Life After Welfare: Report of the Georgia Welfare Leavers Study
Technical Appendices

Prepared for: Georgia Department of Human Resources Division of Family and Children Services
January 2001

Prepared by:

E. Michael Foster, Ph. D., Principal Investigator

Dana K. Rickman, Project Manager

Applied Research Center Andrew Young School of Policy Studies
Georgia State University Atlanta, Georgia

Table of Contents
I. Technical Appendix I Welfare Reform in Georgia II. Technical Appendix II Design of the Georgia State Study III. Technical Appendix III Response Rate and Sample Representativeness IV. Technical Appendix IV ASPE Leavers Studies: Summary of Findings References

p. 1 p. 6 p. 13 p. 26 p. 37

Technical Appendix I -- Welfare Reform in Georgia In this section we review distinguishing features of TANF as implemented in Georgia and describe recent trends in the welfare caseload.
TANF in Georgia Several key features distinguish welfare reform in Georgia, including fairly strict employment standards and program guidelines. TANF recipients must sign both a personal responsibility contract and an employment contract. The personal responsibility contract requires participation in work activities, adherence to child school attendance requirements, participation in life skills or parenting classes, drug or alcohol provisions (if deemed necessary), and attendance at parent/teacher conferences. The applicant also must complete an employability plan that covers work requirements. The only applicants exempt from the employability plan are those with a child under 12 months (State Policy Documentation Project, 1999). The State has opted for a four-year time limit (one year less than the federal guidelines allow) with no exemptions based caring for disabled persons (as allowed in some states). The state also has eliminated the earnings disregard for working recipients. Recipients must have countable assets below $1000. Comparatively, thirty states have asset limits between $2,000 and $3,000, and another six states have limits between $3,500 and $5,000. Child support income is counted as income compared to thirty-one states that have a full disregard for child support earnings. However, in Georgia, housing

Life After Welfare, Technical Appendices (2/19/01)

1

assistance, EITC payments, and Individual Development Accounts (IDAs)1 under $5,000 are exempt. A family member receiving SSI benefits is excluded from the assistance unit (State Policy Documentation Project, 1999).
Unlike some states (e.g., Idaho), Georgia does pay larger amounts to larger families, but Georgia is among the minority of states that have imposed a strict "family cap"welfare payments are not increased for those women giving birth while on welfare after the case has been open for 10 months. As required by federal law, teen mothers must live with their parents, an adult relative or guardian, supervisory adult (group home), or an agency-approved independent living arrangement. All single teen parents under the age of 18 are subject to the school/training requirement (State Policy Documentation Project, 1999).
Finally, an added difference between Georgia and other states involves the formation of childonly cases. Federal guidelines allow states broad discretion in defining and regulating child-only cases. Federal guidelines only require that "[i]f an adult recipient refuses to engage in required work, the State must reduce either aid to the family ``pro rata'' (or more, at State option) with respect to the period of work refusal, or the adult benefit must be ended" (Ways and Means Committee Print WMCP:105-7 1998 Green Book). The termination of adult benefits will produce TANF cases that contain only the children as benefit recipients.2

1IDAs may be used for the following purposes: post-secondary educational expenses, first home purchases, and business capitalization.
2Under federal guidelines, children can only be terminated from a case if they are "[u]nwed mothers under 18 (and their children) unless they live in the home of an adult relative or in another adult-supervised living arrangement; or unwed mothers under 18 without a high school diploma unless they attend school" (Ways and Means Committee Print WMCP:105-7 1998 Green Book).

Life After Welfare, Technical Appendices (2/19/01)

2

These guidelines can be viewed as loopholes for TANF recipients. By removing themselves from the grant, parents could continue to receive benefits for their children while freeing themselves from work requirements and time limits. As a result, the State of Georgia closes the entire case when a parent fails to comply with TANF requirements. For a child to continue receiving support he or she must live in the home of "a specified relative or legal guardian that is not the parent" (Economic Support Services Manual, p. 1045-1). A "specified relative" is defined a grandparent (up to great-great-great), a sibling (half, whole, step), an aunt or uncle (up to great-great), a first cousin, a child of a first cousin, or a spouse of any person named above.3

Caseload Trends in Georgia The tables below highlight trends in Georgia's TANF caseloads and administrative services.4 Note that the Georgia TANF law took effect in January 1997. The interview data presented in this report examines leavers who were interviewed between June 1999 and October 2000. The TANF reform laws have had a dramatic impact on the welfare caseload in Georgia. As figure 1 indicates, the monthly average of active cases has dropped from 139,265 families in 1995 to 65,198 families in 1999 (Georgia Division of Family Services, 1999). This represents a 53% decline over four years. The trends in TANF have been accompanied by dramatic trends in other public assistance programs.

3Georgia law defines the following relationships as a specified relative: grandparents (up to great-greatgreat), siblings (half, whole, step), aunts/uncles (up to great-great), first cousin, child of a first cousin, or spouse of any person named above.
4All case load trend information presented is provided by the Department of Human Resources, Georgia Division of Family and Children Services, Descriptive Data Manual, State Fiscal Year 1999.

Life After Welfare, Technical Appendices (2/19/01)

3

Number of Active Family Cases

Fig. 1: Georgia AFDC and TANF Trends: Monthly Average Families: 1995-1999

160,000 140,000 120,000 100,000
80,000 60,000 40,000 20,000
0

139,265 132,627 114,154 84,525 65,198
1995 1996 1997 1998 1999

Figure 2 presents the dollar amount spent on Food Stamps annually between 1995 and 1999. There has been a decline from $692,666,787 in 1995 to $536,281,477 in 1999. Coinciding with the dramatic decrease in the number of TANF cases, the expenditures on Food Stamps had dropped by approximately 22%.

