and reach thousands of daily visitors
Listen to thisData on job-related stress are reported for personnel who work in and around group homes for young offenders and mentally handicapped persons in one Canadian province in order to illuminate the organizational changes and working conditions which these personnel find most distressing. Based on the assumption that stressful events and working conditions have neither a universal nor a uniform impact on persons, the Organizational Changes Rating Scale and the Stressful Working Conditions Checklist (Adams, 1980) were employed to discriminate between the existence or occurrence of an event or condition and the identification of that event or condition as stressful. Subjects' ratings (N=390) were rank-ordered overall and by job title for examination. The findings are used to support the notion that stressors should be considered for both direct and indirect service personnel in order to illuminate important contextual factors of stress for direct care workers. However, those events and conditions that are perceived to have high impact for all personnel may mask important job stressors for particular groups of personnel if research does not discriminate between the occurrence or presence of an event or condition and its perceived impact on each group.
The purpose of this paper is to examine what job-related stressors are peculiar to the nature of work in group homes and to determine which of these stressors are perceived by the personnel involved as having the most stressful impact upon them.
The findings reported in this paper are part of a larger study in progress (Burford, work in progress) which is predicated on the notion that many attempts to understand the nature of residential group care work have paid insufficient attention to the subjective components of stress evaluation. That is, while there is a strong belief that stress is a necessary and potentially positive element of growth, development and adaptation, it is only in recent years that researchers have come to accept that stressful events and conditions do not have universal impact on people (Gentry & Kobasa, 1984). Exposure to stress cannot be understood without giving careful consideration to the differences in perception between people and between groups of people and to other factors which buffer, cushion or mediate the relationship between stress and performance.
A wide range of factors has been shown to mediate and/or buffer the responses to stress by individuals and work groups including: personality and attitudinal differences; the availability of relevant career and educational ladders I and job enhancement schemes; the quality of reward schemes; the relative position of residential services within the wider service network; the characteristics of the resident group; levels of change and stress; the quality of supervisory, consultative and social supports; organizational and administrative structures and supports; patterns of work schedules; physical environmental conditions; and the individual and collective interpretation or meaning that evolves in relation to an event or condition (Beddoe, 1980; Beehr & Bhagat, 1985; Brendtro & Mitchell, c 1983; Burford & Fulcher, 1985; Fulcher, 1983; Gamer, 1982; Krueger, 1987; Mattingly, 1981; Moos, 1975; Reed, 1977).
In particular, this paper presents findings relative to the differences between reports of actual frequency of occurrence of organizational changes and stressful working conditions, and ratings of their impact by sample subjects. This is consistent with Mattingly's (1981) call for researchers to develop fuller understanding of the situations and conditions that personnel in residential group care find distressing and which aggravate burn-out.
While the supportive literature on these subjects is too broad to summarize in this paper, a few would seem to have particular relevance to the vulnerability or susceptibility of work groups toward adaptive or maladaptive functioning.
In their study of work environment variables related to job satisfaction for group home workers, Connis et al. (1979) offer a number of interesting findings. As suspected, they found high rates of employee turnover. Only a few employees in their study tended to remain in the group homes for more than one year. Varying levels of job satisfaction were associated with different occupational groupings and they noted that those employee groups who spent less time in face-to-face contact with residents reported higher levels of satisfaction. Houseparents had considerably more direct contact than other groups. For all employee groups, satisfaction was positively associated with access to private living space.
Citing the works of Grossbard (1960), Reed (1977), and Mattingly (1981), Clemens (1986) argues that live-in workers are subject to "limited opportunities for withdrawal, psychological repair and personal recovery" (p. 35). On the basis of this observation, we would assume that the provision of these opportunities and the space in which to do them would go some distance toward solving certain problems.
Connis et al. (1979) compare satisfaction rates for two categories of
live-in workers: teaching parents and houseparents. Despite the fact that
both groups live in, teaching houseparents reported much higher work-related
satisfactions than their houseparent counterparts. Teaching houseparents
were twice as likely to have an advanced degree, were ten times as likely to
have received additional formal training and despite the fact that teaching
houseparents worked slightly more hours during the week than houseparents,
considerably fewer of these hours were spent in direct face-to-face contact
with the residents. Perhaps even more importantly, teaching houseparents
were ten times as likely to report that they male exclusive
treatment-relevant decisions and were considerably less likely to be
required to consult superiors or outside professionals, or to participate in
committees in making their decisions than were their houseparent
counterparts. It would appear that the opportunity to withdraw, repair and
recover may be closely related to status differences involving
independence/dependence and complexity and intensity of involvement in both
resident and staff relationships. Again we note that the ability to regulate
the amount of direct contact time, and perhaps the number of relationships
with others, weighs heavily on perceived satisfaction with the job. The
demand for face-to-face teamwork increases the complexity and limits
independence for some workers.
