LS-UI OMB Part A (2014-07-03)

Longitudinal Study of Unemployment Insurance Recipients (LS-UI)

OMB Supporting Statement

Part A: Justification

CONTENTS

PART A: JUSTIFICATION 1

1. Circumstances Necessitating the Data Collection 1

2. How, by Whom, and for What Purpose the Information Is to Be Used 3

3. Use of Improved Technology to Reduce Burden 8

4. Efforts to Identify Duplication 9

5. Methods to Minimize Burden on Small Businesses or Entities 11

6. Consequences of Not Collecting the Data 11

7. Special Data Collection Circumstances 11

8. Federal Register Notice 12

a. Federal Register Notice and Comments 12

b. Consultations Outside of the Agency 12

9. Respondent Payments 12

10. Privacy 14

a. Procedures to Protect the Privacy of the Data Collected as Part of the Study 14

11. Questions of a Sensitive Nature 18

12. Hour Burden of the Collection of Information 18

13. Estimated Total Cost Burden to Respondents and Record Keepers 19

14. Estimated Annualized Cost to the Federal Government 19

15. Changes in Burden 19

16. Publication Plans and Project Schedule 20

17. Reasons for Not Displaying Expiration Date of OMB Approval 21

18. Exception to the Certification Statement 21

REFERENCES 22

Appendix A: 60-Day Federal Register Notice

Appendix B-1: LS-UI FIRst Interview Questionnaire – CATI VERSION

Appendix B-2: LS-UI FIRst Interview Questionnaire – WEB VERSION

Appendix C-1: LS-UI SECOnd Interview Questionnaire – CATI VERSION

Appendix C-2: LS-UI SECOnd Interview Questionnaire – WEB VERSION

Appendix D-1: LS-UI THIRd Interview Questionnaire – CATI VERSION

Appendix D-2: LS-UI THIRd Interview Questionnaire – WEB VERSION

TABLES

A.1 Expected Number of Completed Surveys per Survey Round and MSA 3

A.2 Purpose and Priority for Survey Data Items 5

A.3 Burden Estimates for LS-UI Respondents 19

A.4 Cost to the Federal Government 19

A.5 Schedule for Project Tasks 21

PART A: JUSTIFICATION

1. Circumstances Necessitating the Data Collection

The U.S. unemployment insurance (UI) program was designed to reduce financial hardships for workers who become unemployed through no fault of their own, assist these individuals in finding appropriate reemployment, and ameliorate the negative effects of unemployment on the economy as a whole. The loss of a job poses major hardships for many workers and their families. Not only do unemployed persons need to begin a potentially challenging search for new employment, but they and their families also often need to adjust their spending patterns and seek other sources of income. For qualified unemployed workers, UI benefits can help to reduce the urgency for such adjustments. By providing temporary income support, UI benefits can smooth the transition to new circumstances, reduce financial distress, and provide job seekers with a buffer while they search for work. Furthermore, because UI benefits are time limited and provide only a partial replacement of lost earnings, benefit recipients do not have an incentive to remain permanently unemployed. Regular state UI programs also typically require that UI recipients conduct work search activities weekly while they are receiving UI benefits.

As of mid-June 2014, about 2.4 million unemployed workers were receiving benefits through the regular state UI program, which is financed by each state’s employers and administered as a federal-state partnership.¹ Given the importance of the UI program, the U.S. Department of Labor (DOL) wants to understand how the financial importance of UI benefits in the lives of recipients changes during and after benefit receipt. DOL also would like to learn about how UI recipients’ job search activities and expectations change during and after benefit receipt, as well as recipients’ reemployment outcomes and satisfaction with UI program operations. As the largest state in the nation, California has the largest number of unemployed workers, making its UI system a critical support for U.S. workers. In 2013, for example, California made more than 1.1 million first payments to UI recipients and paid more than $6 billion in UI benefits. Initial claims filed in California during the second week of June 2014 represented 20 percent of all initial claims filed that week nationwide.¹ As a first step, the Longitudinal Study of UI Recipients (LS-UI) will field and analyze a longitudinal survey of UI recipients in California to provide DOL’s Chief Evaluation Office (CEO) with valuable new insights about these issues. This information will be available for use by researchers and policymakers to assess how well the program is serving workers in California and could reveal opportunities for improving the program nationally. The study will also provide methodological insights to DOL should it desire to conduct surveys of UI recipients in the future.