Dollars ($) Allocated in Millions

Fig. 2 Food Stamp Trends: Annual Benefits 1995-1999

800 600 400 200
0
1995

1996

1997

1998

1999

Life After Welfare, Technical Appendices (2/19/01)

4

Figure 3 illustrates the increase in the average number of monthly AFDC/TANF-related Medicaid5 cases from 1995-1999. In 1995, there were 148,764 Medicaid cases compared to 218,911 in 1999. This represents a 32% increase.

Number of Active Cases

Fig. 3: AFDC/TANF Related Medicaid: Monthly Average Cases 1995-1999

300,000 200,000 100,000
0

1995

1996

1997

1998

1999

Finally, figure 4 shows the trend in annual child care expenditures by the State. From 19951998, there was a 46% increase as expenditures rose from $74,623,661 in 1995 to $138,920,826. However, there was a 13% decline in 1999 as expenditures fell to $121,152,111.

Dollar ($) Expenditures in
Millions

Fig. 4: Child Care Annual Expenditures: Fiscal Years 1995-1999
150 100
50 0 1995 1996 1997 1998 1999

5This excludes recipients of aged, blind, and disabled Medicaid.

Life After Welfare, Technical Appendices (2/19/01)

5

Life After Welfare, Technical Appendices (2/19/01)

6

Technical Appendix II-- Design of the Georgia Study The following provides an overview of the Georgia State study and compares it to leavers studies in other states. While there are similarities, this study differs in several crucial ways from these other studies.

Definition of Leavers Like the other ASPE-funded studies, the Georgia leavers study defines leavers as cases not having received cash assistance for two consecutive months. However, unlike the other studies, Georgia is the only study including child-only cases in their population of leavers. It is also important to note that we do not exclude cases that return to the rolls from our analyses. This is also true of the other ASPE studies, except for the state-funded portions of studies in South Carolina and Wisconsin. Including cases that return to the rolls provides a fuller picture of how leavers are faring as a whole.

Data Sources Like the other studies, our study relies on interview data as well as data from various administrative databases. Telephone Interviews As designed, the study involves ongoing telephone interviews with a sample of families having left TANF. We are interviewing approximately 200 women per month over a 24-month period extending from July 1999, through April 2001. Our sample size is quite large compared to the other ASPE studies. Only four other studies have a sample size more than 1,000.

Life After Welfare, Technical Appendices (2/19/01)

7

Use of Administrative Data We are linking survey respondents to various state databases. The initial sample of cases is
drawn from DFCS' database of closed clients. The TANF database provides case and client id numbers, and basic demographic information, including race, age, gender and relationship of respondent to other persons in the household. Additionally, the database provides information on case status, issued payments, address, telephone number, county office, and supervisor number. The initial closed-case file is matched with the TANF Federal Report file that provides supplemental information regarding the case at closure, such as Food Stamp receipt, work eligibility status and work experience, reported earnings, and family structure. Three to four months later, these files are again matched with DFCS' database of active cases to determine recidivism rates and possible location information.
This study also uses the Child Support Enforcement (CSE) database. The CSE database provides two important pieces of information. First, administrative matches are used to verify or update address and telephone information. CSE clients still receive their checks through the mail; therefore, CSE database is more likely to have accurate information than TANF records. Second, the CSE database provides benefit histories and payment information.
Unfortunately, we were unable to link to employment record or ongoing Food Stamp or Medicaid databases. Even so, taken together, the combined administrative databases provide information on employment, Food Stamp use, past use of TANF (both cash payments and other involvement, such as sanctions), current receipt of TANF, as well as receipt of child support.
For a complete comparison of survey and administrative data across studies, see table 1.

Life After Welfare, Technical Appendices (2/19/01)

8

Research Topics

When combined, the administrative and interview data cover a wide range of topics consistent

with those covered in similar leaver studies.

C

Employment and Earnings

The telephone survey asks questions pertaining to the individual's present employment, hours worked,

earnings, and occupation. Questions also probe job search strategies, and length of employment or

unemployment. The TANF Federal Report file also provides information on earned and unearned

income and on work activities before exit.

C

Other Income Supports

In addition to earnings through employment, for a selected number of respondents, the telephone survey

provides further information on income support, including CSE payments, SSI payments, and earned

income credits.

C

Health Insurance

The telephone survey includes questions about Medicaid and employer-provided health insurance for

the respondents and their children.

C

Child Care

We are relying solely on survey data for this information. Detailed questions are asked concerning

types of child care, frequency of use, and satisfaction of child care options.

C

Child Well-being

The combination of questions in the telephone survey is aimed at assessing child-well being. Basic

information about health insurance, school attendance, participation in child care, and disability is

Life After Welfare, Technical Appendices (2/19/01)

9

collected. In addition, we also obtain information about the involvement of absent fathers, home

environment, and domestic violence issues.

C

Barriers to Self-Sufficiency

The telephone survey address such subjects as disability, maternal depression, domestic violence, and

the lack of education/skills. The Mother's Mental Health/Domestic Violence module includes the

Rosenberg self-efficacy scale and the CDI depression scale.

C

Deprivation/Insecurity

The survey asks whether former recipients encounter severe problems in meeting basic needs, including

hunger, access to health care, and housing problems. We address adequacy of food, living

arrangements, and health care.

