Teaching Self-Sufficiency: 30-Month Impacts of a Home Visitation and Life Skills Education Program for Hard-to-Employ TANF Recipients Findings from the Rural Welfare-to-Work Evaluation Alicia Meckstroth Andrew Burwick Quinn Moore Andrew McGuirk APHSA/CBPP Teleconference on Home Visiting in TANF January 9, 2008 Alicia Meckstroth Andrew Burwick Quinn Moore Andrew McGuirk APHSA/CBPP Teleconference on Home Visiting in TANF January 9, 2008

Nebraska’s TANF Policy Context Supportive, work-oriented TANF program Targeted education and training Two-year time limit Various services available in target areas Modest unemployment and poverty Supportive, work-oriented TANF program Targeted education and training Two-year time limit Various services available in target areas Modest unemployment and poverty 2 NOT FOR CITATION

Building Nebraska Families (BNF) Nonexempt, hard-to-employ TANF recipients University extension and state welfare agency Masters’ level educators with very small caseloads 11 service areas throughout Nebraska Nonexempt, hard-to-employ TANF recipients University extension and state welfare agency Masters’ level educators with very small caseloads 11 service areas throughout Nebraska 3 NOT FOR CITATION

Building Nebraska Families (BNF) 4 NOT FOR CITATION Individualized Life Skills Education Through Home Visits Mentoring and Informal Counseling Service Coordination & Advocacy Support Personal Improvement: goal setting, problem-solving, character development, coping skills, relationship- building, communication skills Family Life: child development, parenting, family management Practical Life Skills: money and time management, healthy home, nutrition

Research Questions Program implementation and costs? Effects on employment, earnings, welfare dependence, and well-being? Implications and lessons? Program implementation and costs? Effects on employment, earnings, welfare dependence, and well-being? Implications and lessons? 5 NOT FOR CITATION

Evaluation Methods Experimental design (358 Ts, 242 Cs) Follow-up telephone surveys at 18 and 30 months (87 percent and 83 percent completion rates) Administrative records Program service use and participation data Site visits and focus groups Experimental design (358 Ts, 242 Cs) Follow-up telephone surveys at 18 and 30 months (87 percent and 83 percent completion rates) Administrative records Program service use and participation data Site visits and focus groups 6 NOT FOR CITATION

Subgroup Analysis: “More Disadvantaged” Met 2 or more of these criteria at baseline: - Lack of high school credential - Health condition (self or HH member) - Transportation barrier (no driver’s license or regular access to vehicle) - No earnings in prior year - Received TANF/AFDC for 2+ years in lifetime 43 percent were more disadvantaged Met 2 or more of these criteria at baseline: - Lack of high school credential - Health condition (self or HH member) - Transportation barrier (no driver’s license or regular access to vehicle) - No earnings in prior year - Received TANF/AFDC for 2+ years in lifetime 43 percent were more disadvantaged NOT FOR CITATION

Well-Implemented Program Services delivered successfully Intensive services -Participation over 8 months - 22 contacts -25 hours Program group members more likely to receive skill- building services and mentoring Total cost per participant = $7,200 (approx) Services delivered successfully Intensive services -Participation over 8 months - 22 contacts -25 hours Program group members more likely to receive skill- building services and mentoring Total cost per participant = $7,200 (approx) NOT FOR CITATION

Highlights of Impact Findings for the Full Sample Limited evidence that BNF improved employment status for the full sample No convincing evidence that BNF improved earnings for the full sample Limited evidence that BNF improved employment status for the full sample No convincing evidence that BNF improved earnings for the full sample 9 NOT FOR CITATION

SOURCE: Rural Welfare-to-Work Evaluation's 30-Month Follow-up Survey of BNF sample members. NOTE:The estimates were adjusted using multivariate regression methods and the data were weighted to account for survey nonresponse and to equalize the size of the program and control groups. */**/***Significantly different from zero at the.10/.05/.01 level, two-tailed test. BNF Improved Employment for the More Disadvantaged Group *** Percentage Employed 10 NOT FOR CITATION ** * Program Group Control Group * *

