Organizational consultants and others attempting (and likely succeeding) to make money off this trend are in ready evidence. For example, Creativity Training, uses humor and comedy to help employees work together in teams. The River Network promises teamwork training participants will remember and use. Sensory Systems offers training in systemic leadership and teamwork. RealChange Network specializes in teamwork training for French and American businesses. AGTS uses The Wizard of Oz to train organization members in teamwork techniques. Armstrong's office ceilings, floorings, and space dividers assure teamwork and more effective communication.

Although teams are associated with the postmodern organization and business corporations, teams have been a mainstay in health care for many years. Even so, as the health care industry has become more specialized, sophisticated, complex and technical, the numerous disciplines involved in the delivery of health care have become increasingly interdependent (O'Connor, Hallberg, & Myles, 1999; Sullivan, 1998). Collaboration and cooperation among health care providers, particularly in the interdisciplinary team context, is considered essential to health care system effectiveness (Becker-Reems & Garrett, 1998; Juliá & Thompson, 1994a; Manion, Lorimer, & Leander, 1996). For example, a recent article in Good Health Magazine attributes the health of a heart transplant patient to an interdisciplinary team, including surgeons, cardiologists, and social workers ("Teamwork," 1997). Human Services Consultants, a non-profit health care organization, stresses the need for a teamwork approach based on disciplinary diversity in meeting the needs of persons with disabilities. Yet, despite the widespread use of interdisciplinary health care teams, we know little about how team members communicate and even less about social-cogitive processes which influence health care provider interactions.

The present study focuses on health care team members' cognitions and attempts to persuade other team members. More specifically, we examined team members' social-cognitive processes as antecedents to listener-adaptive persuasive message strategies used to influence other providers on a health care team. The context of the study is the hospice interdisciplinary team, which includes health care workers from several disciplines: nursing, social work, medicine, pastoral care and physical therapy. We investigated two aspects of social cognition and their relationship to the production of persuasive message strategies: (1) individual differences in interpersonal construct systems and (2) individual differences in health care providers' definitions of their work role. We also examined the effects of organizational training on persuasive communication. The following section discusses the characteristics of the hospice interdisciplinary team and the nature of the health care provider role.

CHARACTERISTICS OF THE HOSPICE INTERDISCIPLINARY TEAM

Cline (1990) distinguishes between two types of interdisciplinary health care teams. The first is more like a gathering of individual consultants in which providers from several health care disciplines are "often linked only by a shared client or patient. In this case, the concept is structural and emphasizes efforts organized loosely around a client" (Cline, 1990, p. 71). This type of team is consistent with the traditional medical model of patient care (Lafferty, 1998). The second type of interdisciplinary health care team more closely matches that used in hospice. In this case, the team is "a small group whose members coordinate their efforts explicitly. . . . [T]he goal in the second view is clearly a product of explicit social interaction" (Cline, 1990, p. 71). It is this necessity to coordinate their actions through face-to-face interaction which can lead to influence attempts among hospice interdisciplinary team members.

Effective coordination is not always easily achieved. For example, Corrigan and McCracken (1998) note that interdisciplinary communication in the health care setting is often difficult due to role conflict and overload, lack of understanding among health care professionals, and struggles for autonomy. Cline (1990) argues that, "Like other groups, the health-care team encounters issues and problems related to leadership, role delineation and negotiation, goal setting, problem solving, conflict, power, authority, trust, and support" (p. 72). More specific to hospice, Berteotti and Seibold (1994) found that hospice teams' coordination problems were factors in negative assessments of volunteers' work and volunteers' feelings of alienation. In examining the interdisciplinary team meeting as a coordination mechanism, the researchers identified interpersonal conflicts as severely hindering effective team member coordination. Problems with coordination led to the hospice interdisciplinary team fragmenting into divergent groups.

Thus, while the hospice interdisciplinary team is a clear break from the traditional medical model of patient care, the diversity of perspectives and training team members bring to the task make coordination both essential and difficult. Further, hospice staff often do not make distinct divisions among team members' roles (Zimmermann, 1994). That is, responsibilities and expectations are typically blurred, so that what the nurse is "supposed to do" may overlap with what the social worker is "supposed to do." Yolo Hospice, in Davis, CA, depicts the hospice team in its Hospice Circle of Care. This diagram lists each team member's functions as well as emphasizing patient and family care. Here, differences and similarities in tasks assigned to group members are evident. For example, nearly every team member assesses family and patient needs, although some team members concentrate more on physical needs, others on emotional and spiritual needs, and others on practical/financial needs.

In spite of this role blurring, coalitions and status differences do develop among team members. For example, Vachon (1987) found that hospice team members reported the development of cliques and subgroups within the team. And although Werner (1985), a hospice physician, argues that, "In operation, . . . most hospices do not duplicate the medical model" (p. 345), Berteotti and Seibold (1994) concluded that at least in the hospice they studied, team members took more of a traditional approach than a team approach to patient care. Even though the hospice was structured as a team, a supervisory board essentially directed all team decisions. In her study of stress and burnout among hospice staff members, Vachon (1987) found that conflicts, one source of stress, often "involve power struggles among various disciplines dealing with issues such as credit for the program and control over the services offered" (p. 90).

The hospice interdisciplinary team is unique in its centrality to patient care. The interdisciplinary team is the basis of health care delivery in hospice, both philosophically and legally (National Hospice Organization, 1999). Hospice stresses the importance of treating the family, not just the patient (e.g., Hospice Federation of Massachusetts, 1998). In addition, hospice defines health as including physical, emotional, social, and spiritual needs--needs which can only be met through an interdisciplinary health care team (e.g., Woltermann, 1998). However, it is similar to other health care teams, and small work groups, in its processes and problems. Therefore, while hospice interdisciplinary teams are often described in terms of their collaborative, cooperative, and supportive nature (e.g., Beresford, 1993), attempts to influence other team members should not be unusual and should even be expected as they are inherent in group work (e.g., Boster, 1989; Gastil, 1993; Seibold, Meyers, & Sunwolf, 1996).

In summary, the hospice interdisciplinary teams plays a crucial role in caring for terminally patients and their families. While a cooperative spirit is considered essential to team functioning, the development of cliques, displays of power, and influence attempts are also part of the team decision-making process. In addition, hospice interdisciplinary team members experience role ambiguity and conflict in their interactions, yet need to demonstrate flexibility in peer-related activities. The following section discusses an approach to social cognition and influence employed in the present study to examine social cognitive antecedents and individual differences in team members persuasive communication strategies.