C

Attitudes Toward TANF and Work

We are surveying former recipients about their attitudes toward TANF or welfare reform. We pose

questions concerning their satisfaction with TANF, how they feel now that they have left, their

confidence they will stay off, and their general knowledge of the welfare laws.

In sum, the Georgia State study has features in common with the other ASPE studies but key

differences remain. Like the ASPE studies, the definition of a leaver is based on a two-month period

during which a case receives no cash assistance. The Georgia State and other state studies planned to

utilize similar sources of administrative data and provide information on a similar range of research

topics. Georgia also resembles the other states in terms of the content of the survey interview.6

6Because it is the only study to employ matrix sampling, Georgia's survey interview is among the most comprehensive. There are some topics that we have emphasized less than other states, such as recidivism or detailed neighborhood or community characteristics.

Life After Welfare, Technical Appendices (2/19/01)

10

However, key differences remain between the Georgia State study and the other leavers studies. As noted, this study includes child-only cases and does not exclude cases that have returned to the rolls. In addition, the number of interviews being conducted in this study is generally much larger than in other states. This larger size provides the depth necessary to examine multiple comparisons according to (1) rural-urban location, (2) race, (3) education of leaver; (4) time spent by leaver on welfare and, (5) reason leaver moved off roles (e.g., whether sanctioned). (For an overview of comparisons and those made in other studies, see table 2). We believe this information is critical. It is likely that the effects of welfare reform differ across families of different types. For example, an African-American never-married woman with young children living in an urban area may be affected quite differently than a white divorced woman living in a rural area. Describing such diversity requires a large sample.
There are other differences. The Georgia State study focuses on cases that have closed since the summer of 1999. Therefore, we have no information about those who left under AFDC or immediately after TANF was implemented.

Life After Welfare, Technical Appendices (2/19/01)

11

Table 1: Study Overviews

Population

Administrative Data (All sites include TANF, Medicaid enrollment,
and Food Stamps

Sample Size*

Sample Mode

Adult TANF Leavers Child Only Cases Sanctioned
Eligible not enrolled Diverted
Employment CSE
Child Welfare Foster Care WIC
Housing Asst. SSI
Health Services Mixed Mode Telephone

Georgia Arizona Cuyahoga County District of Columbia

/// / / /

/

/

/

/

/

/

/

7800 completes

/

800 completes

/

300 completes

/

500 sampled

/

375 completes

Florida

/

/

/

15,000 completes

/

Illinois

/

/

/

/

/

800 completes

/

Los Angeles

/

/

/

330 completes

/

Massachusetts

/

/

/

600 completes

/

Missouri

/

/

/

/

4800 completes

/

New York

/

/

/

/

1200 sampled

/

San Mateo

/

//

/

/

950 completes

/

Washington

/

// /

/

1000 completes

/

Wisconsin

/

//

/

/

/

/

2000 completes

/

South Carolina

/

/// /

/

/

/

1000 sampled

/

Total

14 1 3 2 5 11 9

6

2

1

2

1

3

2649 (average) 13 1

* These are planned sample sizes based on grantees original proposals. Some grantees did not achieve the targeted sample size. That information

will be made available in their final reports.

Table 2 Research Topics

Subgroups of Analysis

Urban/Rural Sanctioned Cases Time limit leavers
Race/ethnicity Age of youngest child
Age of recipient Length of Spell Single v. married
Type of exit Recidivism
Number of Children Education and training levels
Housing Assistance AFDC v. TANF leavers

Georgia Arizona Cuyahoga County District of Columbia Florida Illinois Los Angeles Massachusetts Missouri New York San Mateo Washington Wisconsin South Carolina Total

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

10

8

4

12

6

4

3

4

6

12

2

3

2

7

Technical Appendix III -- Response Rate and Sample Representativeness Missing contact information in the administrative data represents a significant challenge to conducting the interviews required for this project. From the data provided by DFCS, approximately 15 percent of the cases have no telephone number (this number does not include those numbers deemed invalid or disconnected), and another 40 percent have incomplete address information. This appendix contains two sections. The first will review our efforts to locate respondents and then will describe the fruits of those efforts to date for each monthly cohort. This section also will outline which location techniques were most effective. The second section will address the representativeness of the sample. We also will compare key outcomes for each monthly cohort as our response rate increased.
Definitions for Selecting Respondents To identify which individual in the case to interview, we combine two different codes in the TANF database: the relationship code and the financial responsibility code. For single-parent cases, we select the individual whose relationship code is "SE" (self, head of household) and their financial responsibility code is "RE" (recipient). We identify respondents in child-only cases based on the same two codes. Those individuals with a relationship code of "SE" (self, head of household) and a financial responsibility code that is not "RE" (recipient) are identified as respondents in child-only cases. This coding was confirmed by the work participation code in the TANF Federal Report file.