The vulnerability of group home
caregivers is further revealed through the sociological analysis of the
structure of professions. Esland (1980) and Benveniste (1987) note that the
power of science--specifically through the practice of medicine – has been a
powerful forum for the allocation of resources, including territory. Unless
other professions compete through this vehicle, they may not grow and
develop. One status-conferring function of profession is to enable members
to regulate the amount of face-to-face contact one has with members of the
client or target group (Esland, 1980).
An analysis of the very nature of caring work supports this view. Waerness (1984) proposes a rationality for caring that she differentiates from scientific rationality and its inherent emphasis on environmental control. She argues that scientific rationales for caring have devalued the components of care and closeness which have typically been labelled as "feminine." She points out that the bulk of care in Western societies is provided by women as an extension of their positions in their own families and represents an expressive quality of care. This type of care is threatened by an overemphasis on instrumental forms of care that grow out of scientific methodology and which underpins much of the emphasis on professionalism, particularly in the bureaucratic context. She makes the case for "more decision-making power to women on the basis of their personal experiences from practical caregiving work in the private sphere and from working class jobs in the public caregiving services" (p. 205).
When we link this back to the findings of Connis et al. (1979) that houseparents tend to be the least autonomous and educated, and consider that it is the housemother who is most often hired in the couple (CWLA Standards for Group Home Care, 1978; Shostack, 1987), we find further explanation for the status of live-in caregivers in group homes. Of interest in this context is Fulcher's (1983) finding that residential group care teams with a higher proportion of males to females and teams with an average age of under twenty-eight were more likely to exhibit dysfunctional patterns of adaptation.
The Sample and Demographic Data
The sample (N=390) consists of houseparents (N=65), careworkers (N= 137), lay board members (N=73), social workers (N=47), and management personnel (N=68) who work in and around group homes for young offenders and mentally handicapped persons in one eastern Canadian province. The rationale for including board members and managers in the study is found in the work of Tosi and Hammer (1974) who advance the notion that all members of the relevant environment who may be able to effect direct decision-making, including policy formulation, should be considered in any examination of organizational behavior. From a practical point of view, this allows one to examine the reports of the direct-service workers in the organizational context in which they occur.
In the province under study, group homes are managed by a sponsoring
volunteer board of directors. The Department of Social Services provides
ongoing policy formulation, funding and licensing; provides social work and
other supportive services; and determines admissions, transfers and
discharges through district and regional offices and the Divisions of Mental
Retardation and Youth Corrections.
The managers referred to in this
study are front-line, middle and senior social services personnel as opposed
to the immediate supervisors in the homes.
The title, houseparent, refers specifically to those persons who live-in during their period of work. While there is great variation amongst this group with respect to their hours of work, their living space within the homes, and their access to private space outside the homes, most houseparents tend to dwell within the home for twenty-four hour periods of at least four days in a stretch. This job category includes senior houseparents and houseparents who are also the supervisors or directors associated with the home. In the latter instance, the houseparent is directly responsible to the board of directors.
The title, careworker, includes child care workers from the homes for young offenders and the front-line staff in the homes for mentally handicapped individuals. Also included in this category are the titles senior care worker and coordinator. Careworkers tend to work shifts of varying lengths, frequently providing overnight coverage, but do not reside in the home.
Social workers are those personnel who perform the field work duties and are employed by the Department of Social Services. Typically they carry the residents in their caseloads as part of their wider responsibilities with the Department, which go well beyond their work with the home(s).
Supervisory staff in the homes are not included in this part of the study be-cause their numbers are so small.
Sex. The majority of houseparents (63 percent), careworkers (66 percent) and social workers (72 percent) are female while board members (53 percent) and managers (71 percent) are male.
Age. Care workers and social workers represent the youngest personnel in the sample with 48 percent and 36 percent, respectively, being under the age of twenty-eight. A much greater number of care workers (27 percent) falls in the higher range of thirty-seven and over than do social workers (6 percent), making social workers, overall, the youngest group.
