TANF Study - Supporting Statement B revised 6 22 15_clean

Temporary Assistance for Needy Families Study (Two-Parent TANF)

OMB Information Collection Request

New Collection

Supporting Statement

Part B

March 2015

Submitted By:

Office of Planning, Research and Evaluation

Administration for Children and Families

U.S. Department of Health and Human Services

7^th Floor, West Aerospace Building

370 L’Enfant Promenade, SW

Washington, D.C. 20447

Project Officer:

Aleta Meyer

B1. Respondent Universe and Sampling Methods

• Respondent Universe.

The universe for the qualitative, telephone interview and site-visit component of the study programs consist of 47 states with TANF or SSF programs identified as extending benefits to any two-parent families. States that fully exclude two-parent families from TANF or SSF programs were excluded from the universe for both the telephone interviews and site visits, in order to better address the key study purpose of providing information on the characteristics of two-parent families on or eligible for TANF.

Of the overall universe of 47 states, the study team has sought to include states that fit into three primary categories: 1) states with largest two-parent TANF caseloads; 2) states that serve two-parent families through SSFs and have large overall TANF caseloads; and 3) states that use federal TANF funds to serve two-parent families but serve relatively few of the families that are eligible. States falling in this latter category were identified by using a proxy measure for rates of participation among eligible families through identifying states with two-parent TANF caseloads that represent a much smaller share of the national two-parent caseload than the state’s share of the overall national caseload. Fourteen states roughly fell in this category and, of these, will visit two states and seek to interview by phone the TANF director from another state for three states total within this group.

• Purposive Sample Selection

From the main universe of 47 states, 10 states will be selected for case studies. Of these focus states, the study team will select five for site visits, with the balance to be engaged via conference calls targeted to each state’s TANF director. The programs will be selected to ensure that they allow the study to develop a systematic description of two-parent families’ characteristics, identify the variety of services these families take up through TANF, and categorize how state policies as well as family or caseworker beliefs affect participation. Because most states provide assistance to two-parent families either through TANF or SSF programs, greatly exceeding the number of study sites and the level of resources available to the study, it is not possible to visit all of these states, nor is it practical to select a representative sample. Thus, site selection will be purposive with the goal of selecting for the study TANF and SSF programs that will provide as much diversity as possible while also providing information that is relevant and applicable elsewhere.

Working in collaboration with ACF, the study team developed the following five criteria to be used in selecting the sample of 10 programs, as follows, roughly organized in order of importance:

1. Operating a two-parent TANF or SSF program.

2. Overall, the study will seek to include states that fit into one of three primary categories:

   1. States with the largest two-parent TANF caseloads.

   2. States with the largest two-parent SSF caseloads.

   3. States that use federal TANF funds to serve two-parent families but serve a disproportionately small share of two-parent families.

3. Ability to readily provide existing state-level administrative data on SSF populations.

4. Diversity of caseload, program, and state characteristics.

To connect with two-parent families that are not receiving benefits, connections will be made through state agencies, and especially, community-based organizations. Although not enrolled in TANF, these families may access other services offered to low-income families. Given that the contractor has no effective way to contact adult members of two-parent TANF families who have not received TANF or interacted with community-based organizations, the perspectives and characteristics of these individuals are likely to be underrepresented in the study findings. Thus, information from the study will be limited in describing the characteristics of this subset of family members; this limitation will be described in publications, as appropriate.

Diversity as a criterion for site selection refers to a range of other caseload and program characteristics, such as the work participation rate, average monthly benefit, share of no-parent families in the statewide TANF caseload, racial composition, and region of the country. Additionally, some states’ TANF programs have different rules for two-parent families than for single parent families, such as eligibility rules, work requirements, or time limits. “Largest” in this case refers to largest in nominal terms.

The selection criteria will be placed in a matrix with the potential two-parent TANF or SSF programs. The study team will fill in the matrix based on the information obtained from program-related documents and secondary data. Site selection will be iterative because the advantages and disadvantages of each site must be determined in relation to other potential selections to generate one or more balanced samples. These criteria are expected to yield a diverse set of states that comprise a large share of TANF caseload and illustrate useful information for identifying TA needs.

