|
[an error occurred while processing this directive]
Return to
OCSE Home Page Return to Table of Contents SOCIAL DEVELOPMENT COMMISSION
Agency: Social Development Commission Executive Director: George Gerharz (Acting) Evaluator: D. Paul Moberg, Ph.D. Director, Center for Health Policy and Program Evaluation University of Wisconsin-Madison 433 West Washington Avenue, Suite 500 Contact Person: George Gerharz Type of Project: Minority Male Employment and Self-Sufficiency Project Title: Minority Male Opportunity and Responsibility Program Model: Demonstration Partnership Program Target Population: Minority Males Between the Ages of 18 and 34 Project Period: November 1, 1991 to August 31, 1994 A. Program Purpose and Specific Objectives The Minority Male Opportunity and Responsibility (MMOR) program was an innovative program targeting minority male unemployment in Milwaukee, Wisconsin. There were two tracks--a Job First approach, which emphasized job placement (this was the control group), and a Basic Skills approach in which the full services of the project were provided. The Basic Skills program provided educational skills development and GED; job search and placement activities; a Job Club, which served as a peer support group; and intensive case management to deal with the wide range of issues which affect employability (e.g., health problems, substance abuse, legal system involvement, family relationships). The anticipated length of stay in the program was relatively open-ended, with completion defined in terms of meeting project objectives. The following specific outcome objectives were developed for the project and served as hypotheses to be tested in the evaluation, both as absolute criteria and relative to control group performance: 183Fifty percent of the participants would obtain and sustain employment over a period of 1 year (defined as employed longer than 12 consecutive weeks during the year). 183Fifty percent of the participants would have increased their reading and math competency levels above the eighth grade. 183Sixty percent of the participants and their families would report a significant improvement in family functioning. 183Sixty percent of the participants would have eliminated one health risk factor.
B. Description of Program The MMOR program combined intensive case management with peer support and family involvement to help unemployed minority males improve their academic skills, find and keep employment, and strengthen family relationships. The original continuation proposal funded two case managers and one half-time job placement specialist to work with a minimum of 60 unemployed minority males. However, the power analysis conducted for the evaluation research design indicated that an N of 80 per group, after anticipated attrition (of 50 percent), was necessary. The number required for evaluation purposes thus significantly increased the number of men to be served over the original budget. After an intake process, which included the collection of basic demographic and needs assessment information as well as an assessment of math and reading skills through the administration of the TABE (Test of Adult Basic Education), staff determined which candidates were appropriate for MMOR. At that time, randomization of candidates into either the treatment or control groups was to occur. (This process is explained fully in Section II. A.) Once accepted into the MMOR treatment group, a participant developed an Individual Development Plan (IDP). The IDP served to articulate clearly participant goals and map out specific strategies for accomplishing those goals. There were two tracks: Basic Skills and Job First. In Basic Skills, the participant was expected to attend the Social Development Commission (SDC) Basic Skills Program for a minimum of 3 hours per day/4 days per week, spend 2 hours twice a week at the Job Club, commit an additional four to six hours to job search, and spend at least 1 hour per week meeting privately with his case manager. Under the Job First option, the client was to spend 20 to 25 hours per week on job search activities.
