WORKING WITH THE HOMELESS:

Introduction

Shelters are the first step in responding to the problems of homeless people in America. The beds, meals and warmth provided by shelters mean the difference between life and death for hundreds of thousands of people who experience homelessness each year.

And shelters are more than emergency support institutions. Many provide medical care, most provide social support to their guests. Some shelters are the site for innovative service programs, including advocacy for affordable housing and welfare benefits, treatment for mental illness or alcoholism, and restoration of family functioning. Yet there is one element in common between all of these activities: it is shelter staff who make them work.

This report presents the results of a comprehensive survey of staff at Boston's Long Island Shelter. It seeks to describe the characteristics and orientations of those who are in the front lines of the response to homelessness and to learn from their experiences about working with homeless persons. The findings cannot be assumed to hold for other cities, even for other shelters in Boston, but the large size and varied programs of the Long Island Shelter do ensure that among its staff are represented a range of backgrounds and experiences.

Methodology

In the spring of 1988, self-administered questionnaires were left in the mailboxes of 48 of the Long Island Shelter's 70 paid staff (excluding maintenance workers and van drivers). These 48 staff were selected in order to yield a representative sample that could be used to draw conclusions about the staff as a whole (it was not possible to survey every employee due to limited resources). Since there were approximately twice as many employees in paraprofessional (counseling) positions as in professional positions (nursing, case management, administration), simply randomly selecting staff for the survey would have resulted in an inadequate number of professional staff for the purpose of generalizing about these employees. In order to include enough professional and paraprofessional staff to allow generalizations about both groups, a stratified random sampling procedure was used: all staff in positions classified as professional were included and a systematic random sample was drawn of one-half of the other staff (counselors).

After one reminder postcard and two follow-up letters with replacement questionnaires, the final response rate was 65 percent (N=31); 78 percent of the staff who actually received their questionnaires responded (some worked nights only and did not pick up the questionnaire in their shelter mailbox). All of the nonrespondents were in counselor (nonprofessional) positions. All results are reported separately in this report for the professional and paraprofessional respondents. This ensures that results are not distorted by the fact that the professionals were more likely to be selected for the study; in addition, the separate figures for professionals and paraprofessionals facilitate inter-group comparisons. Since the response rate was lower for paraprofessionals, all figures for them are more subject to error.

The 15 page survey included questions about the problem of homelessness, about the experience of working at the shelter, about relations with other service providers, and about the socioeconomic background of staff. Some questions were drawn from previous studies of social service employees; many questions made up highly reliable indices of staff orientations.

Staff Background

All of the professional staff had at least some college experience; 43 percent of the paraprofessional staff did not, although all had graduated from high school. (Table 1)[1] The average (mean) year of graduation for those who had completed college was 1979 for the professionals and 1985 for the paraprofessionals; the most common majors were sociology and nursing. Most of both groups had received some additional training about homelessness (86 percent of the professionals, 62 percent of the paraprofessionals). For professionals, this training was likely to have occurred at other agencies; for paraprofessionals it was more likely to have occurred at least in part at the shelter. (Table 2)

TABLE 1

Highest Grade In School Completed

Prof Paraprof

High School Degree 0% 43%

Some College 21 29

College Degree 37 29

Some Graduate Work 10 0

Graduate Degree 32 0

100% 101%^*

(19) (7)

*Sum does not equal 100% due to rounding error

TABLE 2

Special Additional Training

Prof Paraprof

At The Shelter (1) 19% 25%

At Other Agencies (2) 14 0

In Special Courses (3) 5 25

Other (4) 10 0

1 and 2 24 0

1, 2 and 3 0 12

1 and 3 10 0

1, 3 and 4 5 0

None (no answer) 14 38

101%^* 100%

(18) (5)

*Sum does not equal 100% due to rounding error

One out of three professional staff belonged to a professional association, as did about half as many paraprofessionals; trade union membership was equally common for paraprofessionals, but not for professionals. At least two in five professional staff belonged to political or advocacy groups concerned with homelessness, while none of the paraprofessional staff did. Equal proportions of both groups, about 15 percent, were members of religious groups concerned with homelessness. (Figure 1)

