respondents’ perception of the attributes of employment innovations for persons labelled mentally ill, i.e., relative advantage, trialibility, credibility, observability, relevance, ease in understanding, and compatibility. ' It was the same measure used by Glaser & Backer (1977) in their case study of sustained versus nonsustained change in three mental health rehabilitation programs, including the Lodge model (see Appendix C). Please refer to Table 1 for the reliability of this measure and all others. towards em I labelled mental II This measure consisted of 13 Likert type items designed to assess general attitudes towards the employability of persons labelled mentally ill (see Appendix E). This measure was a modified version of an attitude measure developed by MSU researchers for marketing students working with persons labelled seriously mentally ill. It consisted of three areas: ability of the psychiatrically disabled to contribute (items 1, 3, 8, 10); ability to engage in competitive employment (items 2, 4, 6, 7, 11), and; the need of persons labelled mentally ill to feel that they contribute (items 5, 9, 12, 13). W This measure was designed to assess the extent of (un)certainty as outlined by Rogers (1983) and Tomatzky et al (1980), who examined, among others things, uncertainty reduction regarding the Lodge model (see Appendix D). It covered three areas: certainty of 42 knowledge/understanding of employment innovations (items 2, 3); certainty of employment innovation outcomes (items 5, 6); and certainty of the (respondent’s) organization’s ability to implement these employment innovations (items 1, 4). measure was designed to assess the extent of organizational support for employment innovations. Because of the problems associated with organizational support per se, i.e., is it the administrator?, heads of departments?, structures? and so on, here, it was conceptualized as a core group of people within the respondent’s parent organization. This measure utilized five Likert type items to assess knowledge interest, persuasion of adoption, management of implementation and organizational resources (see appendix H). _ mum-.4 This measure was designed to assess the extent to which direct service training participants were interested and motivated to learn about and adopt employment innovations (see Appendix G). It contained eight Likert type items covering three areas: Interest in knowledge of employment innovations (items 1,2); Behavioural indices of interest (items 3, 4, 6, 8), and; Willingness to engage in behaviours associated with adoption and implementation (items 5, 7). BMW The profile embodied a battery of scales designed to measure training participants’ degree of product champion qualities as outlined in the literature (T ornatzky & Fleisher, 1990; Rogers, 1983; Hill, 1982; Tomatzky et al, 1980; Loy, 1969) including 1) cosmopolitan behaviour, 2) 17 43 influence, 3) network linkages, 4) communication potential, and 5) risk taking. Three of the scales, cosmopolitan behaviour (seven items), influence (eight items) and communication potential (five items) were a modified version of scales used by Hill (1982) in her study of the adoption of a university program by insiders". The network linkage scale was developed with ideas from Keams (1989) in mind, Where he attempted to assess administrators’ use of networks regarding computer information (see Appendix F). It contained 12 Likert type items. Risk-taking was measured by a single Likert type item. PROCEDURES mm Administrators were identified from potential adoption units as defined above. Recall that the potential adoption units were blocked on three variables (experience with employment innovations, budget, and Detroit/Wayne CMH sub-divisions) and randomly assigned to one of the four experimental conditions. (Note that voluntary consent was obtained after potential adoption units were randomly assigned.) All administrators were contacted by phone and asked to participate in the experiment by answering a couple of questions immediately, and later, completing a consent form and mailed questionnaire (see Appendix I for approach protocol for administrators; see Hill’s scales yielded reliabilities of: cosmopolitan .61 (alpha),.95 (test-retest, Spearrnan’s rho); influence .95, .80; and communication potential .71, .85. 44 TABLE 1: RELIABIU'RES OF MEASURES EMPLOY. INNOVATIONS EMPLOYABILITY CERTAINTY .75 206 6 SUPPORT .88 198 5 INTEREST .82 54 8 ; COSMOPOLITAN .80 43 7 I l INFLUENCE .72 54 8 COMMUNICATION .57 57 5 : NETWORK LINKAGES .86 55 12 l l- . Appendix L for consent form). If they agreed, they were given instructions appropriate to their experimental condition. Similarly, network people and those identified as direct service training participants were contacted by phone and asked to participate in the experiment by answering a couple of questions immediately, and later, completing a consent form and mailed questionnaire (see Appendix J for approach protocol for network persons; see Appendix K for protocol for direct service training participants; see Appendix L for consent form). 45 If they agreed, they were given instructions appropriate to their experimental condition. To obtain information regarding selection of training participants, administrators (or network persons) were contacted two weeks after they were first approached (see Appendix K for follow-up protocol). 1mm Once a participant agreed to fill out a mailed questionnaire the "Total Design Method" (Dillman, 1978) was followed. This tested method for increased return rate involved mailing an initial cover letter on MSU letterhead with the initial questionnaire, if no response in a week and a half a reminder post card was sent, and when necessary, a new questionnaire and cover letter three weeks later. Although Dillman (1978) recommends mailing additional questionnaires "first class" as a final step in the process, this step was not done after considering the costs and gains involved. All administrators and network persons from potential adoption units, who volunteered for the experiment, were mailed a questionnaire which contained the following measures: 1. Background Check List 2. Manipulation Check 3. Knowledge Seeking Behaviour 4. Perceptions of Employment Innovations for Persons Labelled Mentally Ill 5. Attitudes towards Employability of People Labelled Mentally Ill 6. Uncertainty of Employment Innovations 46 7. Perceptions of Organizational Support for Employment Innovations Similarly, all training participants selected to go to the direct service MSU training by potential adoption units were contacted by telephone. Volunteers were mailed a questionnaire containing the same measures used for the administrators and network people, with the exception of the knowledge seeking behaviour scales. Their questionnaire also contained two additional measures, including: 8. Interest and Motivation concerning Employment Innovations 9. Profile of Product Champions AIM Frequencies were tabulated for all information collected on participants and potential adoption units. Fidelity of the experimental treatment was used to determine inclusion in all other analyses. That is, all adoption units (and their associated participants) that did not allow the provision of the network condition, or, initially said they would not select someone to go but did allow the network condition were not included in the next stage of analyses. Logistic Regression analyses were used to assess the effects of the treatments, blocking variables, and various factors such as location, on two of the dichotomous outcome variables related to sending training participants, i.e., sent direct service training participant/or not; sent training participants to both administrative and direct service training sessions. Because of the potential difficulty raised by correlated dependent variables and an increase in the probability of finding a significant result when performing 47 so many statistical tests, MANOVA was utilized to test for differential effects of the treatment and blocking variables regarding secondary outcome variables. For example, mean scores from the scales regarding interest in attending an administrator training session, attitudes towards employment innovations, attitudes towards the employability of people labelled mentally ill, certainty of employment innovations, and organizational support for employment innovations - which were dependent variables associated with the outcome of facilitating an environment that Is conducive to adoption - were examined via MANOVA to examine the effects of the treatments and blocking variables. This procedure was carried out across the three groups of administrators, network people, and direct service training participants. Similarly, MANOVA was used to test for differential effects of the treatment conditions and blocking variables on the outcome variables associated with profile Of innovative direct service training participants, i.e., interest in adopting employment innovations, cosmopolitan behaviour, influence, network linkages, communication potential and risk-taking behaviour. In addition, because the experiment provided for possible situations whereby more than one person was associated with an adoption unit, all data for network people and direct service training participants were initially analyzed using a nested design to test for an "organizational effect." If no significant effect was detected the nested effect was dropped, and scale scores were analyzed using a normal full factorial experimental design, employing individuals as subjects. 43 A Ioglinear analyses was proposed to examine the effects of the treatment and blocking variables on the categorical knowledge seeking behaviour items. However, for some reason the knowledge seeking behaviour item regarding whether or not they sought information froms other about employment innovations contained in the questionnaire was not often answered. For example, 47% Of the administrators left this question blank, and 54% of the network people did the same. Therefore two chi square tests were used to test for differences between the network/no network groups and between the product champion/no product champion groups. In consideration of the power of this experiment design, typically, 1) analyses began with a full factorial design which included the two blocking variables of budget and experience with employment prOgrams; if no significant results were found, secondary effects such as three-way interactions were sequentially removed from the analysis, followed by the blocking variables and so on, and 2) all test results which yielded a probability of .10 or less were considered as potentially significant results“. "3 The sample size of this study limits the power of this experiment. For example, in some situations n’s within cells were only 1 or 2, or empty. It has been suggested that liberalizing the significance level is a reasonable solution to this problem (i.e., see Bock, 1975). 19 49 RESULTS DEMOGRAPHICS AND PARTICIPATION RATES Sixty of the 76 potential adoption units agreed to participate in the experiment, representing 79% of the total Michigan population of potential adoption units. The demographics of the adoption units are listed in Table 2. A total of 211 people, 51 administrators, 101 network persons, and 59 direct service training participants took part in the entire experiment. These samples represented an 88% questionnaire return rate. The demographics of participants are listed in Table 3. CONTAMINATION AND FIDEIJTY Contamination of the treatment conditions did not appear to be a problem: Only 9% (n = 137) of all the administrators and network participants reported that they spoke with others from potential adoption units regarding the MSU training project. Fifty-seven percent of the administrators and network people reported that they participated in the actual selection process regarding direct service training participants, and 62% reported that their organization participated in the selection process (n = 146)“. Of the administrators and network people who said they participated in the selection process, mean scores on a five point scale suggested that they utilized recommendations regarding qualities of direct service training These figures may actually underrepresent actual participation by the adoption units. For example, administrators, in some cases, delegated the task to someone else. 50 participants at least "somewhat" or "very much" (n = 79; X = 3.4); and 98% (n = 77) talked with others in their organization about the selection process. Table 2: DEMOGRAPHICS OF POTENTIAL ADOPTION UNITS Demographics POTENTIAL ADOPTTON UNITS TREATMENT CONDITIONS 1. no network/no product champion n = 16 ; 2. network/no product champion n = 16 l f 3. no network/product champion n = 12 4. network/product champion n = 16 LOCATION ' Upper Peninsula 13% 5 Detroit 20% 3 Other 67% EMPLOYMENT PROGRAMS Low 53% a Medium 33% ; High 13% 5 BUDGET Low 41% ; Medium 32% , High 27% 51 TABLE 3: DEMOGRAPHICS OF PARTICIPANTS D.S. TMINING ADMINISTRATORS NETWORK PEOPLE PARTICIPANTS T N = 51 N = 101 N = 59 I GENDER female ‘ 26% 53% 63% male 74% 47% 37% EDUCATION h.s. or less 0% 0% 2% some college 0% 5% 10% undergrad deg. 0% 17% 30% some graduate 2% 11% 22% grad. deg. 98% 67% 36% MAJOR mental health 79% 75% 70% rehabilitation 0% 7% 6% both 6% 3% 0% other 15% 15% 24% ETHNICITY asian 2% 0% 0% air. amer. 8% 5% 10% hispanic 2% 0% 0% native am. 0% 1% 2% white 88% 94% 88% AGE 46 yrs. 37 yrs. 36 yrs I JOB f current 7 yrs. I mental health 52 Excluding those four potential adoption units which refused the network treatment and/or said they would not send anyone to the trainings but would allow the treatment resulted in 93% of the 60 potential units being included in the rest of the analyses, along with 94%, 92%, and 99% of the administrators, network people and direct service training participants, respectively. This procedure yielded a sample of 56 potential adoption units, 48 administrators, 93 network, and 58 direct service training participants.2° SENDING TRAINING PMTTCIPANTS 91-01 in EANDADMIN ‘1 " LI, hiNl ' 311 I . LIT» all Fifty- seven percent of the 56 adoption units sent at least one person to the direct service training session, and 46% sent at least one person to each of the direct service and administrative training sessions. A logistic regression was used to predict both the sending of direct service training participants, and the sending of training participants to M the direct service and administrative training sessions. WIth the exception of budget, the two experimental treatments (see Tables 4 and 5 for raw frequencies), blocking variables, and other predictors, such as location and fidelity did not significantly predict either dichotomous outcomes. Actual budgets of adoption units was a significant predictor regarding the sending of direct service training participants, although it accounted for less than 5% of the 2° These n’s could change depending on the analyses and missing values within scales and items. For example, a MANOVA test utilizing 5 dependent variables will only include those participants who have no missing values on all five variables. 