States have broad flexibility in structuring critical features for their regular state UI programs. These features include who qualifies for benefits, the potential number of weeks of benefits that a worker can collect the minimum work search requirements during UI receipt, and the weekly benefit amount (WBA). Thus, program features vary considerably across states. For example, in some states, all unemployed workers who are eligible for benefits can receive up to 26 weeks of benefits. In many other states, including California, the number of weeks of benefits to which an unemployed worker might be entitled could depend on his or her work history, with some workers eligible for as few as 10 to 15 weeks of benefits. In California, unemployed workers might be eligible for 14-26 weeks of benefits.² Regardless of the number of weeks of benefits to which someone is entitled, however, he or she will typically need to collect UI benefits during a one-year period called a “benefit year,” after which he or she will lose entitlement to the benefits. The collection of all of the benefits to which a UI recipient is entitled is often referred to as “exhaustion” of benefits and recipients who have exhausted their benefits are referred to as “exhaustees.”

Understanding how workers and their families adjust to the changes in income during and after UI benefit receipt can support policymakers’ efforts to refine the UI program to better meet the needs of unemployed workers while continuing to encourage their return to work. However, information about UI recipients has generally obtained from surveys that ask about experiences over a period of several years, which may not offer sufficient insight into the dynamic adjustments that unemployed workers make soon after job loss, after exhausting UI benefits, and after end of their benefit year. No longitudinal surveys of California UI recipients of the richness planned for this study have been conducted.

Results from the longitudinal survey will shed light on four main topics which cannot be studied through existing administrative data sets:

1. Financial hardship faced by households after a job loss and their subsequent financial adjustments to respond to the job loss. The survey will support an analysis of the relationship between UI benefits and household income, financial distress, and household adjustments to reduced income. Results from this analysis may provide new insights into the adequacy of UI benefits. The survey also will provide an understanding of whether and how UI recipients use other government assistance programs, draw down their savings, or face hardships such as foreclosure or bankruptcy during and after benefit receipt.

2. Labor market experiences and job search strategies. Survey results will also permit an understanding of how recipients’ job search and labor force participation decisions change over the course of an unemployment spell. For example, the survey will provide information about the earnings levels that recipients would be willing to accept in a new job, how they compared to pre-UI earnings, and whether these levels change over time for recipients who remain unemployed.

3. Reemployment outcomes. LS-UI results will enable DOL to identify whether and when UI recipients become reemployed and to compare the characteristics of pre-UI and post-UI employment, such as wages, fringe benefits, industry, and occupation. If UI recipients who are reemployed have a subsequent job separation within the study time period, they will be asked for their tenure at that job. However, this study will not be able to examine job tenure for individuals after the last wave of the survey.

4. Satisfaction with UI program operations. Survey results will also include information on UI recipients’ experiences with the program from when they file their initial claims to when they exhaust benefits (for those who exhaust their entitlements).

Although research exists on each of these topics in different contexts, it either has been constrained by a lack of detailed data, which makes study findings difficult to interpret, or has become outdated given changes to UI program operations and economic conditions. For example, Gruber’s (1997) study of the consumption-smoothing effects of UI benefits relied on imputed benefit levels because actual benefit levels were not directly available in his data. Similarly, Herkenhoff and Ohanian (2012) make a strong theoretical case that UI receipt reduces mortgage defaults, but the researchers did not have the individual-level data to analyze this connection. Regarding job search and reemployment, numerous studies have focused on a “spike” in exits from unemployment around the time of UI benefit exhaustion, which many interpreted as evidence of recipients’ delay in work search effort and job attainment (Moffitt 1985; Katz and Meyer 1990; Card et al. 2007). However, this interpretation might not be valid because the studies were not able to distinguish among reemployment, labor force exit, and, in some cases, the end of a UI claim. Recent work by Krueger and Mueller (2010; 2012) that examines time use suggests a more complex pattern of job search intensity over an unemployment spell, but these results are difficult to interpret because of limited information about receipt of UI benefits or very low survey response rates.³ The LS-UI addresses these data limitations in the context of two areas in California by explicitly gathering data on benefit amounts and durations, job search intensity, and reemployment outcomes. Furthermore, the most recent study of claimants’ satisfaction with the UI program was conducted about 15 years ago, prior to the widespread use of Internet-based filing (Marcus and Frees 1998). The LS-UI is designed to elicit high response and avoid item non-response by including salient content and minimizing required administration time; it will provide up-to-date information about recipients’ satisfaction with program operations in California.