Women were a majority of both staff groups, but were somewhat more common in the paraprofessional ranks. (Table 3) Few staff were veterans, but those who were more likely to be paraprofessionals. (Table 4) Professionals were somewhat older, with a median age in the '30s--ten years higher than for paraprofessionals. Marital status was similar for the two groups: about two in five were married or living together, about two in five were single; the rest were divorced or separated. (Table 5)

The paraprofessionals were a somewhat more racially diverse group: more than one-quarter were members of minority groups, compared to 14 percent of the professionals. Between four and five of every ten in both groups identified themselves as Roman Catholic (Table 6), but the paraprofessionals reported attending church much more frequently than the professionals: 71 percent of the paraprofessionals attended church at least several times a year, compared to 38 percent of the professionals.

TABLE 3

Gender

Prof Paraprof

Female 57% 71%

Male 43 29

100% 100%

(21) (7)

TABLE 4

Veteran Status

Prof Paraprof

Yes 5% 17%

No 95 83

100% 100%

(21) (6)

TABLE 5

Marital Status

Prof Paraprof

Married 33% 29%

Divorced/Separated 19 14

Living Together 5 14

Single 43 43

100% 100%

(21) (7)

TABLE 6

Religious Preference

Prof Paraprof

Roman Catholic 55% 43%

Fundamentalist Protestant 5 0

Standard Protestant 15 0

Jewish 5 0

Other 10 29

No Religious Preference 10 29

100% 101%^*

(20) (7)

^*Sum does not equal 100 due to rounding error.

Shelter staff were conscientious about voting: between three and four in every five staff members had voted in one of the last three elections, with a slightly higher participation rate for professionals. (Table 7)

TABLE 7

Voting in Elections

Prof Paraprof

1986 Congressional Elections 67%(21) 57%(7)

Local Election in last 4 years 81%(21) 71%(7)

1988 Presidential Election (primary) 60%(20) 57%(7)

Shelter Work

Staff Activities

The distribution of specific jobs appears in table 8 for the two general job strata; the professionals were in assistant administrative or nursing positions; the paraprofessionals were primarily in counseling positions. (Table 8) In both job strata, almost all respondents were full time employees (Table 9); and about three-quarters were paid by the shelter, rather than by another agency. (Table 10)

TABLE 8

Job Title

Prof Paraprof

Director 5% 0%

Assist. Administrator 62 0

Counseling 0 88

Nursing 29 0

Office Staff 0 12

Volunteer 5 0

101%^* 100%

(21) (8)

* does not equal 100 due to rounding error

TABLE 9

Employment Status

Prof Paraprof

Paid, Full Time 85% 88%

Paid, Part Time 10 12

Volunteer 5 0

100% 100%

(20) (8)

TABLE 10

Payment for Work

Prof Paraprof

Paid by the Shelter 80% 75%

Paid by Another 20 25

100% 100%

(20) (8)

Staff worked to assist guests with a range of problems; in each area, staff efforts involved a mix of direct service and referral to other agencies. The most common directly provided service activities were providing food and beds, help with drinking problems, help with personal crises, and help with physical health problems. Help with mental health problems and with family problems was provided somewhat less frequently. Help with job training and placement and with child care were almost never provided directly, on average. (Figure 2)

Professional and paraprofessional staff differed in the frequency with which they provided help to guests in several areas: paraprofessional staff helped with providing food more often than professionals, while professional staff provided help with job training, financial benefits, and all aspects of health (physical, mental, drinking) more often than paraprofessionals.

Referrals were most common for mental and physical health and drinking problems. Referrals were also relatively common for food and beds. Less common were referrals for help with personal crises, financial benefits, and job training/placement. The problems for which referrals were least common were family problems and child care.

Professional and paraprofessional staff differed in the frequency of making referrals for several problems: professionals were more likely to make referrals for help with mental and physical health problems; they were less likely to do so for food and beds than were paraprofessionals.