Examples of the matrix used to determine site selection are shown in the two tables immediately below:

	Willingness to Participate	Operates TANF or SSF 2-Parent Program	Share of National 2-Parent TANF Caseload (SSP-MOE & TANF) FY 2012	Share of National TANF Caseload (SSP-MOE & TANF), all units FY 2012	Average Monthly SSF Caseload FY 2012
Selected States – Category 1
State 1	TBD	√	60.88%	31.04%	0
State 2	TBD	√	7.21%	6.32%	0
State 3	TBD	√	5.89%	2.83%	0

Cat	State	Work Participation Rate (all families, TANF only, FY 2011)	Average Monthly Benefit all TANF families, FY 2012	Percent No-Parent Families (TANF only, FY 2013)	Percent African Americans (TANF only, FY 2012)	Does the state have different rules for 2-parent families?	Region
1	State 1	27.8%	$461.40	46.5%	17.6%	Yes	Pacific
1	State 2	27.3%	$355.60	67.2%	38.7%		East North Central
1	State 3	14.6%	$370.40	36.0%	10.0%		Pacific

Willingness to participate in the study will be one of the criteria used in selecting the states in the sample (both for the larger group of 10 states to be included in phone interviews, and a subset of five states for site visits). All states selected for the site visits are expected to participate. High participation (80 percent or more) is expected among the state officials within states that have agreed to participate. Within each state, a specific county or locality will be invited to participate. Again willingness to participate will be one of the criteria for response, and so all selected localities are expected to participate and high participation (80 percent or more) is expected among the various local officials.

Additionally, the study team will refocus the state selection criteria to select the individual localities/counties for each site visit. Since the study team will be interviewing each state TANF director by phone, the team will not be restricted geographically to selecting a local office close to the state capital for the site visit. The overarching goal of local site selection is to include localities/counties that best represent the most common experiences and characteristics of TANF-eligible two-parent families within each selected state. The above criteria for the state site selection will be adopted on a local/county level so as to provide valuable study data from a sub-state lens. Particularly, the study team will seek localities that have relatively large overall TANF caseloads as well as relatively large numbers of two-parent families served through TANF or state-funded programs, except in states chosen for their relatively small share of two-parent families enrolled in TANF. After applying these objective criteria, the study team will consider any additional information provided by the state TANF director that may be relevant for the local site selection, keeping in mind the risk that this information could bias the selection.

B2. Procedures for Collection of Information

This section describes the plan for collecting qualitative data through telephone interviews and in-person interviews with respondents in states’ TANF Programs. In-person interviews will be conducted in each of the five states selected from the universe of TANF programs providing benefits to two-parent families and from a locality or county within each state. Telephone interviews will be conducted with TANF directors in a larger set of states.

Recruitment. As a primary strategy for recruiting two-parent families, the contractor will reach out to community partners to pass along the invitation and directly recruit clients. We will provide information about the study that is clear both about the purpose of the study and that the discussion itself would be private and led by an independent third party (the research team). See Appendix A-5. This recruitment strategy has proved successful in recent studies because the community partners are trusted intermediaries for the population sought for the focus groups and individual discussions. The location of the community partners has proven a convenient site for holding recent focus groups and interviews because the target population is familiar with the location. If necessary, the study team will use alternative strategies to recruit families for the focus groups and individual discussions. Specifically, local agency staff could post/provide fliers (see Appendix A-5) to clients informing them of the focus groups and directing them to call a 1-800 number to register; the study team would use phone software to manage the calls. Based on previous experience, it is anticipated that at least 50 percent of the adult two-parent family members who agree in advance to participate will do so, but as a contingency, will plan for 50 percent of invited and confirmed respondents not participating. To minimize the burden on focus group and individual discussion participants, the discussions will be scheduled at the respondents’ convenience, such as during the evening, lunch time, or at a time when they are already present at the TANF agency, community organization, or other location appropriate for the discussion.