This study will include UI recipients from two areas in California, the Los Angeles metropolitan statistical area (MSA) and portions of the Central Valley. Both areas are from California because of DOL’s prior experiences receiving high-quality and timely administrative data from the state. Within California, the Los Angeles MSA and Central Valley area were chosen because they are economically and geographically diverse, and they have large populations that can provide enough sample members for the study. The portions of the Central Valley will be selected based on the likelihood that there will be enough UI claimants who received first payments for UI benefits during a calendar week to meet the requirements of the Office of Management and Budget (OMB) for minimum detectable differences. As an example of the diversity captured by the Los Angeles area and the Central Valley area, a comparison can be made of Los Angeles County to Fresno County, which is located inside the Central Valley area. In 2012, Fresno County had the highest value of agricultural production in the U.S. but less than 10 percent of the population of Los Angeles County.⁴ The median per capita income in Fresno County in 2011 was $31,542 compared to $42,564 in Los Angeles. The study will shed light on the experiences that UI claimants from these diverse areas have in seeking reemployment and claiming UI benefits. Even though all respondents will interact with California’s UI program which is consistent state-wide, their needs and their resulting overall satisfaction levels with the UI program might be different. Because these two areas were selected purposively, the statistics from the survey will not be interpreted as generalizable to a broader population of recipients. Furthermore, data on the two areas will not be pooled together for reporting purposes; statistics will be reported separately for each area because statistics on a pooled sample would not have meaning that would be relevant for policymakers.

Depending on results, DOL may want to implement a similar survey in the future to learn about claimants’ experiences in other geographic areas, local economies, or policy contexts, which will require a separate ICR submission. Consequently, the data collection strategies that are described in this request for clearance have been designed to test the effectiveness of these methods and their suitability for future use. In addition, attention will be paid throughout the survey to maximizing efficiency and response and to documenting lessons learned.

2. How, by Whom, and for What Purpose the Information Is to Be Used

The LS-UI will result in a unique data set of information about the experiences of UI recipients during and after benefit receipt. The information will be gathered in a short retrospective window, reducing recall bias. Clearance is being requested for three rounds of survey data collection to be conducted over a 12-month period for up to two cohorts of UI recipients in the Los Angeles MSA and the Central Valley area. Each sample cohort will consist of UI claimants who received their first payment of UI benefits for a week of unemployment during a particular calendar week.⁵ The timing of the survey rounds correspond to individuals’ expected early, middle, and post-UI benefit receipt periods. In each of these areas, 833 individuals are expected to have completed interviews for all three rounds of the survey (Table A.1). The LS-UI study plans to complete interviews with 1,089 individuals at the first round of the survey and 926 of these individuals at the second round in order for the third round to have 833 completed interviews. The information from these interviews will be analyzed, and a public-use file will be produced. The remainder of this section describes the sample frame, the survey topics, and why these topics were selected for inclusion.

Table A.1. Expected Number of Completed Surveys per Survey Round and Area

	Completed Interviews with UI Recipients in Each Area
	Round 1	Round 2	Round 3	Total
Los Angeles MSA	1,089	926	833	2,848
Portions of Central Valley	1,089	926	833	2,848
Both Areas	2,178	1,852	1,666	5,696

MSA = metropolitan statistical area; UI = unemployment insurance

Note: We expect that 85 percent of the individuals who complete surveys in the first round will respond to the second round, and 90 percent of the individuals with complete interviews at the first and second rounds will respond to the third round.