Specific staff service activities ranged from clerical work to housing search. The most common activities were paperwork and answering phones. Contacting agencies and attending staff meetings were somewhat less common. The least common activity was housing search, but training guests and participating in training sessions were also relatively uncommon. All other activities received mean scores about halfway between the extreme values of "never" (1) and "very often" (7): assessment, counseling/therapy, case consultation, crisis management, training other staff, outreach/advocacy. (Figure 3)

Professional and paraprofessional staff differed in the frequency with which they engaged in a number of these activities. Professionals engaged more often in attending meetings, contacting other agencies, assessing clients and providing consultation about cases. Paraprofessionals were more likely to answer phones at the shelter but both groups reported equal involvement in paperwork and in outreach, counseling and crisis management.

These multiple activities created a heavy, demanding workload for professionals. Two-thirds of the professionals reported that their workload was "very heavy," and ninety percent felt the load was at least "moderately heavy"; just 14 percent of the paraprofessionals labeled their work load as very heavy, but almost 60 percent found their work load at least moderately heavy. (Table 11) These multiple work activities also tended to create conflicting demands "very often" or "sometimes" for all of the professionals and for half of the paraprofessionals. (Table 12)

TABLE 11

Current Work Load

Prof Paraprof

Moderately Light 10% 43%

Moderately Heavy 26 43

Very Heavy 63 14

99%^* 100%

(19) (7)

TABLE 12

Frequency of Conflicting Demands

Prof Paraprof

Very Often 40% 0%

Sometimes 60 50

Not Often 0 50

100% 100%

(20) (8)

But these job demands did not stop many staff from participating in other activities on behalf of the homeless. Frequency of participation was measured on a scale ranging from never (1) to very often (7). Reading about the homeless, checking on guests' welfare and attending staff meetings were the most common additional activities, with average scores (after rounding) between five and six. The least common extra activities were volunteering or campaigning for the homeless, helping guests after work and working for a political campaign, with scores of about two. Staff were somewhat more likely to work take work home, work through lunch breaks, and donate to the homeless. (Figure 4)

Several extra work activities were more common among professionals than among paraprofessional employees: attending staff meetings, taking work home and volunteering for the homeless. Paraprofessionals and professionals engaged in the other special activities to a similar extent.

Job specificity, challenge, decision-making

Both professional and paraprofessional employees reported that their jobs were relatively clearly defined: staff reported that they "often" got enough facts and information to work their best, that everyone had a specific job to do and had procedures for dealing with "whatever situation arises." Conversely, staff reported that, on average, it was "seldom" that people made their own rules on the job. (Figure 5)

Levels of participation in decisionmaking varied with specific content areas, but tended to be higher for professional than for paraprofessional employees (Figure 5). On average, professional staff "seldom" participated in decisions to hire staff but "sometimes" participated in decisions to hire new staff and to adopt new policies and programs. Paraprofessionals, on average, "never" participated in decisions to promote staff and "seldom" participated in decisions to hire new staff and to adopt new policies and programs.

Jobs at the shelter tended to be challenging. On average, jobs required staff "somewhat" or "a lot" to learn new things, make a lot of decisions on their own, be creative, work very hard and do a variety of different things. In each of these areas, professional staff reported higher levels of job challenge than did paraprofessional staff. Paraprofessional staff were less likely to report that their job required a high level of skill and somewhat more likely to say that their job required them to do the same things over and over. (Figure 6)

Job satisfaction and commitment

In spite of the high workload and some conflicting demands, job satisfaction among the staff was quite high, particularly among paraprofessionals. (Table 13) About sixty percent of both groups reported that they were very satisfied with their job. Between three-quarters and one hundred percent of the paraprofessionals thought they would strongly recommend their job to a friend, would take it over again with no hesitation, and found it to be an ideal job; between 60 and 80 percent of the professionals reported the same feelings. Feelings that the job was very much like what was expected when respondents took the job were slightly less common. Few staff expected to leave work in the shelter within one year, although such expectations were slightly more common among professionals.

TABLE 13

Job Satisfaction and Commitment

Prof Paraprof

Very satisfied with job 60%(20) 62%(8)

Strongly recommend to friend 67%(18) 75%(8)

Do over, no hesitation 80%(20) 100%(8)

Current job is an ideal job 58%(19) 88%(8)

Job very much up to expection 60%(20) 50%(8)

Moderately likely-leave 1 yr 20%(20) 14%(7)

When asked about their satisfaction with specific aspects of the job, staff again indicated high levels of satisfaction. Feelings were most positive, approaching an average of "very satisfied," with coworkers, supervisors, and the work itself; average ratings of satisfaction with salaries and promotion opportunities were somewhat lower, in the "moderate" range. (Figure 7) Staff in professional positions were slightly less satisfied in each of these areas.