Length of Time in Post. Table 1 summarizes the length of time sample personnel have worked in their respective posts.
TABLE 1
JOB TITLE AND LENGTH OF TIME IN POST
| Job Title | Less than 14 mos. | 15-40 mos. | 41 plus mos. |
| Houseparent | 29% | 51% | 20% |
| Careworker | 53% | 36% | 11% |
| Board Member | 15% | 29% | 56% |
| Social Worker | 34% | 23% | 43% |
| Manager | 26% | 20% | 54% |
Of note is the fact that over half the care workers and close to one-third of the social workers and houseparents have been in their posts for less than fourteen months. Over half the board members and managers have been in their jobs for more than three years and four months. In this context, it is worth pointing out that group homes are a relatively recent development in the province as compared with other parts of North America. The first homes opened in 1977 as part of overall efforts to deinstitutionalize and 60 percent of the homes in the study have opened since 1982.
Hours of Work. As would be expected, houseparents reported that they worked more hours per week than all other groups--59 percent work fifty-six hours or more on average, and 68 percent reported that they had worked in excess of fifty-six hours during the week prior to testing. Of note is the fact that 76 percent of the care workers say they put in forty hours or less during an average week but during the week prior to testing 22 percent reported that they had worked between forty-one and fifty-five hours and 26 percent had worked in excess of fifty-six hours. Social workers and managers both indicated some degree of overwork during the prior week as compared to their usual expectations. Also of note is the fact that 58 percent of managers had worked in excess of forty-one hours during the previous week (22 percent worked in excess of fifty-six) and this is consistent with an average week for them (45 percent work forty-one to fifty, and 12 percent work more than fifty).
Education. Table II identifies subjects' level of education by job title.
TABLE II
JOB TITLE BY LEVEL OF EDUCATION
| Job Title | High School or Less | Business/Technical/Some College | College/University | PostGraduate |
| Houseparent | 40% | 37% | 31% | 3% |
| Careworker | 35% | 25% | 26% | 35% |
| Board Member | 14% | 25% | 26% | 35% |
| Social Worker | 2% | 6% | 85% | 6% |
| Manager | 6% | 21% | 65% | 9% |
Considerable variation exists in the level of education for houseparents and care workers. Social workers and board members are the most highly educated with social workers having the most uniform education—85 percent hold the B.S.W. or a Bachelor's Degree in another discipline. Slightly less than one-third of the managers have less than a university degree. Of importance is the fact that no training program exists in the province for residential, youth or child care work. While a short certification program has been initiated, the content is more specifically directed toward orienting careworkers with the mentally handicapped.
Data Collection and Analysis
Data on organizational changes and chronic stressful working conditions were collected using the Adams (1980) Organizational Changes Rating Scale (thirty items) and The Stressful Work Conditions Checklist (twenty-five) items. In addition to identifying the frequency of pervasiveness of a change or condition, subjects identified which items are the most stressful to them personally.
A frequency distribution was used to identify and rank order the five most commonly reported organizational changes; the five most stressful of these changes; the five most enduring stressful working conditions; and the five working conditions that subjects rank as having the highest stressful impact on them. Frequencies were prepared for sample subjects overall and by job title. Only the overall frequencies are presented in this paper due to the limitations of space.
Tables III through VI present rank orderings of the subjects' responses overall. Note that the percentages are much lower for the categories that report the strength of the changes and conditions. This is because the questionnaires ask subjects to rank the top three stressful items for them personally. For example, while all participants may note that a change occurs, the number of persons who actually rank that item as one of their highest three stressors would naturally be smaller.
TABLE Ill
RANK ORDERING OF TOP FIVE ORGANIZATIONAL
CHANGES BY
OCCURRENCE
(N=390)
TABLE IV
RANK ORDERING OF TOP FIVE MOST STRESSFUL ORGANIZATIONAL CHANGES
(N=390)
TABLE V
RANK ORDERING OF TOP FIVE MOST ENDURING
STRESSFUL WORKING CONDITIONS
(N=390)
TABLE VI
RANK ORDERING OF TOP FIVE MOST
STRESSFUL
ENDURING CONDITIONS OF WORK
(N=390)
Results
Overall, sample subjects share in
common exposure to staff turnover, overwork, conflicted expectations,
frequent changes in instructions, and a crisis-oriented approach to their
work. The item "I have differences of opinion with my supervisors" (Table
VI) is accounted for almost entirely by houseparents, careworkers and social
workers. The item "Decisions or changes that affect me are made without my
knowledge or involvement" (Table VI) is accounted for almost entirely by
managers, houseparents and careworkers.