The sampling frame and sample for the study. The sampling frame for each area will consist of extracts from the California UI administrative data base of individuals who have filed for a first payment of benefits for a specific week for up to 2 reference weeks approximately 6 weeks apart. The extract file for each week will be defined as a cohort. We will select a sample in each cohort of individuals. The sample selected from the first cohort will consist of recipients who filed for a first compensable week of benefits in “week 1.” If a second cohort is needed, the extracts from the CA UI administrative data base will consist of recipients who filed for a first compensable week of benefits six weeks after the first cohort; that is, this second cohort will be selected based on filing for a first compensable week of benefits during “week 7.” An interview will be attempted of sample members in each cohort three times after their first compensable week.

The plan to use up to two cohorts of weekly recipients who enter the UI system over a staggered period of time helps ensures that there will be an adequate sample from each of the two areas to meet the statistical precision needs of the study (detailed in Part B, Section B.1). This is especially important because the number of UI recipients who reside in a particular area and file in any given calendar week for a first compensable week of benefits is unpredictable, even though historical patterns provide some guidance about the number of claims that could be expected in a specific calendar week.⁶

Survey topics. The survey will capture insights about the experiences of UI recipients by timing interviews to coincide with recipients’ early, middle, and post-UI benefit receipt experiences. Each of the three surveys is expected to take an average of 25 minutes to complete. Relative to a longer survey, having a survey of this length is likely to help to keep sample members engaged in the study across survey rounds. Table A.2 outlines the survey topics and data items planned for each interview round. As the table shows, the survey will include questions to determine demographic and household characteristics, pre-UI employment and unemployment experiences, job search activities, job offers, reemployment expectations, participation in reemployment services, reemployment, financial well-being, and customer satisfaction. Given the timing of each survey in relation to sample members’ UI or unemployment cycle, not all questions will be asked of all survey respondents at each survey round. Questions that will be asked at multiple rounds of data collection are indicated as such. Such repetitions are needed to document changes in a respondent’s experience or circumstances since being interviewed in a prior round. Similarly, some data items are included in one or two rounds but not all three rounds because the importance of the items varies across rounds given the survey timing. For example, the second and third survey rounds will capture job training since the job loss and the prior interview, respectively, as respondents will have had more opportunity to engage in post-loss training than they had at the first interview. Furthermore, questions at each round of interviewing will be phrased in a way that both aids survey respondents in providing accurate answers and generates information that is useful for examining changes in behavior and experiences. For example, some questions will be explicitly phrased to prompt respondents to report changes since the most recent interview.