Shelter Operations

Relations Between Staff

Most staff agreed that the shelter's employees generally get along and that there was rarely confusion about who is responsible for what; however, professionals were less in agreement with both of these two positions. In terms of other aspects of inter-staff relations, the opinions of professionals and paraprofessionals were very similar. Respondents were undecided, on average, about whether staff with college degrees saw things differently than other staff or got along with each other better than with other staff, about whether staff often disagreed about procedures for particular guests and whether staff should consult with professionals more often. Staff were also undecided about whether they should bar unruly guests more often. Staff disagreed, on average, that staff without college degrees get along better with the homeless. (Figure 8)

Satisfaction with Shelter Efforts

At the time of the survey, staff were satisfied with the shelter's service efforts in most areas. Satisfaction was highest with the shelter's efforts to provide food and beds and almost as high with respect to the shelter's physical health services; on a scale from 1 to 7, where seven represented "very satisfied," the shelter's efforts in these received an average rating of about six. Satisfaction with the shelter's efforts concerning drinking problems and mental health problems both were rated, on average, at just under five. (Figure 9)

The shelter's efforts to help guests in four areas received an average satisfaction rating of four--exactly in the middle between "not at all satisfied" and "very satisfied": job training/placement, personal crises, financial benefits, and family problems. Only one area, helping with child care, received an average rating below the middle point. In each of these areas but one, the satisfaction levels of professionals and paraprofessionals were similar; with respect to shelter efforts to help guests with family problems, paraprofessionals were more satisfied than professionals.

The overwhelming majority of staff believed that there had been "a lot of change" in service arrangements since the shelter first opened; staff also identified with the shelter as a whole. Most really cared about the fate of the shelter and were proud to tell others that they were part of the shelter. Most respondents also agreed, but somewhat less often, that the shelter inspired their best job performance and that they were willing to put in a great deal of extra effort for it. Respondents were almost neutral, on average, about whether their values and the shelter's values were very similar. In spite of these generally positive feelings about the shelter, professionals were somewhat less positive. (Figure 10)

Service in Boston

External Relations

Staff perceived some problems in the network of agencies delivering services to the homeless. On average, homeless persons requiring multiple services were seen as falling into cracks between agencies and gaps were identified in needed services; in both cases, professionals were more likely to perceive such interagency problems than paraprofessionals. (Figure 11)

On average, staff reported "some" disagreements or disputes between service agencies and some lack of coordination between agencies, but staff also believed that differences between service providers were worked out (to "some" extent). Few believed there was more than a little competition between agencies. Professional and paraprofessional attitudes about cooperation and conflict among service providers were similar.

Resources for the homeless in Boston as a whole were often seen as inadequate. The least adequate resource was transitional housing, receiving an average score close to 1 on a scale ranging from 1, for "inadequate," to 7, for "adequate." Resources for child care, drug abuse and mental health problems also tended to be viewed as inadequate. Average ratings of resource adequacy were slightly higher, though still closer to "inadequate" than to "adequate," for needs related to alcohol abuse, financial benefits, job opportunities, education/training and family counseling. Resources were judged more often as adequate than inadequate in only one area: physical health. Professionals judged resources as less adequate than did paraprofessionals in every area except for physical health care. (Figure 12)

In spite of the inadequacy of resources available for meeting most of the specific needs of shelter guests, staff judged relations with the local agencies concerned with these needs as cooperative. Relations with agencies concerned with physical health, alcohol abuse and mental health were rated as more cooperative than most; relations with agencies providing child care were viewed as less cooperative than others. In each area, professionals viewed relations as somewhat less cooperative than did paraprofessionals.