The criteria for identifying adult members of two-parent families on or eligible for TANF to take part in either focus groups or one-on-one discussions would be the same. As the study team recruits respondents for the focus groups and one-on-one interviews, selection into each interview type would be random and may depend simply on the schedule and availability of the respondent. Each discussion format brings opportunities and challenges, which can be mitigated by conducting both types of discussions. The dynamic nature of focus groups can spur discussion on topics that an individual might not think to raise. What one person says reminds another person of something useful to add to the discussion. On the other hand, more reticent individuals may be less willing to speak up in a focus group than in a one-on-one conversation. Individuals may also be less comfortable discussing sensitive topics in a focus group setting that they would share in a one-on-one conversation.

Quantitative data from client information sheets (see Instrument 5) collected during each focus group will be tabulated so we can summarize the number and characteristics of participants in the focus group and will provide context for the analysis of their comments/answers and will be used to provide important perspective on research questions. No individuals will be identified.

Informed Consent. Verbal consent will be requested from participants. See the topic guides (Instruments 1 through 6) for the verbal informed consent language that will be used in the interviews. Additionally, written consent, in the form of initials, will be obtained from participants in in-person interviews and focus groups. (See Appendix A-4.) Consent with initials allows for a written consent that eliminates the risk of breaching the respondent’s privacy.

Site Visits. When an initial likely sample of 5 TANF programs has been identified, the study team will contact the TANF director in each of the preliminarily-selected states to confirm the information on which the selection was based and to confirm their willingness to participate in the study (See Appendix A-2). After this initial contact, site visit planning will include several steps:

• Contact the designated state and county staff to identify potential dates for the visit and discussion participants.

• Arrange staff travel and accommodations.

• Finalize the agenda at least one week before the meeting time.

• Conduct the site visit.

• Follow up with “thank you” letters to people who gave their time to participate in the interviews (See Appendix A-3).

The fieldwork component will include two- or three-day visits to five states. The primary mode of data and information collection will be a series of semi-structured interviews. These interviews will be conducted with state TANF directors (by telephone in advance of the visit), local TANF directors, local TANF front-line staff, representatives of community-based organizations, and adults in two-parent families on or eligible for TANF (both in focus groups and individual interviews). In the focus groups, only, respondents will be asked to complete a participant information form, which will be used only to summarize participant information. (See Instrument 5). Instruments 1 through 6 are the interview discussion guides for staff and family interviews. The guides are designed to minimize response time and maximize consistency across interviews, but will allow the respondents to elaborate on their answers as needed.

Each site will be visited by a 2-person team, including at least one senior and one junior level staff. The contractor plans to have both site visitors present at each interview whenever possible. The senior staff will lead the interviews. The junior staff will assist with documentation and data collection, ensuring that all topics and subtopics are covered. This approach assures more thorough interviews and notes, allowing one interviewer to pursue additional clarification and follow-up questions while the other is taking notes. Having at least two people listening and recalling the interview is helpful in analysis and interpretation as well. If deemed appropriate, the interviews will be recorded for possible transcription of key quotes or other details.

Telephone Interviews. Prior to each telephone interview with a state TANF director, the following steps will be undertaken. An initial e-mail to each state TANF director will be made to determine interest and availability for participation in the study. The initial e-mail will introduce the purpose of the telephone interview, length of the call and specific information about the topics to be covered during the interview. This e-mail will establish a contact person from the interview team for each director and also provide information about potential available dates for scheduling the calls. (See Appendix A-1)

Each telephone interview will be assigned a team member responsible for facilitating the call, a note taker responsible for capturing responses, and if deemed appropriate, the calls will be recorded for possible transcription. Each call will last approximately 90 minutes and use a structured interview guide. The guide is designed to minimize response time and maximize consistency across interviews, but will allow the respondents to elaborate on their answers as needed. Calls should occur before the site visits.

