The 1995-1996 Transitional Work Study

 

 

 

 

 

 

City of Boston Homeless Services:

EMPLOYMENT & TRAINING

FOR HOMELESS PERSONS

 

 

 

Francine Byrne, MA, Russell K. Schutt, PhD

 

&

 

Lorenzo Boyd, Erica Jablonski, Kelly Kolodny, Shari Mitchell, Yuko Tsutsui

 

Diana Brensilber, James butler, Charles Butts, Mary Harrington,

Richard Hodgson, Charles Kaufmann, John Kiernan, Cheryl Krochmal,

Anne Poster, Stephen Price

 

 

Graduate Program in Applied Sociology

University of Massachusetts-Boston

 

 

 

 

Prepared for and in cooperation with

 

Carol Fabyan, Director of Social Services

&

Beth Grand L’Heureux, Administrator of Services (Long Island Shelter)

Boston Homeless Services

 

We are grateful for the support of Richard Weintraub, Director, City of Boston Homeless Services and the many shelter staff and guests who contributed to the study.


 

The 1996 Transitional Work Study............................................................................................................

Work Programs at Long Island & Woods-Mullen Shelters..................................................

The Data Collection Process.......................................................................................................................

Measures.................................................................................................................................................................

Sampling..................................................................................................................................................................

Analysis...................................................................................................................................................................

Work Involvement................................................................................................................................................

Participation in the Work Programs...................................................................................................

Working Outside of the Work Programs..........................................................................................

Interest in Working........................................................................................................................................

Experience in Other Programs.................................................................................................................

Importance of Job Training.......................................................................................................................

Work Involvement Summary...................................................................................................................

Participant Attitudes Toward Work Programs..........................................................................

Motives for Work Program Participation......................................................................................

Satisfaction with Work Programs.....................................................................................................

Program Attitudes Summary..................................................................................................................

Satisfaction with the Shelters in General......................................................................................

Summary of Shelter Satisfaction.......................................................................................................

Conclusions............................................................................................................................................................

Appendix.....................................................................................................................................................................

References................................................................................................................................................................


EXECUTIVE SUMMARY

·         The 1995-1996 Transitional Work Study was a class project of the Master’s Program in Applied Sociology at the University of Massachusetts-Boston.  Students evaluated employment and training programs at Boston’s Long Island and Woods-Mullen shelters, using in-depth staff interviews, focus group discussions with staff, observations of program activities and meeting, and interviews with samples of work program participants and non-participants.

·         Interviews were conducted with all available work programs participants (N=50) and a stratified random sample of non-participants (N=48).

·         One in five guests participated in work programs; most of them at the Long Island Shelter.

·         Work program participants differed from non-participants in several respects.  Participants tended to be younger and in much better physical health.  They were less likely to believe that getting a job was mostly a matter of luck and were not as likely to evince many signs of depression.  They were more likely to have worked in the past and had been staying at the shelter longer.

·         Work involvement and interests differed markedly between Long Island and Woods Mullen guests who were not work program participants.  At Long Island, just a few of these guests were working or looking for work at the time of the survey, as compared to one in four of the guests at Woods Mullen.  About half of the guests were interested in help with finding work, but again this was more common at Woods Mullen.  Almost half of the nonparticipants at Woods-Mullen had applied to a work program in the past, compared to less than one in five at Long Island. 

·         Job training was seen as extremely important by two-thirds of the work program participants, compared to just 16% of the general shelter population.

·         Better physical health was associated with most aspects of work involvement.  High levels of substance abuse and symptoms of depression were associated with less likelihood of work involvement.

·         Members of minority groups were more likely to participate in shelter work programs than white guests but less likely to be working outside of the shelter.

·         Those working outside of the shelter reported lower levels of overall social support.

·         Men reported more work experience than women, but women were more interested in help with finding a job.

·         Participants were very satisfied with the work programs in general and specific program aspects, including the work experience itself and the program staff.  Satisfaction with pay levels was not as high, particularly at the Woods Mullen shelter . 

·         Work program participants were very interested in improving their skills preparing for a new job.  They tended to be comfortable with searching for a job, although somewhat less so at Woods Mullen

·         Satisfaction with the shelters was high at both shelters and among both program participants and nonparticipants.  Nonetheless, Long Island guests were dissatisfied with the lack of the shelter’s proximity to stores and shopping, and Woods-Mullen guests were very dissatisfied with the lack of privacy at their shelter.

 


            Work serves both direct and indirect functions for persons who are homeless and have been out of the labor force.  Working affirms directly the employee’s participation in an accepted community role.  Indirectly, working enhances independence through increased income and improved functioning.  For many homeless persons, regaining a work role is a foundation for regaining residential and social stability.

Transitional work programs help many homeless persons to reap these benefits even as they provide shelters and other service organizations with an available and economical labor force.  However, in spite of these potential benefits, many homeless persons do not participate in available work programs or do not complete them.  Some transitional work participants begin to see their shelter-based job as adequate in itself, and lose interest in competitive employment.

The 1996 Transitional Work Study

          The 1996 Transitional Work Study was part of Boston’s Homeless Services’ ongoing efforts to evaluate and refine transitional work programs for homeless persons.  The study focused on orientation to work at the Long Island and Woods-Mullen shelters, with orientation to work conceived in terms of work-related attitudes and behaviors.  Primary study goals were to describe guests’ work orientations and to identify guest characteristics associated with these orientations. 

More specifically, we report on the distribution of shelter guests across work programs, the proportion of guests who are working or looking for work, and the level of interest in job training and in help with finding a job.  We describe the degree of satisfaction with the work programs as well as with the shelters in general.  In addition, we examine the association between these aspects of work orientation and guest sociodemographic characteristics, work experience, health problems, and social supports.