21 53 variance (R = .18; p. = .04"). Table 6 illustrates the relationship of the adoption units’ budget to the sending of direct service training participants: the higher the budget, the more likely they were to send at least one person to the direct service training (12 = 11.67; 2 df, p. = .003). Budget was not a significant predictor regarding the sending of training participants to will the direct service and administrative trainings. Although the number of adoption units who sent training participants was lowest in the network/product champion condition all of those same potential adoption units sent training participants to both the administrator and direct service trainings. This was not the case for the other three conditions, where, in all three cases, fewer potential adoption units sent both. Hypothesesconcenmgflteslgnificencecffltenetworkandtheproduct chanplon treatments as predictors regarding the sending ofdlrectservlcetralnlng participants, andflwesendingcftralnlngpu'tlclpantstommthedlrectservlceand ammIstratlvelTainlngswerenotsupported. Thatis,IncII.Islonhthenetworkor fluproductdwnpbncmdifionsdidnctresuflhslgnlfimnflygreatersending' behaviour regarding dkect service or direct service plus edrnlnlstratlve trainings. cher Variables, A correlation matrix indicated that attitudes towards the employability of people labelled mentally ill (r = .17, p. = < .05) and certainty regarding employment innovations (r = .17, p. < .05) were positively related to the sending at least one training participant to the direct service training. Note that as shown above, budget of the potential adoption unit (r = .43, p. < .01) was also shown to be significantly related to sending at least one person to the direct The actual budgets of the organizations were used in the logistic regression analysis, rather than the categories of high, medium and low. 54 service training session. The product champion condition yielded a negative significant correlation with this same outcome variable (r = -.17, p < .05) even though the logistic regression analysis did not detect it as a significant predictor. This discrepancy in results may be a power problem. For example, the cell totals for the product champion treatment depict lower percentages for those in the product champion condition (48%) vs the no product champion condition (65%). suggesting a product champion effect. Correlations revealed that interest in adopting employment innovations by direct service training participants (r = .30, p < .01), attitudes towards the employability of people with mental illness (r = .21, p. < .01), feeling part of the direct service training participant selection process (r = .28, p. < .05), and sending at least one person to the direct service training session (r = .53, p. < .01) were positively related to the sending of training participants to m the direct service and administrative training sessions. Budget (r = .16, p. < .05) and experience with employment programs (r = .21, p. < .01) were also positively correlated with the sending of training participants to bath the direct service and administrative training sessions, even though they were not identified as significant predictors in the logistic regression analysis. The network treatment showed no relationship to feeling part of the direct service training participant selection process. 55 TABLE 4: FREQUENCIES OF ADOPTTON UNITS WHO SENT DIRECT SERVICE TRAINING PARTICIPANT (S) BY NETWORK AND PRODUCT CHAMPION TREATMENTS PRODUCT CHAMPION NO TABLE 5: FREQUENCIES OF ADOPTTON UNTTS WHO SENT MNING PARTICIPANTS TO BOTH DIRECT SERVICE AND ADMINISTRATOR TRAININGS BY NETWORK 8: PRODUCT CHAMPION TREATMENTS - NETWORK [ "HM” M“ _____,____ __ fl __ _____ _.,. _‘ "”"“ m” T _- "T 7 W ‘7 7 ’ l l NO YES 5 PRODUCT CHAMPION 3 NO - 7/15 8/15 I l (44%) (53%) ' YES 6/12 5/13 (50%) (39%) 56 TABLE 6: DIRECT SERVICE TMINING PARTICIPANTS SENT BY BUDGET OF ADOPTTON UNTTS YES 30% 72% 80% I (7) (13) (12) Conduclveness of the Environment to Adopt mm A MANOVA on the mean scale scores related to conduciveness of the environment to adoption yielded a significant interaction between experience with work programs and the productchampion condition (F = 2.15; 10, 36 df; p. = .046, see table 7). Univariate F tests indicated one significant interaction regarding the product champion treatment and interest in attending an administrative training (F = 3.14; 2,22 d1; p. = .06, see table 7). Eyeballing the data, of the 48 administrators, those in the product champion condition with low and high experience with employment programs yielded higher means on "interest in attending an administrator training" than those in the no product champion condition (4.4 vs 3.7; 4.8 vs 2.3, respectively), but the opposite effect was found for those with medium experience. Administrators in the product champion condition with medium experience with employment programs yielded lower means regarding interest in attending an administrator training (3.3 vs 3.6) than those in the no product champion condition (see figure 2). The grand means FIG. 2: ADMINISTRATOR TRG SCALE: WORK EXPERIENCE BY P.C. INTERACTION LOW MEDIUM HIGH EXPERIENCE WITH EMPLOYMENT PROGRAMS ----- ----- NO PROD. CHAMPION —l- PRODUCT CHAMPION 58 for the other four scales were: 3.6 (attitudes towards employment innovations), 4.3 (certainty regarding employment innovations), 4.4 (attitudes towards the employability, of people labelled mentally ill) and 4.0 (organizational support for employment innovation). These means suggested, that in general, administrators felt at least "somewhat" or "very much" that attributes of employment innovations were credible, observable, relevant, easy to understand, etc.,; they felt at least "very certain" that employment innovations worked and could be implemented; they agreed at least "very much" that people labelled mentally ill were employable, had vocational goals and could contribute etc.; and they felt "very much" that there was a core group of people in their organization that were interested in employment innovations and willing to adopt them within their organization. Regarding the two measures of administrators’ knowledge seeking behaviour, no administrators contacted MSU regarding information about the trainings, therefore no test was carried out. A chi square test of significance regarding whether or not administrators sought information about employment innovations revealed no significant differences between groups in the network/no network condition. Similarly, no significant differences were found between groups in the product champion/no product champion conditions. Frequencies indicated that of the 26 administrators who responded to the question 50% sought information about employment innovations. Thehypcttiesesregardlngtl'lemalneffectscftl'lenetworkandpmduct championtreetmentswerenotsupported. Administratorsinthenetworkcondltlon didncthaveslgnlficanflyhighermeammmescalesrdatedtocondudvenessof theerwlronmenttoadoptemploymentlnnovatlons. Similarly,nodlfferenceeon 59 scdeswerefoundbetweenadmlnlshatomhtheproductdiamploncondiflonand thoeenot. TABLE 7: CONDUCNENESS OF THE ENVIRONMENT TO ADOPT: EMPLOYMENT INNOVATION EXPERIENCE BY PRODUCT CHAMPION TREATMENT FOR ADMINISTFMTORS Test Value F Hyp. df Err df Sig. of F WIlks .39 2.15 10.00 36.00 .05 UNNARIATE F-TESTS WITH (2,22) df: Variable MS (TREATMENT) MS ERROR F SIG. OF F EMP. INNOVATIONS .18 .14 1.31 .29 CERTAINTY .02 .13 .16 .85 EMPLOYABILITY .14 .17 .81 .46 ORG. SUPPORT .40 .47 .86 .44 ADM. TRAINING 4.98 1 .59 3.14 .08 mm, Results of a MANOVA, using a nested design, indicated no nested effect. That is, there were no significant differences detected between organizations, therefore the nested factor was dropped. Results of a MANOVA, using a full factorial experimental design indicated that there was a product champion effect (F = 1.92; 5,69 df; p. = .10, see table 8). With the exception of interest in attending an administrative training session, univariate F tests depicted a product champion effect for all scales (see table 8). Of the 88 network people, 60 those in the no product champion condition yielded significantly higher means on attitudes towards employment innovations (3.5 vs 3.4; F = 3.68; 1,73 df; p. = .06), certainty regarding employment innovations (4.1 vs 3.9; F = 3.56, 1,73 df; p. = .06), attitudes towards the employability of people labelled mentally ill (4.2 vs 4.1; F = 4.14; 1,73 df; p. = .05) and organizational support for employment innovations (3.6 vs 3.3; F = 4.62; 1,73 df; p. = .04) in comparison to those in the product champion condition. In general, people in the no product champion condition felt that employment innovations were more credible, observable etc. in comparison to those in the product champion condition. They were also more certain that employment innovations worked; that persons with mental illness were employable and could contribute; and that their organization had a core group of people who were interested in and willing to adopt these employment innovations in comparison to those in the product champion condition. Although means scores on the interest in attending an administrative training session scale were higher for those people in the no product champion condition in comparison to those in the product champion condition (4.2 vs 3.9; grand mean = 4.1), differences were not statistically significant (see figure 3). Regarding the two measures of network people’s knowledge seeking behaviour, only one person sought information from MSU regarding the trainings, therefore a test was not carried out. A chi square test of significance regarding whether or not network people sought information from others concerning employment innovations revealed no significant differences between the product champion/no product champion groups. Frequencies indicated that of the 43 network people who answered the question, 42% (18) sought information regarding employment innovations. 555555 62 Thehypotheslsregardlngthemalneflectoftheproductchamplon teachentonthecondudveneescfflteenvlronmenttoedoptsmleswasnot supported.Thatls,meanscalescoresrelatedtoconduclvenessofthe awkonmmadoptemploymhnovaflomhthepmductdtamploncondlfion werenothlgherthanthosenotlnthecondltlon. Infacttheopposlteeffectwas fumd-wlmmemrcepflonofthehtemsthattendlnganadnflnbuwvetalnlng sessbnscalewheredifierenceswerelnasknilardlrecflonbutnotstaflsficafiy 819M TABLE 8: CONDUCNENESS OF ENVIRONMENT TO ADOPT SCALES: PRODUCT CHAMPION EFFECT FOR NETWORK PEOPLE Test Value F Hyp. df Err df Sig. of F WIIks .88 1.92 5.00 69.00 .10 UNIVARIATE F-TESTS WITH (1,73) df Variable MS (TREATMENT) MS ERROR F Sig. of F EMP. INNOVATIONS .72 .20 3.68 .06 CERTAINTY .79 .22 3.56 .08 EMPLOYABILTTY .58 .14 4.14 .05 ORG. SUPPORT 2.69 .58 4.62 .04 ADM. TRAINING 3.95 3.61 1.09 .30 W. Results of a MANOVA, using a nested design, indicated that there was no nested effect. That is, there were no significant differences detected between organizations, therefore the nested factor was 63 dropped. Results of a MANOVA, using a 2 X 2 factorial experimental design (recall that non-significant factors were sequentially taken out of the analysis), yielded a network effect (F = 3.94; 4,47 cit; p. = .008, see table 9). Univariate F tests indicated significant differences between the network and no network groups on the scales of attitudes towards employability of people labelled mentally ill, certainty of employment innovations and organizational support for employment innovations (see table 10). Of the 54 training participants, those not in the network condition depicted significantly higher means than those in the network condition on the certainty (4.3 vs 4.0; F = 3.47; 1,50 df; p. = .07) and organizational support scales (4.0 vs 3.6; F = 5.18; 1,50 df; p. = .03) - yet a lower mean on the employability of people labelled mentally ill scale (4.2 vs 4.4; F = 4.43; 1,50 df; p. = .04); no significant differences were found regarding attitudes towards employment innovations, although the trend is similar to the certainty and organizational support scales (3.6 vs 3.5; grand mean = 3.5, see figure 4). Although both groups reported that they felt at least "very certain" that employment innovations worked, those not in the network condition felt more certain that employment innovations worked and could be implemented. Similarly, those in the no network condition felt "very much" that their organization had a core group of people who were interested and willing to implement these innovations as compared to those in the network condition who felt only "somewhat" or “very much". Conversely, both groups agreed at least "very much" that the psychiatrically disabled population were employable, but those in the network condition seemed to feel that people labelled mentally ill were employable and could contribute even moreso than those in the no network condition. In 65 general, training participants felt at least "somewhat" or "very much" that employment innovations were credible, observable, compatible etc.. TTTehypclheslsregardingthemalneffectOftheproductchamplon cordfiononthecondudvenesscftheeeronmenttoedoptsceleswasnct supported. Thisls,meanscalescoresdeplclingtheconduslvenesscftl'le environment to adopt employment innovations from direct service training pufldpmlsassodatedwlthflteproductchamploncondiflonwerenothigherh comperlsontothosenctlnthecondltlon. Thedatadldsupportthehypothesls mgardlngfltemalneflectcfbeinghfienetworkcondlflonfortheaflihrdes towardstheemployabllltyofpeoplelabelledmentallylllscale. Thatls,mean scoreswereslgnlficantlyhlgher. Butresultswereopposlteofwhatwaspredlcted forhescabsmgardlngcertalntyofempbynmfilmwaflomandorganizaflonal supportforemploymentlmcvatlonssceles. Thatls,meanscoreswere slgnlflcanflylowerformosetralnlngparfldpantshmenetworkcondlflonh compubontoflnselnfiremnetworkcondltonmndmdlflerencesweredetected betweengroupsontheattlhflestowardsemploymentlnnovafionsscale. 66 TABLE 9: CONDUCNENESS OF THE ENVIRONMENT TO ADOPT SCALES: NETWORK EFFECT FOR TRAINING PARTICIPANTS Test Value F Hyp. df Err df Sig. of F WIIks .74 3.94 4.00 47.00 .01 UNIVARIATE F-TESTS WITH (1,50) df Variable MS (TREATMENT) MS ERR F SIG. OF F EMP. iNNOVATIONS .36 .15 2.39 .13 CERTAINTY .97 .28 3.47 .07 EMPLOYABILITY .59 .1 3 4.43 .04 ORG. SUPPORT 2.72 .53 5.18 .03 PROFILES OF INNOVATTVE DIRECT SERVICE TRAINING PARTTCIPANTS A MANOVA, using a nested design, detected no significant differences between adoption units with regard to the product champion scales, therefore the nested factor was dropped. A MAVOVA, using a factorial experimental design revealed no significant differences between groups on any of the product champion scales. Table 10 displays the grand means for the 39 training participants included in the analysis. All measures reflected five point scales, with the exception of the cosmopolitan behaviour scale which was standardized using 2 scores. In general, these direct service training participants were at least 67 "somewhat" or "very interested" in learning about employment innovations; they were somewhat cosmopolitan, i.e., they read books, journals etc. in their field of TABLE 10: PRODUCT CHAMPION PROFILE: GRAND MEANS OF SCALES I SCALES MEAN E n INTEREST 3.66 39 COSMOPOLITAN .28 39 'l NETWORK LINKAGES 3.12 39 INFLUENCE 3.72 39 ll COMMUNICATION 4.42 39 RISK-TAKING 3.61 39 specialization but little outside their area, they travelled 1-3 times a year outside their local community, but very little outside of their state to learn about mental health issues; they reported that they talked with a variety of others within their organization but much less outside their immediate network; they reported they were at least "somewhat" or "very much" influential within their organization; that they communicated at least "very well" within the context of work; and they were at least "somewhat" or "very much" risk takers. Thehypctheslsregardingtl'lemalneifectoftheproductchamplon treatmentontheproductchamplonscaleswasnctsupported. Thatls,mean 68 scalescoresregerdlngquellflesofa'productdtamplon'fiomdlrectservlce caningparfldpantsmodatedwlmtheproductdiampbncondlflmwerenot slgrificarfiyhlgherthanthosenotlnmecondltlon. SImIIarly,tI'IelTypothesls mga'dingfltemaineflectoffltenetworkmnonmeinterestandmcflvaflm scaleswasnotsupported. Thatls,meanscalescoresregardlnglnterestand motlvalloncfdlrectservlcetralnlngpartldpantswerenotslgnlflcantly higherlnthe networkcondlflonmanfliosenctlntltecondltlon. DISCUSSION Sending Training Partlclpanb W The actual sending of training participants by potential adoption units to the MSU direct service trainings was a salient outcome in this experiment. This "sending" behaviOur is important to the adoption process because, ideally, attendance at the MSU workshops promotes the acquisition of skills and knowledge regarding the adoption and implementation of the employment innovations. Essentially, it is the first step in the adoption process. The hypotheses regarding the product champion