Use and Dissemination of Findings. Based on the analysis of results in all substantive areas, a high-quality report written in plain language will be produced. The report will communicate findings using a variety of methods designed to reach audiences with varying interests. For example, the report will include a stand-alone executive summary appropriate for high-level policymakers, detailed methodological and substantive appendices of interest and use to researchers or program practitioners, and charts, graphs, and maps illustrating study findings. Copies of the report will be sent to each of the states taking part in qualitative data collection. The report’s findings could be used by administrators on the state and federal level to inform how to better serve two-parent families on or eligible for TANF.B3. Methods to Maximize Response Rates and Deal with Nonresponse

Expected Response Rates

Willingness to participate in the study will be one of the criteria used in selecting the states in the sample (both for the larger group of 10 states to be included in phone interviews, and a subset of five states for site visits). All states selected for the site visits are expected to participate. High participation (80 percent or more) is expected among the state officials within states that have agreed to participate. Within each state, a specific county or locality will be invited to participate. Again willingness to participate will be one of the criteria for response, and so all selected localities are expected to participate and high participation (80 percent or more) is expected among the various local officials.

Dealing with Nonresponse

Willingness to participate in the study will be one of the criteria used in selecting the 10 states in the sample (including the five states for site visits). All states selected for the site visits are expected to participate. High participation (80 percent or more) is expected among the state officials within states that have agreed to participate. As a means of reducing nonresponse bias in the client interviews and focus groups, the contractor will give each focus group participant a $40 gift card and each individual interview participant a $30 gift card as a token of appreciation.

At the state level, based on previous studies, at least 80 percent of state and local officials who have agreed to participate follow through. At the family level for the interviews and focus groups, community-based organization where the CBO is an established and trusted community group will invite respondents until the requested amount agree. More participants will be invited than needed in order to ensure there are sufficient numbers within a pre-determined acceptable range. The study team will stay within the estimated burden level for this request (see Table 2, Supporting Statement A). Based on recent experience of conducting focus groups within a similar population in multiple states, 60- to 100-percent participation of the invited families who agreed in advance agreed to take part. Based on that, it is estimated that over 50% of the adult two-parent family members who agree in advance to participate will do so.

Maximizing Response Rates

Once state and local sites are selected, the study team will take a number of steps to minimize the burden to the interview respondents and maximize response rates. First, the site visits will be scheduled in a manner that allows respondents to identify the most convenient time for the interview or visit within the study timeframe. Second, flexibility will be exercised in adjusting the specific order, timing, and location of the on-site interviews to meet the respondents’ needs. Third, this study team relies on trained, experienced interviewers who are skilled at helping respondents feel comfortable and earning their trust and cooperation with the study. Fourth, the study team will conduct specific orientation and training on the particular data-collection instruments for this study to ensure fidelity to the goals and methods of the discussions, and consistent coding, and to address any potential areas for improvement prior to visits to the field. Fifth, for the client interviews and focus groups, the study team will use reminder calls for committed respondents prior to the interviews to help maximize response rates.

B4. Tests of Procedures or Methods to be Undertaken

Given limits on the study timeframe and budget, the study team does not have sufficient time or budget flexibility for an additional pilot of instruments. While we believe the discussion guides will elicit the type of information desired, if, once in the field, changes are found to be necessary, we will submit a request for a nonsubstantive change to make these minor changes.

B5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing Data

The information for this study is being collected by the Urban Institute on behalf of ACF. Dr. Heather Hahn, Urban Institute Principal Investigator, led development of the study design plan and data collection protocols, and will oversee collection and analysis of data gathered through on-site interviews and telephone interviews. Dr. Hahn brings extensive experience designing and implementing complex research studies, expert knowledge of TANF programs, and extensive field and survey research experience and training, including overall design, instrument design, survey and qualitative interviewing, and analysis of qualitative data. The project’s study team, from the Urban Institute, also includes David Kassabian as project manager and site visitor, Eleanor Pratt as research assistant site visitor, Linda Giannarelli as lead for simulations and other quantitative analysis, and Erika Huber as research associate for quantitative analysis.

The agency responsible for receiving and approving contract deliverables is:

The Office of Planning, Research, and Evaluation (OPRE), Administration for Children and Families (ACF)

U.S. Department of Health and Human Services

The Federal project officer for this project is Aleta Meyer. Kathleen McCoy is a contract social science research analyst assisting her on this project.

8