In addition, support services addressed problems which could interfere with the participant's ability to find and keep employment. They included health problems, alcohol and drug abuse, criminal justice problems and other legal difficulties such as child support proceedings, family relationship problems, financial concerns, and other problems which could subvert efforts to attain self-sufficiency. Case managers referred participants to a variety of health and social services as well as provided many key services, such as AODA counseling and health screening, on site. Peer support occurred through the Job Club, which served as a job readiness vehicle as well as a forum for the discussion of personal and family problems. Family involvement occurred through home visits by case managers, family needs assessments and referral services, and family recreation activities. Intensive, flexible, and personalized case management was the core of MMOR. "Intensive" meant nearly daily contact with participants. The MMOR case manager helped the participant identify his goals, design a plan of action, and implement the plan over the course of 7 months. (Anticipated average active length of stay in the program was 7 months.) Case managers were to record carefully the frequency and nature of their interaction with participants, maintain weekly progress notes on each participant, and record participants' involvement in MMOR activities and other support services. C. Target Population The target population for MMOR was minority males between the ages of 18 and 34 who met the following criteria: 183Currently receiving general assistance (work relief) or income eligible for general assistance (unemployed and without legitimate, identifiable income). 183Chronic unemployment (no period of continuous employment of more than three months during lifetime). 183Parent of at least one child. 183Reading and math levels between the third and seventh grades as determined by the Test of Adult Basic Education (TABE). D. Partnerships The Social Development Commission's MMOR program had the following institutional partners: Milwaukee County Department of Social Services; Wisconsin Department of Corrections; Milwaukee Area Technical College; Milwaukee County Combined Community Services Board; and for evaluation research, the University of Wisconsin-Madison Center for Health Policy and Program Evaluation. Additionally, MMOR had cooperative relationships with a variety of public and private, not-for-profit organizations willing to provide needed support services to MMOR participants. II. STUDY APPROACH AND EVALUATION METHODOLOGY
A. Research Design The outcome evaluation used a prospective randomized design in which participants were assigned at random to the full MMOR program or to a job-search-only control track. Assignment occurred only after all baseline data had been collected and staff had determined the participant eligible for the MMOR program. Control subjects were provided a job-search-only track where there was some ongoing contact and some services, but of limited intensity and comprehensiveness. This provided a test of the MMOR model against a far less expensive and very commonly available alternative; all eligible participants were provided some level of job-related services. Data were collected at baseline (including testing and interviews) on all participant and control subjects at three months post-baseline, and at program closure/graduation. Follow-up, planned for 2 months post-baseline, was conducted an average of seven months after intake. A summary of the design is provided as Figure 1. Figure 1: MMOR Evaluation Design and Measurement Summary Intake/Baseline 3 Months Follow-up MMOR Program Data Interview/Test Interview/ Update Test/Staff Randomize Control Progress Update Interview/Test Data Elements: 149 TABE 149 Services to Date 149 TABE 149 Employment History 149 Present Employment 149 Employment Status 149 Family History 149 Present Family Status 149 Present Family Status 149 Education History 149 Present Educ. Involvement 149 Educational Involvement 149 Family Measures (FSI) 149 Progress Rating by Staff 149 Family Measures 149 Health Status Screen 149 Graduation Data** 149 Health Status 149 AODA Screen 149 AODA Screen 149 Participant Satisfaction Notes: *Upon employment, wages and hours worked to be documented at 14, 30, 60, and 90 days. **Graduation/closure data include a complete summary of service utilization, employment status, and case manager rating of case outcomes.