and the network treatments as salient predictors of sending direct service training participants were not supported in this experiment. Instead, actual budgets of potential adoption units emerged as a significant predictor. The larger the budget of the potential adoption unit, the more likely the organization sent at least one training participant to the MSU employment training. Given the zeitgeist of the times - a troubled economy, this finding speaks for Itself. Similar to critics (e.g., Tomatzky 81 Fleischer, 1990) who proposed that size of an organization may be an indicator 69 of other organizational structures, conversations with administrators, network people and training participants led the experimenter to believe that budget was more an indication of whether or not potential adoption units had the personnel to cover for staff who were at the training, rather than strict monetary problems related to the cost of the trainings”. These findings suggest that budget of an bureaucratic organization (or other related indicators) may be paramount in the decision to send training participants to a training workshop concerning innovations. This is supportive of the literature, where slack resources and size of an organization have been shown to be powerful predictors of adoption (T ornatzky & Fleischer, 1990; Kelly & Brook, 1988; Tomatzky et al, 1983; Rogers, 1983; Fairweather et al, 1974; Aiken & Hage, 1970; Cyert & March, 1963). It may be important, however, to consider these findings within the context of the length of the direct service training session (recall the training session was six days). Length may have compounded the effect of budget creating a threshold effect - those organizations with medium and high budgets being able to send, those with low budgets not able to send training participants. It is recommended that future practitioners and researchers concern themselves with the budget of organizations (or similar indicators) if they are interested in sending at least one person from bureaucratic organizations to an innovation adoption training, particularly within the context of the length of the trainings. Other variables which were positively correlated with the sending of a direct service training participant were attitudes towards employment of people labelled mentally ill, and certainty that employment innovations worked. That is, 22 The trainings actually cost only $15 a day, 70 "senders" (people associated with potential adoption units who sent someone to the direct service training session) were not only those with larger budgets, but correlations characterized them as having attitudes which were more positive regarding the psychiatrically disabled’s employability, and more certainty regarding employment innovations, i.e., that these innovations worked and could be implemented, in comparison to the non-senders. These findings need to be interpreted with caution since these correlations reflect data from the participants in the experiment and therefore some groups, is. staff from MHO’s not in the network condition, are not represented. However, these findings are supportive of the literature. Although the logistic regression analysis did not detect the product champion treatment as a significant predictor, as a simple correlation, it was negatively correlated, albeit modestly, with sending direct service training participants. It is assumed that this disparity is primarily because logistic regression analysis takes into account the covariance of multivariate relationships. It could also reflect a chance occurrence. More likely, it is a power problem with the experiment. Recall that the cells for each condition depicted lower percentages of sending behaviour for those in the product champion condition (48%) vs the no product champion condition (65%), implying a product champion effect. Because of the limitations of the real world (i.e., there were only 76 potential adoption units in Michigan) and attrition due to the voluntary nature of this experiment, cell sizes became unequal, resulting, in some instances, in 1 or 2 cases within a cell - or empty cells, and hence lower power. The network treatment did not appear to be a significant factor regarding sending behaviour, therefore, one consideration is to drop this kind of treatment, 71 and concentrate resources on those who can afford to come --relying on a diffusion effect for those unable to. Nonetheless, broad-based participative decision-making has been touted as a reliable factor in the literature with respect to adoption decisions; as such, it is likely that there is a need to develop this kind of behaviour early in the adoption process. Therefore, alternatively, it is recommended that the network treatment be modified. For example, after the network treatment is instituted, researchers could experiment with ongoing prompting. Or, they could vary the network treatment according to budget of the organization: simply calling administrators and sending information to those associated with high and medium budgets; intense networking, rewards, in combination with on sight trainings etc. for those associated with lower budgets. Researchers/practitioners may also wish to tailor the length of the trainings to match budget constraints. W Hypotheses depicting the product champion and network treatments as salient predictors regarding the sending of training participants to m the administrative and direct service training sessions were not supported. These findings were disappointing since the literature identifies the involvement of both top management and front line people as key to the process of adoption and implementation, i.e., ensures the release of resources etc.. However, simple correlations did suggest a positive relationship between sending training participants to both trainings and attitudes towards the employability of the psychiatrically disabled, interest and motivation of direct service training participants, and administrators’ and network people’s "feeling part of" the selection process. That is, based on correlations, "senders of both" (people who 23 72 were associated with potential adoption units who sent 91th) were characterized as feeling more positively concerning the employability of the psychiatrically disabled, as having direct service training participants who were more interested and motivated to implement these innovations, and feeling more a part of the process to select direct service training participants, than were non-senders of both. In essence, these relationships are supportive of the literature, particularly with regard to the process variable. Ideally, feeling part of the selection process is a reflection of broad based participative decision-making, which has been documented as a necessary component in the adoption process, i.e., reduces uncertainty etc.. Interestingly, this variable did not correlate with the network treatment - the very intention of the intervention. This pattern could be a reflection of the fidelity of the experimental treatment, i.e., there is evidence that a large minority (43%) of participants did not involve themselves or were not included in the selection process. It could reflect the strength of the treatment, or simply that it was not effective at inducing these kinds of feelings and so on. In a similar vein, there were some qualitative data to suggest that the network treatment was viewed, in some instances, as intrusive and/or threatening”. Regardless, it appears the network treatment utilized in this experiment was insufficient as an intervention to bring about broad-based participative decision-making and therefore it is recommended that it be abandoned, or modified. In light of the importance that broad-based participative decision-making plays in the adoption In a few instances administrators were irritated at the thought that the experimenter would be calling others in the organization. One labelled it a marketing scheme, and in fact, three administrators would not allow the treatment. Similar sentiments were expressed by a few network people. 73 process, and the correlational support in this experiment for involvement in the selection process with regards to sending both, future researchers could modify the network approach so as to better induce feelings of partaking in the decision- making regarding the adoption process while simultaneously attending to feelings of intrusiveness. Simple correlations also detected budget and experience with employment innovations as having a small, but significant relationship with sending people to both trainings. Like the sending of direct service training participants, this disparity is likely a reflection of analysis differences, but it could be a power or type 1 error problem. As discussed, caution is warranted regarding interpretation of these correlations because of the lack of representation from the no network staff. Although potential adoption units in the produ'ct champion/network condition appeared to be the least likely to send direct service trainings participants (although not statistically significant), all of these same units sent training participants to both trainings as compared to potential adoption units in the other three conditions, where in all cases, not all units sent both. It could be that a combination of both treatments may not entice many to come, but if they do, they go all the way - sending training participants to both the direct service and administrative trainings. This pattern is consistent with Fairweather’s et al (1974) findings that although less active approaches were effective in "getting a foot in the door," Significantly more hospitals in a more intense approach actually agreed to adopt the Lodge program. This conclusion will be explored by MSU during the long term follow-up of adoption and implementation. 74 Conduclveness of the Environment to Adopt Employment Innovations mm The hypotheses concerning the main effects of each of the treatments, i.e., product champion and network, on the conduciveness of the environment to adopt scales were not supported. That is, despite suppositions that means would be higher on the conduciveness ofthe environment to adopt scales for administrators who were in the network or product champion conditions, no differences were detected on scale means of attitudes towards the employability of people labelled mentally ill, attitudes regarding employment innovations, organizational support and certainty towards employment innovations, and interest in attending an administrative workshop between the network and product champion treatment groups. This lack of significant findings was disappointing since such variables are highlighted in the literature as salient to the adoption process, and because there were indications in this study that certainty and attitudes towards the employability of people labelled mentally ill were positively related to sending training participants (direct service and both). Such non-significant findings could be a result of lack of power in the experiment, or a problem with fidelity of the treatment, is. perhaps administrators do not engage with network people, who are outside their peer group. Reliability of the measures may also be a limitation, as, generally speaking, standard deviations of several scale items were small, failing to discriminate between individuals. However, a significant interaction involving the variable "interest in attending an administrator workshop" was detected between the independent variables of experience with employment programs and the product champion treatment. Eyeballing the data, apparently, administrators from potential adoption units with low and high experience, in the product champion condition, were the 24 75 most interested; administrators from potential adoption units who had high experience, in the no product champion condition, were the least interested. Evidently, the product champion treatment interacted with experience in such a way as to effect the outcome of interest, perhaps sensitizing these administrators to different thought processes, priorities etc., related to interest in attending. For example, in a few situations when the experimenter was relaying the qualities of a product champion many administrators keyed in on the word "risk taker'“. But, how important is this finding? Although the product champion condition in concert with experience appeared to affect interest, this interest did not translate into actual attendance. This lack of relatedness parallels findings in the literature, where attitudes have been shown to poor indicators of adoption and implementation behaviour (Rogers, 1983; Tomatzky et al, 1980; Fairweather et al, 1974). Again, it is difficult to evaluate the importance of these differences in interest with regard to long term adoption and implementation without longitudinal data. m The hypothesis pertaining to a product champion main effect was not supported. That is, mean scores involving the conduciveness of the environment to adopt scales for network people in the product champion condition were not significantly higher than those not in the product champion condition - in fact, with the exception of the interest in attending an administrative training For example, administrators said such thing as "there are not risk takers in this organization", "risk takers are not rewarded in this environment", "the mental health boards want a guarantee that no risk taking will take place", “the mental health system filters risk takers out". 76 scale, where no difference was found, the opposite effect occurred. Apparently, the product champion condition had a suppressing effect: network people in the product champion condition felt less certain about employment innovations, less sure about the psychiatrically disabled population’s ability to work, less sure about the compatibility etc. of employment innovations, and less organizational support for these innovations, in comparison to those not in the product champion condition. This effect may have occurred because the focus on "risk taking" was threatening, and/or the product champion treatment sensitized them to the reality of resistance, that there was little support, that they were not that certain that employment innovations worked and so on. In light of the correlations in this study depicting a relationship between the outcome of sending behaviour (direct service and both) and certainty towards employment innovations, and attitudes towards the employability of people labelled . mentally ill, the product champion treatment yielded a negative effect. On the surface it would appear that this pattern is also contrary to the entire process of adoption and implementation. However, because this experiment examined only one component of a multifaceted adoption process, it is difficult to evaluate this outcome without longitudinal adoption and implementation data. For example, this scenario may be ideal in that this "sensitizing" may result in awareness building, then some conflict, more awareness building, then some more conflict etC., leading to an even greater commitment to adoption in the long term. Recall that it has been suggested (Tomatzky & Fleischer, 1990; House 8: Singh, 1987; Mintzberg, Raisinghami & Theoret, 1976) that adoption decisions may be non-linear in nature involving movement back and forth, rich in feedback loops, and sensitive to new information. Or, another interpretation of the decision 77 process might be that factors such as certainty may be somewhat high for people at the beginning (i.e., network people not in the product champion condition were "very certain" regarding employment innovations); but, when people are "sensitized," it results in a decrease in certainty that eventually goes back up until it surpasses the initial feelings - as information is made more available over time. This experiment suggests that, although the product champion treatment is a Simple intervention that only requires asking people to select certain qualities, simple interventions of this kind