Work Programs at Long Island & Woods-Mullen Shelters

          The Long Island Shelter was opened in 1983 and now is Boston’s largest single adult homeless shelter, with beds for 356 (and 25 more in overflow accommodations) guests.  The Woods Mullen Shelter, managed like Long Island by Boston’s Homeless Services, has 160 beds.  At both shelters, 60 beds are reserved for women. In order to stay at Long Island Shelter, which is on an island in Boston’s harbor, clients must first pass through a preliminary intake at the Woods Mullen Shelter (at Boston City Hospital) and then board a bus.

          Four work programs operate out of the shelters. Each is geared toward a different range of client need and functioning.  The Work Experience Program at Long Island and the Work Experience Program at Woods Mullen emphasize rudimentary job readiness and are designed for clients who are motivated to break out of their cycle of homelessness but lack the skills and experience to do so.  These programs are considered the first step toward outside employment and are primarily focused upon providing a structure of employment for the clients by establishing work schedules, assisting in money management, and encouraging clients to confront personal issues.

          The Serving Ourselves Program is the largest work program at the shelters (designed for 32 individuals, with 2 formerly homeless mentors).  It is located at the Long Island Shelter and targets individuals who already have some degree of work skills and experience, allowing them to participate for 6-12 months, with an additional 3 months of follow-up.  The goals of Serving Ourselves are to increase the residential stability, skill level and income of program participants and to create greater self-determination through self-esteem and personal responsibility.  Specialized skills training is offered in the areas of culinary arts, basic kitchen, inventory control, laundry and maintenance.  The program provides food and laundry services for the City of Boston’s emergency shelter system as well as to other institutional clients.

          The Employment Services Program offers job placement services and contract beds for shelter guests who already work outside of the shelter or are actively looking for employment.  It is designed as a system of support for clients who do not necessarily need increased experience or skill and provides numerous services for the various needs of its clients.  For those who have jobs outside of the shelter, it provides contract beds and Individual Service Plans outlining goals and activities to assist clients moving into permanent housing.  For individuals who are not employed but are actively looking for work, it provides job search assistance, contract beds and Individual Service Plans. 

          The Literacy and Adult Education Program at the Island Shelter provides tutoring, classes, assessment, educational counseling and referrals, and assistance with college applications and financial aid.  Classes in adult basic reading, writing and math, preGED, ESL, and computer literacy are available in a special classroom at Long Island.   Tutoring is tailored to the needs of students who are in prevocational programs as well as those with learning disabilities.  Classroom staff seek to maintain a flexible, individualized approach and a supportive environment for all participants. 

          At the time of the survey, the Long Island Shelter had just opened a new transitional housing program for work and training program participants.  Project S.O.A.R. provides transitional supportive housing on Long Island for up to 70 homeless men and women.  The program offers comprehensive services to help program participants move successfully out of the shelter into the community.

The Data Collection Process

 

          All survey participants were interviewed by graduate students with an approximately 45-minute interview schedule.  The interview schedule was designed through the compilation of existing measures as well as the incorporation of information obtained through open-ended interviews with clients and a staff focus group.  The questionnaire sections included: Demographic Information, Service Satisfaction, Educational History and Needs, Residential History, Employment History and Needs, Physical Health, Social Support, Mental Health, Substance Abuse and Legal Issues.  Work program participants were questioned about their level of program satisfaction.  Non-participants were questioned in more detail about their work interests and history.

          Several shelter guests and staff participated in focus group discussions or intensive interviews at an earlier stage of the study.  Comments from some of these discussions and interviews are included in several report sections.

Measures

          Five aspects of work orientation were measured, some of them being appropriate only for work program participants or non-participants.  The perceived importance of job training was measured for all survey respondents as was participation in the transitional work programs.  Work orientation of non-participants was further measured by whether or not they were looking for work.  Non-participants were also asked whether they were currently employed, either full or part-time; for some analyses, these responses were combined with work program participants’ answers to the question of whether they had been working before they joined the work program.  (Specific questions and response choices are in the appendix.)

          Guest characteristics identified as possible influences on work orientation were age, gender, education, race, physical health, work history, substance abuse, social support, length of time since first coming to the shelter, feelings of distress, and belief that luck is not the primary reason for obtaining a satisfying job.  (Specific questions and response choices are in the appendix.)  These measures are individually cross-tabulated with the measures of work orientation. Only relationships that are statistically significant at the .1 level (using Pearson’s chi-square test) are reported. 

Sampling

          All available work program participants were interviewed, with a total of 50 participating (see figure 1).  This number represented 54% of the individuals registered in these programs during the study period; others could not be reached for interviews.  The rate of participation was highest for the Work Experience Program participants (73% at Long Island and 62% at Woods-Mullen), but relatively low for the Serving Ourselves Program (41%).  An almost equal number (48) of non-work program participants were selected randomly from nighttime bed lists.  Responses were then weighted by race, sex, shelter and program participation to increase the correspondence of the sample characteristics to those of the actual shelter population during this period.

          Because the work programs target different populations, the general characteristics of the participants vary.  However, due to the small sample size, distinctions between the programs are not made in most of the analyses reported here.  Other programmatic boundaries also influence the characteristics of the transitional work program population.  In order to participate in any of the shelters’ work programs clients are prohibited from using drugs or alcohol.  If substance abuse is an issue for them, they are required to attend Alcoholics Anonymous or Narcotics Anonymous meetings.  As many of the jobs require physical activity, guests with significant physical disabilities or poor health cannot participate.  Due to concerns that the guests may become overly comfortable in the shelter setting, clients who are very young or very new to the shelter system are generally prohibited from participating.