Life After Welfare, Technical Appendices (2/19/01)

14

Locating Respondents In order to locate individuals, we have implemented a step-by-step process that utilizes all of
the resources available to us. This process represents a significant improvement over what was included in our original proposal.
The process of locating respondents begins with contact letters that are mailed (at the beginning of the survey month) to all individuals eligible to be interviewed. The letter announces that we will be contacting potential respondents over the next month, explains the study, and describes how their names were selected. The letter also provides a toll-free telephone number for individuals to call the survey lab if they feel we will have difficulty locating them over the telephone. The letter also offers $25 to any individual who calls the Applied Research Center to complete the interview.
After the letters are mailed, we begin our efforts to contact respondents by phone. The Applied Research Center's telephone survey lab utilizes a variety of techniques to increase our response rate. First, numbers are called up to seven times until they can be verified as disconnected or invalid. Calls are made to hard-to-reach households at various times throughout the day and week. For numbers with answering machines, messages providing the toll-free number and explaining the incentive are left. Disconnected and invalid numbers are re-called after two weeks to allow for the reconnection of telephone service. As we locate additional phone numbers through our search efforts (described below), contact information is updated, and the process continues for up to three months or more.
For individuals for whom DFCS has no phone number or for whom the DFCS number is disconnected or otherwise invalid, we begin a series of steps to locate valid contact information. Once

Life After Welfare, Technical Appendices (2/19/01)

15

a telephone number is deemed invalid, or if there is no number listed in DFCS records, we then use directory assistance and reverse directory look-ups to identify new numbers. (These look-ups are done using a software program called Pro-CD. That contains nationwide addresses and telephone numbers.) Numbers are then matched against the Child Support Enforcement (CSE) database. CSE mails checks directly to their clients. Therefore, the addresses in their database tend to be more accurate than the food stamp information.
When no valid phone number can be located, ARC staff interviewers go to the last-known address of the respondent and start their field tracking procedures. In-person tracking is limited to the Atlanta metropolitan area defined as inside the I-285 perimeter. Once individuals are located, the trackers provide them with a cellular phone to complete the interview over the phone. Multiple visits are made until the respondent's address is verified. If the respondent is not home, a card notifying them of the tracker's visit and the lab's phone number is left. If the respondent is busy, an appointment to complete the phone interview is scheduled.
Finally, the names of any individuals we are still unable to locate are sent to the local DFCS offices. DFCS employees provide the ARC with any additional location information they may have on the respondent.

Life After Welfare, Technical Appendices (2/19/01)

16

Response Rates Reflecting ongoing improvements in our tracking techniques, the study's response rate has
improved steadily. Below is a detailed break down of our monthly response rates.

Survey Cohort

Total Sample Size

Response Rate (%)

Valid Sample Size*

Adjusted Response Rate (%)**

June `99

600

26.0

293

53.2

August `99

600

34.1

370

55.1

September `99

600

37.3

413

54.2

October `99

600

52.6

355

89.0

December `99

600

50.6

362

83.9

February `00

450

54.6

246

87.2

March `00

450

63.6

286

87.5

April `00

450

67.8

323

94.4

May `00

450

66.9

336

89.6

June `00

450

70.1

378

83.3

August `00

400

69.3

328

84.1

* Excludes those where no working telephone number could be established. ** Response rate for valid sample.

Refusal Rate (%)
1.7 0.5 2.2 0.9 2.8 7.5 2.5 2.7 1.3 0.4 2.3

Efforts to Locate Individuals As stated earlier, respondents can be located in any of seven ways. Below we present which
location techniques were most effective. This information is provided separately for the "beltway" cases in table 3. Only individuals in these areas were eligible for the in-person tracking. Given the high percentage of individuals calling in to complete the interview, one can surmise that the incentive

Life After Welfare, Technical Appendices (2/19/01)

17

payments greatly increased our response rate. One can also see that the response for cases in and out of the beltway were roughly comparable. In early months, the rate for the beltway cases was higher; by the May cohort, the rate for the other cases was somewhat higher. Finally, one can see that in the last few months of data collection, in-person tracking was especially valuable as it served to narrow the gap between beltway and other cases. One can see that DFCS personnel generally accounted for a higher percentage of non-beltway cases.

Life After Welfare, Technical Appendices (2/19/01)

18

Table 3

Oct. `99

Dec. `99

Feb. `00

March `00

April `00

May `00

June `00 August `00

Beltway* Other Beltway* Other Beltway* Other Beltway* Other Beltway* Other Beltway* Other Beltway* Other Beltway* Other

Percent Completed By Type

GSU interviewer contact

32.2 41.0 53.7 62.9 32.1 40.1 32.8 44.9 22.0 26.2 17.0 34.3 19.5 34.1 23.9 28.6

Respondent call-in

37.6 53.7 31.7 33.0 37.7 45.1 43.1 50.9 26.3 30.7 26.0 32.6 31.9 30.2 29.3 33.4

Directory assistance

2.2 2.2 6.1 4.1 7.5 4.4 1.7 1.7 33.9 4.8 3.0 2.6 2.7 2.7 4.3 3.6

Pro-CD look-ups

1.1 2.2 0

0

0

0

0

0 16.9 4.2 3.0 0.8 2.7 0.9 2.2 3.2

CSE

1.1 0.9 0

0

0

0

0

0

0

0

0

0

0

0

0

0

In-person tracking

25.8 NA 8.5 NA 20.8 NA 20.7 NA 10.2 NA 10.0 NA 10.6 NA 4.3 NA

Local DFCS personnel

N/A NA N/A NA 1.9 1.6 1.7 2.6 0.8 2.7 0.1 2.8 5.3 0.9 1.0 2.6

Response rate before field interviews and local DFCS personnel

44.8 51.3 52.1 48.5 38.3 47.5 46.4 64.6 53.4 65.9 49.0 69.7 56.6 68.5 59.8 68.8

Final response rate by area 60.3 51.3 56.9 48.5 49.5 53.1 59.8 66.3 64.4 68.6 60.0 71.2 72.5 69.4 65.2 71.4

Final response rate (total)

53.6

50.5

55.6

63.6

67.8

66.9

70.1

69.3

* this reflects the percent completed of eligible respondents (those who live in the Atlanta metropolitan area)