Table A.2. Purpose and Priority for Survey Data Items

	Data Item Purpose			Survey Round(s) with Data Item
Data Item	Validation	Control Variables	Description of Experiences	1st Interview	2nd Interview	3rd Interview
Demographic Characteristics
Date of birth/agea	X	X		X	X	X
Marital status/cohabitation		X		X		X
Veteran status		X		X
Highest degree completed		X		X
Household Characteristics
Household size		X		X
Number of children		X		X
Home ownership		X		X
Household income			X	X	X	X
Household sources of income			X	X	X	X
Household income from TANF (Yes/No)			X	X	X	X
TANF income amount			X	X	X	X
Household income from SNAP (Yes/No)			X	X	X	X
SNAP income amount			X	X	X	X
Household income from SSDI (Yes/No)			X	X	X	X
SSDI income amount			X	X	X	X
Household income from Medicaid (Yes/No)			X	X	X	X
General health status			X	X	X	X
Pre-UI Employment (separating job only) and Unemployment
Earnings		X		X
Occupation		X		X
Industry		X		X
Job tenure		X		X
Health insurance benefits available		X		X
Paid sick leave		X		X
Retirement or pension plan		X		X
Hours worked		X		X
Seasonality of job		X		X
Reason for job separation		X		X
Current claim filing method		X		X
History of UI receipt		X		X
Job Search and Job Offers
Time spent looking for work			X	X	X	X
Methods used to look for work			X	X	X	X
Reasons for not looking for work			X	X	X	X
Contacts with potential employers			X	X	X	X
Visited an American Jobs Center			X	X	X	X
Registered with Employment Services/public workforce system			X	X	X	X
Services received at American Jobs Center (resource room; workshops; tests or assessments; job clubs or other peer support; individual counseling; other types of reemployment services received)			X	X	X	X
Usefulness of AJC Services			X	X	X	X
Applied for job requiring relocation			X	X	X	X
Received job offer			X	X	X	X
Accepted job offer			X	X	X	X
Reason for rejecting offer			X	X	X	X
Benefits included in offer			X	X	X	X
Offered pay rate			X	X	X	X
Relocation required			X	X	X	X
Change in occupation/industry			X	X	X	X
Participated in job training			X			X
Reemployment Expectations
Recall expectations			X	X
Expected time to reemployment			X	X	X	X
Reemployment outlook			X	X	X	X
Reservation wage (minimum wage someone is willing to accept for a job)			X	X	X	X
Fringe benefits sought			X	X	X	X
Anticipates relocation will be necessary			X	X	X	X
			X	X	X	X
Reemployment
Ever held a job after UI claim			X	X	X	X
Currently working more than 35 hours			X	X	X	X
Number of jobs held			X	X	X	X
For each job:
Retention (start and stop dates)			X	X	X	X
Earnings			X	X	X	X
Hours worked			X	X	X	X
Availability of fringe benefits			X	X	X	X
Financial Well-Being
Spouse/Partner earnings		X	X	X	X	X
Pre-UI savings, investments and debt			X	X	X	X
60+ days late on bills			X	X	X	X
Savings and investments drawn down			X	X	X	X
Changes in other household members’ employment status			X	X	X	X
New government benefit application and amounts			X	X	X	X
Perceived role of UI benefits in financial well-beingb			X	X	X	X
Customer Satisfaction
Experience filing initial claim			X	X
Ease of filing and following instructions			X	X
Clarity of explanations of rights and responsibilities			X	X
Explanation of benefits and services			X	X
Time/effort required to file			X	X
Speed or timeliness of payment			X	X	X	X
Staff helpfulness in the filing process			X	X
Staff knowledge of laws and policies			X	X
Level of respect and courtesy of staff			X	X
Overall program satisfaction			X		X	X
Contact Information
Confirm/update respondent’s address	X			X	X	X
Email address	X			X	X	X
Primary phone	X			X	X	X
Second phone	X			X	X	X
Alternate contact information	X			X	X	X

Notes: Data items that have a “Validation” purpose will be used to confirm or update information in the administrative data or the locating process for future interviews. Data items with a “Control Variables” purpose are factors that are plausibly not influenced by benefit receipt and may be correlated with the experiences of UI recipients. These data items, such as earnings at a prior job, may be used to define subgroups of interest in the analysis. Data items that have a purpose of “Description of Experiences” will be used to examine the experiences of recipients during and after benefit receipt.

Date of birth and/or age will be available in the administrative data, but it will be asked in the survey for verification purposes..

^bTo learn about sample members’ perceptions of the role of UI benefits in their financial well-being, the plan is to rely on survey respondents’ reports of whether or not they received UI benefits, rather than on administrative data. This is because the data extract that will be received prior to the fielding of the surveys might not include accurate information about whether, or the extent to which, a sample member received UI benefits.

UI = unemployment insurance

3. Use of Improved Technology to Reduce Burden

At each round of data collection, sample members will have two advanced technology options by which to complete the LS-UI. They will be able to complete the survey either (1) through self administration using the web or (2) with an interviewer using computer-assisted telephone interviewing (CATI).

Use of the web is convenient for most UI recipients. In 2010, 75 percent of all households had Internet access, and the proportion continues to increase (U.S. Census 2010). Those that do not have home-based Internet access often have access elsewhere. Many of our sample members will conduct job searches online and have Internet access through resource rooms at American Job Centers. These resource rooms also often provide computers for customers to apply for jobs online. Further, experience suggests that the UI administrative data extracts will contain email addresses that can be used to alert sample members about the study and remind them to complete the web survey. Email reminders will contain a URL and password so the sample member can easily “click through” to the questionnaire. Both the invitation letters and emailed reminders will contain a toll-free number for sample members who have questions or cannot complete the survey by web. Surveys will be programmed in both English and Spanish and bilingual interviewers will be available to conduct CATI interviews in Spanish.