B. Operational Definitions of Outcome Measures The key outcome indicators are employment status, educational attainment, family functioning, and health risk factor reduction. All items were to be measured at baseline (prior to randomization) as part of the screening and assessment process, and at follow-up via participant survey/interview. In addition, a subset of the outcome indicators was collected on a graduation/closure form, which also summarizes services received. 183Employment Measures--Number of hours worked, type of position (full- or part-time, temporary/permanent), nature of work, hourly wage, duration of employment, number of positions over given time frame. This information was confirmed with employer when possible (with appropriate consent). 183Educational Measures--TABE, published by McGraw-Hill, was used to establish reading, mathematics, and language skills grade equivalence levels. It was administered to all subjects at baseline and repeated at the follow-up interview. 183Family Functioning--At baseline and follow-up, subjects were asked to complete family cohesiveness and conflict scales and a community support inventory. We also ascertained participant's living situation, number and living situation of children, participant's involvement in parenting, and marital status at each interview. Program staff also rated participants on family functioning at closure. 183Risk Factors--Data on use of (or at least referral for) health services and known changes in health risk factors were obtained by staff and reported on case closure forms. Additionally, as part of the baseline and follow-up interviews, the brief General Health Survey from the Medical Outcome Studies (MOS), a short alcohol abuse screen (the CAGE), and an alcohol consumption measure were administered. Baseline and follow-up data collection was done by staff from SDC's Planning and Research Department independently of the program. Staff intake and closure data were collected as part of the project screening, assessment, and case management process by MMOR staff (including data on control group members). Baseline and follow-up data were collected via a combination of self-reported survey and paper and pencil tests (i.e., TABE) of all potential participants, and in-person interviews by staff. The follow-up testing and interviews were conducted at the same central location as the intakes, as well as at set times at the MMOR offices. Participants were paid $15 for their time and as an incentive to participate. Reminder wallet cards, multiple mailings, and phone contacts were attempted with all participants to enlist their cooperation in the follow-up. Process evaluation sought to develop a descriptive qualitative analysis of the implementation of the program, and to gain an understanding of the program as experienced by the frontline staff and clients. A participant observer attended seven meetings of the Job Club between February and April of 1993, and interviewed several program participants and all staff members. In addition, program staff maintained records of client participation and progress; these records provided service utilization data, client counts, and progress ratings.
III. EVALUATION FINDINGS A. General Discussion of Findings 1. Process Evaluation Process evaluation was conducted using participant observation of Job Club and other counseling sessions and intensive interviews with clients and staff. Results of this qualitative analysis point to the strength of the peer interaction in the Job Club as a critical factor in developing bonds between the men in the program, but also indicate very sporadic and transitory participation by all but a small group of regulars. Family relationships received little emphasis, and there were serious limitations in the job development aspects of the program. There was a perception of the female case manager as a mother figure to the men in the program, and genuine concern for the men from the staff involved. Men in the control group expressed the opinion that it was designed for failure, and there were some. It appears that this group was presented in a negative light to at least some of the men, emphasizing it as a control rather than as an intervention in its own right. There were criticisms by staff of the program design because of its exclusion of men over age 34, its emphasis on numbers served rather than on the quality of service provided, and a concern about the nature of participants referred through the primary referral source, the County Department of Health and Social Services. There was general agreement among staff that many men referred through this mechanism were interested more in using the program to beat the system and keep benefits flowing than to use it to gain employment. The overall implementation of the program and the services provided were far less intensive than what was described in the initial grant application. This was due to the larger than planned number of participants, as well as to minimal compliance from a number of the participants. 2. Outcome Evaluation Ultimately, a fairly rigorous experimental outcome evaluation was in fact implemented, albeit over a truncated time frame. The two groups of participants were similar in demographic characteristics, educational background, and work history. While in high need of employment-related programming, they did not all strictly meet the criteria initially stated for MMOR program eligibility (between third and seventh grade equivalence on the TABE, less than 3 months continuous employment, under age 34). Participants assigned to the case management track received much more extensive services in the program than did those in the job track condition, but overall, participation fell short of the plans for the program. Employment outcomes were significantly more favorable for the case management clients than for those in job placement, although both groups showed similar changes on the TABE and no significant changes in the areas of family relationships and health status. The use of a controlled design was critical in this evaluation, since the positive effects on employment would not have been apparent using only a before/after design in which MMOR's success was judged only against the pre-established program goal of 50 percent employment.