can be powerful - creating perceptual differences between groups of people by an W This has implications for outsiders trying to affect change in bureaucratic organizations. Since the product champion treatment yielded a negative effect with regard to sending behaviour it is recommended that the product champion treatment either be dropped or modified. Because of the documented evidence outlining the importance of a "product champion" in the adoption process, and because such a simple intervention can be powerful from the context of an outside change agent, researchers might experiment with softening or modifying the treatment presentation. For example, researchers or practioners might choose another word for "risk-taker." Note, however, that within the context of long term adoption it may be that change agents want the "negative" effect because of its potential implications for adoption, long term, as discussed. W WIth the exception of attitudes towards employability of persons labelled mentally ill, the hypotheses regarding the main effects of the network and product champion treatments were not supported. That is, mean scale scores involving the conduciveness of the environment to adopt were not significantly higher in the product champion group or the network group; the one 78 exception was scale scores related to attitudes towards the employability of persons labelled mentally ill. In fact, similar to the above, the network treatment produced the opposite results on two of the scales. Direct service training participants sent from potential adoption units in the network condition yielded lower scores pertaining to organizational support and certainty towards employment innovations in comparison to those not in the network condition. Within the context of sending training participants (direct service or both) the network approach gets mixed reviews. On the one hand, it appears to have raised attitude scores involving employability of people labelled mentally ill; on the other hand, it lowered certainty. Recall that in this experiment both variables were correlated with sending at least one direct service training participant. If "sending" is the goal, it is recommended that the network approach be dropped or modified to attend to feelings associated with certainty and instrusiveness. Once again, though, it is difficult to evaluate the effects of this snap shot on long term adoption without longitudinal follow-up data for the reasons mentioned earlier, i.e., the network intervention may initially lower scores on such variables as certainty, but over time their scores may rise until they surpass those people scores who had no intervention. Similar to the product champion treatment, the network treatment was a simple intervention instigated by an outsider which resulted in differential perceptions, and as such has implications for outside change agents. Product Champion Profiles The hypotheses regarding main effects of the network and product champion treatment were not supported. That is, neither product champion scale 79 means from direct service training participants who came from potential adoption units who were in the network condition nor those associated with the product champion condition were significantly higher than those in the no conditions. In fact, no differences at all were found. Training participants in this experiment appeared to be somewhat average or slightly above average regarding product champion qualities, although they were high on the communication scale. This lack of difference could be related to the nature of the mental health system. Recall that. some administrators claimed that there were few risk takers in the system. Also, since potential adoption units with low budgets were significantly less likely to come to the MSU trainings, the range of scores on these measures might have been restricted. This non-significant finding needs to be interpreted with caution because sample size was very small for the analysis, affecting the power. Part of the problem seemed to be with the cosmopolitan scale: 16 of the 58 respondents left this entire section blank. This measure needs to be modified for future research of this nature. Summary In summary, budgets of potential adoption units was a predictor regarding sending at least one direct service training participant to trainings involving employment innovations for people labelled mentally ill. There was some evidence, although not statistically significant, that suggested that the product champion treatment was also a predictor; however the network treatment was not detected as a significant predictor. Correlated with this sending outcome were attitudes towards employability of people labelled mentally ill and certainty towards employment innovations. These results are supportive of the literature. Similarly, the experimental treatments of network and product champion were not significant 80 in predicting the sending of training participants to :2th administrative and direct service participants. Correlations suggested that attitudes regarding employability of the psychiatrically disabled, interest and motivation of training participants, and feeling part of the selection process were positively related to sending both. This pattern is also supportive of the literature. Although not investigated in this study, there was some evidence indicating that a combination of the product champion and network conditions may not entice many organizations to send direct service trainings participants; but, if they do, they also send someone to administratrive trainings. Correlations did not show a relationship between the network treatment and feeling part of the selection process, which was contrary to what the treatment was trying to accomplish. Although there were no treatment effects for administrators with regards to the conduciveness of the environment to adopt scales, a significant interaction between experience with employment programs and the product champion treatment was detected on the variable "interest in attending". However, in the context of sending training participants, this finding may be moot, since interest did not translate into attendance. A product champion effect was found for network people on the conduciveness of the environment to adopt scales. Apparently the product champion condition had a suppressing effect on all the variables related to this outcome, with the exception of "interest in attending an administrative training session" where no differences were found. In the context of sending training participants this would be interpreted as a negative effect, since attitudes concerning the employability of people labelled mentally ill and certainty of 81 employment innovations were correlated with sending at least one direct service training participant, and with sending both. However, this is a snap shot of the adoption process, therefore this finding cannot yet be evaluated In the context of adoption and implementation until follow-up data are collected. In a similar vein, a network effect for training participants suppressed scale scores involving organizational support and certainty towards employment innovations. However, it did positively affect scales scores related to attitudes of employability for persons labelled mentally ill, which was positively correlated with sending at least one direct service training participant. It was suggested that factors related to the adoption process, such as certainty, may actually vary over time as information is provided. No experimental effects were detected concerning the procurement of training participants who resembled product champion characteristics. It was suggested that this may be because of the nature of the mental health system, I.e., it filters out risk takers, or a restriction of range problem, or a power problem. Evidence suggested that the network and product champion treatments were simple, yet powerful with regard to changing perceptions and/or attitudes. This may be an important finding for those "outsiders" trying to induce change In I bureaucratic organizations. Caution is warranted regarding the interpretation of these findings. Because of small sample sizes and unequal n’s the power of the experiment to detect differences may have been hampered. In a related manner, the significance level of the experiment was raised to .10, increasing the probability of committing Type 1 errors. Also, correlations were obtained from samples that were not equally represented. Finally, the hazards of interpreting correlational data apply- 82 It was recommended that: 1) Particularly given the economic times, budget (or other similar indicators) of an organization should be considered when attempting any Intervention to persuade bureaucratic units to send direct service training participants to trainings involving the adoption of new innovations. This consideration might mean that a stronger treatment, and/or a variation of treatments according to budget may be needed. Alternatively, researchers and/or practitioners could concentrate their efforts and resources on those bureaucratic organizations that can afford to go, relying on the diffusion effect. 2) There is some indication that the interaction of product champion and network treatments may not entice many organizations to send direct service training participants, but If they do they also send someone to a management training. Although not Investigated in this experiment, this tentative finding may be worth exploring in future dissemination research. 3) Greater understanding of the dynamics of "feeling part of" the adoption process would be useful In advancing dissemination theory and related interventions. 4) A bureaucratic organization’s experience with the targeted innovation may need to be taken Into account as experience may interact with other variables. 5) Within the context of "sending" behavior, the network and product champion treatments should be dropped or modified so that they are more effective in inducing certainty, feelings of being part of the adoption process, instrusiveness and so on. 6) Evidence suggests that even though the network and product champion treatments were simple and were directed by an outside change agent, they can have an impact on attitudes and/or perceptions. This finding has Implications for "outsiders" who are attempting to induce organizational change. 7) Early findings In the adoption process are difficult to evaluate without longitudinal data. Despite the arduous task of this type of research, the utility of longitudinal data are formidable. 83 REFERENCES Anthony, W. A. & Blanch, A. (1987). Supported employment for persons who are psychiatrically disabled: An historical and conceptual perspective. WWW 11(2): 5-22. Anthony, W. A., Cohen, M.R. & Vitalo, R. (1978). The measurement of rehabilitation outcome. sultan, 4.- 365-383. Anthony, W. A., Buell, G., Sharratt, S., & Althoff, M. (1972). The efficacy of psychiatric rehabilitation. WM 78: 447-455. Avison, W. R. & Nixon-Speechley, K. (1987). The discharged psychiatric patient: A review of social, social-psychological and psychiatric correlates of outcome. WWJW 144: 10-18. Backer, T.E., Liberman, R. P. & Knehnel, T.G. (1986). Dissemination and adoption of innovative psychosocial interventions. lanai J Canary all MW 54.- 111-118. Becker, M.H. (1970). Factors affecting diffusion of innovations among health professionals. m mqmm 60.- 294-305. Benvignati, A. M. ( 1982) . Interfirm adoption of capitol-goods innovation. Wafmrfamlm 64: 330-335. Barman, P. & McLaughlin, M. w. (1978). WWW WW W “may W(Final Report, R-15898-H.E.W.) Santa Monica, CA: Rand Corporation. Berelson, B., & Freedman, R. (1964). A study in fertility control. W aura-rear, 210: 29-37 . Beyer, J.M. & Trice, H.M. (1982). The utilization process: A conceptual framework and synthesis of empirical findings. WWMZR 591- 622. Bilvosky, D. & Matson, J. ( 1977) . Community colleges and the developmentally disabled. MC. Washington, DC Co O 34 Black, B. J. (1986). WafiafiaqyarlmfidmfirfhmauQ/fl New York, NY: Altro Health 5 Rehabilitation Services. Black, B. J. (1970). mqmwpmma New York, NY: Grune & Stratton. Blakely, C., Mayer, J., Gottschalk, R., Roitman, D., Schmitt N., Davidson, 11., a Emshoff, J. (1984). 5am mwwgwm East Lansing, MI: Final Report, Grant No. ISI-7920576-01, NSF. Bock, 12.0. (1975). WWWHWW New York, NY: McGraw-Hill Book Company. Bond, G. (1987). Supported work as a modification of the transitional employment model for clients with psychiatric disabilities. W W Janna; 11(2) : 55-71 . Bond, G. & Boyer, S.L. (1988). Rehabilitation programs and outcomes. In Ciardiello, J.A., Bell, M.D. (Eds). WWfirmMWWW Baltimore, MD: lJohn Hopkins Press, pp. 231-263. Burns, T. & Stalker, G. M. (1961). WWJW London: Tavistock Publications. Chakrabarti, A. K. (1974). The role of product champion in product innovation. WWW17:58-62 . Chappanis, A. (1971). Prelude to 2001: Exploration in human communication. WW”: 949-961. Cnaan, R. A., Blankertz, L., Messinger, K.W., Gardner, J.R. (1988). Psychosocial rehabilitation: Toward a definition. WMWWMMCHM): 2-15. Coleman, J. S., Katz, E., 5 Menzel, H. (1966). WWW aafiium NY: Bobbs-Merill. Collignon, F.C., Noble, Jr., J.N. & Toms-Barker, L. (1987). Early lessons from the Marion county demonstration in integrating vocational and mental health services. WWJWC Cyert, 12.14. a March, J. c. (1953). awmqwfim Englewood Cliffs, NJ: Prentice Hall. 85 Dillman, D. A., (1978). mummwmrmm NY: John Wiley 5 Sons. Downs, G. & Mohr, L. (1976). Conceptual issues in the study of innovations. WWMZIflOO-7l4. Fairweather, G. W., Fergus, E. 0., Bryant, B.D., Horthrop, J., Gonzalez, M., (1988). magenta; 1M1: m a s m feat My. East Lansing, MI: Michigan Lodge Dissemination Project. Fairweather, G. W. (ed.) (1980) . Mwiimtmfornmrtafficaltfiseww: WWW. ‘25me 7. San Francisco: Jossey-Bass. Fairweather, G. W. & Tornatzky, L. G. (1977). WWW [01'ch Mm NY: Pergamon Press. Fairweather, G. W., Sanders, D. H. a Tornatzky, L. G. (1974). CW aw in alarm! in“ We. NY: Pergamon Press. Fairweather, c. w. (1972). Mange: wmwanmt Morristown, N.J.: General Learning Press. Fairweather, G. W., Sanders, D. R., Maynard, R., & Cressler, D. L. (1969). Wflfifwffiemaulgfll Chicago: Aldine. Fergus, E. 0., Bryant, 8. D. a Balzel, A. (1989). melanoma!) in W a Mud-lg #14 Paper presentation, American Psychiatric Association, NY. Fernandez-Sandin, M.I. (1986). ‘Ifie fimadfacwfna MWTWW‘TWW East Lansing, MI: Unpublished Dissertation, MSU. Festinger, L. (1954). A theory of social comparison processes. 111mm, 7: 117-140. Fleigel, F. C. et al, (1967).:brrrowflarfufnfla: mama fu 108 Min “my". Hyderabad, India: National Institute of Community Development, Research Report 9. Gailbraith, J.R. (1973). Wrmqufmamflam Reading, MA: Addison-Wesley. 86 Gerwin, D. (1988). A theory of innovation processes for computer-aided manufacturing technology. m WMWW M3512): 90-100. Glaser, E., Abelson, H. 8 Garrison, K. (1983). MW farm. San Francisco: Jossey-Bass. Glaser, E. M. & Backer, T. E. (1977). Case studies of sustained vs nonsustained change. .fanHMn 6(June): 1-46. Haige, J. a Aiken, M. (1970). WMMMW NY: Random House. Havelock, R. G. (1975;69). wpmmmu manual q M Ann Arbor: Institute for Social Research. Hill, G. W. (1982). WWWMMMTDW hmanflht East Lansing, MI: Unpublished Doctoral Dissertation, MSU. House, R.J. & Singh, J.V. (1987). Organizational behavior: some new directions for I/O psychologists. In M.R. Rosenweig & L.W. Porter (Eds). WW 6% W33). Paulo Alto, CA, Annual Reviews. Kearns, K. P. (1989). Communication networks among municipal administrators: Sharing information about computers in local government. W Croatia: mm mm 10(4): 250-279. Kelly, M. R. 8 Brooks, H. (1988). Gianni! Whammy Cambridge, MA: Center for Business & Government, JFK School of Government, Harvard University. La Piere, R. T. (1966). MW NY: McGraw-Hill. Lazarsfeld, P.F. & Merton, R. K. (1964). Friendship as a social process: A substantive and methodological analysis. In M. Berger 5 others (eds.), frufiuarlmuffnrmlannfdy. NY: Octagon. Lee, 8. a. (1977). WWQ‘MMWMWW Ann Arbor: unpublished Doctoral Dissertation, University of Michigan. 