SOCIAL COGNITION AND PERSUASIVE COMMUNICATION

Several constructivist researchers have examined communicators' persuasive message strategies (e.g., Applegate & Woods, 1991; Leichty & Applegate, 1991; O'Keefe, Murphy, Meyers, & Babrow, 1989; Wilson, Cruz, & Kang, 1992). The focus for these researchers is the link between individuals' social-cognitive structures and the features of messages produced, such as person-centered and listener-adaptive message qualities (O'Keefe, 1990). This research is characterized by an attention to messages and bases analysis on concrete communicative behaviors (Applegate, 1990). We take a constructivist approach to persuasive communication in this study.

Our interest was in the adaptive quality of persuasive message strategies produced by hospice team members when attempting to influence other members of the team. Eisenberg and Goodall (1997) suggest that in the team-based, postmodern organization, "adapting messages to one's listeners takes precedence over individual eloquence" (p. 183). Assessing listener-adaptiveness of persuasive message strategies involves determining the degree to which a message recognizes and adapts to the target's perspective (Applegate, 1982; Delia et al., 1979). In an earlier study, we found that "hospice interdisciplinary team members produced comforting messages that generally explicitly recognized the perspective of the other" (Zimmermann & Applegate, 1992, p. 254). This finding is supported in the hospice literature. Basile and Stone's (1987) research on hospice team member characteristics revealed that the effective team member is able to distinguish the patient as a unique individual rather than as simply another case, and able to communicate with that patient in a person-centered manner. Given the nature of hospice, with its emphasis on understanding others' perspectives both in its philosophy (e.g., Boling & Lynn, 1998; Siebold, 1992) and training (e.g., Babler, 1997), we can reasonably expect that hospice team members are also able to employ message strategies in influence situations which reflect listener adaptation. This led to the first hypothesis:

H1: Hospice team members will produce persuasive messages which explicitly recognize and adapt to the target person's perspective.

In examining the link between individual difference variables in social cognition and message features in the present study, we were concerned with the influence of cognitive differentiation, training, and role differentiation on team members' production of listener-adapted messages. Constructivist research has established a significant and positive relationship between cognitive differentiation and listener-adapted (e.g., Delia et al., 1979; Applegate, 1982) and person-centered message strategies (see Applegate, 1990, and Burleson & Caplan, 1998, for thorough reviews). Our earlier study of hospice team members' comforting communication found that cognitive differentiation was positively but not significantly associated with sophistication of message strategy (Zimmermann & Applegate, 1992). Given these results, our second hypothesis was:

H2: Cognitive differentiation will be positively associated with higher levels of listener- adaptiveness in message strategies produced by hospice team members when attempting to influence a coworker.

The weak relationship between cognitive differentiation and degree of message strategy sophistication found in our previous analysis merits attention. The finding may have been due to the study's small sample size, or other variables, such as organizational socialization processes or employee training, may influence the relationship between general interpersonal construct system structure and message production (Applegate, Coyle, Seibert, & Church, 1989). For example, in a study of police officers, Applegate et al. (1989) found a negative correlation between length of service and the means of (a) construct system development and (b) person-centeredness of persuasive message strategy. These researchers suggest that organizational or professional socialization factors may influence interpersonal construct system structures which in turn affect message features. However, while police officer experiences tend to emphasize interacting with others from a position-centered perspective (Applegate et al., 1989), hospice training and practice focuses on adapting to others as persons. For example, in a training program discussed by Wilkinson and Wilkinson (1987), participants received instruction in listening skills and understanding the perspective of the dying individual and her or his family. Yet, in our previous study (Zimmermann & Applegate, 1992) we found a negative relationship between amount of training in hospice practices and level of person-centeredness when comforting another hospice team member. Although this appears counter-intuitive, this finding is consistent with some research in clinical psychology which suggests a negative impact of training on therapist effectiveness (see the reviews of Lambert, DeJulio, & Stein, 1978; Truax & Mitchell, 1971). Several clinical researchers argue that desired therapist training effects can only be expected when the training identifies specific dimensions of perceptual and behavioral functioning clearly tied to specific client needs. Further, positive results can occur only when specific client needs and therapist skills are correctly matched (Truax & Carkhuff, 1967; also the reviews of Lambert et al., 1978; Truax & Mitchell, 1971). Thus, the third hypothesis was:

H3: Formal training in hospice principles and skills will be negatively associated with degree of message strategy listener-adaptiveness produced when attempting to influence another team member.

Hospice interdisciplinary team members are not immune from the conflicts experienced by other health care team members and teams in general. In their study of teamwork, Larson and LaFasto (1989) found that while clear roles are essential to effective teamwork, team members must strike a balance between differentiation and integration. That is, team members need to specialize and collaborate at the same time. Distinct boundaries between disciplines, leading to inflexible team member roles, can be problematic for health care teams (Juliá & Thompson, 1994b). In a study of medical students' person-centered communication in the medical interview context, Kline and Ceropski (1984) found that how the students defined their role in the medical interview was positively and significantly associated with messages strategies which acknowledged and elaborated on the patient's perspective. This finding held cross-situationally, when medical students attempted to regulate patient behavior, advised a patient, and conducted a patient admission interview. The results of this study suggest that role differentiation is associated with the ability to produce messages which recognize and adapt to the target's perspective.

The use of such situation-specific differentiation measures also is suggested by the work of several researchers who argue that while the Role Category Questionnaire (RCQ) is useful for assessing cognitive structures about persons, it may be less useful for gaining insight into how individuals organize information concerning other phenomena, such as conversations (Daly, Bell, Glenn, & Lawrence, 1985), relationships (Martin, 1991, 1992), and persuasive situations (Wilson et al., 1992). For example, Daly et al. (1985) found conversational differentiation to be a better predictor of involvement and comprehension in dyadic communication than interpersonal cognitive differentiation. Similarly, Martin's (1992) findings in his study of married couples suggest an association between the male partner's relational cognition differentiation and the couple's interaction style. Wilson et al. (1992) suggest that construction of a "measure of 'situational differentiation' might provide additional insight into how Interpersonal Construct Differentiation affects attribution processes" (p. 363). This work led to the fourth hypothesis:

H4: Team members who have a more differentiated conceptualization of their role on the team will produce more listener-adaptive message strategies when attempting to influence a coworker.

While Kline and Ceropski's (1984) findings suggest that an individual's beliefs about her or his role in a particular context influence message strategies employed, we know little of the content of those beliefs and how it is related to message production. For example, Zimmermann, Vaules, and Opperman (1990) identified five themes, impression management, responding skills, attending to status issues, being persuasive, and other-centered communication, in subjects' descriptions of their role as an interviewee in a selection interview. These researchers found that participants who stressed the interviewee's responding skills in defining their role were rated as more likely to be hired for the job than those who did not define their role in this way. While this context is considerably different from the hospice interdisciplinary team, Zimmermann et al.'s findings do suggest that examining the content of individuals' role definitions is important in understanding interaction phenomena. We therefore posed the following research question:

R1: How do hospice team members define their role on the interdisciplinary team? How is this related to general interpersonal cognitive differentiation? How is this related to the persuasive message strategies produced by team members?

