Five Year Effects

Views:
 
Category: Education
     
 

Presentation Description

No description available.

Comments

Presentation Transcript

Five-year Effects of an Anti-Poverty Program on Marriage Among Never-Married Mothers : 

Five-year Effects of an Anti-Poverty Program on Marriage Among Never-Married Mothers Anna Gassman-Pines and Hirokazu Yoshikawa New York University OPRE Welfare Research and Evaluation Conference June 5, 2006 The views expressed in written conference materials or publications and by speakers and moderators at HHS-sponsored conferences, do not necessarily reflect the official policies of the Department of Health and Human Services; nor does mention of trade names, commercial practices, or organizations imply endorsement by the U.S. Government.

Recent and upcoming publications: 

Recent and upcoming publications Gassman-Pines, A. andamp; Yoshikawa, H. (2006). Five-year effects of an anti-poverty program on marriage among never-married mothers. Journal of Policy Analysis and Management, 25, 11-30. Gassman-Pines, A., Yoshikawa, H., andamp; Nay, S. (in press). Can money buy you love? The relationship between dynamic employment characteristics, the New Hope Program, and entry into marriage. In H. Yoshikawa, T. S. Weisner, andamp; E. Lowe (Eds.). Making it Work: Low-wage employment, family life, and child development. New York: Russell Sage.

Recent attention to marriage: 

Recent attention to marriage Policymakers concerned about 'retreat from marriage' among low-income individuals Lower marriage rates Rates of marriage are declining

Trends in marriage rates: 

Trends in marriage rates Source: Author calculations from U. S. Census data.

Why concern over marriage?: 

Why concern over marriage? Research on family structure and children’s outcomes Children in two-parent families fare better than single-parent families Higher levels of school achievement Fewer behavior problems Married two-parent families better than cohabiting two-parent families Single-parent family significant risk factor

Policies to encourage marriage: 

Policies to encourage marriage Current welfare reform efforts Types of policies and programs Marriage education programs Relationship counseling Change tax code eliminate 'marriage penalties' Nearly exclusive focus on dyadic context

Marriage promotion programs: 

Marriage promotion programs Are marriage promotion program models effective for low-income individuals? Different concerns about marriage? Different reasons for entering into marriage? Marriage and childbearing decisions de-coupled

Data from ethnographic research: 

Data from ethnographic research Support a link between economic conditions and marriage New Hope Ethnographic Study Participants not asked specifically about marriage Often talked spontaneously about marriage and relationships Fragile Families Study (Gibson andamp; Edin, 2002)

New Hope study participants described that before marriage they wanted: 

New Hope study participants described that before marriage they wanted Financial stability and accumulation of assets 'She said that she can’t see getting married any time soon. She said that she would like to be more financially secured before she got married. . . She said she would have to have a regular check that was the same every two weeks, and knows where her money is coming from before she ever thought about marriage.'

New Hope study participants described that before marriage they wanted : 

New Hope study participants described that before marriage they wanted Financial stability and accumulation of assets Overall stability/reduction in chaos 'She still would like to be married, but she seemed to think that their lives were too complicated to plan a wedding.'

New Hope study participants said that marriage requires major commitment: 

New Hope study participants said that marriage requires major commitment Participants attach more significance to marriage than to cohabitation 'Heather said that marriage tells you that this commitment is for real and is not just for show for right now.'

Both theory and mothers themselves link economic circumstances and marriage: 

Both theory and mothers themselves link economic circumstances and marriage Importance of increased financial stability Key question: Do programs that increase employment and income also increase rates of marriage?

Effects of employment policies on marriage among single mothers: 

Effects of employment policies on marriage among single mothers Little evidence overall of 1996 welfare reform impact (Blank, 2002; Murray, 2001) Short-term effects of pre-1996 welfare programs (Gennetian andamp; Knox, 2003) Mixed Few effects overall

Effects of employment policies on marriage among single mothers: 

Effects of employment policies on marriage among single mothers Little evidence overall of 1996 welfare reform impact (Blank, 2002; Murray, 2001) Short-term effects of pre-1996 welfare programs (Gennetian andamp; Knox, 2003) Meta-analysis found small, significant positive effect of generous earnings-supplement programs when not combined with a time limit

Effects of employment policies on marriage among single mothers: 

Effects of employment policies on marriage among single mothers Little evidence overall of 1996 welfare reform impact (Blank, 2002; Murray, 2001) Short-term effects of pre-1996 welfare programs (Gennetian andamp; Knox, 2003) Meta-analysis found small, significant positive effect of generous earnings-supplement programs when not combined with a time limit Few long-term follow up data available

New Hope Project: 

New Hope Project Anti-poverty program that aimed to increase both employment and income Earnings supplements Other services Operated outside the welfare system Collected long-term follow up data Collected mixed-methods data: quantitative and qualitative

Research Questions : 

Research Questions Did New Hope affect rates of entry into marriage among never-married mothers? What mediating factors might explain why New Hope affected entry into marriage?