SOURCE: Rural Welfare-to-Work Evaluation's 30-Month Follow-up Survey of BNF sample members. NOTE:The estimates were adjusted using multivariate regression methods and the data were weighted to account for survey nonresponse and to equalize the size of the program and control groups. */**/***Significantly different from zero at the.10/.05/.01 level, two-tailed test. BNF Substantially Increased Earnings for the More Disadvantaged Group **** Average Earnings ** 11 NOT FOR CITATION ** Program Group Control Group ** * *

Program Group Members Held Better Jobs 12 NOT FOR CITATION Percentage SOURCE: Rural Welfare-to-Work Evaluation's 30-Month Follow-up Survey of BNF sample members. NOTE:The estimates were adjusted using multivariate regression methods and the data were weighted to account for survey nonresponse and to equalize the size of the program and control groups. The sample for these variables includes working and nonworking members. */**/***Significantly different from zero at the.10/.05/.01 level, two-tailed test.

BNF Reduced Welfare Receipt NOT FOR CITATION SOURCE: Administrative records data from the State of Nebraska, compiled by Mathematica Policy Research, Inc. NOTE:The estimates were adjusted using multivariate regression methods and the data were weighted to account for survey nonresponse and to equalize the size of the program and control groups. */**/***Significantly different from zero at the.10/.05/.01 level, two-tailed test. Percentage Receiving TANF Control Group Program Group ** *** ** *** 13 * *** * **

BNF Increased Household Income 14 NOT FOR CITATION Total Household Income at 30- Month Follow-up Living in Poverty at 30-Month Follow-up Percentage 2004 dollars SOURCE: Rural Welfare-to-Work Evaluation's 30-Month Follow-up Survey of BNF sample members. NOTE:The estimates were adjusted using multivariate regression methods and the data were weighted to account for survey nonresponse and to equalize the size of the program and control groups. */**/***Significantly different from zero at the.10/.05/.01 level, two-tailed test.

Fewer Health-Related Hardships, More Household Hardships NOT FOR CITATION 15 Overall health fair or poor Self-reported depression or mental health issue Spousal or partner abuse Had utility turned off Ever had serious housing problem Often or sometimes not enough money to buy food Percentage SOURCE: Rural Welfare-to-Work Evaluation's 18-Month Follow-up Survey of BNF sample members. NOTE:The estimates were adjusted using multivariate regression methods and the data were weighted to account for survey nonresponse and to equalize the size of the program and control groups. */**/***Significantly different from zero at the.10/.05/.01 level, two-tailed test.

What Might Influence BNF’s Impacts on Employment and Earnings? Services complement existing employment- related assistance Home visits allow for individualized support— especially valuable to more disadvantaged Well-developed life skills curriculum Highly-qualified, professional staff Low caseloads Services complement existing employment- related assistance Home visits allow for individualized support— especially valuable to more disadvantaged Well-developed life skills curriculum Highly-qualified, professional staff Low caseloads 16 NOT FOR CITATION

Questions and Implications What consideration should be given to BNF’s cost? How might BNF operate within context of new work and participation requirements? Could the BNF program model transfer to other environments? What additional supports might be needed for clients who become employed? What consideration should be given to BNF’s cost? How might BNF operate within context of new work and participation requirements? Could the BNF program model transfer to other environments? What additional supports might be needed for clients who become employed? 17 NOT FOR CITATION

Rural Welfare-to-Work Strategies Demonstration Evaluation For more information: Mathematica Policy Research Alicia Meckstroth, (614) , Previous evaluation reports available at U.S. DHHS, ACF Michael Dubinsky, (202) , Karl Koerper, (202) , University of Nebraska Cooperative Extension Marilyn Fox, (308) , (BNF curriculum) For more information: Mathematica Policy Research Alicia Meckstroth, (614) , Previous evaluation reports available at U.S. DHHS, ACF Michael Dubinsky, (202) , Karl Koerper, (202) , University of Nebraska Cooperative Extension Marilyn Fox, (308) , (BNF curriculum) 18