Staff felt a need for training about most of the problems found among homeless persons. Training was viewed as most needed, by both professionals and paraprofessionals, with respect to drug abuse and transitional housing. Professionals tended to desire more training, relative to paraprofessionals, in the health-related areas: mental health, physical health and alcohol abuse, while paraprofessionals were particularly concerned with training in child care issues. Needs for training were similar for the two occupational groups in the areas of drug abuse, transitional housing, financial benefits, job opportunities, education/training and family counseling.

Recent Changes in Boston

Most citizens learn of the problems of the homeless through occasional stories in the press or on television, or from a chance encounter with a homeless person. Shelter employees have much more frequent and intimate contact with the homeless; they are uniquely situated to comment on the conditions of the homeless and the quality of available services.

Shelter employees were asked six questions about changes in the number of homeless persons and in the availability of services for them over the preceding year (Spring, 1987-Spring, 1988). In general, the staff perceived that the problem of homelessness was increasing at the same time that services for the homeless were improving. (Figure 13)

More than three-quarters of the respondents, both professionals and paraprofessionals, believed that the number of homeless people in Boston had increased and that housing opportunities had worsened during the preceding year. However, most also believed that efforts to help homeless persons had increased. Of the three types of services mentioned, changes in physical health services were viewed the most favorably: none reported that physical health services had gotten worse and over one quarter found that physical health services had improved. More than two-thirds of the staff believed that opportunities for receiving social services and financial benefits had stayed about the same in the preceding year; the rest were about equally split between those believing service opportunities had gotten better and those believing that they had gotten worse.

Perceptions of change in two areas differed between professionals and paraprofessionals: proportionately more professionals viewed mental health services and publicity about homelessness as having changed for the worse than did paraprofessionals.

Beliefs about the Homeless

There is much disagreement about the nature of homelessness both among service providers and in the larger community: Are the homeless responsible, somehow, for their situation? Do the homeless need special transitional housing or just regular apartments? Do shelters need to hire professionals with college degrees to work with the homeless?

Perceptions of the Homeless

Staff estimated that many of their guests suffered from serious health problems. Alcohol abuse was seen by both professional and paraprofessional staff as the most common problem, occurring among half of the guests; drug abuse was believed to be only slightly less common. Professional staff viewed chronic mental illness as equally common, but the average estimate by paraprofessional staff of chronic mental illness was only one-third. Both professional and paraprofessional staff identified physical illness or injury as a problem for slightly less than one-third of the shelter's guests. (Figure 14)

Beliefs about Services

Staff at the Long Island Shelter were asked for their opinion of the type of services or service approach homeless people need with a series of eight statements; each statement presented two alternative perspectives. Staff responses to these statements indicated a perception that the homeless need special services, not just more housing opportunities; however, opinions were mixed about the need for staff with professional training and about the nature of homeless people. (Figure 15)

Professional and paraprofessional staff differed in their beliefs about the nature of homeless people. Paraprofessionals were slightly more likely to agree that "the homeless are responsible to some extent for their homelessness; they're not just victims of social problems" and that "the homeless are just regular people, like the rest of us; they're not so different."

The most widely shared belief about services for

the homeless, agreed to by more than nine out of every ten respondents and about as often by paraprofessionals as professionals, was that "the homeless need a lot of social services to get back on their feet; just giving them financial benefits is often not enough." Most respondents also agreed that "the homeless need supervised, transitional housing, that they're not ready to live on their own," but this belief was more common among professionals than among paraprofessionals.

Professional staff were more likely to support a formal professional approach in shelter work than were paraprofessional staff: although most staff in both occupational groups disagreed that staff should avoid intrusive procedures like intake interviews, such disagreement was more common among professional staff. Similarly, professional staff were more likely to disagree that staff should have experienced poverty--that a college education is not so important--than were paraprofessional staff, although most paraprofessionals also rejected this proposition. Overall, opinions were divided about whether staff needed to be professionally trained and whether staff should be somewhat detached in their work, rather than getting personally involved; but professionals were more likely to agree with both of these ideas.