The New Hope Project: 

The New Hope Project Milwaukee, WI Mid-1990s Random Assignment: 1994-1995 Intervention Period: 1994-1997 Five-Year Follow-up: 1999-2000

New Hope Project: “If you are working, you should not be poor”: 

New Hope Project: 'If you are working, you should not be poor' For those working 30+ hours/week: Earnings supplements Health insurance Child care subsidies Community service jobs Program representatives Neighborhood-based store-front offices

New Hope program eligibility: 

New Hope program eligibility Live in one of two target low-income neighborhoods Age 18 or over Income at or below 150% of poverty threshold Willing to work 30 hrs/week

Individuals who volunteered for the New Hope program: 

Individuals who volunteered for the New Hope program Randomly assigned to either: receive New Hope benefits and services control group, not eligible for New Hope New Hope benefits provided on top of other government programs Test: New Hope plus other government aid vs. other government aid only

New Hope Survey Sample: 

New Hope Survey Sample 337 never married mothers Black: 63% Latino: 24% Mean Age: 27 (SD 6) HS diploma / GED: 61% Receiving welfare or Food Stamps: 85% Ever worked full time: 84% Earnings in prior year: $0 38% $1 to $4,999 43% $5,000 or more 19% Youngest child less than 2: 55% Three or more children: 38%

Baseline covariates: 

Baseline covariates Race/ethnicity 2 or more children in household Youngest child under 2 years old Working full-time Earnings in year prior to random assignment Receiving government assistance High school diploma / GED Have a car Maternal age Child gender Child age

Research Question 1: 

Research Question 1 Did New Hope affect rates of entry into marriage among never-married mothers? Hypothesis program that increases employment and income will increase entry into marriage

Data: 

Data Survey 5 years after random assignment Self-reported marital status

Results: 

Results Examine all women who were never married at baseline

New Hope program effect on marriage at 5 years: Increase of 9 percentage points: 

New Hope program effect on marriage at 5 years: Increase of 9 percentage points 11.8% 20.7%

New Hope program effect on entry into marriage among never-married mothers: 

New Hope program effect on entry into marriage among never-married mothers Robust to tests for: Potential attrition bias Marriage defined as with or without cohabitation With and without adjustments for variety of baseline characteristics

New Hope program effect on entry into marriage among never-married mothers: 

New Hope program effect on entry into marriage among never-married mothers Robust to tests for: No effect on cohabitation

New Hope program effect on entry into marriage among never-married mothers: 

New Hope program effect on entry into marriage among never-married mothers Robust to tests for: No effect on cohabitation Effect concentrated among never-married sample members Different effects for ever married group Difference due to initial marital status and not other characteristics

How might New Hope have affected marriage?: 

How might New Hope have affected marriage? Increases in employment and income Theory and empirical evidence linking increased financial resources with marriage

How might New Hope have affected marriage?: 

How might New Hope have affected marriage? Increases in employment and income New Hope might also affect well-being Increased efficacy Decreased depression

How might New Hope have affected marriage?: 

How might New Hope have affected marriage? Increases in employment and income New Hope might also affect well-being People more likely to stay married when they have higher levels of well-being If program improves well-being, might also increase likelihood of marriage

Research Question 2: 

What mediating factors might explain why New Hope affects rates of entry into marriage? Economic: employment; income; wage growth Psychological: goal achievement; well-being Relationship quality Research Question 2

Mediators: 

Mediators 2 year follow up Administrative records Quarterly employment Average total annual income Income from earnings, welfare, Food Stamps, earnings supplements

Mediators: 

Mediators 2 year follow up Administrative records Survey Average job length Wage growth Perceived goal efficacy (6 items; α = .82) Depression (CESD) (20 items; α = .90) Parenting stress (3 items; α = .59) Material hardship (6 items)

Results: Impacts on mediators: 

Results: Impacts on mediators Among never-married mothers, New Hope significantly increased: Employment Income Wage growth Goal efficacy

Results: Mediation: 

Results: Mediation Include mediators in models predicting marriage 1st block: Economic mediators 2nd block: Well-being mediators Including economic mediators in model reduced size and significance of New Hope impact on marriage Adding well-being mediators did not further reduce New Hope coefficient