Figure 1

 

 

Analysis

The report is organized in four sections: work involvement, reactions to the work programs, satisfaction with the shelters in general, and characteristics of shelter users.  The first two sections are the focus of the report.  The work involvement section describes the extent of participation in the work programs, the prevalence of working outside of the programs, frequency of job hunting, and interest in job training.  Graphs summarize the distributions of each of these measures for work program participants and other guests at the two shelters.  Crosstabulations are then used to show the relationship between these measures and other personal characteristics.  Relationships are examined with indicators of guest social backgrounds, orientations, and health.  Specifically, these indicators are gender, education, race, age, physical health, substance abuse, social support, depression, time since arrival at the shelter, and belief that finding a satisfying job is a matter of luck (“locus of control”).  Only those relationships judged to be statistically significant are presented in the report.  Because of concerns expressed by shelter management, special attention is given in these analyses to the role of gender.  Results of these analyses involving gender are presented in the summaries for the first two major sections.

The next section also uses graphs to describe the distributions of responses to the program satisfaction questions asked of work program participants.  In the final section, graphs display the distribution of satisfaction with the shelters for the work program participants and for other guests using the two shelters.

Work Involvement

            Work involvement is described in terms of participation in work programs, working outside of the programs, self-reported job search, and interest in working.

Participation in the Work Programs

            The weighted sample approximates the distribution of shelter guests across the work programs and the two shelters.  Two-thirds of the guests stayed at the Long Island Shelter (figure 2).  One in five guests participated in one of the work programs, most at Long Island.

Figure 2

 

 

 

            Serving Ourselves was the largest work program at either of the shelters, with 8% of the guests (figure 3).  Three percent of the guests at either shelter were enrolled in the Long Island Work Experience Program, while less than 2% were in this program at Woods-Mullen.  while only 4.7% reported participation in Employment Services.

Figure 3

 

Compared to other shelter guests, work program participants were more likely to be of minority ethnicity, younger, and in much better health (table 1).  They also were less likely to believe that getting a satisfying job was mostly a matter of luck.  They did not differ in terms of gender, education, substance abuse and level of perceived social support.

Table 1.  Participation in Work Programs by Age, Locus of Control, Health      

 

                        Age**                                                                  JOB DUE TO LUCK           PHYSICAL Health

Work training programs

20-29

 

30-39

40-49

 

 

50+

No

Yes

Very good

Good

Fair

Poor

No

 

70.7%

61%

89.8%

97.8%

59.5%

89.4%

24.4%

54.4%

71.4%

77.8%

Yes

 

29.3

39

10.2

2.2

40.5

10.6

75.6

45.5

28.6

22.2

Count

Total

16

100%

45

100%

24

100%

13

100%

54

100%

33

100%

41

100%

33

100%

14

100%

9

100%

                           x 2=22.75,  p=.000                           x 2=7.10,  p=.008       x 2=15.95,  p=.001             

 

 

Work program participants were also more likely to have worked full time in the past, to have low (but not very low) symptoms of depression, and to have been at the shelter longer (table 2).

Table 2.  Participation in Work Programs by Job History, Depression and Time Since Arrival at the Shelter

 

                                Job history              Depression Scale**                                    Time since arrival*

Participation in Work Training Programs

No

Yes

Very Low

Low

High

Very High

0-3 mos.

4 mos-1 year

1+ to 2 years

Over 2 years

No

 

96.3%

79.9

94.6%

56.5%

79.1%

94.3%

95.6%

67.5%

88.3%

76.9%

Yes

 

3.7

20.1

5.4

43.5

20.9

5.7

4.4

32.5

11.7

23.1

Count

Total

8

100%

90

100%

17

100%

21

100%

45

100%

12

100%

16

100%

32

100%

17

100%

33

100%

                             x 2=5.75,  p=.016        x 2=16.68,  p=.001                                  x 2= 8.49,  p=.037

 

*              p<.05

**            p< .01

           

Working Outside of the Work Programs

            Work experience differed markedly between the shelters and program participants.  Half of the work program participants had been working before they started the program (figure 4).  One in four of those not in a work program at Woods-Mullen were working outside of the shelter, compared to just 5% of those who were not in a work program at Long Island.

Figure 4

 

            Among shelter guests who were not work program participants, those who were white or Hispanic, and those in good physical health, were more likely to be working.

Table 3. Employment by Race and Physical Health: General Shelter

 

                   Race*                                 Physical Health*

Currently Employed

Black

White

Other

Very good

Good

Fair

Poor

No

 

92.1%

79%

20%

100%

40%

75%

100%

Yes

 

7.9

21

80

0

60

25

0

Count

Total

14

100%

17

100%

5

100%

5

100%

15

100%

8

100%

7

100%

                 x 2=7.21,  p=.027                  x 2=111.33,  p=.01

 

Guests who reported some symptoms of substance abuse were more likely to be working than those who reported many symptoms or no symptoms of substance abuse (table 4).  Working was also more common among those who reported a lower level of social support.

Table 4. Employment Status by Substance Abuse, Social Support:

            General Shelter Population

 

                                Substance abuse**              Social Support             *                                             

Currently Employed

None

Some

Many

Low

Med.

High

No

 

98.7%

60.3%

84%

64%

82.7%

100%

Yes

 

1.3

39.7

16

36

17.3

0

Count

Total

12

100%

13

100%

11

100%

12

100%

14

100%

9

100%

             x 2=12.05,  p=.002                      x 2=6.16,  p=.046                    

*              p<.05

**            p<.01

Interest in Working

Among those guests who were not current work program participants, interest in working was more common at Woods Mullen than at Long Island.  Most striking, more than 40% of the Woods-Mullen guests who were not in a work program were looking for work, compared to fewer than 15% who were looking for work among the non-work program guests at Woods-Mullen (figure 5).  Interest in help with finding work was more widespread, expressed by about half of the guests, but again more common at Woods-Mullen.  About one-third of the guests at both shelters reported that they would like to get into a job training program.