METHODS

Data for the present study were generated in two hospice programs through observation, focus group interviews and a questionnaire. The participant organizations, subjects, data collection procedures, measures and coding systems are described below.

Organizations and Study Participants

Two hospice organizations located in different large Midwestern cities, agreed to participate in the study. Hospice A, classified as a home health agency hospice and the larger of the two, provided inpatient and home care services. When the study began, the hospice employed 90 full-time and 100 part-time staff members and also relied on the aid of 500 volunteers. For the two years prior to data collection, this hospice cared for over 900 patients and their families each year. Hospice B, a community-based home care program, had 29 full-time and 23 part-time paid positions and almost 200 volunteers. The program had been in existence for seven years when the study began, with the number of patients served growing from just under 50 the first year to over 300 the sixth year.

Participants and Data-gathering Procedures

Participants. Seven hospice interdisciplinary teams participated in the study, five at Hospice A and two at Hospice B. The total number of hospice team members was 39 for Hospice A and 14 for Hospice B. Disciplines represented on the teams included nursing, social work, pastoral care, medicine and physical therapy. In addition, medical secretaries, home health aides and volunteers also attended meetings and were included in the sample.

Data collection procedures. The first author conducted separate focus group interviews with four hospice teams, two at each site, six weeks after observations began. The interviews were used to identify key situations for the team members in which persuasive message strategies would be appropriate. The interviewer asked team members: "What are the kinds of situations in which team members need to influence the behaviors of other team members?" and "Describe those situations." The interviews were audiotaped with the permission of the participants and transcribed later for analysis.

The first author distributed a questionnaire packet to all team members 10 weeks after the focus group sessions. Subjects completed the packets on their own time and returned the questionnaires in sealed envelopes to a member of the hospice administrative staff (who was not a team member). A total of 36 team members responded to the questionnaire (response rate=68%), 26 from Hospice A (response rate=67%) and 10 from Hospice B (response rate=71%).

Coding Systems and Measures

The questionnaire included both forced-choice and free-response items which were chosen to examine the relationship between individual social-cognitive differences and the production of persuasive message strategies.

Cognitive differentiation. Cognitive differentiation was measured using Crockett's (1965) Role Category Questionnaire (RCQ) which asks respondents to describe two people, one liked and one disliked. Burleson and Waltman (1988) argue that "[c]onsiderable research experience with the RCQ has indicated that reliable estimates of construct system properties can be obtained from as few as two interpersonal impressions" (p. 5). Previous researchers have demonstrated the measure's test-retest reliability (e.g., D. O'Keefe, Shepherd, & Streeter, 1982) and internal consistency (e.g., Burleson, Applegate, & Neuwirth, 1981). Further, research indicates that the RCQ as a measure of cognitive differentiation is independent of verbal ability and general intelligence measures (e.g., Applegate et al., 1985; Burleson et al., 1981; Shepherd & Trank, 1992). In their thorough review of constructivist research, Burleson and Caplan (1998) observe: "Although the reliability and validity of measures derived from the RCQ have been questioned by several researchers, the balance of evidence weighs strongly in favor of RCQ-based measures" (p. 244).

The RCQ does not elicit all the constructs individuals employ in their social construal processes, but rather samples respondents' interpersonal knowledge (Burleson, Applegate, & Delia, 1991; Burleson & Caplan, 1998; Burleson & Waltman, 1988). As Burleson et al. (1991) argue, a respondent's impression "provides a sampling of the types and number of constructs readily available for use in understanding other people" (p. 116). Moreover, the free-response nature of the RCQ "provides a way of tapping a construct's natural range and focus of convenience and, thus, the relative availability of constructs for use in forming impressions" (Burleson et al., 1991, p. 116). Fundamental changes in the instrument's administration (e.g., Allen et al., 1990; Beatty & Payne, 1985) and scoring procedures (e.g., Allen, Burrell, & Kellermann, 1993) have produced results inconsistent with previous findings. Leichty (1997) notes that altering the motivation for the task (such as extra credit in the Beatty and Payne study and requesting a specific number of constructs in Allen et al.'s 1990 study) will undoubtedly lead to different results. These results underscore the importance of consistent methods and do not call into question the validity of the RCQ (Burleson et al., 1991; Leichty, 1997).

In our study, we evaluated cognitive differentiation according to the guidelines provided by Crockett, Press, Delia, and Kenny (1974) and Burleson and Waltman (1988). Participants' responses to the RCQ were scored for number of constructs used to describe a liked and disliked other. Reliability was tested by using independent coders to evaluate participants' responses (intraclass correlation=.92). Cognitive differentiation scores ranged from 10 to 34, with a mean score of 19.17, a standard deviation of 6.19, and median of 18.5.

Training. One item on the questionnaire was included to determine respondent's training in hospice. Respondents were asked to indicate their training by responding to the question, "What type(s) of training have you had to prepare you to work in hospice?" Participants were provided with the following categories: volunteer training, special professional training, workshop/ conference and other. Respondents marked all categories which applied to them. Because increased experience in various types of training would likely provide respondents with greater depth and breadth in hospice principles and philosophy, the number of training categories each respondent checked were added together to form an overall "training index." Nine of the 36 respondents (25%) reported receiving no training whatsoever in hospice.

Role differentiation. Following Kline and Ceropski (1984), team members were asked to respond to the following question: How would you define your role as a member of the hospice interdisciplinary care team? Role differentiation responses were scored according to Kline and Ceropski's (1984) procedures. Responses were "scored for the number of discrete activities, intentions or attitudes" ( Kline & Ceropski, 1984, p. 138) used to describe the role of the team member in the interdisciplinary team context. The total number was then used as a quantitative score for each participant. Independent coders scored the data from all respondents to assess coder reliability (intraclass correlation =.90). Scores ranged from 3 to 15, with a median of 8.0, a mean score of 8.08, and a standard deviation of 3.70.

The responses for role differentiation were also coded by two coders in terms of the content of the descriptions. All elements attributed to the role of the hospice interdisciplinary team member were first recorded by each coder. Themes or categories which emerged from the data were then identified by each coder independently. Five thematic categories were identified: position-related responsibilities (e.g., "responsible for updating patient charts" and "identify patient/family problems and develop a plan of treatment"), task-related interaction with other team members (e.g., "inform the social worker of patient/family psychosocial needs that need follow up" and "listen to other team members' comments that have had contact with my patients"), maintenance-related interaction with other team members (e.g., "general concern for other team members who may be under unusual stress" and "touch team members in support"), interaction with hospice staff outside the team (e.g., "talking with floor nurses who have had a bad day" and "advocate for my team members with upper-level management" ), and patient/family concerns (e.g., "provide information and instructions to patient and family" and "support patients emotional, spiritual and physical needs"). Categories identified were the same or similar with intercoder agreement 91% on the total of 291 descriptors. Initial differences in categorizing descriptors were then discussed to achieve agreement.