Slide39: 


Results: Mediation: 

Results: Mediation Run separate models breaking down income by source Earnings and income from earnings supplements significantly related to marriage

Results: Mediation: 

Results: Mediation Including economic mediators in model reduces size and significance of New Hope impact Suggests possible mediating role of economic factors Income, earnings, earnings supplements, and material hardship significantly related to entry into marriage Well-being factors do not appear to mediate New Hope impact on marriage

Another possible mediator: 

Another possible mediator Relationship quality What makes relationships good or bad? Not asked about in survey Use data from ethnographic study

New Hope Ethnographic Study: 

New Hope Ethnographic Study 45 sample members randomly selected from larger survey sample Generalize to survey sample Took place between 2-year and 5-year survey Participant-observation visits by fieldworkers every 2-3 months

Ethnographic study data collection: 

Ethnographic study data collection Gathered information through Semi-structured interviews Informal conversations Visits to important places (ie, workplace, children’s school) Each contact with family summarized in fieldnotes

Coding fieldnotes: 

Coding fieldnotes Extract all fieldnotes relating to topic of interest Initial reading and identification of themes Develop and define codes What to identify Inclusion and exclusion criteria Coding by team of researchers Establish inter-rater reliability

Dynamic analysis of relationships: 

Dynamic analysis of relationships Identify each relationship occurring over 3-year ethnography Fieldnotes describing each relationship coded Positive quality codes: partner involvement with children, childcare and household tasks, financial contributions, emotional support Negative quality: conflict in interactions, drain on household resources, lack of involvement with children, failure to help with household responsibilities, jealousy or stalking, infidelity, domestic abuse, drug use, involvement with criminal justice system

Dynamic analysis of relationships: 

Dynamic analysis of relationships Identify each relationship occurring over 3-year ethnography Fieldnotes describing each relationship coded Based on codes, rated quality of each relationship

Dynamic analysis of relationships: 

Dynamic analysis of relationships Identify each relationship occurring over 3-year ethnography Fieldnotes describing each relationship coded Based on codes, rated quality of each relationship For those in more than one relationship, examined change in quality over time

Results: Relationship quality: 

Results: Relationship quality Overall, most sample members relationships mixed in terms of quality Included both positive and negative features Quality of relationships improved over time Improvement concentrated among those who experienced wage growth in prior two years

Slide50: 


Quality of relationship at end of study: 

Quality of relationship at end of study At the end of the ethnography, most women in relationships of mixed quality More women who had experienced wage growth in positive relationships Fewer women who had experienced wage growth in negative relationships

Relationship quality and New Hope: 

Relationship quality and New Hope Few differences between New Hope and control groups No one in New Hope group experienced decline in relationship quality Slightly larger percentage of New Hope group experienced improvement in relationship quality

Summary: Relationship quality: 

Summary: Relationship quality Overall, most relationships mixed in quality Relationship quality improved over time, especially for those who experienced wage growth Suggests link between increase financial resources and relationship quality Relationship quality might play a role in explaining New Hope impact

Summary of findings: 

Summary of findings New Hope program substantially increased rates of entry into marriage among never-married mothers Increases in income may help explain New Hope impact on marriage Especially income from earnings and earnings supplements

Limitations: 

Limitations Sampling particular to New Hope eligibility criteria May not generalize to welfare population

Limitations: 

Limitations Sampling particular to New Hope eligibility criteria No survey data and limited ethnographic data from men Difficult to disentangle role of earnings supplement from other NH services Potential selection bias in non-experimental analyses Mediators  marriage

Policy Implications: 

Policy Implications Current marriage promotion policy is not testing efforts to increase both employment and income

Policy Implications: 

Policy Implications Policymakers should consider efforts to increase employment and income Evaluate alone or in combination with relationship-focused marriage promotion efforts Random assignment to multiple conditions: Relationship enhancement alone Increase employment and income alone Relationship enhancement and increase employment and income

Thanks to: : 

Thanks to: National Institute of Mental Health predoctoral fellowship APF Koppitz Child Psychology Fellowship William T. Grant Foundation Scholars Award The David and Lucile Packard Foundation John D. and Catherine T. MacArthur Foundation NYU students: Sandra Nay, JoAnn Hsueh, Erin Godfrey, Amanda Roy, Frank Gaytan, Maria Ramos Comments on prior versions: J. Lawrence Aber, Lindsay Chase-Lansdale, Kathryn Edin, Lisa Gennetian, Robinson Hollister, Julie Kerksick, Edward Lowe, Sara McLanahan, Cynthia Miller, Rashmita Mistry, Thomas Weisner