Figure 5

 

            Among those in the general shelter population who were unemployed, looking for work was more likely for those who had been staying in the shelter for at least one year, rather than for more recent arrivals.  None of the recent arrivals were looking for work, compared to one-third of those who had stayed at the shelter for at least one year (table 5).

Table 5. Propensity to Look for Work by Job History, Substance Abuse, Social Support, Depression and Time Since Arrival at Shelter:

            Unemployed General Shelter Population

 

                            Time since arrival

Looking for Work

0-1 year

Over 1 year

No

 

100%

66.4%

Yes

 

0

33.6

Count

Total

13

100%

10

100%

 x 2=2.85,  p=.092

 

*              p<.05

**            p<.01

           

            Among those in the general shelter who were already employed, looking for work was more common among those who had at least a high school, those who reported being in good physical health, and those who had been staying at the shelter for at least one year.  In fact, none of those with less than a high school education, who were in poor health, or who had first arrived at the shelter within the past year  said they were looking for work (table 6).

Table 6.  Propensity to Look for Work by Education, Physical Health, Time since arrival

            Employed General Shelter Population

 

Education**          Physical Health*                       Time since arrival

Looking for Work

Less than

H.S.

H.S.

or more

Good

Poor

0-1 year

Over 1 year

No

 

100%

5.5%

75%

0%

100%

39.3%

Yes

 

0

94.5

25%

 

0

60.7

Count

Total

7

100%

5

100%

9

100%

2

100%

5

100%

7

100%

                                 x 2=8.40,  p=. 004.                         x 2=4.27,  p=.038                               x 2=4.28,  p=.038

 

 


Experience in Other Programs

            Shelter guests who were not participating in a work program were asked whether they had used the educational services at the Long Island Shelter.  Just two percent of the guests at Long Island, and 8% of those at Woods-Mullen, said that they had (figure 6).  (Many Woods-Mullen guests have stayed previously at the Long Island Shelter.)

Figure 6

 

            About half of the guests at both shelters who were not currently in a work program said that were familiar with the work programs (figure 7).  At Woods Mullen, about half of the guests who were familiar with the work programs had in the past been program participants.  However, at Long Island almost none of the guests who knew about the work programs had previously been in a work program.

Figure 7

            A substantial minority of shelter guests who were not participating in a work program had applied to one in the past—almost one in five at Long Island and one in three at Woods-Mullen (figure 8).

Figure 8

 

 


Project SOAR, the residential program at Long Island, had only begun several months before the survey.  No one in the Woods-Mullen sample had heard of it (figure 9).  Almost two-thirds of the Long Island guests were aware of SOAR, as were fully 80% of the work program participants.  A majority of the work program respondents were also SOAR participants.

Figure 9

Importance of Job Training

Job training was rated as extremely important by almost two-thirds of the work program participants, but by just 15.6% of the general shelter population.  About half of the general shelter population rated job training as not at all important, compared to about 12% of the work program participants.  On average, work program participants said that help with finding a job and with job training was considerably important (figure 10).  Most disagreed that finding a satisfying job was mostly a matter of luck. However, members of the general shelter population only attached moderate importance to help with finding a job and not much than slightly importance, on average, to job training.  Many believed that finding a satisfying job was a mostly a matter of luck.

Figure 10

            Importance attached to job training by work program participants was lower for those who were high school graduates than for those with more or less education (table 7).  More importance was attached to job training by those who were black than by those who were white or Hispanic.

Table 7

Perceived Importance of Job Training by Education, Race

            Work Program Participants

                                                                       

           Education *                            Race                                     

Importance of job training

Less than

H.S.

H.S.

Grad

More than H.S.

Black

White

Hisp.

Not at all

 

14.3%

7.3%

11.4%

13.2%

12.5%

0%

Somewhat

 

16.7

48.9

0

11.8

58.3

0

Extremely

 

69

43.7

88.6

75

29.2

100

Count

Total

23

100%

15

100%

12

100%

35

100%

11

100%

4

100%

    x 2=11.49,  p=.021                    x 2=7.858,  p=.096             

*              p<.05;  **                p<.01

            Importance attached to job training by shelter guests who were in a work program was higher among those who thought getting a satisfying job was not just a matter of luck, among those in better health, and among those who felt they had less social support (table 8).

Table 8

 Perceived Importance of Job Training  by  Gender, Education, Race, Age, Locus of Control and Physical Health :

            General Shelter Population

     Locus of Control**            Physical Health*                                      Social Support**

Importance of job training

Internal

External

Very good

Good

Fair

Poor

Low

Med

High

Not at all

 

68.9%

25.8%

8.3%

57.1%

40%

100%

16.8%

67.5%

85.7%

Somewhat

 

7

59.4

66.7

24.1

60

0

64.5

19.2

0

Extremely

 

24.1

14.8

25

21.4

0

0

18.8

13.3

14.3

Count

Total

17

100%

17

100%

10

100%

18

100%

6

100%

3

100%

13

100%

18

100%

6

100%

                                x 2=12.9,  p=.002                   x 2=12.81  p=.046             x 2=20.15,  p=.000.

 

Work Involvement Summary

Many shelter guests participated in one of the work programs, but a much larger number did not.  Would more guests participate if the capacity of the work programs was expanded?  Do guests who currently are uninvolved want to participate?  Would they be able to participate? 

Roughly half of the shelter guests who were not in a work program at the time of the survey were interested in working or a work training program.  There was a considerable difference between the two shelters.  Many of the Woods-Mullen guests who were not in a work program were working at the time of the survey, compared to few of the Long Island guests who were not in a work program.  Almost half of the Woods-Mullen guests who were not in a work program were looking for work, compared to a small percentage of the Long Island guests.  Interest in help with finding a job was also higher at Woods-Mullen than at Long Island, although about a third of the guests at both shelters rated job training as very important.  Clearly there is room for expansion of work programs at both shelters and, particularly at Woods-Mullen, for increasing assistance with job hunting.