When we examine the top five
organizational change and stress items for occurrence and impact by job
title we note much similarity in the reporting, with the following
exceptions.
Management
The highest change item for management is "Experiencing a sudden decrease in the number of positive recognitions of my accomplishments" (67 percent) although it is not indicated as an item that is perceived to be troubling for them. It is almost as if managers do not feel they should expect positive recognitions. "Acquiring new subordinates" does appear on both lists, however. Of particular note is the fact that for managers themselves the item "Management expects me to interrupt my work for new priorities" appears as one of the five most enduring characteristics of their jobs overall (25 percent) and as one that they find to be most stressful (19 percent). They also rate "I am cautious about what I say in meetings" (25 percent) and "I must attend meetings to get my job done" (75 percent) in the top five on both occurrence and impact. A full 23 percent say that a high stress item for them is "Decisions or changes that affect me are made without my knowledge or involvement." These reports may be surprising to other personnel who attribute far more power to managers than many in this group would seem to experience. At a minimum, some managers would appear to feel inhibited in the use of their power. The responses of the managers are particularly illuminating when we consider that
Managerial jobs ... (are) ... particularly stressful if they are demanding in time requirements and decision-making activities yet the incumbent has little freedom of action and must continually clear decisions with the immediate boss or with other supervisors (Quick & Quick, 1984, p. 21).
This predominantly male group (71 percent) has been in their jobs considerably longer, on average, than any of the other groups and nearly one-third do not meet the present educational requirements for promotion to or within management. If this group is to reach its potential for supporting the work of the homes, ways must be found to value their experience and reduce their feelings of intimidation, thereby enabling them to manage.
Board Members
The items for board members look much the same as the overall picture except that board members identify that "The people I work with closely are trained in a field that is different from mine" (23 percent) is an enduring characteristic of their roles and "I am cautious about what I say in meetings" is both an enduring characteristic (25 percent) and a high stress item (20 percent). As with managers, board members apparently feel they must hold back in their meetings which, again, is compounded by the fact that "I must attend meetings to get my job done" is high on both lists (71 percent and 22 percent, respectively). If the use of lay boards is to fulfill its potential, this group must be helped to feel that what they have to offer is valuable. This group contains many highly educated individuals who, for the most part, are trained in fields other than the social sciences.
Social Workers
Again the picture for social workers is much like the overall one presented in Table IV with certain exceptions. The most notable ones include "Management expects me to interrupt my work for new priorities" (24 percent) as an enduring characteristic of the job and "I lack confidence in management" (23 percent) as both an enduring characteristic and a high impact item. Additionally, social workers identify "I do not receive the right amount of supervision" (22 percent) and "I get feedback only when my performance is unsatisfactory" (16 percent) as conditions that impact highly on them. A full 49 percent say that "The demands of others for my time are in conflict" is a high stress item impacting on them. Overall, social workers tend to identify more items as stressful and occurring more frequently than any of the other groups; in other words, their additive levels of reported stress could be presumed to be the highest of all groups. This young, predominantly female, and relatively uniformly educated group appears to experience the most conflict and perhaps resentment in their roles when compared to the other groups. One is tempted to speculate that members of this group believe they know what they should be doing but feel prevented from doing it.
Of importance in this study is the question of the extent to which some social workers' particular frustrations may be communicated to and impact upon other personnel, especially the residential workers. There is a need for careful design of the social work role to reduce the level of stress and conflict for social workers and to ensure that their impact on other groups will be constructive.
Care Workers
The difference between the overall reports and those of the care workers are important ones. In terms of organizational changes, the careworkers share with other groups the point that they find it particularly stressful when they are required to work more hours due to crisis (37 percent), when frequent changes in instructions occur (27 percent), and when they acquire a new boss (24 percent). The latter appears to have occurred with some frequency with 49 percent of the careworkers reporting that this is a frequently occurring organizational event.
Unlike the other personnel, careworkers say that "Encountering a major change in my work schedule" is one of the most frequently occurring changes (51 percent) and is also in the top five most stressful items for them (27 percent). Also of relevance is the fact that while 65 percent say they have acquired new co-workers, this does not appear as a change that is particularly stressful for them. What is stressful is "Experiencing the transfer, resignation, termination, or retirement of a close friend or valued colleague" (21 percent). Apparently it is not the mere fact of turnover that careworkers find troubling but the quality of the loss.