Relations Between Education, Work Type and Orientation

Since professional employees differ from paraprofessional employees in their average level of education as well as in the extent of their involvement in direct service delivery, these differences could explain their different orientations. The next tables indicate how some employee orientations varied in relation to employee education and/or involvement in direct service delivery, without regard for occupational classification.[2]

Education was associated with what can be called a professional view of the problems of the homeless. Staff with more education were more likely to agree that homeless persons needed supervised (transitional) housing and that intrusive client interviews are useful and to disagree that the homeless are just like regular people. More educated staff were more likely to agree that shelter staff need professional training but do not need to have experienced poverty; however, staff without college degrees were more likely to believe that staff should be emotionally detached in their jobs. (Table 14)

TABLE 14

Orientations to Homeless By Education

Education

Less Than College College Degree

Homeless need transitional housing 46%(10) 95%(17)*

Homeless responsible for situation 15 (10) 35 (16)

Homeless just like regular people 86 (11) 48 (17)

Staff should be professionally trained 31 (10) 81 (17)

Staff should have experienced poverty 14 (11) 5 (17)

Staff should avoid intrusive procedures 15 (10) 0 (17)

Homeless need many social services 86 (11) 95 (17)

Staff should be emotionally detached 85 (10) 52 (17)

* Percent Agree or Strongly Agree (Number of cases, after weighting)

Perceptions of the quality of relations among staff also varied with education: staff with more educational experience were less likely to believe that staff get along and were more likely to believe that staff disagree over procedures, but staff with college degrees were less likely to believe that degreed staff are different from other staff. Greater involvement in service delivery was also associated with perceptions of a lower level of staff harmony. (Table 15)

TABLE 15

Staff Relations By Education, Service Involvement

Education Service Involvement

LT college college low high

Staff get along 100%(11) 90%(16) 100%(13) 89%(14)*

Rarely confused 77 (10) 75 (16) 88 (13) 59 (14)

Degreed staff different 36 (11) 20 (16) 25 (13) 33 (14)

Staff disagree/procedures 14 (11) 35 (16) 0 (13) 50 (14)

Staff should consult more 14 (11) 25 (16) 12 (13) 28 (14)

Degreed staff get along 14 (11) 10 (16) 19 (13) 11 (14)

Staff without degree get

along better with poor 0 (11) 5 (16) 6 (13) 0 (14)

Bar unruly guests 14 (11) 20 (16) 12 (13) 22 (14)

* Percent Agree (Number of cases after weighting)

Although satisfaction with the shelter was relatively high overall, satisfaction was not uniform across the staff. Staff with more involvement in direct service delivery were less satisfied with the shelter and with their jobs; they were also more likely to put in extra time at work. Greater participation in decision-making on the job did not seem to influence level of satisfaction, although those who were more involved in making decisions were less likely to be thinking of leaving their jobs and were more likely to put in extra time at work. (Table 16)

TABLE 16

Work Orientation By Service Involvement,

Participation in Decisions

Service Participation

Involvement in Decisions

low high low high

More Satisfied with Shelter 62%(13) 11%(14) 32%(15) 41%(14)

More Satisfied with Job 62 (13) 19 (17) 35 (16) 53 (14)

Moderately Likely to Leave 25 (13) 22 (14) 25 (16) 19 (13)

Less Extra Work 68 (13) 50 (14) 65 (16) 44 (13)

Staff more directly involved in service delivery were less satisfied with the level of resources and of cooperation in the local service delivery network. (Table 18)

TABLE 18

Network Functioning By Service Involvement

Service Involvement

Low High

Less Resource Adequacy 21%(11) 68%(15)*

Less Agency Cooperation 23 (10) 67 (14)

* Percent Agree (Number of cases after weighting)

Conclusions

Work at the Long Island Shelter in 1988 was challenging, but also rewarding for many staff. Paperwork and other bureaucratic activities required much staff time, but networking with other agencies and responding directly to guests' needs were also common. Staff generally agreed with the shelter's goals and were satisfied with its efforts for homeless persons. Many staff spent time outside of work as advocates or volunteers on behalf of the homeless.

The mix of staff activities reflects a shelter that is first and foremost oriented to emergency needs--providing food and beds, helping with crises and physical health problems--but that also includes services for guests in need of help with drinking or psychiatric problems.

According to the staff, service arrangements at the shelter had changed a lot since it first opened; much of this change involved developing a broader range of services at the shelter. Staff responses highlighted several areas in which further expansion of services should be considered: Little staff time was devoted to providing job training or placement services to guests, to aiding in housing searches, and to helping with child care.