Different aspects of work involvement varied with a variety of guest characteristics.  Although several of these correlates of work involvement, age, race, physical health, and substance abuse were associated with each other, their relationships with work involvement were mostly independent of these interrelations.

Self-reported physical health was the most consistent correlate of work involvement.  It was associated with participation in work programs, working outside of the work programs, looking for work, and the importance attached to job training.  Of course the shelter screened applicants for work programs to make sure that they were physically able to do the work required, so this could explain the association between physical health and work program participation.  However, the associations between each of the other work involvement variables and physical health suggest that this is a more generally important influence on work involvement and should be viewed as a potential barrier to work involvement for many shelter guests.  Further research is needed to determine how well self-reported health captures variation in illness and disability and to identify those aspects of physical health that interfere with working.  In any case, expectations for work involvement must take into account the problems created by physical health problems.

In addition to physical health problems, other health problems, substance abuse and mental illness, appear to diminish work involvement.  Guests who had many symptoms of substance abuse were less likely to be employed.  Those who were more psychologically distressed were less likely to participate in a work program.  When attention is given to health problems that inhibit employment, strategies for treating mental illness and substance abuse should also be considered.

Race was associated with several aspects of work involvement, in a way that may reflect larger labor market patterns.  Black and Hispanic guests were more likely to participate in shelter work programs and to rate job training as important.  However, white guests were more likely to be employed outside of the shelter.  These different relations might reflect greater employment opportunities for homeless persons who are white, either due to more prior training or to labor market openings.  The result could be that black and Hispanic guests turn to the work programs to increase their employability or, perhaps, to take advantage of an opportunity for employment that would be less attractive if they had more opportunities in the labor market outside of the shelter.  More intensive analysis of this issue should take account of prior job training, job offers received by shelter guests, and the association between race and shelter of residence (as statistics presented later indicate, interviewees at the Woods Mullen shelter, which is more convenient to jobs in the city, were more likely to be white).  It is also important to take into account the relative attractiveness of residing at Long Island compared to guests’ home community.  For some, Long Island can provide an escape from problems of crime and substance abuse. 

The implementation of SOAR, the shelter’s new transitional residence program, has already taken steps to reduce these problems.  Direct transportation is provided to and from the shelter for SOAR participants who are in job training or community jobs (one-half of the SOAR participants work in the city).  This reduces the problem of isolation on the Island and may spur more effective community job search activities.

Guests who had only recently arrived at the shelter were less likely to be in the work programs, but were also less likely to be looking for work.  Current shelter policy is to try to tailor the job search assistance given to recent arrivals to their interest in working.  Perhaps a more proactive policy should be considered to stimulate work interest among these guests short after they arrive.  

Employment outside of the shelter and importance attached to job training were associated with lower levels of perceived social support.  This may reflect fewer social opportunities for employed guests or those actively looking for work.  It may also be that high levels of social support indicate a longer history of shelter use and a degree of “settling in” to shelter life.  Some efforts to create opportunities for social interaction for those employed outside of the shelter setting might help to counteract such a tendency.

Observations in one of the computer classes conducted at the Long Island Shelter indicate the value of this resource and suggest its appeal.

The students in the computer class seemed comfortable and motivated.  I wondered if this motivation stemmed from their interactions with their supportive teacher….  [T]he students supported each other.  It was almost as if a women’s network or support group had formed.  Perhaps this is because the students are working together towards a common goal, to learn how to use computers.  Perhaps it is because they have become like a family to each other.

It may be that this training meets particular needs of the group of women who were observed in this particular class, but it seems likely that many men should also benefit from this type of experience.  Although Serving Ourselves participants are required to participate in shelter educational programs, among the general population, none of the men reported having participated in an educational program at the shelter, compared to 38% of the women. 

There were several other gender differences in work involvement.  Men were more likely to have worked full time in the past (84%) than were women (60%).  The types of jobs they reported having had were also different, with women mentioning office work more than did men.  In addition, men were more likely to have heard of the work programs from graduates of the work program (45%) than were women (29%).  However, women were more likely to be familiar with the work programs (69%) than were men (47%), they were more interested in help with finding a job (90%) than were men (43%), and they were more likely to want information about the SOAR program.

In spite of these gender differences, the similarities were more striking.  Male and female guests were as likely to have used these programs in the past and they were as likely to have had some job experience.  On balance, the survey provides little basis for designing different programs for men and women, although if it is to take maximum advantage of prior experience, the particular types of skills training offered should not be identical for men and women.  It seems clear that women are as motivated, perhaps even more motivated than men to participate in work and training programs.

Participant Attitudes Toward Work Programs

            Almost nine in ten work program participants rated the quality of work training they received as good or excellent (figure 11).  There was some room for improvement, as “excellent” was the choice of 25%.

Figure 11


 

Motives for Work Program Participation

 

            Work program participants reported keen interest in improving their employability.  Half strongly agreed that they wanted to learn a new skill and almost 90% either agreed or strongly agreed (figure 12).

Figure 12

 

           


            On the other hand, almost 80% reported that they were not in the program for the money (figure 13). 

 

Figure 13

 

 

            Three-quarters of the work program participants stated that they were somewhat or extremely comfortable with conducting a job search (figure 14).

Figure 14

 

            On average, work program participants agreed that the training they were receiving in the work program was preparing them for their next job (figure 15).  They also agreed, on average, that the shelter staff get along with the program’s workers and that the workers in the program get along with each other. 

Figure 15

 


 

Satisfaction with Work Programs

 

            Overall, 75% of the participants were satisfied or very satisfied with the work program (figure 16).