Life After Welfare, Technical Appendices (2/19/01)

19

Representativeness of the Sample The response rate for this project is less than ideal but is in the range of response rates among comparable projects. The quality of a study, however, depends not only on the response rate but on the extent to which respondents and non-respondents differ. The response rate could be rather high (80%), but the study might be very misleading if the 20% who do not respond differ substantially from those who complete interviews. At the same time, a study with a lower response rate might describe the population of leavers accurately if respondents and non-respondents are similar. Generally, any bias induced by non-response is difficult to diagnose. In many instances, one knows very little about these individuals. In the case of leavers studies, however, we know a great deal about these individualsbasic demographics, history of welfare use at the time they left the rolls and whether they have returned to the rolls. This information is extremely valuable and provides important insights into the nature of any biases induced by non-response. In Table 47 we consider differences among individuals for whom the project was and was not able to complete telephone interviews. The table presents the results of a logit model, a form of regression well-suited to dichotomous outcomes. In that model, we consider whether one could have predicted who would complete interviews based on various characteristics (taken from administrative data). We estimated statistical models using the roughly 3000 individuals we tried to contact for the study between October and May. The table has five columns. The first is the estimated beta

7 Table 4-A uses data from the June 1999 interview cohort through the February 2000 interview cohort. For these cohorts, the TANF Emergency data file was made available to us for analysis and the variables used in the logit regression are from that file. Table 4-B provides limited analysis on the March 2000 through May 2000 interview cohorts as the TANF Emergency file was not available.

Life After Welfare, Technical Appendices (2/19/01)

20

coefficient from the regression model; and the second is the corresponding standard error. The third and fourth columns present t-statistics and the corresponding p-value. These gauge the statistical significance of the relationship between a given characteristic and the likelihood of response. Finally, because the beta coefficient has no straightforward interpretation, we translate that coefficient into a 'marginal effect'the impact of the characteristic of interest on the likelihood that an individual was interviewed. For example, individuals who lived in rural growth region were 8.2 percentage points less likely to complete an interview.

Life After Welfare, Technical Appendices (2/19/01)

21

Table 4-A
Demographics age household size presence of Child <5 race
Characteristics of Case when Leaving TANF case type8 amount of last benefit # of months received TANF received food stamps intercept
Interview Cohort (relative to omitted category: July 1999 ) August 1999 September 1999 October 1999 December 1999 February 2000
Region (relative to omitted category: Urban ) Suburban Rural Growth Rural Decline
Observations Log Likelihood+
8Child-only cases relative to single-parent cases
Life After Welfare, Technical Appendices (2/19/01)

Beta
0.01 -0.02 -0.01 -0.23
-0.09 0.00 0.00 0.04 0.27
-1.10 -0.98 -0.06 -0.19 -0.24
-0.08 -0.35 0.15

SE

T-value

0.01

1.70

0.05

-0.45

0.07

-0.08

0.12

-1.80

0.16

-0.53

0.00

0.29

0.00

1.70

0.11

0.42

0.33

0.82

0.23

-4.90

0.23

-4.50

0.13

-0.48

0.28

-0.69

0.22

-1.10

0.13

-0.69

0.13

-2.70

0.18

0.85

1816 -1189.52

p-value
0.09 0.65 0.93 0.06
0.59 0.77 0.09 0.67 0.41
0.00 0.00 0.63 0.49 0.41
0.48 0.01 0.39

Marginal Effects
0.002 -0.001 -0.001 -0.056
-0.021 0.000 0.001 0.012 0.067
-0.259 -0.238 0.036 0.046 0.058
-0.022 -0.086 0.037

22

Psuedo-R squared Table 4-B
Demographics age race
Characteristics of Case when Leaving TANF case type amount of last benefit
Interview Cohort (relative to omitted category: March 2000) April 2000 May 2000 Region (relative to omitted category: Urban ) Suburban Rural Growth Rural Decline
Observations Log Likelihood+ Psuedo-R squared

.04

Beta

SE

T-value

p-value

Marginal

Effects

0.01

0.21

0.51

0.61

0.002

-0.18

0.11

-1.50

0.12

-0.054

0.34

0.14

2.30

0.01

0.078

0.00

0.00

-0.10

0.92

0.000

0.09

0.14

0.67

0.50

0.021

0.18

0.14

1.30

0.19

0.042

-0.09

0.15

-0.61

0.54

-0.021

-0.19

0.14

-1.40

0.17

-0.044

0.04

0.20

-1.40

0.17

0.011

1350 -874.10
.05

Life After Welfare, Technical Appendices (2/19/01)

23

Finally, we compare the results average monthly income, education, race and employment as our response rate increased. The results are presented below.
Figure 5 shows that as our response rate increased the reported mean earnings decreased. The response rate for cohort 1 was 26% and the corresponding mean earnings was $1,058. Conversely, the response rate for Cohort 9 was 67% and the mean earnings fell to $544.

Response Rate (%)
Avg. Monthly Income

Fig. 5 Percent Complete by Average Monthly Earnings

80%

$1,500

60% 40% 20%

$1,000 $500

0%

$0

1 23 4 56 78 9

Cohort

Response Rate

Avg. Monthly Income

Figure 6 shows that even though our response rate increased over time, the rates of employment stayed relatively stable, averaging around 60%.