B. Operational Issues--Process Evaluation Findings Difficulties in implementation, related to the requirements of the evaluation design and evaluation data collection, were such that the first year of the project is treated as an unevaluated pilot effort in this report. Data on approximately 100 participants entering the project prior to June of 1992 were discarded due to design contamination and inconsistent data collection and recording. The client and program descriptions and data which follow are based on implementation during the final year (Summer 1992 through June 1993) of the project, during which a relatively successful rigorous evaluation effort was incorporated. This necessitated truncating the follow-up period to an average of seven months after baseline. Thus, the discussion and presentation of data focus only on that final period of program operation and evaluation. 1. The MMOR Program as Implemented Case Management Program (Experimental Condition)--This program track included an initial assessment of strength and problem areas, the construction of an Individual Development Plan (IDP) for each participant, in-house "basic skills" classes, referrals for a variety of training and education programs, referrals for human services, individual counseling, and participation in the weekly Job Club. The plan to have separate Job First and Basic Skills tracks was not consistently implemented, and the two tracks are indistinguishable in both the observational qualitative data and in the quantitative process and outcome data. Overall, the observational data, as well as case records, indicate that the primary service vehicle of the program was the Job Club. Individual counseling, AODA service, job search, recreation, family issues, etc., were apparently dealt with largely through the Job Club or as an adjunct to it. Staff weekly service logs tended to incorporate other services into Job Club attendance, as opposed to recording them as separate discrete events. Typically, about 10 men attended each Job Club session, which lasted up to three hours and usually included both didactic presentations by MMOR staff members and guided group discussion. The following topics were included during the period in which process observations were conducted: 183Role playing to enhance employment-related skills 183Cultural heritage, in particular African-American history 183Personal responsibility 183Dealing with authority figures 183The importance of education 183Personal grooming and hygiene as factors in the employment search 183Positive support for job search and employment skills 183The impact of substance abuse and crime on the community and employability 183The economic structure of the inner-city community 183Family responsibility 183Racism as a factor in minority male unemployment
The typical case management client had an average of 9.8 (+13.5) documented in-person contacts with the MMOR program (38% had two or fewer contacts). Major services received (see Table 1) included case management (63%); Job Club (59%); AODA counseling (32%); family counseling (26%); and health services (22%). External services checked on program records as having been received included transportation assistance to 48 percent, medical services for 40 percent, vocational training (34%), and apprenticeship program (33%). Job Placement (Control Condition)--The control track consisted primarily of weekly two-hour group sessions with a job developer who provided lists of possible openings and coached participants in job-related issues such as filling out applications, interview skills, and transportation assistance. The job developer during the final 9 months of the program strictly interpreted his role as assisting the client in gaining employment, as opposed to the broader case management and individual participant support goals implemented by the case management program staff.
The typical job placement client had a mean of 5.2 (+6.4) in-person contacts with MMOR documented in his case file (39% had two or fewer contacts). Major services documented included job placement (55%), Job Club (16%), and transportation assistance (32%)-- see Table 1. The Job Club provided to the control participants was apparently a group session limited to job-seeking skills. Process evaluation data from case files, participation logs, and staff interviews indicate a number of problems with implementation of the control track, including staff turnover and minimal record keeping. MMOR written materials sometimes refer to the control condition as the Job First track, while in other places it is overtly called the control group. Thus, there was confusion between the original plans for Job First and Basic Skills within the overall experimental program, as well as potential demoralization of clients and staff who considered themselves relegated to a control condition which was designed to fail. 2. Implementation Problems Many of the program implementation problems encountered were due to evaluation issues and concerns. The number of clients needed for evaluation purposes (160 per group initially, anticipating 50% case loss) based on the power analysis exceeded the capacity of the program (budgeted for 60 in 1 year in the program, with no service budget for controls). The two case managers were expected to provide and document intensive services to all MMOR clients; an additional staff person served as a job placement specialist for the control group. The case managers were unable to provide significant in-house substance abuse services, family counseling, or mental health services on a one-on-one basis because of the caseload size. Clients were reluctant to make use of "outside" support services offered in the community, and thus did not receive the level of service initially envisioned in the proposal. Case documentation that was needed for the County Department of Social Services (DSS) and for the evaluation was a large burden. Program feasibility was threatened by the evaluation's needs for relatively large numbers, which became even more critical when the evaluation period was truncated to 1 year.