Figure 16

 

 

            Work program participants reported a high level of satisfaction with the program staff and with the program overall (figure 17).  There was slightly less satisfaction with the type of work respondents did on the job training program.  Dissatisfaction was more evident with pay levels, with the average satisfaction level falling between satisfied and dissatisfied.

Figure 17

 

 

            Levels of satisfaction with the type of work and the program staff were similar across the four specific programs examined (figure 18).  Pay generated the most dissatisfaction across all four programs, but particularly among the Woods-Mullen Work Experience participants.

Figure 18

            Impressions of the quality of training and of the job program overall were similar across the four programs (figure 19).  Work Experience participants at Long Island were somewhat less satisfied with their ability to get a job than participants in the other programs.

Figure 19

Feelings About Specific Work Programs

 

            Motivations for the program and perceptions of its value were similar across programs (figure 20).  However, WEP participants at Woods-Mullen were less confident in their ability to secure a job than were participants in other programs.

Figure 20


 

Program Attitudes Summary

            Across all four work programs, participants were satisfied with the program itself, with the training they were receiving, and with social relations in the program.  In addition, they felt prepared for a job.  They wanted to learn new skills in the program and rejected the notion that they were in the program only for the money.  Pay levels were the only source of some dissatisfaction.

            Observations in a Work Experience Program meeting provided additional evidence of participant motives. 

The one thing that I heard repeatedly was that no one felt as if they were owed anything.  Most of them were there because they needed and sincerely wanted to put their lives back together.  They all wanted to regain their dignity.

            Intensive interviews conducted six work program participants indicate the diverse bases for satisfaction with the work programs.  From cleaning the shelter to serving food or janitorial work, everyone responded positively.  Working made one participant “feel more responsible now.”  One respondent had trouble getting along with others and was pleased to work alone.  Some enjoyed receiving a paycheck, having a bank account and an ATM card.  Another liked the work because “it is easy”; another liked giving orders.  One emphasized the training he received and the greater self-esteem he felt.  Another liked his job because he was able to “focus on people’s level…[and to see] where they’re coming from.” 

An interview with a work program supervisor emphasized how the programs mix personal support and rehabilitation efforts with work training.  One example mentioned was of a participant in the Culinary Arts program (Serving Ourselves) who had “mental health issues that made it difficult to interact with people.”  “The loose environment [in the kitchen’ along with the teamwork and positive bonds that the other staff members portray have helped this clients speak more, interact and even laugh along with the joking of other kitchen workers.”  Work programs thus met different needs for different guests.

            Staff comments in a focus group give more insight into the basis for guest satisfaction with the work programs.

“Saving lives is number one…  We offer stability to those who have never had it in their lives.  We acknowledge their pain for the first time.  We treat them justly.  You know when it makes a difference.  Most of the clients have been abused.  It helps and is rewarding to say that what was done to them was wrong.  It makes a world of difference to them that their behaviors are understandable.

            Participants in the Work Experience Program at the Woods-Mullen Shelter were distinctive in some of their attitudes toward work.  They were more dissatisfied than others with their pay and less optimistic about their ability to get a job after leaving the program.  Since employment outside of the shelter was much more common at Woods-Mullen than at Long Island, these more negative attitudes may be a consequence of comparing themselves to employed guests, rather than to unemployed shelter users.   These attitudes may also result from the narrower array of work and service programs at Woods Mullen compared to Long Island.  A support group for work program members and for employed guests might help to share job search and retention information, thereby turning the presence of many employed guests into an advantage for work program participants. 

            Staff comments in a focus group suggested other bases for more pessimistic attitudes among work program participants at Woods-Mullen.  “In the city…, there is more negativity and negative influences:  substance abuse, female abuse, estrangement from children, a lot of pressures.”  In response, the program staff try to connect clients to services and get them back into the community, so they can [among other goals] reestablish their connection with their children.

Again, gender differences were explored but none were identified.  Men and women were roughly as satisfied with the work and educational programs and reported similar motivations for program participation.  Of course, some possibilities for direct comparison are limited by the exclusion of women from the Work Experience Program at Woods Mullen.

Satisfaction with the Shelters in General

            Overall, guests were satisfied with the shelters (figure 21).  Half of the sample rated their shelter as better than others and few rated them it as worse.  About one in five said they did not know about other shelters.

Figure 21

                        Only a small fraction of the guests expected to be at the shelter more than one year, although just one in ten expected to leave within a week or two (figure 22).      

Figure 22

 

            Work program participants and Woods-Mullen guests felt their shelter compared more favorably to other shelters than did Long Island guests who were not in a work program (figure 23).  There were no differences in likelihood of leaving the shelter across the two shelters or between work program participants and others.

 

Figure 23


            Satisfaction with the shelters was generally high.  Work program participants were particularly satisfied with staff and particularly dissatisfied with their proximity to stores and shopping (most were on Long Island and would have had the opportunity to go into the city less often than the other guests) (figure 24). 

Figure 24

                       Work program participants were also somewhat more satisfied with physical conditions, space, and furnishings (figure 25).  There was a high level of dissatisfaction with the level of privacy at the Woods-Mullen shelter.

Figure 25

Summary of Shelter Satisfaction

            Overall, satisfaction with the shelters was high, as it was with most specific shelter features.  However, many guests at Long Island reported they had no basis for comparison to other shelters.  Most guests planned to stay at the shelter between one month and one year, suggesting that programs requiring enrollment for this length of time could have a steady enrollment.