In terms of working conditions, careworkers share with other groups that they find it particularly stressful when there is too much to do and too little time in which to do it (24 percent), when the demands of others for their time are in conflict with one another (23 percent); when they have differences of opinion with supervisors (20 percent); and when decisions that affect them are made without their knowledge or involvement (18 percent).
What sets the careworkers apart is the item "I am unclear what is expected of me" which 24 percent say is their most stressful condition of work. It is important to note that this item does not appear in the careworkers' list of most enduring conditions, hence, it would appear that this is not a frequently troubling issue with this group of workers, but when it does happen the impact is high. Unclear role expectations have been associated with a variety of negative outcomes (Van Sell, Brief, & Schuler, 1981). Beddoe (1980) points out that poor communication is perhaps the greatest cause of stress amongst residential workers and that the failure to maintain clear communications, including expectations, renders uninformed groups powerless to make decisions, and make them feel alienated from other members of the helping team.
Of interest, is the fact that while 44 percent rate "I must attend meetings to get my job done" as an enduring characteristic of their jobs, and 26 percent say they must exercise caution about what they say in meetings, they do not rate these items as ones that cause high stress. Apparently they want to have the meetings. This is consistent with Kane's (1975) observation that it is often the personnel who have the least power and status who advocate for teamwork. Perhaps some workers are seeking direction or clarification of expectations in meetings.
It is also possible that attending meetings is one of the few ways that careworkers have to regulate their face-to-face contacts with residents. In any event, the most obvious antidote to the careworkers' chief sources of stress would be to emphasize their involvement in clarifying expectations and to protect their work schedules from major or frequent changes.
Houseparents
In terms of organizational changes, houseparents share with the total sample (Table IV) that the most stressful ones are the requirement to work more hours due to crisis (56 percent); experiencing an increase in the pace of their work (29 percent); and undergoing frequent changes in instructions (21 percent).
On the other hand, one-third of the houseparents report "Experiencing an increase in the number of positive recognitions of my accomplishments" and "Encouraging a major change in my work schedule" as having occurred but neither of these items is reported as stressful. However, "Undergoing a major reorganization" (18 percent); "Experiencing a sudden decrease in the number of positive recognitions of my accomplishments" (15 percent); and "Experiencing an in-crease in status" (15 percent) are reported as stressful. It would appear that the role of live-in houseparents is one which places the staff member in a position where they are particularly sensitive or vulnerable to changes in recognition and/or status, both positive and negative.
The working conditions reports for houseparents are also much like the overall reports except that "Getting feedback only when performance is unsatisfactory" is not an issue for them (Table V). Instead, "I do not receive the right amount of supervision" is fifth of the five most enduring working conditions (17 percent) but is not reported as a stressful item. Unlike the overall group (Table VI) houseparents do not feel stressed because they are "fighting fires" but they do identify "I am cautious about what I say in meetings" (18 percent) as troubling for them. Yet houseparents, like the careworkers, do not say that meetings are particularly stressful even though they describe attending meetings as an enduring characteristic of their jobs (44 percent). Over one-third (38 percent) identifies "The demands of others for my time are in conflict" and 28 percent identify "I have differences of opinion with my supervisor" as having high stressful impact upon them.
The reports of the houseparents support the notion that workers who live in (i.e., have the most limited opportunities for withdrawal, repair and recovery) are most vulnerable to having their personal feelings on the line from moment-to-moment. This would seem to be even more problematic when events and conditions lend themselves to the kind of overwork, crisis orientation, and conflicted expectations to which these houseparents are exposed, and which they also report as stressful. Supervision, including support, for this group must be geared toward keeping closely in touch with how they are interpreting events.
Discussion
From this preliminary analysis support is found for the inclusion of all relevant personnel in an assessment of stress. While the themes of overwork, crisis and conflict are pervasive ones for all sample subjects these may be somewhat universal characteristics associated with the sort of "bedlam" we know exists in many human service settings (Gamer, 1982). Perhaps of greater immediate importance is the different perceptions held by various groups. Subsequent analysis of the qualitative and quantitative data in this study will serve to further illuminate the situations and conditions that the various groups of personnel find distressing and that aggravate teamwork.