After the fact, the primary evaluator has concluded that attempting an experimental design was inappropriate given the level of funding involved for the program, let alone for the evaluation. The reasons for a randomized design were not understood by staff and participants, and the design was resented and (possibly) sabotaged at first. Both the program and the evaluation research component suffered from trying to do too much with too few resources. 3. Interorganizational Operational Issues The major source of referrals for the MMOR program was the county DSS, which requires basic education, job readiness training, and/or active job search from individuals receiving general assistance support. The initial referral mechanism from DSS was an inconsistent and often- missed recruitment message provided by in-house DSS staff. These messages were eventually replaced by a thrice weekly session where the job developer/control group facilitator for MMOR made a 4-5 minute presentation. As a condition of receiving general assistance, participants were required to check in at least weekly, and reports back to DSS were required (for both tracks). There is little evidence that compliance was strictly enforced, particularly given the high dropout rate. Interviews with program staff and clients, and observation of Job Club sessions, suggest that many of the referrals from this source were "more interested in beating the system" than in obtaining employment. There was also some controversy in the program regarding the rightful place of corrections clients who were also working on job development/employability training and job placement within SDC, under funding from the state Department of Corrections. There was some overlap of staff and clients between programs. C. Findings 1. Impact on Clients Client Characteristics--Useable data were obtained on 168 participants in the program, of which 79 were assigned to the full experimental track (case management) and 89 were assigned to the control track (a standard job placement program). Most participants were referred to MMOR as a condition of receiving general assistance funds. The two groups of young male African Americans were similar demographically, with an average age of 29, mostly never married (83%) with only 5 percent presently married, typically living either alone or with parents/relatives. Only 16 percent were living with a partner, spouse, and/or children. Participants reported having an average of two children, most not in the participant's care. The typical participant had not completed high school, although about one-fourth had obtained GED/HED degrees. The average number of years of education reported was 11.2. Of those reporting job histories (25% did not, either because they had none, or due to data collection problems), the average number of months worked on the longest job ever held was about 18 months. About one-third had worked from 3 to 12 months, and another third, over 1 year. Thus two-thirds did not meet the stated criterion for involvement of "less than 3 months of continuous employment during one's lifetime."
The hourly wage on the most recent job averaged $5.54 per hour. In answer to the question of how many weeks they had worked in the past year (prior to intake), about one-fourth reported none, and one-third did not answer the question. For those reporting having worked in the past year, the average duration of the employment was about 13 weeks. Fourteen percent of participants were veterans. Only 17 percent possessed valid driver's licenses. Only 23 percent reported never having been arrested (19% failed to answer this question); for the remaining 57 percent of the participants, an average of about four arrests each was reported. One in five admitted to a history of Alcohol and other Drug Abuse treatment; staff assessed 26 percent as having alcohol problems and 19 percent as having drug problems on the intake forms. However, interviews with case managers suggest that substance abuse is far more common in this population than appears in the structured data;"almost all participants" were believed to have substance abuse issues. The TABE was administered to most participants as part of the program assessment and enrollment process. The reading subscale yielded average grade equivalents of 9.1 for the case management participants and 8.0 for the job placement group (a significant difference), with 36 percent and 52 percent, respectively, scoring below the eighth grade level. Mathematics and language skills scores did not differ between groups. Grade equivalent averages in mathematics were about 7.4 (with 65% below the eighth grade attainment) and in language skills were about grade 5.4 (85% below the eighth grade equivalent). Thus, these participants are experiencing serious educational deficits, particularly in the critical (for employment) area of language skills. Even those with high school degrees or GEDs scored at a deficit on these measures--the mean grade equivalent was about ninth grade in reading, eighth grade in mathematics, and only sixth grade in language skills for those with high school diplomas or HEDs. Services Provided--Services received from MMOR differed considerably for the two groups of participants (see discussion of operational issues, above). The case management track included general support and linkage services typical of case management programs; an in-depth weekly Job Club in which wide-ranging but relevant topics were discussed and mutual support among the participants was sought; a full range of counseling (individual, AODA, family, and vocational); educational programs and referrals; and referrals for medical, vocational, housing, and legal services. The job placement track most often involved providing the participant with referrals for potential jobs (usually in batches of 10), along with what was at times called a job club in which job-seeking issues were discussed. Because of the low percentage of MMOR participants who had driver's licenses, transportation assistance (usually bus tickets) was provided to clients in both tracks. The services provided, as documented in case files, are summarized in Table 1. The two groups of participants differed significantly in number of contacts with MMOR which were documented, with the case management participants averaging 3.5 telephone and 9.8 in-person contacts, about twice the number of contacts as was documented for the participants who had been assigned to the job placement track. Much time was spent in trying to contact clients, and in documenting no-shows; in the job placement track in particular, "no show" was the primary service documented on the weekly reporting logs. In the case management track, there seemed to be a small core of regulars and many on the roles who did not participate. This may explain the large standard deviations in number of sessions. There was far less participation than planned.