            Work program participants were more satisfied with staff and with physical conditions than others, but were much less satisfied with the proximity of the shelter to stores and shopping.  There clearly is a tradeoff between the advantages of participating in work programs at Long Island and losing convenient access to urban amenities.  Some regular plan for transit to the city, or perhaps a commissary on the island might reduce some of these complaints [by the time this report was finished, a new van was already in service to make it easier for SOAR participants to commute to the city].  The major complaint at Woods-Mullen shelter was the lack of adequate privacy.  Greater attention to possibilities for creating private spaces in this very densely populated shelter could result in more satisfied guests.


Conclusions

            Transitional work programs at Long Island and Woods-Mullen shelters offer a vital service to a large number of guests.  Guests who participate in these programs are very satisfied with the work programs, with program staff, and, more generally, with shelter services.  Work program participants report a high level of interest in improving their skills, believe that their ability to get a job is not just a matter of luck, and feel well prepared to search for employment.  Program staff are viewed very positively by program participants.

            The benefits of the work programs extend beyond the immediate impact of employment.  The support they provided for many participants helped to increase confidence in achieving longer term goals and using some of their unique experiences in productive ways.  One intensive interview provides an illustration: 

[He] confided that he would  like to be a counselor, particularly for troubled/at risk kids who had not yet dropped out of school.  … [He] feels that he has a message to convey.  Here, he speaks from experience.  He told us that he had hung out with a group of older kids, and that doing so pulled him away from school prematurely….  Although he’s a father, [he] missed out on a lot of his kids growing up, so working with kids would be like another chance at parenthood.  …[He] feels that although he’s not old, that he has done a lot of things wrong is his life and that he doesn’t have much time to do things right to keep from being [defined] a failure.

There is enough interest in the work programs to warrant their expansion.  As one program participant argued in an intensive interview,

The only thing she would change about the program would be to increase the opportunities.  ….She says the shelter contracts out some services and she thinks it would be better if they got the clients involved in providing those services instead.  She said even if they have the [clients] paint the walls at least that would be something.

Physical health problems seem to be the most important barrier to work program participation for many guests.  Further investigation of the nature of health problems in the general shelter population are called for, as are efforts to develop vocational rehabilitation programs that are suited to individuals with physical health problems.  Providing services to substance abusers and those experiencing  high levels of psychological distress should also bring some guests into work programs who would otherwise be unable or unwilling to participate.  Nonetheless, flexible management of the work programs can help them to meet the diverse needs of guests, as indicated in one intensive interview:

He was very happy with the program.  He had started out working in the kitchen.  However, he said he had trouble with that job.  He says he has a very bad temper and that he doesn’t work well with other people, so he asked to be changed to a different job.  Now he works as a janitor.  He very recently had used a floor buffer for the first time and was very excited about learning this skill.  He prefers this job because he gets to work alone and no one gives him orders.

Long Island and Woods-Mullen shelters have developed somewhat different social patterns which may influence the work involvement of shelter guests.  Woods-Mullen facilitates entry of guests into competitive employment, while Long Island supports many more work program activities on site and has few other employed guests.  Because of Long Island’s distant location from the city, it may be hard to change this pattern.  However, it is important to prevent the development of a self-sustaining ethos of service dependence that could undermine efforts to return shelter guests to regular employment.  More job search assistance efforts at Long Island and greater monitoring of guests’ job search efforts may help to overcome these liabilities.  Staff members participating in a focus group suggested some changes to overcome the isolation of Long Island Shelter work program participants--integrating the program with the outside community and having a support group to relieve job stress, especially during the winter.

At the same time, the greater exposure of Woods-Mullen work program participants to guests working outside of the shelter may be the reason for somewhat greater pessimism about their own prospects for finding a suitable job.  More job counseling or support group activities may be useful.


Appendix

Variables

Variable

Name

Description

Answer choices

Statistics*

SOURCE01

Client work program participation status

 

0= General Population

1= Work Training Participant

0 = 82%

1 = 18 %

WORKNOW

Recoded answer to, “ Would you say that you are currently not working, working at a part-time job, working at a full time job or working at odd jobs’?

 

0 = Not working

1 = Part-time job

3 = Full-time job

4 = Odd jobs

Recoded to:

No = 75%

(not working)

Yes = 25%

Working at 1-4

LOOKWORK **

Answer to, ‘Are you looking for work now?’

0 = No

1 = Yes

No = 78.7%

Yes= 21.3%

IMPTRN

Index comprised of average of two questions:

1)‘ How important to you now is job training?’ and 2)‘How important to you now is help with finding a job?’

1 = Not at all

2 = Slightly

3 = Moderately

4 = Considerably

5 = Extremely

Alpha = .75

EVERWORK

Answer to question , ‘Have you ever had a full or part-time job?’

0 = No

1 = Yes

No = 11.4%

Yes= 88.6%

GENDER

Question not asked of client, but was coded by interviewer.

1 = Male

2 = Female

Male = 79%

Female = 21%

AGE

Computed from ’When were you born?’

Date

Mean = 43.16

sd = 11.3

RACE

 

Combined responses to two questions: 1) ‘Would you say that you are Black, White, Native American, Asian, Pacific Islander or something else?’ and 2) ‘Are you of Hispanic or Latino origin?’

1)  1 = Black/African American

    2 = White

    3 = Native American

    4 = Asian/Pacific Islander

    5= Other

2)    0 = No

    1 = Yes

Black = 47%

White = 35 %

Hispanic=15%

Other  = <1%

EDUCATION

 

Response to , ‘ What was the highest year of school that you completed?’