Life After Welfare, Technical Appendices (2/19/01)

24

Response Rate (%)
Employment Rate (%)

Fig. 6 Response Rate by Employment

80%

70%

60%

60% 50%

40%

40% 30%

20%

20% 10%

0%

0%

12 3 4567 89

Cohort

Response Rate

% Employed

Figure 7 shows that the average education level decreased with an increase in response rates. Only 19% of those in cohort 1 did not have a high school education or GED, compared to 27% of those in cohort 9.

Response Rate % With Less Than
High School Diploma

Fig. 7 Response Rate by Education

80%

35%

70% 60% 50% 40% 30% 20% 10%

30% 25% 20% 15% 10% 5%

0%

0%

12 3456789

Cohort

Response Rate

% With Less Than a High School Diploma

Life After Welfare, Technical Appendices (2/19/01)

25

Finally figure 8 shows the percentage of African-Americans responding to the survey remained relatively stable, fluctuating between 74% and 81%.

Response Rate (%) AfricanAmerican
(%)

Response Rate by Percentage of AfricanAmericans Interviewed

80%

85%

60% 40% 20%

80% 75%

0%

70%

1234 56789

Cohort

Response Rate

% African-American

Life After Welfare, Technical Appendices (2/19/01)

26

Technical Appendix IV ASPE Leavers Studies: Summary of Findings This section summarizes the findings from the other ASPE leaver studies.9 While our study differs from other studies in important ways, our findings inevitably will be compared with those in other states, and so it is relevant to note the major findings of other states. First, the review focuses on other states who received ASPE funding since the Georgia State project received supplemental funding from the Assistant Secretary of Planning and Evaluation (ASPE) of the US Department of Health and Human Services; Arizona, Illinois and Washington State. We also include other leavers studies from other funding sources; Cuyahoga County, Ohio, and Virginia. Based on their final reports, we present a brief outline of these studies, and highlight their findings. Arizona The Arizona study (Westra and Routely, 2000) examined the experiences of cash assistance cases that closed during the three-month period, January 1998- March 1998 (10,657 cases). These cases were matched with information from the Arizona Department of Economic Security (ADES) unemployment insurance, cash assistance, Food Stamps, Medicaid, childcare, child support, child protective services, community services automated systems to determine wages, recidivism, and use of ADES benefits. A random sample (821 respondents) was selected for follow-up interviews. The survey included information about reasons for leaving cash assistance, use of other services, and to assess respondents' perceptions of their experiences and current status.

9 The studies selected where those who had made their reports available by July, 2000.

Life After Welfare, Technical Appendices (2/19/01)

27

Key findings of the report are summarized below.



Leaving cash assistance

Survey respondents identified employment (54%) and sanctions (11%) as the top two reasons

for leaving cash assistance. Administrative data indicated that 20% of the cases were closed due to a

sanction. The study found that individuals from these cases were different than their non-sanctioned

counterparts. Adults in the sanctioned cases were less likely to have completed high school (45% v.

54%) and more likely to have reached the 24-month time limit (27% v. 9%). The sanctioned

individuals also had less income from reported quarterly wages ($1,090-$2,216 v. $1,359-$2,717),

were more likely to say the were worse off now than when they were on cash assistance (23% v.

13%), and were more likely to have never been married (55% v. 50%).



Recidivism

According to the administrative data, 35% of all closed cases returned to cash assistance within

twelve months of case closure. The top three reasons for returning to cash assistance were loss of

employment or decreased wages (54%), divorce, separation, or moving away from partner or family

member (11%), and change in household circumstances, such as an eligible child moved into the home,

or a disability was certified (10%).



Continuation of government services

Approximately one year following exit, individuals were still receiving ongoing services and

support according to administrative and survey data: 14% were utilizing childcare services, between

13% and 15% were receiving child support dollars, 51% received the Earned Income Tax Credit, and

44% were receiving Food Stamps.

Life After Welfare, Technical Appendices (2/19/01)

28



Health Care

Survey respondents indicated that they were more likely to go without needed medical attention

when they were no longer receiving cash assistance. Twenty-three percent of the respondents said they

had one or more children in their home with a chronic health condition. A little more than one year

following exit, 40% of the adults and 26% of the children had no health insurance.



General Well-being

While individuals generally reported improvements in well-being following exit, many still

needed assistance. For example, 37% got behind in rent or housing payments. Twenty-four percent

reported not having enough to eat at times, and 21% received food from shelters or food banks.

Illinois

Using administrative and survey data, the Illinois examined the experiences of all cases with at

least one adult and one child that closed between July 1997 and December 1998 (the Institute for

Public Affairs and the University of Illinois 2000). TANF leavers were defined as cases that remained

closed for at least two consecutive months. For all exit cohorts, there were 131,967 exits for single-

parent cases and 13,513 exits for two-parent cases, summing to a total of 145,480 case closings during

the study period.

Key findings are summarized below.



Demographics

Almost 92% of the leavers were female, and a majority (62%) had never been married. The

median age of youngest child was just over 4 years old (4.2). Most leavers (63%) had one or more

children under 6 years old, while few leavers (11%) had one or more children under 1 year old.

Life After Welfare, Technical Appendices (2/19/01)

29



Recidivism

By three months after exit, 16% had returned to cash assistance at some point. This cumulative

recidivism rate increased to 23% six months after exit. Younger women were more likely to return to

TANF than older women. African-American women were about 31% less likely than their white

counterparts to remain off TANF for 12 months. Women without highschool diplomas were about

21% less likely than their more educated counter parts to remain off TANF for 12 months.