Since the majority of households have internet access, completion by web is easy for most respondents. Also, web administration, on average, costs less than CATI, enabling more completions with cost savings. For these reasons, LS-UI sample members will be encouraged to complete surveys using the web and will be offered an additional incentive for doing so (see Section A.9). Both web and CATI interviews will be programmed in Blaise, which is well-established survey software that enables embedding of skip pattern logic and preloading of data into survey instruments. To ease completion of the LS-UI, administrative data such as UI initial claim date and prior employer name will be preloaded into the instrument before the first interview. Similarly, questions in rounds 2 and 3 will refer to information provided in prior rounds. Thus, the software will enable easy navigation for respondents, particularly compared with paper instruments, and will minimize the need for respondents to repeat information across rounds. In addition, respondents will be prompted to correct answers when responses are deemed invalid based on pre-programmed checks for ranges and logical consistency. The web survey is designed to prompt respondents for a response whenever one is expected but not provided. Respondents will have to select the “continue” button located at the top of the page to move to the next question (without responding) instead of the convenient “next” button on the bottom of the page. Other web surveys conducted by Mathematica have not encountered substantial click through problems. The ability to program these features into the software virtually eliminates the needs for costly callbacks to respondents that are commonly required for paper surveys. Finally, web surveys also allow respondents to quit and return to the survey at their convenience.

Even with the option of a higher incentive for web completion, some respondents will prefer to talk with a person to complete the interview. Offering a CATI option typically leads to higher response rates while leveraging technology to reduce burden in the same ways as a web survey. Except for language changes that are necessary to accommodate self-administration versus being asked by an interviewer, the content of web and CATI survey versions are identical. (Copies of both versions for each round of interviewing are included as Appendices A through C of this volume.) CATI offers the option to speak to a person and the convenience of answering from any location with a telephone. CATI dialing errors will be almost completely eliminated through the use of a preview dialer whereby the interviewer presses one button to dial the number after reviewing the case (this is akin to one-touch or speed dialing). The dialer also enables the interviewer to preview case history notes and the history of dispositions so that respondent preferences (such as interview time or language) provided in earlier rounds can be addressed proactively. During the call, interviewers can correct errors in response to edit checks, minimizing the need for callbacks. An automated call scheduler will simplify scheduling and rescheduling of calls to respondents at convenient times. It can also be used to assign cases to specific interviewers such as those who are trained in refusal conversion techniques or those who are fluent in Spanish.

4. Efforts to Identify Duplication

As described in Section A.1, the LS-UI will provide evidence from two areas in California to assess (1) financial hardship faced by households after a job loss and their subsequent adjustment, (2) labor market experiences and job search strategies, (3) reemployment outcomes, and (4) satisfaction with UI program operations. The measures of key study outcomes about these topics are not reliably obtained or measured through other sources, such as administrative data or other surveys.

For each of the main topics covered by the LS-UI discussed in Section A.2, the longitudinal survey fills a gap where California’s and DOL’s administrative data either do not collect information or collect incomplete information. For example, the surveys will gather information on financial hardships faced by UI recipients and their families. Although this topic is an outcome of substantial interest to DOL to determine adequacy of UI benefits, UI administrative data contain no information about it. In addition, the LS-UI survey data will provide measures of other important outcomes for this study, such as the characteristics of post-UI jobs. DOL does not collect individual-level data that could be used to examine this issue, and state administrative data do not have extensive information about job characteristics; they typically contain information only on the total earnings for the job in a calendar quarter, the industry of the employer, and the employer name. Therefore, California’s administrative data are considerably less rich in detail about the post-UI jobs than what will be obtained through the survey data, which will be able to provide information about the start and stop dates of the post-UI jobs, the earnings or wage rates, the typical numbers of hours worked per week, the reasons for separations from jobs, the occupations of the jobs, and the availability of fringe benefits.