87 Louis, M.R. (1980). Surprise and sense making: what newcomers experience in entering unfamiliar organizational settings. Wm M 25: 256-251 . Loy, J.W. (1969). Social psychological characteristics of innovators . WWW 34: 73-82 . Malamud, T.J. (1986). Community adjustment: Evaluation of the clubhouse model for psychiatric rehabilitation. W 11196 9: 1-4 . Mansfield, E. (1977). WMWMMMM papersforacoflbqrdwnontfierehtimufirpsfietwearxdmand economrzgrowtliprorfuctim'ty. Washington, D. C.: NSF.. Mansfield, E., (1968). wmuwm An and: NY: W.W. Norton, for the Cowles Foundation for Research and Economics at Yale University. Merton, R. K. (1957). Patterns of influence: Local and cosmopolitan enf luent1als . WMGIIWW NY: Free Press of Glencoe. MSU (1989) . WWW Funded by Department of Education: RSA Regional Training Project. Mintzberg, H.D., Raisinghani, D., and Theoret, A. (1976) . The structure of "unstructured" decision processes. Wm my, 21: 245-275. Noble, Jr., J.R. & Collingnon, F. (1987). Systems barrier to supported employment for persons with chronic mental illness. WWW 11(2): 25-44. Norwood, L a Mason, M. (1982). cm WWW“ m: m angels, flaw all ma Texas Dept. of Mental Health & Mental Retardation. Parkinson, S. T. & Avlonitis, G. J. (1986). Economic, technical and management influences on the adoption of flexible manufacturing systems: decision processing model. WWW 5, 243-251. Pincus, J. (1974). Incentives for innovation and the public schools. W q‘WW 44: 113-144 . 88 Placek, P.J. (1975). Welfare workers as family planning change agents and the perrential problem of heterophily with welfare clients. WW m2: 298-316. Repetto, R. (1969). India: A case study of the madras vasectomy program. 5W in M Thing, 31: 8-16 . Rees, C. (1989) . Manny. Coalition for Improved Care of the Mentally Ill. Roethlisberger, F. J. & Dickson, w.J. (1964). Wall“: “fir. Cambridge, MA: Harvard University Press. Rogers, E. M. (1983). omqmm NY: Free Press. Rogers, E. M. & Kincaid, D. L. (1981). WWW: W amfigWfirm NY: Free Press. Rogers, E. M. & Shoemaker, F.F (1971). Wafm 5! WWW NY: Free Press. Rogers, E. M. & Svenning, L. (1969). WWW '15: W 9" mm NY: Holt, Rinehart & Winston. Rogers, 13.14. (1962). Wofm NY: Free Press of Glencoe. Ryan, B. & Gross, N.C (1943). The diffusion of hybrid seed corn in two Iowa communities. MW 8:15-24. Schacter, S. (1959). WWVW Stanford, CA: Stanford University Press. Schalock, R.L. (1983). Staff Development. In Wfir WM Baltimore, MD: University Park Press. Stein, L.I. & Test, M. A. (1979). From the hospital to the community: A shift in the primary locus of care. In Lamb, R. (ed.). MWfirWW m,m1. San Francisco: Jossey-Bass Inc.. Stroul, B. A. (1986). Mold: WWW:WUW mwmmmnfiz‘. NIMH. Szilvazgi, S. (1991). Personal communication. 89 Tashjian, M.D., Hayward, B.J., Stoddard, S., Kraus, L. (1989). mmmqwmmw {Em RSA, US Department of Education: Policy Studies Assoc. Inc., Washington, D.C, April. Test, M.A. & Stein, L. I., (1978). Community treatment of the chronic patient: Research overview. mun: 350-364. Tornatzky, L.VG. & Fleischer, M. (1990). ‘5:ij mm Chicago: Lexington Books. Tornatzky, L.G. & Klein, K.J. (1982). Innovation characteristics and innovation adoption- implementation: A meta analysis of findings. WWW was: 28-45. Tornatzky, L.G., Fergus, E.O., Avellar, J.W., Fairweather, G.W. & Fleischer, M. (1980). mumblpm: AWWmWWW NY: Pergamon Press. Tushman, M. L. and Nadler, D.A. (1978). Organizing for innovation. WWWZJQ): 74-92. Weber, M. (1958). (translated by Parsons, T.) WWW: “15¢ij NY: Scribner's. Wilson, J. Q. (1966). Innovation in organizations: Notes toward a theory. In J.D. Thompson (ed.), wWIayn. Pittsburgh: University of Pittsburgh Press. Yin, R.K. (1977) . Product effectiveness vs bureaucratic self- interest: two innovative processes? WW, 8: 381-399. Zaltman, 6., Duncan, R., & Holbek, J. (1973). Wm! M NY: John Wiley & Sons. 9O BACKGROUND CHECK LIST APPENDIX A 91 Background Information THE FOLLOWING IS A SHORT INVENTORY CONCERNING YOURSELF. PLEASE FILL IN THE BLANKS OR PUT A CHECK BESIDE THE APPROPRIATE ANSWER. 1. Name, Work.Place, and Associated Community'Menta1.Health Center: 2. Title and Job Description: 3. Date of birth: 4. Gender: 1. 2. 5. Education: 1. 2. 3. ‘0 5. FEMALE MALE HIGH SCHOOL OR LESS SOME COLLEGE/UNIVERSITY COLLEGE/UNIVERSITY UNDERGRADUATE DEGREE SOME GRADUATE COLLEGE/UNIVERSITY GRADUATE COLLEGE/UNIVERSITY DEGREE 6. Ethnic Origin: 1. 2. 3. 4. s. ASIAN/PACIFIC ISLANDER BLACK/AFRICAN AMERICAN HISPANIC NATIVE AMERICAN 'HITE/CAUCASIAN 7. Length of time in your current job position? (MONTHS OR YEARS) 8. Length of time in the mental health system? (MONTHS OR YEARS) THANK YOU VERY MUCH FOR YOUR PARTICIPATION. YOUR PERSISTENCE THROUGH OUT THIS QUESTIONNAIRE IS MUCH APPRECIATED. 92 MANIPULATION CHECK: MEASURES APPENDIX B 93 Administrator and Network Questionnaire THIS QUESTIONNAIRE.ASKS ABOUT YOUR INVOLVEMENT CONCERNING THE MICHIGAN STATE UNIVERSITY (MSU) LONG TERM TRAINING CERTIFICATE PROGRAM FOR DIRECT SERVICE STAFF. PLEASE PUT A CHECK BESIDE THE MOST APPROPRIATE ANSWER. 1. Did you or your organization in any way attempt to select someone to go to the training sessions? 1. NO (PLEASE GO TO QUESTION 33) 2. YES 3. DO NOT KNOW (PLEASE PROCEED TO QUESTION 33) The following questions concern the selection process. 2. Did you participate in any way in the selection process of direct service training participants? 1. NO (PLEASE PROCEED TO QUESTION 33) 2. YES If yes....a) did you talk with others in your organization regarding the decison to select someone? 1. NO 2. YES b) did you tilize the recommendations concerning the kinds of people who would be appropriate for the training (as suggested by MSU)? 1. NOT AT ALL 2. VERY LITTLE 3. SOMEWHAT 4. VERY MUCH 5. USED ENTIRELY 3. How'much did you feel a part of the decision-making process regarding the selection of someone to go? 1. DID NOT FEEL AT ALL A PART OF 2. FELT VERY LITTLE 3. FELT SOMEWHAT 4. FELT VERY MUCH 5. COULD NOT FEEL MORE A PART OF 94 4. How satisfied were you concerning your felt participation in the decision-making process? 1. VERY UNSATISFIED 2. UNSATISFIED 3. SATISFIED 4. VERY SATISFIED 5. When considering to select someone, 5A) did you (or co-workers) seek out sources of information regarding new employment programs and practices? 1. NO 2. YES 3. DON'T KNOW 5B) did you (or co-workers) talk with other CMHC about the MSU training program. 1. NO 2. YES 3. DON'T KNOW 7. What was the outcome of the selection process? 1. DECIDED TO NOT SELECT SOMEONE 2. DECIDED TO SELECT SOMEONE 3. DON'T KNOW 8. How satisfied were you with the outcome of the selection process? 1. VERY UNSATISFIED 2. UNSATISFIED 3. SATISFIED 4. VERY SATISFIED 9. Why did you/the organization decide to select/not select someone to go? (PLEASE LIST THE REASON( S) FOR THE DECISION) 10. How interested are you in attending a training program for administrators regarding new employment programs and practices for persons with mental illness? 1. NOT INTERESTED 2. NOT VERY INTERESTED 3. SOMEWHAT INTERESTED 4 . VERY INTERESTED 5. PLANNING TO ATTEND 6. ALREADY ATTENDED 96 MSU RESEARCHERS MANIPULATION CHECK 1 . O R G A N I Z A T I O 2. CODE AND TREATMENT CONDITION: Budget Work score Area (Detroit/non Detroit) Condition 3.ASSOCIATED CMH: 4. ACTUAL TREATMENT STRATEGY: A) TELEPHONED ADMINISTRATOR AND ASKED THEM TO ATTEMPT TO SELECT SOMEONE 1) YES ii) NO RESULT: i) AGREED ii) DID NOT AGREE COMMENTS: B) TELEPHONED ADMINISTRATOR AND ASKED THEM TO ATTEMPT TO SELECT SOMEONE BASED ON SELECTION CRITERIA i) YES ii) NO RESULT: i) AGREED ii) DID NOT AGREE COMMENTS: 97 C) TELEPHONED ADMINISTRATOR AND ASKED THEM TO GIVE TWO NAMES ii) iii) iv) COMMENTS: YES ii) NO AGREED TO GIVE NAMES AGREED TO TALK WITH OTHERS DID NOT AGREE TO TALK WITH OTHERS DID NOT AGREE TO GIVE NAMES TELEPHONED NAMES 1) RESULT: 1) YES ii) NO AGREED TO GIVE TWO MORE NAMES ii) iii) iv) COMMENTS: AGREED TO TALK WITH OTHERS DID NOT AGREE TO TALK WITH OTHERS DID NOT AGREE TO GIVE NAMES TELEPHONED NEXT NAMES 1) RESULT: YES ii) NO 98 i) AGREED TO TALK WITH OTHERS ii) DID NOT AGREE TO TALK WITH OTHERS COMMENTS: D) TELEPHONED ADMINISTRATOR AND ASKED THEM TO GIVE TWO NAMES WITH SELECTION CRITERIA 1) YES ii) NO RESULT: i) AGREED TO GIVE NAMES ii) AGREED TO TALK WITH OTHERS iii) DID NOT AGREE TO TALK WITH OTHERS iv) DID NOT AGREE TO GIVE NAMES COMMENTS: TELEPHONED NAMES 1) YES ii) NO RESULT: i) AGREED TO GIVE TWO MORE NAMES ii) AGREED TO TALK WITH OTHERS iii) DID NOT AGREE TO TALK WITH OTHERS iv) DID NOT AGREE TO GIVE NAMES COMMENTS: TELEPHONED NEXT NAMES 1) YES ii) NO RESULT: i) AGREED TO TALK WITH OTHERS ii) DID NOT AGREE TO TALK WITH OTHERS COMMENTS: 5. INFORMATION SENT OUT As PART OF TREATMENT A) ADMINISTRATORS: i) YES ii) NO B) FIRST TWO NAMES 1) YES ii) NO C) NEXT TWO NAMES 1) YES ii) NO COMMENTS: 100 6. CONTACT WITH MSU (PROACTIVE, IE., THEY SOUGHT US OUT) A) ADMINISTRATORS: 1) YES KIND ADM/STAFF AMOUNT MSU RESPONSE ii) NO B) FIRST TWO NAMES i) YES KIND ADM/STAFF AMOUNT MSU RESPONSE ii) NO C) NEXT TWO NAMES i) YES KIND ADM/STAFF AMOUNT MSU RESPONSE ii) NO COMMENTS: 101 7. RESULTS AND FOLLOW-UP A) SELECTED PERSON(S) 1. NO 2. YES B) SENT MEASURES TO ADMINISTRATOR 1. NO 2. YES RECEIVED: C) SENT MEASURES TO TRAINING PARTICIPANT(S) 1w NO 2. YES RECEIVED: D) SENT MEASURES TO NETWORK STAFF 1. NO 2. YES RECEIVED: 102 EMPLOYMENT INNOVATIONS: PERSONAL PERCEPTIONS APPENDIX C 103 To what.degree do you feel that employment innovations such as the Clubhouse, the Lodge, transitional and supported employment have the following characteristics? (CHECK ANSWER) 1. Credibility, soundness of evidence for its value or espousal by highly respected persons or institutions. 1. 2. 3. 4. 5. NOT CREDIBLE NOT VERY CREDIBLE SOMEWHAT CREDIBLE VERY CREDIBLE COULD NOT BE MORE CREDIBLE 2. Observabiiity, the Opportunity for you to see a demonstration of the innovation or its results in operation. 3 . Relevance I 1. 2. 3. ‘. 5. 1. 2. 3. 4. 5. NOT OBSERVABLE NOT VERY OBSERVABLE SOMEWHAT OBSERVABLE VERY OBSERVABLE COULD NOT BE MORE OBSERVABLE to your goals/problems in rehabilitation. NOT RELEVANT NOT VERY RELEVANT SOMEWHAT RELEVANT VERY RELEVANT COULD NOT BE MORE RELEVANT 4. Ease in understanding and instaiiatiog, as contrasted with difficulty in putting it into operation, or transplanting it to different settings. 1. 2. 3. 4. 5. NOT EASY TO UNDERSTAND NOT VERY EASY TO UNDERSTAND SOMEWHAT EASY TO UNDERSTAND VERY EASY TO UNDERSTAND COULD NOT BE MORE EASY TO UNDERSTAND 5. Compatibility, with your values, norms, procedures, facilities 1. 2. 3. ‘. 5. concerning rehabilitation. NOT COMPATIBLE NOT VERY COMPATIBLE SOMEWHAT COMPATIBLE VERY COMPATIBLE COULD NOT BE MORE COMPATIBLE 104 6. Iniainniligy on Revgzsibi iigy, which permits a pilot tryout of employment innovations one step at a time, and does not call for an irreversible commitment by your organization. 1. 2. 3. ‘. 5. NOT TRIALABLE NOT VERY TRIALABLE SOMEWHAT TRIALABLE VERY TRIALABLE COULD NOT BE MORE TRIALABLE 7 . Bgiative advantage, over existing rehabilitation practices. 1. 2. 3. 4. 5. NO RELATIVE ADVANTAGE NOT MUCH RELATIVE ADVANTAGE SOME RELATIVE ADVANTAGE VERY MUCH RELATIVE ADVANTAGE COULD NOT HAVE MORE RELATIVE ADVANTAGE 105 UNCERTAINTY OF EMPLOYMENT INNOVATIONS APPENDIX D 106 PLEASE CONSIDER THE FOLLOWING STATEMENTS ABOUT EMPLOYMENT INNOVATIONS. PUT A CHECK BESIDE THE ANSWER THE BEST EXPRESSES HOW MUCH YOU AGREE OR DISAGREE WITH THE STATEMENT. 1. I feel certain that.my organization is capable of mustering up the needed resources to adopt employment innovations. Iam I am 1. 2. 3. 4. 5. STRONGLY DISAGREE DISAGREE NEITHER AGREE NOR DISAGREE AGREE STRONGLY AGREE certain I can learn how employment innovations work. 1. 2. 3. 4. 5. STRONGLY DISAGREE DISAGREE NEITHER AGREE NOR DISAGREE AGREE STRONGLY AGREE certain I can learn how to implement employment innovations within my organization. 1. 2. 3. 4. 5. STRONGLY DISAGREE DISAGREE NEITHER AGREE NOR DISAGREE AGREE STRONGLY AGREE I feel certain that my organization can implement employment innovations within its rehabilitation practices. STRONGLY DISAGREE DISAGREE NEITHER AGREE NOR DISAGREE AGREE STRONGLY AGREE I feel certain that employment innovations can benefit persons with mental illness. 1. 2. 3. 4. 5. STRONGLY DISAGREE DISAGREE NEITHER AGREE NOR DISAGREE AGREE STRONGLY AGREE 107 6. I feel certain the including employment innovations in our rehabilitation practices would result in significantly greater competitive employment outcomes for persons with mental illness. 1. STRONGLY DISAGREE 2. DISAGREE 3. NEITHER AGREE NOR DISAGREE 4. AGREE 5. STRONGLY AGREE 10S ATTITUDES TOWARDS THE EMPLOYABILITY OF PERSONS LABELLED MENTALLY ILL APPENDIX E 109 THE STATEMENTS BELOW PRESENT SOME OPINIONS ABOUT EMPLOYMENT FOR PERSONS WITH MENTAL ILLNESS. PLEASE PUT A CHECK BESIDE THE BLANK THAT BEST EXPRESSES YOUR OPINIONS. 1 . Persons 1. 2. 3. 4. 5. with mental illness can contribute to society. STRONGLY DISAGREE DISAGREE NEITHER AGREE NOR DISAGREE AGREE STRONGLY AGREE Competitive employment (work in the private sector) is a viable goal for persons with mental illness. Persons Persons 1. 2. 3. 4. 5. STRONGLY DISAGREE DISAGREE NEITHER AGREE NOR DISAGREE AGREE STRONGLY AGREE with mental illness are usually less intelligent "normal" people. 1. 2. 3. 4. 5. 1. 2. 3. 4. 5. 1. 2. 3. 4. 5. STRONGLY DISAGREE DISAGREE NEITHER AGREE NOR DISAGREE AGREE STRONGLY AGREE with mental illness cannot be trusted with a STRONGLY DISAGREE DISAGREE NEITHER AGREE NOR DISAGREE AGREE STRONGLY AGREE with mental illness are not motivated to work. STRONGLY DISAGREE DISAGREE NEITHER AGREE NOR DISAGREE AGREE STRONGLY AGREE 10. 110 Persons with mental illness cannot work in the competitive job market because they will show symptoms of their illness. 1. STRONGLY DISAGREE 2. DISAGREE 3. NEITHER AGREE NOR DISAGREE 4. AGREE 5. STRONGLY AGREE Given adequate support, persons with mental illness can work in a competitive job. 1. STRONGLY DISAGREE 2. DISAGREE 3. NEITHER AGREE NOR DISAGREE 4. AGREE S. STRONGLY AGREE A major reason why persons with mental illness have difficulty in the competitive job market is because of society's intolerance for deviant behaviour. 1. STRONGLY DISAGREE 2. DISAGREE 3. NEITHER AGREE NOR DISAGREE 4. AGREE 5. STRONGLY AGREE Persons with mental illness have vocational goals. 1. STRONGLY DISAGREE 2. DISAGREE 3. NEITHER AGREE NOR DISAGREE 4. AGREE 5. STRONGLY AGREE Persons with mental illness are capable of furthering their education to obtain their vocational goals. 1. STRONGLY DISAGREE 2. DISAGREE 3. NEITHER AGREE NOR DISAGREE 4. AGREE 5. STRONGLY AGREE 11. 12. 13. 111 Persons with mental illness are capable of running a business. 1. STRONGLY DISAGREE 2. DISAGREE 3. NEITHER AGREE NOR DISAGREE 4. AGREE 5. STRONGLY AGREE Employment is an important component of rehabilitation for persons with mental illness. 1. STRONGLY DISAGREE 2. DISAGREE 3. NEITHER AGREE NOR DISAGREE 4. AGREE 5. STRONGLY AGREE Like "normal" persons, persons with mental illness have the need to feel productive. 1. STRONGLY DISAGREE 2. DISAGREE 3. NEITHER AGREE NOR DISAGREE 4. AGREE 5. STRONGLY AGREE 112 PROFILE OF PRODUCT CHAMPIONS APPENDIX F 113 THE STATEMENTS BELOW CONCERN WORK ACTIVITIES AND PERCEPTIONS ABOUT YOURSELF. lF YOU HAVE NO OBJECTIONS, PLEASE PUT A CHECK BESIDE THE ANSWER THAT BEST DESCRIBES YOUR ACTIVITIES AND SELF-PERCEPTIONS. THERE ARE NO RIGHT OR WRONG ANSWERS. W 1. I subscribe to professional journals. 2. I read professional journals almost every month. 3. I belong to professional organizations/societies. 4. I read, almost every month, from the following sources of information.in my area of specialization. (CHECK ALL THAT APPLY) 5. I read, 1. 2. 3. 4. 5. 6. 7. JOURNALS BOOKS UNPUBLISHED REPORTS/GOVERNMENT REPORTS PROFESSIONAL MAGAZINES/NEWSPAPERS NEWSLETTERS OTHER (PLEASE SPECIFY) NONE OF THE ABOVE almost every month, from the following sources of information outside of my area of specialization. (CHECK ALL THAT APPLY) 1. 2. 3. 4. 5. 6. 