Table 1 Summary of Services Received from MMOR and Closure Status Case Job Management Placement Total
Number: 79 89 168 Percent Documented as Receiving: Job Placement 08% 55% 33% Basic Skills Training 09% 0 04% GED Training 01% 0 01% Health Services 22% 0 10% Job Club 59% 16% 36% Cultural/Recreational Services 11% 0 05% Family Counseling 26% 02% 14% AODA Counseling 32% 02% 16% Case Management 63% 04% 32% Mean (+s.d.) Number of Contacts: By Telephone 3.5 (5.0)a 1.1 (1.9)a 2.2 (3.9) In Person 9.8 (13.5)b 5.2 (6.4)b 7.3 (10.6) Mean Days Intake to Closure 238 (121)a 146 (75)a 189 (109) Services Received Outside of MMOR: Vocational Training 34% 04% 18% Apprenticeship Program 33% 08% 20% Legal Services 06% 01% 04% Housing Assistance 14% 01% 07% Financial Assistance 05% 02% 04% Transportation Assistance 48% 32% 40% Medical Services 40% 06% 22% Closure Status: Successfully Completed Program 24% 09% 16% Client Withdrew/Refused to Cooperate 61% 73% 68% Program Ended 08% 14% 11% Other 07% 03% 05% a Significantly different, p < .001. b Significantly different, p < .01.
MMOR staff rated only 16 percent of participants as having successfully completed the program--24 percent of the case management clients and 9 percent of the job placement clients. About two-thirds of all participants were documented as having withdrawn or refused to cooperate (Table 1).
Client Outcomes--The most important outcome sought by MMOR was employment. At closure, 28 percent (22) of the case management participants, and 10 percent (9) of the job placement participants were documented as being employed. The status of more than one-third of all participants was unknown. Comparing the two groups' employment status at closure (assuming only those known to be employed were and that all others were unemployed), we find a significant difference between groups (p=.003) (Table 2). Table 2 MMOR Employment Outcomes Case Job Management Placement Total Number of Participants: 79 89 168 Employment at Closure: Employed 28%a 10%a 18% Student 01% 00% 01% Odd Jobs/Sporadic 05% 01% 03% Unemployed 33% 44% 39% Not Ascertained/Missing 33% 45% 39% Number with Follow-up Data: 30 (38%) 33 (37%) 63 (38%) Employment at Follow-up*: Employed 47%b 12%b 28% Student 07% 00 03% Odd Jobs/Sporadic 13% 06% 10% Unemployed 33% 82% 59% *Follow-up results based only on those with follow-up data. a Comparing employed to all others, chi-square = 8.75, p=.003. b Comparing employed to all others, chi-square = 9.19, p=.002. Further data on employment was recorded where available. There was little difference between the case management and job placement groups in the nature of employment. Of those employed at closure, nearly one-half had obtained the job on their own without direct assistance from MMOR. Most (60%) were working as laborers, in food service, or in custodial work. The longest job held since entering MMOR averaged 10 weeks, for those who had worked at all. The average hourly wage documented was $5.95. No information on fringe benefits was obtained, but typically jobs of this nature provide few or no benefits. Slightly over one-third of all participants (38%) returned for the follow-up interview and testing. The only attrition bias in this sample was a history of significantly fewer self-reported arrests among those participating in the follow-up (2.1 versus 5.9), and a higher likelihood of presenting for follow-up among those who had been employed at closure (58% of the employed, and 34% of the unemployed, presented for follow-up). No differential attrition between groups was detected.