The LS-UI will also provide data about job search behaviors and the use of reemployment services and training. Although some administrative data exist about these items at the national level, the data pertain to participation in activities and services that are funded by specific funding sources, such as the Workforce Investment Act. No such administrative data exist for UI recipients in the Los Angeles and the Central Valley areas. If administrative data were used in lieu of survey data, there would be significant gaps in the coverage of job search, reemployment services, and training activities that could be conducted or used by UI recipients. Thus, any analysis that would be conducted using those data would not be able to provide a very useful picture of the totality of recipients’ experiences.

Finally, the longitudinal survey data will contain richer information about the characteristics of UI recipients than what would be available through administrative data. For example, administrative data will not be able to provide much information about a sample member’s household structure or detailed data on his or her pre-UI employment status.

The LS-UI does not duplicate any data collections within California. No surveys gather information about the experiences of UI recipients in the Los Angeles MSA and Central Valley areas, or in California more broadly. The longitudinal survey also does not duplicate the efforts of other large, national surveys, whose respondents might include California UI recipients as part of a nationally representative sample. For example, the Current Population Survey (CPS) is a large data set that is designed to calculate nationally representative estimates of employment and poverty.⁷ Respondents in the CPS are asked about UI receipt, so CPS data have been used to examine research topics related to the LS-UI, such as financial hardship for families of UI recipients (Congressional Budget Office 2010). However, the CPS was not designed to calculate representative estimates of the Los Angeles MSA and the Central Valley area. In addition, for three reasons, the CPS is not adequate to address the research questions in the LS-UI, at either the national or local level. First, the CPS does not collect longitudinal information during and after benefit receipt, so it cannot be used to track UI experiences throughout and after benefit receipt. Second, it collects limited information or does not collect any information on several key topics for the LS-UI. For example, the CPS collects limited information on prior jobs and no information about reemployment expectations or satisfaction with UI program operations. Third, it uses respondents’ self-reports of UI receipt and benefit amount, which may be less accurate than administrative data to identify UI recipients. In 2009, the UI benefit amounts recorded in the CPS implied that national UI benefit amounts were 30 percent smaller than DOL’s administrative data on total UI payments (Congressional Budget Office 2010).

Other national surveys share some of the same limitations as the CPS for answering the research questions in the LS-UI, and they gather less detailed or reliable information on key variables like UI receipt. For example, the American Community Survey (ACS) is a large, national survey that collects information on employment and income. However, the ACS is cross-sectional and does not collect information about UI receipt as a separate source of income.⁸ Consequently, identifying UI recipients and accurately characterizing their experiences is infeasible with the ACS. For example, Martinez-Schiferl et al. (2011) used models from the CPS to simulate UI benefit amounts from 2008 ACS data in three states. They estimated that total UI benefit amounts in the ACS were 75 percent lower than the total UI amounts in states’ administrative data.

5. Methods to Minimize Burden on Small Businesses or Entities

This data collection does not involve small businesses or other small entities.

6. Consequences of Not Collecting the Data

The longitudinal survey of UI recipients in two areas of California will provide the only source of reliable estimates of the financial adjustments that UI recipients in these areas make after they have lost their jobs. Furthermore, the survey data will provide unique information about the labor market experiences and job search strategies of UI recipients in the Los Angeles MSA and the Central Valley area at distinct points during and after UI benefit receipt, as well as information about recipients’ satisfaction with the UI program in California. Without this information, DOL and policymakers in California would not have strong, recent evidence to identify ways in which California’s UI program, and potentially other state UI programs, could be adapted to better balance the goal of providing temporary financial support to unemployed workers with the goal of encouraging their quick return to work. The information that will be collected through the LS-UI is not available through either other surveys or administrative data. The most recent prior study that provided similar information was conducted in Arizona in the 1970s—at a time when both the labor market and the UI system were dramatically different than they are today (see, for example, Kingston and Burgess 1978). No other national, state-specific, or locality-specific studies have been conducted that are similar to the LS-UI.