7. JOURNALS BOOKS UNPUBLISHED REPORTS/GOVERNMENT REPORTS PROFESSIONAL MAGAZINES/NEWSPAPERS NEWSLETTERS OTHER (PLEASE SPECIFY) NONE OF THE ABOVE 6. During the year; I gnnygi.tolgain information about mental in the following ways... health 1 . CONSULTING I MEETINGS I CONFERENCES WITH COLLEAGUES 3. 4. CONSULTING/MEETINGS[CONFERENCES WITH REPS OF REGIONAL OR STATEWIDE ORGANIZATIONS CONSULTING/MEETINGS[CONFERENCES WITH REPS OF NATIONAL ORGANIZATIONS CONSULTING/MEETINGS[CONFERENCES WITH REPS OF INTERNATIONAL ORGANIZATIONS OTHER (PLEASE SPECIFY) NONE OF THE ABOVE 114 7. On the average, I leave my local community times a year to gain knowledge about mental health issues. 1. 7 OR MORE TIMES A YEAR 2. 5 OR 6 TIMES A YEAR 3. 3 OR 4 TIMES A YEAR 4. 1 OR 2 TIMES A YEAR 5. NONE Nggwgnks. SUPPOSE YOU .ARE SEEKING .ADVICE, INFORMATION OR.‘WISHC TO PROBLEM-SOLVE REGARDING YOUR PROFESSIONAL PRACTICES PROGRAMS, SERVICES. CONSIDER YOUR NETWORK, OR PERSON(S) THAT YOU WOULD gm TALK TO. PLEASE CHECK THE ANSWER THAT BEST DESCRIBES WHAT YOU WOULD DO. I would talk to a person(s) witnin my organization.... 8. who is similar or near to my position at work. 1. NEVER 2. RARELY 3. SOMETIMES 4. FREQUENTLY 5. ALWAYS 9. but outside of my department or work area. 1. NEVER 2. RARELY 3. SOMETIMES 4. FREQUENTLY 5. ALWAYS 10. that are "higher up" (job position) than me. 1. NEVER 2. RARELY 3. SOMETIMES 4. FREQUENTLY 5. ALWAYS 11. that are "lower" (job position) than me. 1. NEVER 2 . RARELY 3 . SOMETIMES 4. PREQUENTLY s . ALWAYS 115 I would talk to a person(s) in fine nengai neaitn nrena nut who in gunning ny onganigation.... 12. and who is similar or near to my position at work. 1. NEVER 2. RARELY 3. SOMETIMES 4. FREQUENTLY 5. ALWAYS 13. who's department or work arena would not be similar to mine. 1. NEVER 2. RARELY 3. SOMETIMES 4. FREQUENTLY 5. ALWAYS 14. and "higher up" (job position) than me. 1. NEVER 2. RARELY 3. SOMETIMES 4. FREQUENTLY 5. ALWAYS 15. and "lower" (job position) than me. 1. NEVER 2. RARELY 3. SOMETIMES 4. FREQUENTLY 5. ALWAYS I would talk to a person(s) oggside the ngntal benign ngnn. . . . 16. who is similar or near to my position at work. 1. NEVER 2. RARELY 3. SOMETIMES 4. FREQUENTLY 5. ALWAYS 116 17. who's department or work arena would not be similar to mine. 1. NEVER 2. RARELY 3. SOMETIMES 4. FREQUENTLY 5. ALWAYS 18. and "higher up" (job position) than me. 1. NEVER 2. RARELY 3. SOMETIMES 4. FREQUENTLY 5. ALWAYS 19. and "lower" (job position) than me. 1. NEVER 2. RARELY 3. SOMETIMES 4. FREQUENTLY 5. ALWAYS Wigs... 20. When a staff person wants to meet with me, he or she can usually do so... 1. AFTER A WEEK OR SO 2. WITHIN A WEEK 3. WITHIN A DAY OR TWO 4. SOMETIME THAT DAY 5. WITHIN A FEW HOURS 21. In committee meetings, I express my opinions to the group 1. VERY SELDOM 2. ONCE IN A WHILE 3. OCCASIONALLY 4. FAIRLY OFTEN 5. VERY OFTEN 22. I would say that approximately % of my co-workers consider me a skilled communicator. 1. 0-20% 3. 41-60% 4. 61-80* 5. 81-100% 117 23. I would say that approximately % of my co-workers consider me a skilled writer. 1. 0-20% 2. 21-40% 3. 41-60% 4. 61-80% 5. 81-100% 24. I am involved in tasks requiring persuasion of co- workers... 1. ONCE A MONTH OR LESS 2. EVERY FEW WEEKS 3. EVERY WEEK 4. EVERY FEW DAYS 5. ALMOST EVERY DAY F ue ce. PLEASE CHECK THE RESPONSE THAT BEST DESCRIBES YOUR ESTIMATE OF YOUR INFLUENCE IN YOUR WORKPLACE. 25. I have a considerable amount of inggznni influence in my workplace. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 26. I have a considerable amount of ggnnni influence in my workplace. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 27. I am among the first to adopt new ideas which are later accepted in my work place. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 28. 29. 30. 31. 32. 118 I would say that approximately % of my co-workers have contacted me in the last year to discuss mental health issues. 1. 0-20% 2. 21-40% 3. 41-60% 4. 61-80% 5. 81-100% I would say that approximately % of my co-workers have contacted me in the last year for advice, information concerning new rehabilitation programs/procedures. 1. 0-20% 2. 21-40% 3. 41-60* 4. 61-80% 5. 81-100% I would say that approximately % of my co-workers would describe me as credible in my job. 1. 0-20% 2. 21-40% 3. 41-60% 4. 61-80% 5. 81-100% I would say that approximately % of my co-workers would say that I have high professional prestige within my my place of work. 10 0-20* 2. 21-40* 4. 61-80% 5. 81-100% I have been a member of some of my organization's most influential committees. 1. VERY SELDOM 2. ONCE IN A WHILE 3. OCCASIONALLY 4. FAIRLY OFTEN 5. VERY OFTEN 119 Sgifi-ngncention. Please rate how true the following characteristics are of you. I am.... 33. sociable 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 34. imaginative 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 35. dominant 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 36. self-sufficient 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 37. venturesome 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 38. persevering 39. sensitive 40. a risk 1. 2. 3. 4. 5. NOT TRUE NOT VERY TRUE SOMEWHAT TRUE VERY TRUE TRUE NOT TRUE NOT VERY TRUE SOMEWHAT TRUE VERY TRUE TRUE taker 1. 2. 3. 4. 5. NOT TRUE NOT VERY TRUE SOMEWHAT TRUE VERY TRUE TRUE 120 121 INTEREST AND MOTIVATION MEASURE APPENDIX G 122 PLEASE PLACE A CHECK BESIDE THE ANSWER THAT BEST EXPRESSES YOUR INTEREST AND MOTIVATION CONCERNING EMPLOYMENT INNOVATIONS. 1. I am very interested in learning about employment innovations for persons with mental illness. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 2. I am very interested in learning about how to implement employment innovations into current rehabilitation practices within my organization. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 3. I have already sought out general information concerning employment innovations. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE llll 4. I have already sought out information concerning the possibility of implementing employment innovations within my organization. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 5. I am willing to try to persuade my organization to adopt employment innovations within its rehabilitation practices. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 6. 123 I have already attempted (I am currently trying) to persuade my organization to adopt employment innovations. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE I am interested in managing the implementation of employment innovations into my organization's rehabilitation practices. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE I have already attempted (I am currently trying) to manage the implementation of employment innovations into my organization's rehabilitation practices. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 124 ORGANIZATION SUPPORT FOR EMPLOYMENT INNOVATIONS APPENDIX H 125 PLEASE PLACE A CHECK BESIDE THE ANSWER THAT BEST EXPRESSES YOUR OPINIONS OF ORGANIZATIONAL SUPPORT FOR EMPLOYMENT INNOVATIONS. 1. There is a core group of people within my organization that is interested in information regarding employment i innovations 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 2. There is a core group of people within my organization that recognizes that employment innovations can significantly increase work outcomes for persons with mental illness. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 3. There is a core group'ofjpeople‘within‘my organization that is willing to persuade my organization to adopt employment innovations. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 4. There is a core group of people within my organization that is willing to manage the implementation of employment innovations into my organization. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 126 5. There is a core group of people within my organization that feel that my organization can muster up the resources to adopt employment innovations. 1. NOT TRUE 2. NOT VERY TRUE 3. SOMEWHAT TRUE 4. VERY TRUE 5. TRUE 127 APPROACH PROTOCOL: ADMINISTRATORS APPENDIX I 128 we Co dition- Pr no on Ch u es Co 0 Hello WW I am from the MSU Long Term Training Project. About two weeks ago, we mailed you a flyer about upcoming training sessions that cover materials on employment innovations in the field for persons with serious mental illness That is, training sessions that will show how mental health and rehabilitation staff can work together to encourage vocational exploration and employment with people who have serious mental illness. These trainings are offered to both administrators and direct service personnel in the field, although. they‘ will be held at different times. . My reason for calling, is to alert you to our training offerings that have been made possible with federal grant support, and the support of both the Michigan Rehabilitation Services and the Department of Mental Health. Our goal is to ensure that people in the field are exposed to information that is current about how to include work components into current rehabilitation practices for people with serious mental illness. We are also interested in finding out the most effective and useful ways to share this information with the field. Would you be willing to help us by agreeing to participate in an experiment that will evaluate the best ways to initially introduce this curriculum to the field? This will involve completing a questionnaire that will be mailed to you in about three weeks, and answering a couple of questions right now. Participation.is strictly voluntary, and.all.data will be kept in the strictest confidence and anonymous. ILNQ. Thank you for your time. I hope that you will be able to attend our training session and select some interested and motivated staff to go to our direct service personnel training sessions. We will mail you a brochure which gives more details of the training. END 0 V . Iz_x3§y Thank you for your time. I will send you a consent form, with the questionnaire, that we would like you to sign when you complete the questionnaire. We would also like to mail you a brochure today, which gives more details of the training. We are interested in identifying staff who might have direct contact with serving people with mental illness, who might be interested in strengthening the employment aspects of consumer's lives, and motivated to go to these trainings. We would like to send them a brochure as well. Would you give us two names of staff who might be able to help us identify people in your agency, who are interested and.motivated, like 129 this, to come to our training sessions? Please understand that this involvement would not commit you to sending people, just the possibility of sending people. 1. NO (THANK YOU FOR YOUR TIME) END QF INIEBYIEE 2. YES 11_1§§y Thank you. 1. (NAME) 2. (NAME) After we have talked with these staff people we will ask them for two more names. Would you consider talking with all of them about selecting motivated and interested people to attend these training sessions? 1. NO (THANK YOU FOR YOUR TIME) Enn_gg_inzznyizn 2. YES Great. I will back in touch with you in about two weeks to find out what your decision was. You may want to choose a contact person within the group. Is this your correct address and phone number? Thank you again for your time. END OF INIEBYIEE Name of Organization Address Telephone 130 No Network Condition; No Production Champion Qualities CO d . Hello Enegntive Directgr's Name, I am from the MSU Long Term Training Project. About two weeks ago, we maileduyou.a flyer about upcoming training sessions that cover materials on employment innovations in the field for persons with serious mental illness That is, training sessions that will show how'mental health and rehabilitation staff can work together to encourage vocational exploration and employment with people who have serious mental illness. These trainings are offered to both administrators and direct service personnel in the field, although they ‘will be held at different times. My reason for calling, is to alert you to our training offerings that have been made possible with federal grant support, and the support of both the Michigan Rehabilitation Services and the Department of Mental Health. Our goal is to ensure that people in the field are exposed to information that is current about how to include work components into current rehabilitation practices for people with serious mental illness. We are also interested in finding out the most effective and useful ways to share this information with the field. Would you be willing to help us by agreeing to participate in an experiment that will evaluate the best ways to initially introduce this curriculum to the field? This will involve completing a questionnaire that will be mailed to you in about three weeks, and answering' a question right now. Participation is strictly voluntary, and all data will be kept in the strictest confidence and anonymous. I: N . Thank you for your time. I hope that you will be able to attend our training session and select some interested and motivated staff to go to our direct service personnel training sessions. We will mail you a brochure which gives more details of the training. nan OF INTERVIEW. I£_X£§; Thank you for your time. I will send you a consent form with the questionnaire, that we would like you to sign when you complete the questionnaire. We would also like to mail you a brochure today, which gives more details of the training. We are interested in identifying staff who have direct contact with serving people with mental illness, who might be interested in strengthening the employment aspects of consumer's lives, and motivated to go to these trainings. Would you consider trying to select motivated and interested people, like this, to come tO‘thOSO training sessions? .Please 131 understand that this involvement would not commit you to sending people, just the possibility of sending people. 1. NO (THANK YOU FOR YOUR TIME) END OF INTERVIEW 2. YES Iz_x3§y Great. I will back in touch with you in about two weeks to find out what your decision was. Is this your correct address and phone number? Thank you again for your time. END_QE_IHI§EYIEE Name of Organization Address Telephone 132 w Co dit - oductio c an on a ties Co d t Hello Executive Di reggor'g Name. I am from the MSU Long Term Training Project. About two weeks ago, we mailed.you.a:flyer about upcoming training sessions that.cover materials on employment innovations in the field for persons with serious mental illness That is, training sessions that will show how mental health and rehabilitation staff can work together to encourage vocational exploration and employment with people who have serious mental illness. These trainings are offered to both administrators and direct service personnel in the field, although. they’ will be held at different times. My reason for calling, is to alert you to our training offerings that have been made possible with federal grant support, and the support of both the Michigan Rehabilitation Services and the Department of Mental Health. Our goal is to ensure that people in the field are exposed to information that is current about how to include work components into current rehabilitation practices for people with serious mental illness. We are also interested in finding out the most effective and useful ways to share this information with the field. Would you be willing to help us by agreeing to participate in an experiment that will evaluate the best ways to initially introduce this curriculum to the field? This will involve completing a questionnaire that will be mailed to you in about three weeks, and answering a couple of questions right now. Particpation is strictly voluntary, and all data will be kept in the strictest confidence and anonymous. IF_N_Q. Thank you for your time. I hope that you will be able to attend our training session and select some interested and motivated staff to go to our direct service personnel training sessions. We will mail you a brochure which gives more details of the training. END OF INTEEV;EW, IF YES. Thank you for your time. I will send you a consent form with the questionnaire, that we would like you to sign when you complete the questionnaire. We would also like to mail you a brochure today, which gives more details of the training. We are interested in identifying staff who might have direct contact with serving people with mental illness, who might be interested in strengthening the employment aspects of consumer's lives, and motivated to go to these trainings. In particular, we are looking for persons with certain qualities that we think would benefit from these trainings. These qualities include: a risk taker, a good communicator, someone 133 who networks alot with staff in and outside your organization, someone who travels or seeks out information about mental health outside of your organization, and who has influence within your organization. We would like to send them a brochure as well. Would you give us two names of staff who might be able to help us identify interested and motivated people in your agency, with these special qualities, to come to our training sessions? IPlease understand. that this involvement would not commit you to sending people, just the possibility of sending people. 