In this small follow-up sample (total useable n=63), 47 percent of the case management group and 12 percent of the job placement participants reported being employed at the time of the follow-up. Average hourly wages reported were $6.40 (Table 2). TABE scores were also compared for those who returned for follow-up. Reading scores deteriorated significantly from baseline to follow-up in both groups. This may be a testing problem, perhaps a more difficult version of the TABE was used at follow-up. Scores on all other scales increased. Since the changes tended to be uniform in both groups, it is unlikely that this is a true treatment effect--it may be a function of time and test familiarity. Data on health status outcomes for the participants in the follow-up show no statistically significant changes for either group of participants on any of the subscales of the MOS health survey. Measures of family relationships were also examined, comparing baseline and follow-up survey responses for those clients who completed the follow-up. The only significant change was an increase on the family conflict scale for the case management group. Measures of alcohol use and problems showed no significant changes in either group. Because of the importance of employment, a final set of exploratory analyses was conducted to examine which variables were associated with employment at closure. First, a series of bi-variate analyses were run. Variables which emerged as significantly related to employment were treatment condition, number of MMOR contacts, longest job ever, years of education, age, CAGE score, TABE scores, number of children, and family cohesiveness. These were then included in a multivariate regression analysis. The final results indicate that only a few of these variables remain significant when they are examined simultaneously. The findings confirm the effectiveness of the MMOR program as having an independent effect on employment, even when other predictors are controlled for. Being in the MMOR case management track was significantly and positively related to being employed. In addition, the number of MMOR contacts was a significant predictor of employment, for both groups of participants. The bi-variate analysis showed that employed individuals averaged 11.9 (+15.1) sessions, compared to 6.3 (+8.8) for those who were not employed. Past employment was also a major predictor of employment at closure from MMOR. The employed participants had reported an average of 47.4 weeks as the longest job they had ever held, compared to 29.6 weeks for the unemployed. Higher scores on the CAGE were predictive of employment; this counterintuitive result could be explained if the CAGE scores reflect a recovery process and lack of denial, rather than active abuse. While those who were employed at closure scored significantly higher on the TABE at intake, this relationship could probably be accounted for by the "longest prior job" measure. Those individuals with relatively higher levels of basic education were more likely to have longer prior job histories. Because of this, TABE scores were not significant predictors in the regression model.