Persuasive message strategies. Problems associated with collecting naturally-occurring data in influence situations have led to the use of hypothetical situations in the majority of persuasive research (O'Keefe, 1990). Although the use of hypothetical situations has its critics (e.g., Seibold, 1988), Burleson and Caplan (1998) argue that the use of carefully designed hypothetical situations based on participants' own experiences is one strategy for making those situations more concrete for respondents. Further, research suggests that there is a high correspondence between individuals' levels of competence demonstrated in response to hypothetical situations and actual performance in "real-world" situations (Applegate, 1980; Carter, 1993; Kline & Ceropski, 1984; O'Keefe, 1990). For example, in the medical context Kline and Ceropski (1984) found a significant and positive relationship between message strategy person-centeredness in actual interviews with patients and regulative and interpersonal hypothetical situations. Similarly, Carter's (1993) observations of identity management strategies used by the elderly in face-to-face conversations were consistent with subjects' responses to hypothetical situations.

In the present study, participants provided written responses to a hypothetical persuasive situation developed from analyses of focus group interview responses. The situation was presented as a dyadic, face-to-face conversation in which respondents were to write exactly what they would say to another hospice team member to convince that person not to quit working at the hospice.¹ Responses to the persuasive situation were coded in accordance with Delia, Kline, and Burleson's (1979) guidelines which follow a nine-level hierarchy identifying the degree to which messages are listener-adapted or take the perspective of the other. "[T]he principle animating the hierarchy reflects the extent to which an individual's persuasive strategies accommodate to the perspective of the target" (Delia et al., 1979, p. 248). Reliability was tested by using three independent coders to assess participants' responses. One coder evaluated all responses, another all the responses from Hospice A, and the third coded the responses from Hospice B (intraclass correlations=.80 and .82 respectively). In addition, because there was concern that team members who had participated in the focus groups might be "sensitized" to the persuasive situation and thus produce more listener-adaptive message strategies than those who were not interviewed, a t test was performed to compare the two groups. No significant differences were found in level of message strategy produced (t[34] = .76, ns).

RESULTS

The first hypothesis proposed that hospice team members would produce persuasive message strategies which explicitly recognized and adapted to the target's perspective when attempting to influence another team member. The scores for the persuasive message strategies produced by hospice team members ranged from level four to nine, with the mode 8.0, median 7.0, mean 6.80, and standard deviation 1.83, supporting the hypothesis. The median strategy level clearly falls within the upper division of Delia et al.'s (1979) nine-level hierarchy, with the mean nearly within this division. Messages produced at these levels both express recognition of and adaptation to the target individual's perspective. Examples of the persuasive messages produced by the hospice team members are provided below in Table 1.

Table 1

Persuasive Message Strategies Produced by Health Care Providers

1. No discernible recognition of and adaptation to the target's perspective
   1. No statement of desire or request; no response given:
      No messages produced at this level.

   2. Unelaborated request:
      No messages produced at this level.

   3. Unelaborated statement of personal desire or need:
      No messages produced at this level.

2. Implicit recognition of and adaptation to the target's perspective
   1. Elaboration of the necessity, desirability or usefulness of the persuasive request:
      "Why do you want to leave? I think you are a valuable asset and we would really miss you."

   2. Elaboration of persuader's or persuasive object's need plus minimal dealing with anticipated counter-arguments:
      "You know if you decide to leave we are really going to miss you around here. You are a very good hospice nurse. We really need good nurses like you. You know you really don't want to go."

   3. Elaborated acknowledgment of and dealing with multiple anticipated counter-arguments:
      "I feel you are an asset to Hospice and feel your leaving would leave a real void. The way you relate to patients and families is important as well as the way you relate to co-workers. I understand your reasons for leaving, but want you to know I appreciate what you have done to help me."

3. Explicit recognition of and adaptation to the target's perspective
   1. Truncated efforts to demonstrate relevant consequences to the target of accepting (or rejecting) the persuasive request:
      "Don't go. We'd miss you. . . I think you should make a list of what's good and what's bad here. There's a lot that's wrong, but you'll see--the things that are good, the things that brought you here and made you stay for this long--they haven't changed."

   2. Elaboration of specific consequences of accepting (or rejecting) the persuasive request to one with characteristics of the target:
      "Sometimes it does get pretty tough. I know that it feels as though all you do is give, give, give, and maybe nobody gives back to you? I know you do good work. But I also know that you can't keep giving at this pace. Let's take a look at how we can make it more manageable for you. We want to keep you around."

   3. Demonstrable attempts by the persuader to take the target's perspective in articulating an advantage or attempts to lead the target to assume the perspective of the persuader, another person or the persuasive object:
      "You are so good at what you do I think you should consider staying here. Compassion flows from you like a water faucet turned fully open. These families and patients need people like you whom they see giving care so tenderly and lovingly. This is such a stressful time of their lives that a team member like you is a must--a STAT order!"

Coding system adapted from: Delia et al. (1979) pp. 248-249.
Note: Roman numerals correspond to major levels of the coding system. Arabic numerals correspond to the sublevels at which specific messages were scored.

Hypotheses 2, 3 and 4 were concerned with the listener-adaptiveness of persuasive message strategy and individual differences variables. First, we computed intercorrelations (see Table 2 below) to identify the relationships among the individual differences variables. Cognitive differentiation and role differentiation were positively and significantly correlated (r=.52, p‹.001), although clearly measuring different cognitive variables. That is, the correlation coefficient is less than the threshold of .60 needed to establish that two instruments are measuring the same construct (Beatty & Payne, 1984). Training was not related to either interpersonal construct system differentiation (r=-.07, p›.30) or role differentiation (r=.09, p›.20).

Table 2

Correlation Matrix of Antecedent Variables and Persuasive Message Strategy Listener-Adaptiveness

The above correlation matrix represents simple correlations that do not parcel out combined influences on a particular variable, such as listen-adaptiveness. To examine the separate influences of the individual differences variables on persuasive listener-adaptiveness, second-order partial correlations of the measures used to assess cognitive differentiation, role differentiation, training and listener-adaptiveness were computed. Hypothesis 2, that cognitive differentiation would be positively associated with listener-adaptive persuasive communication, was supported, although the relationship was marginally significance (r=.28, p‹.05). Hypothesis 3, that training in hospice would be negatively related to sophistication of persuasive message strategy, was supported (r=-.29, p‹.05), although the correlation again was weak. The fourth hypothesis, which posited that role differentiation would be positively associated with sophistication of message strategy produced, was supported (r=.40, p‹.01). Multiple regression was employed to examine the combined influences of the individual differences variables, with cognitive differentiation, role differentiation, and training as the independent variables and listener-adaptiveness as the dependent variable. Results from the analysis suggest an excellent fit of the model (R²=.396, F (3, 32)=7.00, p‹.001), with the combined independent variables explaining 40% of the variance in the dependent variable.