Also the rationale for moving away from linear, cause-and-effect models of assessment is supported. The impact of stress on these personnel is neither universal nor uniform.
Experience suggests that this type of information can be used to promote respect, understanding and positive alignments between these groups of personnel. This type of assessment moves away from models that overemphasize individual or personal stress management only, and attempts to locate both common and unique stressors which can be addressed by all members of the team. Coaching workers to increase their personal tolerance or resistance to stress without addressing organizational and interpersonal issues is a little like combatting pollution by picking up pop bottles in the park-it is certainly necessary and makes things more attractive but it will not reduce pollution.
References
Adams, J.D. (1980). Understanding and Managing Stress: A Workbook in Changing Lifestyles. San Diego: University Associates.
Beddoe, C. (1980). Stress in Residential Work. A Report to the Residential Care Association: Platform for the 80's. Middlesex: Dare Graphic Printers.
Beehr, T., & Bhagat, R. (1985). Human Stress and Cognition in Organizations: An Integrated Perspective. New York: John Wiley.
Benveniste, G. (1987). Professionalising the Organization: Reducing Bureaucracy to Enhance Effectiveness. San Francisco: Jossey-Bass.
Brendtro, L.K., & Mitchell, M.L. (1983). The Organizational Ethos: From
Tension to Teamwork. In L.K. Brendtro & A.E. Ness (Eds.), Re-educating
Troubled Youth: Environments for Teaching and Treatment. New York: Al-dine.
Burford, G. (in progress)What Works? A Comparative Study of Staff Team Functioning in Group Homes in Newfoundland and Labrador. Ph.D. Thesis, University of Stirling, Stirling, Scotland.
Burford, G., & Fulcher, L. (1985). Resident Group Influences on Team Functioning. In L. Fulcher & F. Ainsworth (Eds.), Group Care Practice With Children. London: Tavistock.
Child Welfare League of America (1978). Standards for Group Home Service for
Children. New York: Author
Clemens, D.W. (1986). Staffing Patterns for
Child Care Workers in Residential Group Care. (6):
35-47.
Connis, R.T., et al (1979). "Work Environment in Relation to Employee Job Satisfaction in Group Homes for Youths." Child Care Quarterly, 8(2): 126-142.
Esland, G. (1980). Professions and Professionalism. In G. Esland and G. Salaman (Eds.), The Politics of Work and Occupations. Toronto: University of Toronto Press.
Fulcher, L.C. (1983). Who Cares for the Caregivers? A Comparative Study of Residential and Day Care Teams Working with Children. Ph.D. Thesis, University of Stirling, Stirling, Scotland.
Gamer, H.G. (1982). Teamwork in Programs for Children and Youth: A Handbook for Administrators. Springfield: Charles C. Thomas. Stressful Events and Working Conditions
Gentry, W.D., & Kobasa, S.C. (1984). Stress and Psychological Resources Mediating Stress-Illness Relationships in Humans. In W.D. Gentry (Ed.), Handbook of Behavioral Medicine. New York: Guilford Press.
Grossbard, H. (1960). Cottage Parents: What They Have to be, Know and Do. New York: Child Welfare League of America.
Kane, R. (1975). Interpersonal Teamwork. Syracuse University School of Social Work: Division of Continuing Education and Manpower Development. Manpower Monograph Number 8.
Krueger, M. (1987). Making the Team Approach Work in Residential Group Care. Child Welfare, 66(5): 447-457.
Mattingly, M.A. (1981). Occupational Stress for Group Care Personnel. In F. Ainsworth and L.C. Fulcher (Eds.), Group Care for Children: Concept and Issues. London: Tavistock.
Moos, R. (1975). Evaluating Community and Correctional Settings. New York: John Wiley.
Reed, M.J. (1977). Stress in Live-In Child Care. Child Care Quarterly, 6(2): 114-120.
Shostack, A.L. (1987). Group Homes for Teenagers: A Practical Guide. New York: Human Science Press.
Tosi, H.L., & Hammer, W.C. (1974). Organizational Behavior and Management: A Contingency Approach. Chicago: St. Clair Press.
Van Sell, M., Brief, A.P., & Schuler, R.S. (1981). Role Conflict and Role Research. Human Relations, 34(1): 43-71.
Waerness, K. (1984). The Rationality of Caring. Economic and Industrial Democracy, 5:185-211.
This article is reprinted from The Journal of Child Care , Vol.3 No.6, 1988, pages 95-109