2. Institutional Impacts The MMOR project solidified working relationships between SDC and the Milwaukee County Department of Health and Social Services in the area of employment services to minority males on general assistance. The project also increased awareness of the local Private Industry Council (PIC) regarding programming for severely disadvantaged minority males. It was helpful in the PIC's initiation of the Food Stamp Training and Employment Program that provides basic skills training and job search assistance. 3. Expected versus Actual Outcomes
The program did not reach the stated or expected goals regarding employment, education, family, or health. However, using a randomized design, the evidence showed positive program results in the area of employment. Without the controlled research design, we would not have reached this positive conclusion, since the program would have been judged only against its stated objectives. The expected outcomes were: 183Hypothesis: Fifty percent of the participants will obtain and sustain employment over a period of one year (defined as employed longer than 12 consecutive weeks during the year). Result: Twenty-eight percent of the case management participants were employed at closure, with an average of 10 weeks as the "longest employment" since beginning MMOR. In contrast, only 10 percent of the control participants were employed at closure. 183Hypothesis: Fifty percent of the participants will have increased their reading and math competency levels above the eighth grade. Result: For those completing the TABE at follow-up (38%), 83 percent scored below the eighth grade level in reading (scores declined from baseline for both groups); in mathematics, 59 percent scored above the eighth grade level at follow-up (up from 20 percent, but no different from controls); in language skills, 24 percent scored above eighth grade at follow-up (up from 7%, but slightly worse than controls). 183Hypothesis: Sixty percent of the participants and their families will report a significant improvement in family functioning. Result: No evidence of change in family functioning, and little indication that the program's efforts affected family functioning. 183Hypothesis: Sixty percent of the participants will have eliminated one health risk factor. Result: No evidence of impact on health risk factors as measured by MOS short-form health survey and alcohol consumption measures.
D. Research Significance Most significant, from a research perspective, is the value of using a controlled design in program evaluation. Even with serious implementation problems, a less intensive than desired program, and many tradeoffs to obtain a sufficient number of cases in a short time period, the design allowed us to draw positive conclusions which otherwise could not have been drawn. The results of our regression analysis confirm the need to invest in human capital--little has been invested in this population. This is particularly the case where we see the value of past successful work experience as a predictor of future employment, along with bi-variate analysis indicating that educational achievement is predictive of employment. However, it is also critical to avoid a "victim-blaming" interpretation of these results. Our observational and interview data suggest that many of the men in this program would accept employment, if it were available to them "at a family-supporting" wage. Efforts to make employment available must consider transportation as well as other factors, because very few of the participants have a driver's license. Our results also raise the issue of the TABE's cultural specificity, especially of the language skills scores. Even if the low language skills scores are a result of cultural bias, this bias is likely to generalize into the work setting and needs to be dealt with in educational programs to improve employment prospects. Further research is needed in all of these areas. In particular, replication of the program, with sufficient resources to provide the intended intensive level of service, is called for. Particular attention also needs to be paid to mechanisms to enhance participant compliance with the program. E. Replication Issues The data suggest that the provision of intensive case management services, within the social context of a Job Club, even when done on this (less than intended) level of intensity, can have a significant effect on employment outcomes. Other sites can learn from our experience and attempt to implement similar projects. We suspect that, with more intensive implementation, positive outcomes in the other areas targeted (education, family, health) would also emerge.
We recommend against trying to replicate the rigorous evaluation unless sufficient staff, financial resources, and technical expertise is in place. The needs of the evaluation for a relatively large number of cases, coupled initially with inadequate program management, overwhelmed the service delivery aspects of the program. Any replication should have an evaluation staff member on site almost constantly, assuring that implementation of the program and data collection follows intended protocols. The control condition should be presented by staff as a potentially effective alternative approach, and not as an inferior option implemented only for research purposes.
F. Final Recommendations Our overall conclusion is that the MMOR case management and Job Club approach to providing services to this population is significantly more effective than a limited job placement approach in increasing rates of employment. Thus, the model deserves further careful and rigorous development and evaluation, with adequate time and resources to implement the model intensively. The above narrative, and our detailed report, are full of implicit and explicit recommendations which should be consulted in doing so. A final caveat, however, is to avoid an approach which neglects structural issues and focuses solely on the unemployed individual. A truly effective approach to the problem of minority male unemployment needs to focus comprehensively on issues of racial bias in the educational and employment sectors, family structures with an increasing absence of positive male role models, structural features of the economy in which well-paying industrial jobs have disappeared, transportation systems in which individuals without access to private vehicles have limited access to employment, and failures of our education and training systems to develop alternative models to the traditional high school. Job availability is a critical underlying factor. These structural issues, as well as individual training and educational approaches, are critical to solving the problem of minority male unemployment. |