1. I“) (THANK )KHJ FOR )MHHI TIME) lflEL_QE IHIEBYIEE 2. YES m Thank you. 1. (NAME) 2. (NAME) After we have talked with.these staff people, we*will ask them for two more names. Would you consider talking with all of them about selecting motivated and interested people, with these special qualities to attend these training sessions? 1. NO (THANK YOU FOR YOUR TIME) ENQ_Q£_INIEBEIEW 2. YES Great. I will back in touch with you in about two weeks to find out what your decision was. You may want to choose a contact person within the group. Is this your correct address and phone number? Thank you again for your time. m INIEBYIEE Name of Organization Address Telephone 134 No Ne wor Cond tion- Product n C io alities Cond tion. Hello Execngive Qizeggog'g Nnng. I am ' from the MSU Long Term Training Project. About two weeks ago, we mailed.you.a flyer about upcoming training sessions that.cover materials on employment innovations in the field for persons with serious mental illness That is, training sessions that will show'how mental health and rehabilitation staff can work together to encourage vocational exploration and employment with people who have serious mental illness. These trainings are offered to both administrators and direct service personnel in ‘the field, although. they' will be held at different times. My reason for calling, is to alert you to our training offerings that have been made possible with federal grant support, and the support of both the Michigan Rehabilitation Services and the Department of Mental Health. Our goal is to ensure that people in the field are exposed to information that is current about how to include work components into current rehabilitation practices for people with serious mental illness. We are also interested in finding out the most effective and useful ways to share this information with the field. Would you be willing to help us by agreeing to participate in an experiment that will evaluate the best ways to initially introduce this curriculum to the field? This will involve completing a questionnaire that will be mailed to you in about three weeks, and answering' a question right now. Participation is strictly voluntary, and.all data will be kept in the strictest confidence and anonymous. i! N . Thank you for your time. I hope that you will be able to attend our training session and select some interested and motivated staff to go to our direct service personnel training sessions. We will mail you a brochure which gives more details of the training. NNN OF INNNNVINN. IN 1E8. Thank you for your time. I will send you a consent form with the questionnaire, that we would like you to sign when you complete the questionnaire. We would also like to mail you a brochure today, which gives more details of the training. We are interested in identifying staff who might have direct contact with serving people with mental illness, who might be interested in strengthening the employment aspects of consumer's lives, and motivated to go to these trainings. In particular, we are looking for persons with certain qualities that we think would benefit from these trainings. These qualities include: a risk taker, a good communicator, someone 135 who networks alot with staff in and outside your organization, someone who travels or seeks out information about mental health outside of your organization, and who has influence *within.your organizationm'Would you consider trying to select motivated and interested people with these special qualities to attend these training sessions? Please understand that ‘this involvement would not commit you to sending people, just the possibility of sending people. 1. NO (THANK YOU FOR YOUR TIME) END OF INTEBVIEE 2. YES I: NEE. Great. I will back in touch with you in about two weeks to find out what your decision was. Is this your correct address and telephone number? Thank you again for your time. END OF INTERVIEW Name of Organization Address Telephone 136 APPROACH PROTOCOL: NETWORK NAMES APPENDIX J 137 CO 0 ' O C C S EINSI 1N9 NAMES: Hello Network Penson's Name. I am from the MSU Long Term Training Project. About two weeks ago, we mailed your organization a flyer about upcoming training sessions that cover materials on employment innovations in the field for persons with serious mental illness That is, training sessions that will show how mental health and rehabilitation staff can work together to encourage vocational exploration and employment with people who have serious mental illness. These trainings are offered to both administrators and direct service personnel in the field, although they will be held at different times. We approached. your' administrator about sending motivated and interested staff to participate in the training sessions. He/she gave us your name. My reason for calling, is to alert you to our training offerings that have been made possible with federal grant support, and the support of both the Michigan Rehabilitation Services and the Department of Mental Health. Our goal is to ensure that people in the field are exposed to information that is current about how to include work components into current rehabilitation practices for people 'with serious mental illness. We are also interested in finding out the most effective and useful ways to share this information with the field. Would you be willing to help us by agreeing to participate in an experiment that will evaluate the best ways to initially introduce this curriculum to the field? This will involve completing a questionnaire that will be mailed to you in about three weeks, and answering a couple of questions right now. Participation is strictly voluntary, and all data will be kept in ‘the strictest confidence and anonymous. ;£_NQ. Thank you for your time. I hope that you or another staff will be able to attend our training session. We will mail you a brochure which gives more details of the training. EED_QI_IETERYIEEI 1£_XE§; Thank you for your time. I will send you a consent form with the questionnaire, that we would like you to sign when you complete the questionnaire. We would also like to mail you a brochure today, which gives more details of the training. We are interested in identifying staff who might have direct contact with serving people with.mental illness, who might be 138 interested in strengthening the employment aspects of consumer's lives, and motivated to go to these trainings. In particular, we are looking for persons with certain qualities that we think would benefit from these trainings. These qualities include: a risk taker, a good communicator, someone who networks alot with staff in and outside your organization, someone who travels or seeks out information about mental health outside of your oganization, and who has influence witin your organization. We would like to send them a brochure as well. WOuld you consider being involved in the process of selecting someone from your organization who is interested and.motivated with these‘qualities to attend these trainings? Please understand that this invovlement would not commit you to sending people, just the possibility of sending people. 1. NO (THANK YOU FOR YOUR TIME) ENQ_Q£_1NIER¥IEE 2. YES I: 1E8. Great. We will be talking to others in your organization. Would you.be willing to give us two more names of other staff who might be able to help us identify staff in your agency who are interested and motivated to attend these trainings? 1. NO (THANK YOU FOR YOUR TIME) END OF INIENVIEN 2. YES (NAME) (NAME) After we have talked with.these people, we will send them some information as well. Would you consider talking with all of them.and your administrator about the selecting'motivated and interested people to attend these trainings? 1. NO (THANK YOU FOR YOUR TIME) END OE INTERVIEE 2. YES Great. I will back in touch with you or your organization in about two weeks to find out what your decision was. You may want to choose a contact person within the group. Is this your correct address and phone number? Thank you again for your time. ENQ OE iNTENVIEN Name of Organization Address Telephone 139 O CO D ON° O PRO UC C ON UALI ES EIESI 1N9 NAMES: Hello Nggwgng Egzson's NgNe. I am from the MSU Long Term Training Project. About two weeks ago, we mailed your organization a flyer about upcoming training sessions that cover materials on employment innovations in the field for persons with serious mental illness That is, training sessions that will show how mental health and rehabilitation staff can work together to encourage vocational exploration and employment with people who have serious mental illness. These trainings are offered to both administrators and direct service personnel in the field, although they will be held at different times. We approached. your administrator about sending motivated and interested staff to participate in the training sessions. He/she gave us your name. My reason for calling, is to alert you to our training offerings that have been made possible with federal grant support, and the support of both the Michigan Rehabilitation Services and the Department of Mental Health. Our goal is to ensure that people in the field are exposed.to information that is current about how to include work components into current rehabilitation practices for people with serious mental illness. We are also interested in finding out the most effective and useful ways to share this information with the field. Would you be willing to help us by agreeing to participate in an experiment that will evaluate the best ways to initially introduce this curriculum to the field? This will involve completing a questionnaire that will be mailed to you in about three weeks, and answering a couple of questions right now. Participation is strictly voluntary, and all data will be kept in the strictest confidence and anonymous. EE_NQ. Thank you for your time. I hope that you or another staff will be able to attend our training session. We will mail you a brochure which gives more details of the training. E!2_QZ_I!T§REIE!1 iE YES, Thank you for your time. I will send you a consent form with the questionnaire, that we would like you to sign when you complete the questionnaire. We would also like to mail you a brochure today, which gives more details of the training. We are interested in identifying staff who might have direct contact with serving people with mental illness, who might be 140 interested in strengthening the employment aspects of consumer's lives, and motivated to go to these trainings. Would you consider being involved in the process of selecting interested and motivated people from your organization to attend these trainings? Please understand that this invovlement would not commit you to sending people, just the possibility of sending people. 1. NO (THANK YOU FOR YOUR TIME) END OF INIEBVIEW 2. YES ML Great. .We will be talking to others in your organization. Would you be willing to give us two more names of other staff who might be able to help us identify staff in your agency who are interested and motivated to attend these trainings? 1. NO (THANK YOU FOR YOUR TIME) END OF INTERviEN 2. YES (NAME) (NAME) After we have talked with these people, we will send them some information as well. Would you consider talking with all of them and your administrator about the selecting motivated and interested people to attend these trainings? 1. NO (THANK YOU FOR YOUR TIME) END OE lNTERVlEfl 2. YES Great. I will back in touch with you or your organization in about two weeks to find out what your decision was. You may want to choose a contact person within the group. Is this your correct address and phone number? Thank you again for your time. ENE QF iNIERVIEN Name of Organization Address Telephone 141 NETWO CON TION' RODU C O U T S NEE: TWO NNNES: Hello e wo Person's e. I am from the MSU Long Term Training Project. About two weeks ago, we mailed your organization a flyer about upcoming training sessions that cover materials on employment innovations in the field for persons with serious mental illness That is, training sessions that will show how mental health and rehabilitation staff can work together to encourage vocational exploration and employment with people who have serious mental illness. These trainings are offered to both administrators and direct service personnel in the field, although they will be held at different times. We approached your administrator and some staff about sending motivated and interested staff to participate in the training sessions. They gave us your name. My reason for calling, is to alert you to our training offerings that have been made possible with federal grant support, and the support of both the Michigan Rehabilitation Services and the Department of Mental Health. Our goal is to ensure that people in the field are exposed to information that is current about how to include work components into current rehabilitation practices for peOple with serious mental illness. We are also interested in finding out the most effective and useful ways to share this information with the field. Would you be willing to help us by agreeing to participate in an experiment that will evaluate the best ways to initially introduce this curriculum to the field? This will involve completing a questionnaire that will be mailed to you in about three weeks, and answering a couple of questions right now. Participation is strictly voluntary, and all data will be kept in. the strictest confidence and anonymous. E: N . Thank you for your time. I hope that you or another staff will be able to attend our training sessions. We will mail you a brochure which gives more details of the training. EN2_QZ_I!T£R!I§!1 LE_EEE; Thank you for your time. I will send you a consent form with the questionnaire, that we would like you to sign when you complete the questionnaire. We would also like to mail you a brochure today, which gives more details of the training. We are interested in identifying staff who might have direct contact with serving people with.mental illness, who might be 142 interested in strengthening the employment aspects of consumer's lives, and motivated to go to these trainings. In particular, we are looking for persons with certain qualities that we think would benefit from these trainings. These qualities include: a risk taker, a good communicator, someone who networks alot with staff in and outside your organization, someone who travels or seeks out information about mental health outside of your organization, and who has influence within your organization. We would like to send them a brochure as well. Would you consider being involved in the process of identifying people from your organization who are interested and motivated, with these special qualities, to attend these trainings? 