Research question 1 focused on the content of health care providers' descriptions of their role in the hospice interdisciplinary team. To examine this question, the 291 descriptors identified in the responses to the role definition question were coded according to the themes previously discussed. Team members placed the greatest emphasis on the responsibilities associated with their position (40.6%). This category included descriptors such as "make an individualized plan for each patient," "order supplies and equipment," "send forms to physicians for their signatures" and "assess relationship between patient and caregiver." Interacting with team members about task-related concerns was also emphasized (32.3%) and included descriptors such as "refer to each discipline to get their input [on patient/family assessment]," "keep the rest of the team filled in on patient's conditions and family members," and "call on other team members for help with [patient and family care] areas that need their input." Patient/family concerns (11.3%), team maintenance interaction (11.0%), and interacting with other staff members outside the hospice team (4.8%) were considered less important than the two categories previously discussed. Descriptors which focused on patient/family concerns included "providing friendship to hurting people [patients and families]," "front line in fielding problems/concerns with the patient and family," and "help out in any way patients and families with anything." Team maintenance descriptors included "facilitator of the team meeting,"support to other team members," and "serve as a motivator to the team members." The final category, interacting with other hospice staff, included "communicate any pertinent information to other staff members," and "take back to my department suggestions or criticisms of my department's actions."

To assess the relationship between the content of team members' role descriptions and cognitive differentiation, participants were first divided into two groups at the median differentiation score, with scores of 18 and below in the low differentiation group and high differentiation scores 19 and above. Chi square analyses revealed no significant differences in role descriptions based on cognitive differentiation. In a similar manner, listener-adaptiveness scores in response to the persuasive situation were divided at the median to assess the relationship of message strategy level and role descriptions. High listener adaptiveness included scores of eight and nine; low listener adaptiveness included scores of seven and below. Significant differences were found in two areas. Hospice team members who emphasized the responsibilities associated with their position produced more listener-adaptive strategies than those who included this category of descriptors with less frequency (X² [10, N =36] = 22.45, p‹.02). Respondents who more frequently used descriptors associated with maintaining team member relationships also produced more listener-adaptive persuasive messages (X² [5, N =36] = 11.93, p‹.03).

DISCUSSION AND IMPLICATIONS

This study was concerned with individual social-cognitive differences and persuasive message strategies among health care providers. As hypothesized, hospice interdisciplinary team members produced persuasive messages which generally explicitly recognize the perspective of the other (median=7.0; mode=8.0), supporting previous research which has found hospice staff members to be concerned and caring about others, mature, flexible and interpersonally sensitive (Bené & Foxall, 1991; Reese & Brown, 1997; Vachon, 1986, 1987). Interestingly, the Great Ideas section of the Hospice Homepage includes several strategies for adapting to other team members' perspectives. Suggestions include role exchange, rotating team leadership, and changing the typical seating arrangement (a sort of musical chairs).

Effective persuasive skills are essential in organizational life, particularly in organizations that are diverse and based on teamwork. Organizations that incorporate more decentralized structures, such as the hospices in this study, are particularly in need of employees who are able to design appropriate persuasive messages in decision-making contexts (Ekhouse, 1999; Neher, 1997). Eisenberg and Goodall (1998) suggest that organization members who use persuasive strategies adapted to the target's perspective are more apt to achieve their communicative goals. Influencing others in ethical ways that recognize others' perspectives is crucial for effective group work and fulfilling leadership functions (Gastil, 1993; Larson & LaFasto, 1989). This study suggests that hospice team members are especially adept at taking the perspectives of their coworkers and communicating in ways which recognize those perspectives.

The persuasive situations that emerged from the focus group interviews also merit attention. Although influence attempts are often associated with power and status issues in organizational work groups, the persuasive situations described by these hospice team members seem to focus on influence attempts that are designed to help others. Of the four top situations cited by the members of the two hospices, three were concerned with relationship rather than task issues. This is consistent with Thornton, McCoy and Baldwin's (1980) finding that the majority of interaction among health care team members focuses on maintenance rather than task issues. The nature of persuasive situations experienced in these hospices also points to the value of participant-generated hypothetical situations in research concerned with cognition and conversation. For example, Berteotti and Seibold (1994) describe a hospice in which subgroups clash, volunteers feel alienated from professional staff, and team interaction seems based more on competition than collaboration. For team members in this hospice, salient persuasive interactions may focus more on "us versus them" situations, rather than the more "we are all in this together" situations described by members of the hospices who participated in the present study. Asking hospice team members to generate examples in which influence attempts would occur within the team also revealed the dyadic nature of persuasive communication for the participants. The situations described occurred not within the team context, as within an interdisciplinary team conference, but when two team members were engaged in a private face-to-face conversation.

Although the relationship between amount of training and level of listener-adaptiveness of persuasive message strategy did emerge as predicted, it still seems counter-intuitive that these two variables would be negatively associated. Lack of information concerning the specific content and methods of the training programs in which respondents for the present study participated makes it difficult to interpret this finding and suggests an area for further examination. Previous studies concerned with hospice training programs indicate that training focuses, as would be expected, primarily on patient and family care while only briefly touching on interdisciplinary team communication (Wilkinson & Wilkinson, 1987). Further, Basile and Stone's (1987) research concerning the characteristics of the effective hospice team member identified 26 competencies associated with direct patient/family care responsibilities and just three which addressed working with the hospice team. With their apparent emphasis on delivery of patient and family care, hospice training programs may encourage hospice workers to focus on patient and family issues to the neglect of team communication and coordination. This focus is certainly understandable given the nature of the hospice staff's task, but also may result in a lack of attention to skills needed to deal with other team members and team issues.

Role differentiation was found to be most positively and significantly related to sophistication of persuasive strategy, supporting Kline and Ceropski's (1984) finding in the use of person-centered communication in regulatory, advisory, and informational situations. This suggests that those who have a more encompassing view of their role on the team are better able to identify and adapt to the perspectives of other team members. This also suggests that role differentiation may provide a more context specific measure of complexity. Because respondents in the present study were simply asked to describe a liked and disliked other for the RCQ, rather than a coworker, RCQ scores may reflect a more generalized measure of their interpersonal construct systems, rather than one tied more specifically to the work environment. Role differentiation may be a better predictor of communication abilities in work situations.