Income

The percent of leavers with earned income increases from 49% in the quarter before exit to

over 55% in the quarter after exit. The mean quarterly earnings increased from $1,576 in the quarter

before exit to $2,176 in the quarter of exit. Median quarterly earnings continued to rise in the quarters

after exit.



Continuation of government services

Although 88% of leavers received food stamps in the month prior to exit, only 20% received

them in the month of exit. Almost all leavers (99%) participated in Medicaid in the month prior to exit,

but this rate decreased to 43% in the month of exit.

Washington State

Using administrative and survey data, this study examines the experiences of cases closed

during the fourth quarter of 1998 (Fogarty and Kraley 2000).

A summery of the findings are presented below.



Employment/ Income

According to administrative data, 58% of leavers were employed in the second quarter of 1999

Life After Welfare, Technical Appendices (2/19/01)

30

(two quarters after exit) compared to 68% during the quarter they left TANF. Most leavers (61%) left

TANF due to an increase in earnings. Among the employed leavers, 69% reported working full time.

Average reported hourly wages were $7.70. Quarterly wages averaged $2,585 in the quarter of exit,

and rose by about %200 over the next two quarters.



Food and other resources

Ninety percent of leaver received food stamps during the exit quarter, compared to only 42%

two quarters later. Around 40% of leavers cut their meal size sometimes or often because there was

not enough money for food, and 14% reported experiencing hunger.



General well-being

Sixty percent of leavers indicated that they were better off since leaving welfare. However, the

share of TANF adults who had children with abuse or neglect referrals has increased over time. Two

quarters after exit, 1.2% of 1996 leavers had referrals, compared to 1.4% of 1997 leavers and 2.0% of

1998 leavers. There was also increased referral rates for ongoing recipients.



Medical care

Ninety-eight percent of leavers had Medicaid coverage during the exit quarter, compared to

only 56% two quarters later. Overall, 74% of leavers reported having medical coverage. Of those

without insurance, 51% claimed that it was because their medical coupons had been stopped.



Leaving TANF and recidivism

After leaving TANF in the fourth quarter of 1998, 23% returned to TANF within nine months

according to administrative data. This percentage compares to 20% of TANF leavers who returned

within nine months after leaving in the fourth quarter of 1997. Of leaver who returned to assistance,

Life After Welfare, Technical Appendices (2/19/01)

31

25% mentioned being laid off or fired as the primary reason. Of those still off TANF, 85% felt it was

unlikely they would return to TANF in the next six months.

New Mexico

The New Mexico study examines the experiences of 2,500 cases randomly selected from cases

that were receiving TANF benefits between July 1998 and June 1999 (Maximus 2000).

A summary of the findings are presented below.

Recividism

Of the individuals who were no longer receiving TANF, two-thirds thought it was unlikely that

they would be returning to welfare in the near future. Individuals who were most "at risk" of returning

to welfare were unemployed, had not completed high school, had been sanctioned off the rolls, or not

currently married.



Employment/ Earnings

Of the leavers, approximately 62% were employed. Another 12% were unemployed, but living

with an employed spouse or partner. Only 21% were unemployed and not living with an employed

spouse or other adult.

Of those who were employed, 55% were working 40 hours per week or more, and 79% were

working at least 30 hours per week. Slightly more than half (52%) were earing more than $1,000 per

month.



Continuation of government services and child support

Approximately 11% of unmarried leavers were receiving child support. Also among leavers,

30% were still receiving food stamps and 58% were still receiving Medicaid.

Life After Welfare, Technical Appendices (2/19/01)

32



General well-being

About 58% of leavers thought they were better off than when on welfare. Approximately 62%

thought their family's income and benefits were adequate to meet their needs. About 18% said there

had been times since leaving welfare when they needed medical care for a family member but could not

afford it. Twenty-five percent reported that there were times when they had no money to buy food.



Health Care

Seventy-four percent of the leavers reported having some type of health care coverage for their

children. Among employed respondents, 24% were participating in an employer provided health plan.

Another 32% were working for an employer with a health plan, but were not enrolled, and 44% were

working for an employer with no health plan.

Cuyahoga County, Ohio

The Cuyahogo County study was designed to monitor the status of families leaving cash

assistance at multiple time points after exit, compare the experiences of exit cohorts drawn at quarterly

intervals throughout the gradual process of reform implementation, and describe differences in exiting

families' experiences before and after time limits go into effect (Coulton et all 2000). The results

presented here cover exit cohorts from quarter 4, 1998, and quarter 1, 1999. A total of 8,263 cases

left cash assistance during this time period. A survey interview was conducted with 198 individuals.

A summary of the findings are presented below.



Leaving cash assistance and other government programs

More than half (55%) left cash assistance on their own. Of total leavers, 79% left due to an

increase in individual or family income. As recipients left TANF, the percent maintaining other benefits

Life After Welfare, Technical Appendices (2/19/01)

33

increased over time. Of those who left in Q4 of 1998, 48% retained their food stamps compared to

55% of Q1 leavers in 1999. Similarly, 57% of Q4 leavers in 1998 continued Medicaid for their

children, compared to 63% of Q1 leavers in 1999.



Income/Employment

It is estimated that 55% of families leaving cash assistance had incomes below the poverty line

six months after leaving TANF. Approximately 28% lived between 101 and 150% of poverty, and

14% lived below 50% of poverty. The predominate source of income for leavers was employment,

either that of the former recipient or other adults in the household.

Approximately 87% had held at least one job six months after leaving TANF. Forty-eight

percent of the respondents had continuous employment, defined as working at least 20 hours per week

in all the months.