1-16

Mean = 10.49

sd = 2.65

MOSINCE

 

Time since first arrival at shelter

1= 0-3 months

2=  4- 12 months

3 =13- 24 months

4=over 24 months

1 = 27.2%

2 =23.6%

3 = 17.3%

4 = 32%

*          Statistics reflect weighted percentages

**         Asked only of non program participants.


Variable Name

Description

Answer choices

Statistics**

LUCK

Response to, ‘indicate the degree to which you agree or disagree with the following statement: “If I find a satisfying job, it will mostly be a matter of luck”

0 = Strongly disagree

1 = Disagree

2 = Uncertain

3 = Agree

4 = Strongly agree

Mean = 2.29

mode = 3

sd = 1.32

SAINDEX

Index counting affirmative answers to 4 alcohol and drug questions: 1)‘Have you drunk any alcoholic beverages’, 2)‘Have you felt guilty about things you have done while drinking alcohol’, 3)‘ Have you ever used illegal drugs like crack, cocaine, heroin, speed, or marijuana ?’ and 4)‘Did you think your problems are at least partly due to your drug use?’

0 = No

1 = Yes

Alpha = .72

CESD

Index calculated by counting affirmative answers to the following questions: At any time during the past week …? 1)‘ Was your appetite so poor that you did not feel like eating?’, 2)‘Did you feel so tired and worn out that you could not do anything?’ 3)‘Did you feel depressed?’4) ‘Did you feel unhappy about the way your life is going? ‘ 5)‘ Did you feel lonely?”

1 = Never

2 = Some of the time

3 = Most of the time

Alpha = .64

SSINDEX

Index computing mean response to the following  questions: 1) indicate the amount of support you get from the following sources: friends in the shelter, friends outside of the shelter, family, shelter staff, people at other agencies.  2) “I feel like I’m not always included by my circle of friends”, “I think that my friends feel that I am not very good at helping them solve their problems”, and “ There are several different people I enjoy spending time with” * would you say this statement is(are)…3) If you needed to get in touch with your family, would you be able to?, and Did you visit with any relatives outside of the shelter in the last month?,

1)   1= None

      2=Some

      3= a good amount

      4= a great deal

2)   1= Definitely true

       2= Probably true

       3 = Probably false

       4 = Definitely false

3)   recoded 1 = No

                    4= Yes

                   2.5 = don’t know

Alpha =.75

PHEALTH

Index computing mean response to the questions, 1) “In general, would you say your health is...?” 2) Compared to six months ago, how would you rate your health in general now?  3) During the past 4 weeks, to what extent has your physical health and/or emotional problems interfered with your normal social activities with family, friends, neighbors, or groups? 4)  During the past 4 weeks, how much did pain interfere with your normal activities with family, friends, neighbors, or groups?   

 

 

1)  1 = Excellent     4= Fair

    2 = Very Good     5= Poor

    3 = Good 

2) 1 =Much better

   2 =Somewhat better

   3 =About the same

   4 =Somewhat worse

   5 =Much worse

3 , 4) 1 = Not at all   4 = Quite a bit

­­­   2 = A little bit         5 = Extremely

   3 = Moderately

 

*     “There are several different people I enjoy spending time with”: Answer choices were reversed for a consistent direction in index

**    Statistics derived from weighted sample


One question measured presence of an external and internal orientation toward control.  The respondent was asked: ‘If I find a satisfying job, it will be mostly a matter of luck.’  Respondents who agreed with the statement displayed an external locus of control while those who disagreed displayed a more internally oriented locus of control (Caplan, 1967). For crosstabular analysis, LUCK was collapsed into two categories: ‘internal’ indicating those who disagreed with statement, and ‘external’ corresponding to responses of agreement.

The Center for Epidemiological Studies Depression (CES-D) Scale measured feelings of distress. This measure has been shown to have high internal and test-retest reliability in homeless research (Rossi, 1989; Susser, Conover, & Struening,1990).  An abbreviated version was used which consists of five different questions related to depressed feelings in the last week: lonely, depressed, too tired to do anything, unhappy  or unable to eat.  Answer choices consisted of ‘never’, ‘some of the time’ and ‘most of the time’.   

          An index was constructed to measure physical health by computing the mean response to four different questions regarding self reported health status and the degree to which pain or poor health interferes with daily activity (Schutt et al, 1994).  A Cronbach’s alpha of .64 indicates that the reliability of this measure needs to be improved.

          The number of months since first arrival at the shelter was computed by subtracting the month and year indicated as the date of first arrival at the shelter from the month of the interview.  As the interview did not contain questions regarding total length of time homeless, this measure was used as the closest indication possible of chronic homelessness.

          A substance abuse index was constructed through counting all of the yes statements to four questions regarding alcohol and substance abuse (Schutt et al, 1994). An alpha level of .72 in the reliability analysis of these variables indicates that this index is a reliable measure of substance abuse.

Level of social support was measured as the mean response to several different questions regarding perceived level of social support.  These questions were derived from Cohen and Syme’s (1985) 40-item Interpersonal Support Evaluation List (ISEL) (alpha = .75).


References

 

Caplan, G. (1981).  Mastery of stress: Psychological aspects.  American Journal of Sociology, 138, 413-420

 

Cohen, S. & Syme, S.L. (1985).  Issues in the study and application of social support.  In Social Support and Health, edited by S. Cohen and S.L. Syme, pp. 3-22.  New York: Academic Press.

 

Rossi, P.H. (1990).  The old homeless and the new homeless in historical perspective.  American Psychologist,  45, 954-959.

 

Schottenfield, R., Pascale, R. & Sokolowski, S. (1992)  Matching services to needs: vocational services for substance abusers.  Journal of Substance Abuse Treatment, 9, 3-8.

 

Schutt, R.K. & Garrett, G. (1992).  Responding to the Homeless: Policy and Practice.  New York: Plenum Press.

 

Schutt, R.K., Tatjana, M. & Rierdan, J. (1994). Distress, suicidal thoughts and social support among homeless adults.  Journal of Health and Social Behavior, 35:134-142.

 

Susser, E., Conover, S. & Struening, E. (1990).  Mental illness in the homeless: Problems of epidemiological method in surveys of 1980.  Community Mental Health Journal, 26, 198-217.