Shelter staff reported high levels of satisfaction with their work and with the shelter. Shelter jobs were very challenging, relations among staff were positive and commitment to the work was high. Although some staff were concerned with pay levels and promotional opportunities, the staff survey provides persuasive evidence that the Long Island Shelter is managed well.

Staff generally shared what can be called a "professional" approach to work with the homeless: believing in the need for detachment, formal needs assessments and special services. There were some differences of opinion about whether the shelter's guests can be viewed as "regular people" not unlike the staff; in addition, there were some indications that staff occasionally disagreed about procedures for specific guests and about the importance of consulting with professionals. Shelter leaders may want to consider whether more discussion about these issues is needed to develop a flexible approach that most staff will view as appropriate.

The shelter's basic mission of providing food and beds and responding to physical health problems was carried out to the satisfaction of most staff. The staff reported little involvement with families and children relative to single adults, but staff indicated that there should be improvements in efforts to help this group.

Staff also indicated there was room for improvement in the network of agencies delivering other services to the homeless. Gaps in services between agencies need to be closed and resources need to be increased in several areas--particularly transitional housing, drug abuse services and mental health services. Staff estimates of the frequency of drug abuse among the shelter's guests highlight this area of concern, while the staff's perception that housing opportunities in Boston are declining emphasizes the need for more attention to housing problems.

Staff experiences and orientations cannot be understood apart from their backgrounds and jobs. Staff in positions classified as professional had, overall, higher levels of education and training than those classified as paraprofessionals; the professionals also tended to be older. Paraprofessional staff were more racially diverse and attended church more often, while professionals were more likely to participate in advocacy and other activities on behalf of homeless persons outside of work.

Professionals and paraprofessionals differed in the focus of their shelter work. Paraprofessionals were more involved with the provision of food and beds, while professionals were more involved in providing health care and referring guests to other agencies; and professionals were more likely to report problems in relations with service agencies, to perceive resources in the city as inadequate, to describe their workload as "very heavy" and to experience conflicting demands in their work.

The organization of shelter work also differed for professionals and paraprofessionals. Professionals were more likely to report that they were involved in decisionmaking on the job and that their jobs were challenging. Nonetheless, professionals expressed somewhat lower levels of satisfaction with their jobs than did paraprofessionals. In addition, professionals perceived somewhat more problems in relations between employees and identified somewhat less with the shelter. But, overall, the attitudes of professionals about shelter work were more similar to than different from the attitudes of paraprofessionals.

Perceptions of homeless people by professional and paraprofessional staff were generally similar. However, professionals tended to perceive a higher level of mental illness among shelter guests and to believe that homeless people needed more services and a more professionally oriented service approach.

Variation in educational backgrounds and service delivery involvement among staff had influenced work orientations in ways that could account for some of the effects of job classification: a college education tended to be associated with the belief that the homeless need more services, while staff who were more directly involved in service delivery were somewhat less satisfied with their work, with the shelter and with relations between other staff; they were much less satisfied with the surrounding service network. Direct efforts to respond to the grave needs of the shelter's guests thus appear to generate stress and more criticism of service arrangements. This suggests the need for ongoing evaluation of the effectiveness of service arrangements both within the Shelter and in the surrounding community; the greater stress experienced by direct service staff also suggests that some variation in work assignments may be appreciated by some staff. On the other hand, efforts to decrease turnover among the staff could be made more effective by involving staff more in decisionmaking.

The staff survey documents the Long Island Shelter's multi-faceted service approach and its success in engaging a wide range of staff in service delivery; survey findings also suggest directions for further development of the shelter's approach. The assistance of the shelter's directors and staff with the survey is yet another indication of the Long Island Shelter's commitment to developing innovative and effective approaches to the needs of homeless persons.

[1]Since paraprofessional and professional employees are treated separately in the following sections, all figures in these sections are based on the actual number of cases sampled.

[2]Since paraprofessional and professional employees are combined in this section, all figures presented are based on the weighted number of cases. This represents more accurately the actual prevalence of paraprofessionals and professionals in the shelter workforce.