This claim is consistent with Wilson et al.'s (1992) suggestion that the RCQ be adapted to specific communicative contexts. For example, in discussing their notion of conversational complexity, Daly et al. (1985) argue that individuals "vary in their cognitive sophistication about social interactions in terms of their abilities to differentiate along, and organize their representations of, conversations" (Daly et al., 1985, p. 33). Further, measures of cognitive complexity focus on the organization of knowledge about people, and while this is related to conversations, complex ways of thinking about persons does not necessarily translate into complex ways of thinking about conversations. Martin (1991) provides a similar argument for the development of his Relational Cognition Complexity Instrument (RCCI). As Martin (1991) notes, "an individual might have a rather complex or sophisticated cognitive system for understanding other persons, yet be rather unsophisticated when conceptualizing on a relationship level" (p. 469). While the purpose of the present study was not to develop a measure of complexity for influence attempts in coworker conversations, the results do suggest an area of further research. Although role differentiation and cognitive complexity were positively and significantly correlated (r=.52, p<.001), as discussed previously, the correlation coefficient does not exceed the limit of .60 or greater needed to establish that two instruments are measuring the same construct (Beatty & Payne, 1984). The lack of a relationship between cognitive differentiation and the content of hospice team members' role definitions further suggests that the two measures differ in important ways.

Examining the content of participants' responses in defining their role as a hospice interdisciplinary team member provides insight into how health care providers view themselves in relation to their colleagues. The prototypic hospice interdisciplinary care team member performs those duties associated with her/his position, interacts with team members both in terms of task accomplishment and relationships among team members, expresses concern for patient and family welfare, and acts as a bridge between the team and other hospice staff. Moreover, identifying how participants described their role on the team provided additional insight into how those beliefs are linked with message production. Hospice team members who stressed the importance of attending to team maintenance functions produced more listener-adaptive persuasive messages than those who placed less of an emphasis on this area. This finding may be due to the situation, attempting to influence a team member to not leave the hospice, which is clearly a relationship concern.

Team members who emphasized job-related duties in their descriptions also produced more listener-adaptive messages than those who did not. This may be due to the development of a comprehensive understanding of the responsibilities associated with a position and a possible corresponding lessening of role ambiguity and tendencies toward disciplinary territoriality. That is, health care providers who are more certain of their own job responsibilities experience less uncertainty about their role on the team and the function of their discipline in hospice patient care. This security in knowing their own job-related duties may allow for greater flexibility as a team member and thus an increased ability to adapt to others' perspectives.

The weaker correlation between general interpersonal cognitive differentiation and listener-adaptive persuasive message strategy in the workplace suggests that we need to examine other factors that may either directly or indirectly influence message strategy selection in specific conversational contexts. As Coopman, Hart, Allen, and Haas (1997) argue, environmental conditions such as organizational culture and social norms may act as a filter through which interpersonal construct systems function. In the present study, hospice team members' beliefs about their role on the team may be a more powerful resource in developing a particular message strategy when attempting to influence a coworker than a more generalized set of interpersonal construct systems. Still, the relationship of cognitive differentiation to communication behavior evidenced here is generally consistent with previous constructivist research. Theoretically, we should expect that interpersonal differentiation generally should facilitate creation of more complex context-specific beliefs. These beliefs then serve as direct antecedents to communication behavior (Applegate et al., 1989; Applegate, Burleson, & Delia, 1992; Coopman et al., 1997; O'Keefe & Delia, 1982).

There are obvious limitations to the study. The small sample size makes it difficult to generalize the findings to other hospices, health care teams, or interdisciplinary teams. Also, the unique characteristics of hospice may produce teams that are more of an anomaly than the norm for teamwork. The use of a hypothetical situation does tap into what respondents are able to do, possibly not what they actually do in everyday interactions. Moreover, the use of a single hypothetical situation may make the results situation specific; that is, respondents may react differently to persuasive situations with parameters unlike the one used in the study. Finally, we did not measure other cognitive variables that may be salient in persuasive contexts. As Waldron and Applegate (1994) note, the link between cognitive complexity and message production may be indirect, rather than direct.

In summary, the present study successfully applies an individual differences model to identify important social cognitive (interpersonal complexity, role differentiation) and contextual (organizational training) antecedents to listener-adaptive persuasive communication in a health-care setting. The ability to influence team members in such a context is clearly an important skill affecting decisions that impact coworkers and clients. Previous research has shown that the persuasive abilities studied here are related to a variety of important outcome variables, including upward mobility in organizations (Sypher & Zorn, 1986), peer acceptance and evaluation (Burleson, Applegate, & Delia, 1992; Woods & Applegate, 1993), and compliance-gaining in a health care setting (Kasch, 1984; Kline & Ceropski, 1984).

We might expect that differences in listener-adaptive, person-centered persuasive communication skills could have similar relations to effectiveness and others' perceptions on a hospice team. If so, this argues for inclusion of features of training programs that go beyond what current practices seem to call for, especially given the negative correlation between training and adaptive persuasive communication evidenced here. Using principles suggested by Clark, Willinghanz, and O'Dell (1985) in their work with children, professionals could be encouraged to differentiate their schemes for conceptualizing people and their own role in the organization. The specific communication behaviors indexed in the coding systems used also should be useful in modeling adaptive persuasive communication in specific contexts. Moving beyond hospice interdisciplinary teams, organizations should closely examine current training programs and the degree to which such training promotes listener-adaptive communication. This is particularly important in teamwork training, which tends to focus on the application of specific skills rather than facilitating more complex and flexible cognitive processes.

Beyond these practical implications, this study offers another successful implementation of the individual difference model developed within constructivist theory that ties general and context-specific social-cognitive antecedents to important differences in communication behavior. The situation used here to generate communication strategies was ecologically valid based on outcomes of our focus group work. Responses to such situations have been shown again and again to provide valid ability measures that reflect communication behavior in face-to-face settings in reliable and expected ways (Applegate, 1980, 1982; Carter, 1993; Kasch, 1984; Kline & Ceropski, 1984; Woods & Applegate, 1993). That the model works in health care settings is especially gratifying given the increasing attention such settings are now justifiably receiving from communication research.

NOTES

¹ The four hospice teams generated several persuasive situations. However, there was only one situation that received strong support across all four teams. To make the hypothetical situation as true to the team members' own experiences as possible, we elected to use the one situation they expressed sharing in common. Although this reduces the reliability of our measure to some degree, we believe that the situation's validity was more important.

Back to text

REFERENCES

Allen, M., Mabry, E., Banski, M., Carter, P., & Stoneman, M. (1990). A thoughtful appraisal of measuring cognition using the Role Category Questionnaire. Communication Reports, 3, 49-57.

Allen, M., Burrell, N., & Kellermann, K. (1993). A comparison of observer and actor coding of the Role Category Questionnaire. Communication Reports, 6, 1-7.