1. NO (THANK YOU FOR YOUR TIME) END OF lNTEBVIEW 2. YES 12.138; Great. Would you consider talking with these staff and your administrator about selecting interested and motivated people, with these special qualities, to attend these trainings? 1. NO (THANK YOU FOR YOUR TIME) ENE QE INIEBVIEW 2. YES Great. I will back in touch with you or your organization in about two weeks to find out what your decision was. You may want to choose a contact person within the group. Is this your correct address and phone number? Thank you again for your time. END OF INEERVIEW Name of Organization Address Telephone 143 O O D ION' NO ODUCT C ION U ITI S O N S: Hello Ngtwonk Person's Nnme. I am from the MSU Long Term Training Project. About two weeks ago, we mailed your organization a flyer about upcoming training sessions that cover materials on employment innovations in the field for persons with serious mental illness That is, training sessions that will show how mental health and rehabilitation staff can work together to encourage vocational exploration and employment with people who have serious mental illness. These trainings are offered to both administrators and direct service personnel in the field, although they will be held at different times. We approached your administrator and some staff about sending motivated and interested staff to participate in the training sessions. They gave us your name. My reason for calling, is to alert you to our training offerings that have been made possible with federal grant support, and the support of both the Michigan Rehabilitation Services and the Department of Mental Health. Our goal is to ensure that people in the field are exposed to information that is current about how to include work components into current rehabilitation practices for peOple with serious mental illness. We are also interested in finding out the most effective and useful ways to share this information with the field. Would you be willing to help us by agreeing to participate in an experiment that will evaluate the best ways to initially introduce this curriculum to the field? This will involve completing a questionnaire that will be mailed to you in about three ‘weeks, and answering a question right now. Participation is strictly voluntary, and all data will be kept in the strictest confidence and anonymous. IF NO. Thank you for your time. I hope that you or another staff will be able to attend our training sessions. We will mail you a brochure which gives more details of the training. END OF lNTERVIEW. IF NEE. Thank you for your time. I will send you a consent form with the questionnaire, that we would like you to sign when you complete the questionnaire. We would also like to mail you a brochure today, which gives more details of the training. We are interested in identifying staff who might have direct contact with serving people with mental illness, who might be 144 interested in strengthening the employment aspects of consumer's lives, and motivated to go to these trainings. Would you consider being involved in the process of trying to identify people from.your organization.who are interested and motivated to attend these trainings? Please understand that this invovlement would not commit you to sending people, just the possibility of sending people. 1. NO (Thank you for your time. END OF INTERVIEW) 2. YES IF 1E8. Great. Would you consider talking with these staff and your administrator about selecting interested and motivated interested people attend these trainings? 1. NO (THANK YOU AGAIN FOR YOUR TIME) E OF iNTERVIEfl 2. YES Great. I will back in touch with you or your organization in about two weeks to find out what your decision was. You may want to choose a contact person within the group. Is this your correct address and phone number? Thank you again for your time. END OF INTERVIEW Name of Organization Address Telephone 145 FOLLOW-UP PROTOCOL APPENDIX K 14 6 FOLLOW-UP PROTOCOL : ADMINI STRATORS CONDITION: DID ADMINISTRATOR AGREE PREVIOUSLY TO PARTICIPATE IN STUDY: 1. NO 2. YES Hello Executive Director's Nnme. This is from the MSU Long Term Training Project. Remember that I (or initial contact's name) spoke with you on (initial contact date) over the phone about the possibility of selecting someone to come to our training sessions concerning work components for persons with mental illness. I am calling to find out what your decision was. Did you/your organization (depending on condition) select someone to attend? 1. NO (COMMENTS: (IE NO, ANQ NOT A EMICIEANI) . THANK YOU FOR YOUR TIME. I HOPE YOU WILL CONSIDER ATTENDING OUR TRAINING SESSIONS FOR .ADMINISTRATORS. ENE QF iNTERVIEW IF NO B G E O C P T‘ THANK YOU FOR YOUR TIME. I UNDERSTAND THAT YOU WERE KIND ENOUGH TO AGREE TO PARTICIPATE IN THIS STUDY. I WILL BE MAILING YOU A QUESTIONNAIRE TO COMPLETE, AS WELL AS A CONSENT FORM. PLEASE WATCH FOR IT IN THE MAIL, AND RETURN IT TO MSU AS QUICKLY AS POSSIBLE. YOUR TIME AND EXPEDIENCY IS MUCH APPRECIATED. THANKS AGAIN. END OF INIQVIEE 2. YES (IE XE§, MD N01 A EmIClPMI) . THANK YOU FOR YOUR TIME. I HOPE YOU WILL CONSIDER ATTENDING OUR TRAINING SESSIONS FOR ADMINISTRATORS. ENE 92: NT VEW. (lF XES, AND AGREED IO PmICIPATE) . GREAT. WHO IS THIS PERSON OR PERSONS (IF THEY DO NOT KNOW ASK WHO WOULD KNOW I UNDERSTAND THAT YOU WERE KIND ENOUGH TO AGREE TO PARTICIPATE IN THIS STUDY. I WILL BE MAILING YOU A QUESTIONNAIRE TO COMPLETE, AS WELL AS A CONSENT FORM. PLEASE WATCH FOR IT IN THE MAIL, AND RETURN IT TO MSU AS QUICKLY AS POSSIBLE. YOUR TIME AND EXPEDIENCE IS MUCH APPRECIATED. 147 3. DO NOT KNOW (IF THE DO NOT KNOW D NOT PARTICIPANT YOU FOR YOUR TIME. I HOPE YOU WILL CONSIDER ATTENDING OUR TRAINING SESSIONS FOR ADMINISTRATORS. END OF iNIERVIEN. (IE IflEY Q0 NQT EOW, EUT AQREED TO EARTLCIEATE) . WHO WOULD KNOW I UNDERSTAND THAT YOU WERE KIND ENOUGH TO AGREE TO PARTICIPATE IN THIS STUDY. I WILL BE MAILING YOU A QUESTIONNAIRE TO COMPLETE, AS WELL AS A CONSENT FORM. PLEASE WATCH FOR IT IN THE MAIL, AND RETURN IT TO MSU AS QUICKLY AS POSSIBLE. YOUR TIME AND EXPEDIENCE IS MUCH APPRECIATED. 148 FOLLOW-UP PROTOCOL: CONTACT PERSON CONDITION: DID PERSON AGREE PREVIOUSLY TO PARTICIPATE IN STUDY: 1. NO 2. YES 3. NO PREVIOUS CONTACT Hello Contact's Name. This is from the MSU Long Term Training Project. Remember that I (or initial contact's name) spoke with you on (initial contact date) over the phone about the possibility of selecting someone to come to our training sessions concerning work components for persons with mental illness. OR I understand from speaking with your administrator that you are the person who would know about selecting someone to come to MSU training sessions regarding work components for persons with mental illness. I am calling to find out what your decision was. Did you/your organization (depending on condition) select someone to attend? 1. NO (COMMENTS: (IF NO, AND DID NOT WISH TO PARTICIEATE) . THANK YOU FOR YOUR TIME. END OF INIERVIEW (IF NO, BUT AQREED 20 PARTICIPAIE). THANK YOU FOR YOUR TIME. I UNDERSTAND THAT YOU WERE KIND ENOUGH TO AGREE TO PARTICIPATE IN THIS STUDY. I WILL BE MAILING YOU’A.QUESTIONNAIRE TO COMPLETE, AS WELL AS A CONSENT FORM. PLEASE WATCH FOR IT IN THE MAIL, AND RETURN IT TO MSU AS QUICKLY AS POSSIBLE. YOUR TIME AND EXPEDIENCY IS MUCH APPRECIATED. THANKS AGAIN. END OE INIEEVIEE (IE NO, BQI HAD NO EBEZIOQ§ QONIAQI). THANK YOU WE ARE ALSO INTERESTED IN FINDING OUT THE MOST EFFECTIVE.AND USEFUL WAYS TO SHARE THIS INFORMATION WITH THE FIELD. WOULD YOU BE WILLING TO HELP US BY AGREEING TO PARTICIPATE IN AN EXPERIMENT THAT WILL EVALUATE THE BEST WAYS TO INITIALLY INTRODUCE 149 THIS MSU TRAINING CURRICULUM TO THE FIELD? THIS WILL INVOLVE COMPLETING A QUESTIONNAIRE THAT WILL BE MAILED TO YOU? 1. NO THANK YOU FOR YOUR TIME. END OF INTERVIEW 2. YES GREAT. I WILL MAIL YOU THIS QUESTIONNAIRE IMMEDIATELY, WITH A CONSENT FORM TO FILL OUT. PLEASE WATCH FOR IT IN THE MAIL, AND RETURN IT TO MSU AS QUICKLY AS POSSIBLE. YOUR TIME AND EXPEDIENCE IS MUCH APPRECIATED. END OE INTERVIEW. 2. YES (IE YES, AND DID NOT WANT TQ EmICIEAIE) . GREAT. WHO IS THIS PERSON(S) THANK YOU FOR YOUR TIME. END OF INTERVIEE. ( YES DAG E O T IPA ‘. GREAT. WHO IS THIS PERSON OR PERSONS I UNDERSTAND THAT YOU WERE KIND ENOUGH TO AGREE TO PARTICIPATE IN THIS STUDY. I WILL BE MAILING YOU A QUESTIONNAIRE TO COMPLETE, AS WELL AS A CONSENT FORM. PLEASE WATCH FOR IT IN THE MAIL, AND RETURN IT TO MSU AS QUICKLY AS POSSIBLE. YOUR TIME AND EXPEDIENCE IS MUCH APPRECIATED. (IF YES. AND NO PREVIOU§ CONTACT). GREAT. WHO IS THIS PERSON(S) THANK YOU FOR YOUR TIME. WE ARE ALSO INTERESTED IN FINDING OUT THE MOST EFFECTIVE AND USEFUL WAYS TO SHARE THIS INFORMATION WITH THE FIELD. WOULD YOU BE WILLING TO HELP US BY AGREEING TO PARTICIPATE IN AN EXPERIMENT THAT WILL EVALUATE THE BEST WAYS TO INITIALLY INTRODUCE THIS MSU TRAINING CURRICULUM TO THE FIELD? THIS WILL INVOLVE COMPLETING A QUESTIONNAIRE THAT WILL BE MAILED TO YOU? 1. NO THANK YOU FOR YOUR TIME. END OF INTERVIEW 150 2. YES GREAT. I WILL MAIL YOU THIS QUESTIONNAIRE IMMEDIATELY, WITH A CONSENT FORM TO FILL OUT. PLEASE WATCH FOR IT IN THE MAIL, AND RETURN IT TO MSU AS QUICKLY AS POSSIBLE. YOUR TIME AND EXPEDIENCE IS MUCH APPRECIATED. 151 Follow-up Protocol: Training Participant CONDITION: DID PERSON AGREE PREVIOUSLY TO PARTICIPATE IN STUDY: 1. NO 2. YES 3. NO PREVIOUS CONTACT Hello Training Pagticinang's Nnme. This is from the MSU Long Term Training Project. Remember that I (or initial contact's name) spoke with you on (initial contact date) over the phone about the ' possibility of selecting someone to come to our training sessions concerning work components for persons with mental illness. I understand.that.you are the person selected to come to our MSU training sessions concerning work components for persons with mental illness. Welcome. OR I understand from speaking with your administrator that you are the person selected to come to MSU training sessions regarding work components for persons with mental illness. Welcome. Do you still plan to attend? 1. NO (COMMENTS: (IF NOI AND REASON IS NONE!) WOULD YOU CONSIDER COMING IF WE COULD GET A SCHOLARSHIP TO PAY YOUR WAY? IF NO THANK YOU FOR YOU TIME END F INTERVIEE IF YES THANK YOU FOR YOUR TIME. I WILL GET BACK TO YOU. (IF NO, END DID NQI WISH IO PflTIQIRATE) . THANK YOU FOR YOUR TIME. EN F ERVIEW (IE NQ, EDI AGREED IO RARIICIPAIE). THANK YOU FOR YOUR TIME. I UNDERSTAND THAT YOU WERE KIND ENOUGH TO AGREE TO PARTICIPATE IN THIS STUDY. I WILL BE MAILING YOU A.QUESTIONNAIRE TO COMPLETE, AS WELL.AS A CONSENT FORM. PLEASE WATCH FOR IT IN THE MAIL, AND RETURN IT TO MSU AS QUICKLY AS POSSIBLE. YOUR 152 TIME AND EXPEDIENCY IS MUCH APPRECIATED. THANKS AGAIN. END OF INIERVIEE (IF NOl BUT H_AD NQ PREVIOUS CONTACT) . THANK YOU WE ARE ALSO INTERESTED IN FINDING OUT THE MOST EFFECTIVE AND USEFUL WAYS TO SHARE THIS INFORMATION WITH THE FIELD. WOULD YOU BE WILLING TO HELP US BY AGREEING TO PARTICIPATE IN AN EXPERIMENT THAT WILL EVALUATE THE BEST WAYS TO INITIALLY INTRODUCE THIS MSU TRAINING CURRICULUM TO THE FIELD? THIS WILL INVOLVE COMPLETING A QUESTIONNAIRE THAT WILL BE MAILED TO YOU? 1. NO THANK YOU FOR YOUR TIME. mD OF INIERVIEW 2. YES GREAT. I WILL MAIL YOU THIS QUESTIONNAIRE IMMEDIATELY, WITH A CONSENT FORM TO FILL OUT. PLEASE WATCH FOR IT IN THE MAIL, AND RETURN IT TO MSU AS QUICKLY AS POSSIBLE. YOUR TIME AND EXPEDIENCE IS MUCH APPRECIATED. W 2. YES (IF YES . AND DID NOT WAN: To EARIIQIEAIE). GREAT. THANK YOU FOR YOUR TIME. N OF ( F ES AND ED T CIPATE . GREAT. I UNDERSTAND THAT YOU WERE KIND ENOUGH TO AGREE TO PARTICIPATE IN THIS STUDY. I WILL BE MAILING YOU A QUESTIONNAIRE TO COMPLETE, AS WELL AS A CONSENT FORM. PLEASE WATCH FOR IT IN THE MAIL, AND RETURN IT TO MSU AS QUICKLY AS POSSIBLE. YOUR TIME AND EXPEDIENCE IS MUCH APPRECIATED. (IF YEE, MD NO RREZIODE CONIACI) . GREAT. THANK YOU FOR YOUR TIME. WE ARE ALSO INTERESTED IN FINDING OUT THE MOST EFFECTIVE AND USEFUL WAYS TO SHARE THIS INFORMATION WITH THE FIELD. WOULD YOU BE WILLING TO HELP US BY AGREEING TO PARTICIPATE IN AN EXPERIMENT THAT WILL EVALUATE THE BEST WAYS TO INITIALLY INTRODUCE THIS MSU TRAINING CURRICULUM TO THE FIELD? THIS WILL INVOLVE COMPLETING A QUESTIONNAIRE THAT WILL BE MAILED TO YOU? 1. NO THANK YOU FOR YOUR TIME. END DE TERVIEW 153 2. YES GREAT. I WILL MAIL YOU THIS QUESTIONNAIRE IMMEDIATELY, WITH A CONSENT FORM TO FILL OUT. PLEASE WATCH FOR IT IN THE MAIL, AND RETURN IT TO ~MSU AS QUICKLY AS POSSIBLE. YOUR TIME AND EXPEDIENCE IS MUCH APPRECIATED. 154 CONSENT FORM APPENDIX I 155 CONSENT FORM FOR RESEARCH PARTICIPANTS Purpose of the Project To explore the most effective ways to disseminate information concerning employment innovations for those people that work with persons with serious mental illness. artici ant 'n 's 0 55 I agree to complete a mailed questionnaire, and to consider sending someone to the MSU training. I understand that the information I give you will be kept strictly confidential. All information will be tabulated in aggregate form only. I understand that I may refuse to answer any question and/or withdraw from this project at any time, without penalty. The MSU research team will provide me with a copy of the results of this project upon request. I understand that if I have any questions I can call Dr. Esther Fergus or Brenda Bryant, at MSU, 1 (313) 355-0166. Date Signature of ResearCh Participant 156 APPENDIX M POTENTIAL ADOPTION UNITS: BUDGETS AND WORK SCORES 157 COMMUNITY MENTAL HEALTH BOARDS WORK SCORE BUDGET HIGH WORK SCORE (N = 8) Alger-Marquette 4 5,411,540 (M) Clinton-Eaton-Ingham 1 4 25,463,971 (H) Goodwill Industries (D) 4 1 , 6 9 6 , 7 6 l (L) Genessee County 4 32,873,151 (H) Jackson-Hillsdale 6 10,738,486 (H) Kalamazoo County 6 20,254,493 (H) Monroe County 4 9,485,574 (H) Oakland County 6 18,847,151 (H) MEDIUM WORK SCORE (N = 21) Allegan County 2 4,545,742 (M) Branch County 2 1,149,552 (L) Calhoun County 2 10,308,125 (H) Delta County 2 3,275,583 (M) Detroit Central (D) 2 2,058,253 (L) Detroit East (D) 2 3,464,457 (M) Fairlane (D) 2 1,149,741 (L) Lenawee County 2 5,039,463 (M) Livingston County 2 5,723,653 (M) Macomb County 2 1,349,660 (L) Muskegon County 2 15,048,425 (H) New Center (D) 2 2,309,828 (L) North Central (D) 2 1,680,010 (L) North Central Michigan 2 5,088,156 (M) North East Guidance (D) 2 2,987,177 (L) Renaissance West (D) 2 2,519,463 (L) St. Clair County 2 12,810,846 (H) Suburban West (D) 2 1,270,410 (L) Tuscola County 2 1,362,391 (L) Van Buren County 2 4,608,016 (M) Represents sub-units within Detroit Wayne CMHC 158 LOW WORK SCORE (N = 21) Antrim-Kalkaska Aurora Community Center (D) Barry County Bay-Arenac County Berrien County Cass County Copper County Community Care Services (D) Family and Neighbourhood (D) Grand Traverse Lee County Huron County Lapeer County Luce County Dickinson Iron Eastern Upper Peninsula Gratoit County Ionia County Lake County Mason County Menominee County Midland-Gladwin County Montcalm County Newaygo County Northeast Michigan County Northern Michigan County Ottawa County Saginaw County St. Joseph's County Sanilac County Shiawassee County Schoolcraft Washtenaw County WORK SCORE OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO BUDGET 1,418,177 1,735,479 1,149,552 10,107,996 10,583,705 1,130,175 5,529,969 2,793,210 3,348,779 8,725,794 2,289,760 3,888,933 1,349,660 3,628,362 3,315,942 799,537 3,098,396 494,221 2,749,883 1,938,126 4,965,117 2,050,536 2,537,761 3,647,140 5,612,185 7,952,660 14,375,942 3,610,863 1,230,175 4,364,666 1,188,211 16,701,822 (L) (L) (L) (H) (H) (L) (M) (L) (M) (H) (L) (L) (L) (M) (M) (L) (M) (L) (L) (L) (M) (L) (L) (M) (M) (M) (H) (M) (L) (M) (L) (H) 159 APPENDIX N INTERCORRELATIONS OF CONDUSIVE ENVIRONMENT AND PRODUCT CHAMPION MEASURES INTERCORRELATIONS 160 CONDUSIVE ENVIRONMENT FOR ADOPTION: MEASURES EI CERTAINTY MI ORGSUP EI 1.00 .45** .34** .36** CERTAINTY .45** 1.00 .41** .56** MI .34** .41** 1.00 .26** ** ** ** _.2.§2.SUP 32.2... -55 43 1191.; ** p<.01 PRODUCT CHAMPION PROFILE: MEASURES INTER COSMO NETWORK COMM INFLUE RI SK-T INTER l.00 .28 .24 .24 .47** .33* COSMO .28 1.00 .37* .08 .36* .38* NETWORK .24 .37* 1.00 .08 .20 .05 COMM .24 .08 .08 1.00 .59** .34** INFLUE .47** .35* .20 .59** 1.00 .37** RISK .33* .37* .05 .34** .37** 1.00 * p<.05 ** p<.01 W.