General well-being

Three quarters of the respondents had experienced at least one hardship10 by the sixth month

after exit. Sixty percent paid more than 1/3 of their monthly income for housing expenses, and 15%

reported losing their utilities at least once.



Recidivism

Approximately 23% of the respondents returned to cash assistance at least once between exit

and the six month follow-up interview. This rate is lower than that in among cases closing in 1996,

which found that 28% were back on cash assistance the second quarter after leaving. Of the

10Hardships were defined by: paid >1/3 of income for housing, adults missed/cut meals, missed dentist visits, missed doctor visits, lost utilities, overcrowded housing, worse relationships with children, evicted, worse neighborhood, lived in a shelter/homeless.

Life After Welfare, Technical Appendices (2/19/01)

34

respondents that returned to cash assistance, 49% reported doing so because of job or earnings related

reasons. Pregnancy, child-birth, or custody change was another important reason for returning to cash.

Virginia

The Virginia leavers study was a multi-faceted evaluation of the Virginia Independence Program

(VIP) and the Virginia Initiative for Employment not Welfare (VIEW) (Kuhns et all. 1999). The

Closed Case Study analyses data collected from a telephone survey of former VIP/VIEW clients

approximately one year after their cases closed. The final report describes findings from the telephone

survey and analysis of administrative records from a randomly drawn sample of 779 cases that closed

between July, 1997 and October 1997.

A summery of the findings are presented below.



Reason for leaving

Most cases closed because the client found a job, and very few closed due to sanctions.

Respondents reported VIEW provisions helped them leave TANF primarily through work

requirements but also through job placement assistance.



Income/Employment

Sixty percent of the respondents were employed when their case closed or within a month after

leaving TANF. Nearly 50% worked steadily in the year after leaving TANF. Former TANF

recipients earned and average of $1,067 per month. More than 3/4 of the respondents were working

more than 30 hours per week. Half the jobs offered health benefits, and 1/3 offered sick leave.

Families were 10% more likely to receive child support one year after leaving TANF than

before their case closed. Three-fourths of the families received food stamps the month their case

Life After Welfare, Technical Appendices (2/19/01)

35

closed compared to 54% a year after leaving TANF. The average household income increased by

40% after leaving TANF. Twenty-eight percent of households had incomes above the federal poverty

level at the time of the interview compared with 13% during the month they left TANF.



Health Care

More than 80% of former recipients had health insurance coverage, most often through

Medicaid. Three-fourths of the families without health insurance reported using the emergency room to

meet health care needs. Overall, respondents reported using multiple strategies to pay for medical fills,

including not going to the doctor, doing without other things, borrowing money, or using a free clinic.



Children's experiences

Children's living arrangements were stable. Households most often included a single parent,

and an average of two other individuals. Two-fifths were in public housing or received a housing

subsidy. Few families (3%) were homeless after leaving TANF.

One-half of the children had contact with their non-custodial parent and nearly one-third of the

respondents received child support for children in their household. Non-custodial parents providing

assistance (other than child support) were more likely to have contact with their children.



General well-being

Nearly half of the respondents believed their situation was better after leaving TANF, and less

than one-fifth believed it was worse. Financial problems were the most frequently reported difficulty.

However, few families reported sever hardship, such as not being able to buy enough food. More

families reported more positive than negative experiences for their children since leaving TANF.

Life After Welfare, Technical Appendices (2/19/01)

36

References Center for Law and Social Policy and the Center on Budget and Policy Priorities. 1999. State Policy
Documentation Project. [Online]. Available http://www.spdp.org/index.htm. Coulton, Claudia, Cara Pasqualone, Neil Banjia, Toby Martin, Nina Lilich, Margaret Fernando, and
Fang Li. 2000. "How Are they Managing? A Six Month Retrospective of Cuyahoga County Families Leaving Welfare."[Online]. Available http://aspe.os.dhhs.gov/hsp/leavers99/reports.htm Department of Human Resources: Georgia Division of Family and Children Services. 1999. 1999 Descriptive Data. Evaluation and Reporting Section: Atlanta. Ga. Fogarty, Debra and Shon Kraley. 2000. "A Study of Washington State TANF Leavers and TANF Recipients: Findings from Administrative Data and the Telephone Survey."[Online]. Available http://aspe.os.dhhs.gov/hsp/leavers99/reports.htm Institute for Public Affairs, University of Illinois at Springfield and School of Social Work, University of Illinois as Urbana-Champaign. 2000. "Illinois Study of Former TANF Clients" Analysis of Administrative Data." [Online]. Available http://aspe.os.dhhs.gov/hsp/hspwelfare.htm. Kuhns, Carole, Anne Gordon, Roberto Agodini, and Renee Loeffler. 1999. "The Virginia Case Closed Study: Experiences of Virginia Families One Year After Leaving TANF." [Online]. Available http://aspe.os.dhhs.gov/hsp/leavers99/reports.htm Maximus. 2000. "Survey of Current and Former Welfare Recipients." [Online]. Available http://aspe.os.dhhs.gov/hsp/leavers99/reports.htm

Life After Welfare, Technical Appendices (2/19/01)

37

Ways and Means Committee Print WMCP:105-7 1998 Green Book [Online]. Available http://www.access.gpo.gov/congress/wm001.html.
Westra, Karen l, and John Routley. 2000, "Arizona Cash Assistance Exit Study: First Quarter 1998 Cohort Final Report." [Online] Available http://aspe.os.dhhs.gov/hsp/leavers99/reports.htm.

Life After Welfare, Technical Appendices (2/19/01)

38