Applegate, J. (1980). Person- and position-centered communication in a day care center. In N. Denzin (Ed.), Studies in symbolic interaction (Vol. 3, pp. 59-96). Greenwich, CT: JAI Press.

Applegate, J. (1982). The impact of construct system development on communication and impression formation in persuasive contexts. Communication Monographs, 49, 277-289.

Applegate, J. (1990). Constructs and communication: A pragmatic integration. In G. Neimeyer & R. Neimeyer (Ed.), Advances in personal construct psychology (Vol. 1, pp. 203-230). Greenwich, CT: JAI Press.

Applegate, J., Burke, J., Burleson, B., Delia, J., & Kline, S. (1985). Reflection-enhancing parental communication. In I. Sigel (Ed.), Parental belief systems: The psychological consequences for children (pp. 107-142). Hillsdale, NJ: Lawrence Erlbaum.

Applegate, J., Burleson, B., & Delia, J. (1992). Reflection enhancing parenting as an antecedent to children's social cognitive and communicative development. In I. Sigel, A. McGillicuddy-DeLisi, & J. Goodnow (Eds.), Parental belief systems: The psychological consequences for children (Vol. 2, pp. 3-40). Hillsdale, NJ: Lawrence Erlbaum.

Applegate, J., Coyle, K., Seibert, J., & Church, S. (1989). Interpersonal constructs and communicative ability in a police environment: A preliminary investigation. International Journal of Personal Construct Psychology, 2, 385-400.

Babler, J. (1997). A comparison of spiritual care provided by hospice social workers, nurses, and spiritual care professionals. The Hospice Journal, 12(4), 15-27.

Basile, J., & Stone, D. (1987). Profile of an effective hospice team member. Omega, 17, 353-362.

Beatty, M., & Payne, S. (1984). Loquacity and quantity of constructs as predictors of social perspective-taking. Communication Quarterly, 32, 207-210.

Beatty, M., & Payne, S. (1985). Is construct differentiation loquacity? A motivational perspective. Human Communication Research, 11, 605-612.

Becker-Reems, E., & Garrett, D. (1998). Testing the limits of teams: How to implement self-management in health care. Chicago: American Hospital Publishing.

Bené, B., & Foxall, M. (1991). Death anxiety and job stress in hospice and medical-surgical nurses. The Hospice Journal, 7(3), 25-41.

Beresford, L. (1993). The hospice handbook: A complete guide. Boston: Little, Brown.

Berteotti, C., & Seibold, D. (1994). Coordination and role-definition problems in health-care teams: A hospice case study. In L. Frey (Ed.), Group communication in context: Studies of natural groups (pp. 107-131). Hillsdale, NJ: Lawrence Erlbaum Associates.

Boling, A., & Lynn, J. (1998). II. Hospice: Current practice, future possibilities. The Hospice Journal, 13(1/2), 29-32.

Boster, F. (1989). Group argument, social pressure, and the making of group decisions. In J. Anderson (Ed.), Communication yearbook 13 (pp. 303-312). Newbury Park: Sage.

Burleson, B., Applegate, J., & Delia, J. (1991). On validly assessing the validity of hte Role Category Questionnaire: A reply to Allen et al. (1991). Communication Reports, 4, 111-119.

Burleson, B., Applegate, J., & Delia, J. (1992). Effects of maternal communication and children's social-cognitive and communication skills on children's acceptance by the peer group. Family Relations, 41, 264-272.

Burleson, B., Applegate, J., & Neuwirth, C. (1981). Is cognitive complexity loquacity? A reply to Powers, Jordan, and Street. Human Communication Research, 7, 212-225.

Burleson, B., & Caplan, S. (1998). Cognitive complexity. In J. McCroskey, J. Daly, M. Martin, & M. Beatty (Eds.), Communication and personality: Trait perspectives (pp. 233-286). Cresskill, NJ: Hampton Press.

Burleson, B., & Waltman, M. (1988). Cognitive complexity: Using the role category questionnaire measure. In C. Tardy (Ed.), A handbook for the study of human communication: Methods and instruments for observing, measuring, and assessing communication processes (pp. 1-35). Norwood, NJ: Ablex.

Burleson, B., Wilson, S., Waltman, M., Goering, E., Ely, T., & Whaley, B. (1988). Item desirability effects in compliance-gaining research: Seven studies documenting artifacts in the strategy selection procedure. Human Communication Research, 14, 427-486.

Carter, D. (1993). Identity management strategies among the elderly. Unpublished doctoral dissertation, University of Kentucky, Lexington.

Clark, R., Willinghanz, S., & O'Dell, L. (1985). Training fourth graders in compromising and persuasive strategies. Communication Education, 34, 331-342.

Cline, R. (1990) Small group communication in health care. In E. Ray & L. Donohew (Eds.), Communication and health: Systems and applications (pp. 69-91). Hillsdale, NJ: Lawrence Erlbaum.

Cooper, M. (1986). Self selection. In R. Turnbull (Ed.), Terminal care (pp. 177-181). Washington: Hemisphere Publishing Corporation.

Coopman, S., Hart, J., Allen, M., & Haas, J. (1997). Detecting cultural knowledge in organization members' personal construct systems. Journal of Constructivist Psychology, 10, 321-338.

Corrigan, P. & McCracken, S. (1998). An interactive approach to training teams and developing programs. In P. Corrigan & D. Giffort (Eds.), Building teams and programs for effective psychiatric rehabilitation (pp. 3-12). San Francisco: Jossey-Bass.

Crockett, W. (1965). Cognitive complexity and impression formation. In B. Mahler (Ed.), Progress in experimental personality research, (Vol. 2., pp. 47-90). New York: Academic Press.

Crockett, W., Press, A., Delia, J., & Kenny, C. (1974). Structural analysis of the organization of written impressions. Unpublished manuscript, Department of Psychology, University of Kansas, Lawrence.

Daly, J., Bell, R., Glenn, P., & Lawrence, S. (1985). Conceptualizing conversational complexity. Human communication research, 12, 30-53.

Delia, J., Kline, S., & Burleson, B. (1979). The development of persuasive strategies in kindergartners through twelfth-grade. Communication Monographs, 46, 241-256.

Eckhouse, B. (1999). Competitive communication: A rhetoric for modern business. New York: Oxford University Press.

Eisenberg, E., & Goodall, H. (1997). Organization communication: Balancing creativity and constraint, 2nd ed. New York: St. Martin's Press.

Gastil, J. (1993). Democracy in small groups: Participation, decision making, and communication. Philadelphia, PA: New Society Publishers.

Hospice Federation of Massachusetts. (1998). Website. Available at: http://www.hospicefed.org

Juliÿ, M., & Thompson, A. (1994a). Group process and interprofessional teamwork. In R. Casto & M. Juliÿ (Eds.), Interprofessional care and collaborative practice (pp. 35-41). Pacific Grove, CA: Brooks/Cole Publishing.

Juliÿ, M., & Thompson, A. (1994b). Essential elements of interprofessional teamwork: Task and maintenance functions. In R. Casto & M. Juliÿ (Eds.), Interprofessional care and collaborative practice (pp. 43-57). Pacific Grove, CA: Brooks/Cole Publishing.

Kasch, C. (1984, April). Interpersonal competence, compliance, and person-centered speech: A study of nurses' and para-professionals' communicative strategies. Paper presented at the Central States Speech Association Convention, Chicago.

Kline, S., & Ceropski, J. (1984). Person-centered communication in medical practice. In J. Wood & G. Phillips (Eds.), Human decision-making (pp. 120-141). Carbondale: Southern Illinois University Press.

Lafferty, C. (1998). Transformational leadership and the hospice R.N. case manager: A new critical pathway. The Hospice Journal, 13(3), 35-48.

Lambert, M., DeJulio, S., & Stein, D. (1978). Therapist interpersonal skills: Process, outcome, methodological skills, and recommendations for future research. Psychological Bulletin, 85(3), 4657-489.

Larson, C., & LaFasto, F. (1989). Teamwork: What must go right, what can go wrong. Newbury Park, CA: Sage.

Larson, D. (1998). The Hospice Home Page. Available at: http://www.scu.edu/Hospice/.

Leichty, G. (1997, April). Making sense of the Role Category Questionnaire. Paper presented at the annual meeting of the Southern States Communication Association, Savannah, GA.

Manion, J., Lorimer, W., & Leander, W. (1996). Team-based health care organizations: Blueprint for success. Gaithersburg, MD: Aspen.

Martin, R. (1991). Examining personal relationship thinking: The Relational Cognition Complexity Instrument. Journal of Social and Personal Relationships, 8, 467-480.

Martin, R. (1992). Relational cognition complexity and relational communication in personal relationships. Communication Monographs, 59, 150-163.

National Hospice Organization. (1999). Website available at: http://www.nho.org/

Neher, W. (1997). Organizational communication: Challenges of change, diversity, and continuity. Boston: Allyn and Bacon.

O'Connor, R., Hallberg, S., & Myles, R. (1999). Organizational communication and health care administration. In L. Jackson & B. Duffy (Eds.), Health communication research: A guide to developments and directions (pp. 103-124). Westport, CT: Greenwood Press.

O'Keefe, B., & Delia, J. (1982). Impression formation and message production. In M. Roloff & C. Berger (Eds.), Social cognition and communication (pp. 33-72). Beverly Hills, CA: Sage.

O'Keefe, B., Murphy, M. Meyers, R. & Babrow, A. (1989). The development of persuasive communication skills: The influence of developments in interpersonal constructs on the ability to generate communication-relevant beliefs and on level of persuasive strategy. Communication Studies, 40, 29-40.

O'Keefe, D. (1990). Persuasion: Theory and research. Newbury Park: Sage.

O'Keefe, D., Shepherd, G., & Streeter, T. (1982). Role category questionnaire measures of cognitive complexity: Reliability and comparability of alternative forms. Central States Speech Journal, 33, 333-338.

Reese, D., & Brown, D. (1997). Psychosocial and spiritual care in hospice: Defferences between nursing, social work, and clergy. The Hospice Journal, 12(1), 29-41.

Seibold, D. (1988). A response to "Item desirability in compliance-gaining research." Human Communication Research, 15, 152-161.

Seibold, D., Meyers, R., & Sunwolf. (1996). Communication and influence in group decision making. In R. Hirokawa & M. Poole (Eds.), Communication and group decision making, 2nd ed. (pp. 242-268). Thousand Oaks, CA: Sage.

Siebold, C. (1992). The hospice movement: Easing death's pains. New York: Twayne.

Shepherd, G., & Trank, D. (1992). Construct system development and dimensions of judgment. Southern Communication Journal, 57, 296-307.

Shulman, A. (1996). Putting group information technology in its place: Communication and good work group performance. In S. Clegg, C. Hardy, & W. Nord (Eds.), Handbook of organization studies (pp. 357-374). London: Sage.

Sullivan, T. (1998). Concept analysis of collaboration: Part I. In T. Sullivan (Ed.), Collaboration: A health care imperative (pp. 3-42). New York: McGraw-Hill.

Sypher, B., & Zorn, T. (1986). Communication abilities and upward mobility: A longitudinal investigation. Human Communication Research, 12, 420-431.

Sypher, H., Witt, D., & Sypher, B. (1986). The comparative validity of three measures of cognitive differentiation as predictors of written communication ability. Communication Monographs, 53, 376-382.

"Teamwork gives Julia a new heart." (Jan./Feb., 1997). Good Health Magazine. available at http://www.goodhealth.com/gh_mag/janfeb97/Teamwork.html

Thornton, B., McCoy, E., & Baldwin, D. (1980). Role relationships on interdisciplinary health care teams. In D. Baldwin, B. Rowley, & V. Williams (Eds.), Interdisciplinary health care teams in teaching and practice. Reno, NV: New Health Perspectives.

Truax, C., & Carkhuff, R. (1967). Toward effective counseling and psychotherapy. Chicago: Aldine.

Truax, C., & Mitchell, K. (1971). Research on certain therapist skills in relation to process and outcomes. In A. Bergin & S. Garfield (Eds.), Psychotherapy and behavior change (pp. 299-344). New York: Wiley.

Vachon, M. (1986). Myths and realities in palliative/hospice care. The Hospice Journal, 2(1), 63-79.

Vachon, M. (1987). Team stress in palliative/hospice care. The Hospice Journal, 3(2/3), 75-103.

Vincent, P., & Garrison-Peace, H. (1986). Do hospice nurses differ from non-hospice nurses? American Journal of Hospice Care, 3(1), 41-42.

Wilkinson, H., & Wilkinson, J. (1987). Evaluation of a hospice volunteer training program. Omega, 17, 263-275.

Wilson, S., Cruz, M., & Kang, K. (1992). Is it always a matter of perspective? Construct differentiation and variability in attributions about compliance gaining. Communication Monographs, 59, 350-367.

Woltermann, M. (1998, Feburary). Modern hospice here. Amarillo Business Journal. Available at: http://businessjournal.net/stories/020698/here.html

Woods, E., & Applegate, J. (1993, August). An integrated analysis of verbal and nonverbal components of person-centered argument and partner response. Paper presented at the Alta Conference on Argumentation, Alta, Utah.

Zimmermann, S., & Applegate, J. (1992). Person-centered comforting in the hospice interdisciplinary team. Communication Research, 19, 240-263.