7. Special Data Collection Circumstances

The data collection effort is intended to gather information from survey respondents at three points in time over an approximate nine-month period after they begin UI benefit collection: (1) near the start of the respondents’ UI claim periods, (2) in the middle of the claim periods, and (3) after they are likely to have stopped benefit collection. This frequency and timing is by design, because a major goal of the study is to examine how the role of UI benefits, sample members’ job search, employment, and other outcomes change over time. The study results are expected to produce valid and reliable results that can be generalized to each of the areas in the study. In all respects, the data will be collected in a manner consistent with federal guidelines.

8. Federal Register Notice

a. Federal Register Notice and Comments

As required by 5 CFR 1320.8 (d), a Federal Register Notice, published on July 22, 2013 (FR/Vol. 78, No. 140, pp. 43929-43930), announced DOL’s intention to conduct a longitudinal study of unemployment insurance recipients and provided the public an opportunity to review and comment on the planned data collection and evaluation. Comments from this notice were due within 60 days of the publication (September 20, 2013), in accordance with the Paperwork Reduction Act of 1995. No comments were received. A copy of this 60-day notice is attached as Appendix A to Part A of this clearance request. The 30-day Federal Register Notice was published on February 10. 2014 (79 FR 7705, pp. 7705 -7706). Comments from this notice were due within 30 days of the publication (March 12, 2014), in accordance with the Paperwork Reduction Act of 1995. No comments were received.

b. Consultations Outside of the Agency

All expert consultations to ensure the technical soundness of the study and the relevance of its findings and to verify the importance, relevance, and accessibility of the information sought in the study were conducted by project staff from the Mathematica Policy Research evaluation team, as well as members of a Technical Working Group (TWG) that was convened about the study design and survey instruments. Their names and telephone numbers are listed below.

Mathematica Policy Research Evaluation Team Staff
Ms. Julita Milliner-Waddell	Project/Survey Director	(609) 275-2206
Dr. Karen Needels	Senior Researcher	(541) 753-0201
Dr. Frank Potter	Senior Fellow	(239) 558-5956
Dr. Walter Nicholson Dr. Joanne Lee	Senior Fellow Researcher	(239) 774-3693 (510) 830-3727
Dr. Stephen Wandner	Visiting Scholar at the Urban Institute and W. E. Upjohn Institute for Employment Research	(301) 785-6670
Dr. Randall J. Olsen	Center for Human Resource Research at the Ohio State University (CHRR)	(614) 442-7348
TWG Members
Dr. Rich Hobbie	National Association of State Workforce Agencies	(202) 434-8020
Dr. Christopher O’Leary	W. E. Upjohn Institute for Employment Research	(269) 385-0407
Polly Phipps	Bureau of Labor Statistics (BLS), U.S. Department of Labor	(202) 691-7513

9. Respondent Payments

For the LS-UI, the contractor will pursue many strategies to maximize survey response. These include developing a survey that is easy to navigate and administer within a reasonably short time, offering optional modes for response, using DOL letterhead, and employing persistent and rigorous efforts to locate and contact sample members. To complement these efforts and further encourage sample members to complete the survey at each round of data collection, a monetary incentive is also needed.

A combination of pre- and post-paid incentives is proposed for the LS-UI. At round one, a $5 bill will be included in the advance mailing that invites sample members to participate in the study. The advance mailing will be sent using regular mail service from the United States Postal Service. Since the study will only follow first interview completers at subsequent rounds of data collection, rounds two and three will use a post-pay incentive only. The rationale for this is that the researchers will have established a relationship and trust with sample members and the pre-payment will not be needed.

In an experiment conducted for the Trade Adjustment Assistance (TAA) study conducted for DOL, Mathematica tested the effect of a nominal cash pre-pay followed by a postpaid check. In that experiment 20 percent of the sample received a $5 pre-payment and was eligible for a $20 interview completion post-payment compared to a post-payment of $25. The prepaid incentive had a small effect on interview completion rates, with an overall response rate of about 43 percent for the $5 prepayment group, compared to 39 percent for the post-payment-only group. The effectiveness of a prepaid cash incentive was also tested for the 2007 Health Tracking Household Survey (HTHS) and was shown to encourage faster response, have a lower initial refusal rate, and reduce the effort needed to complete an interview. Interviewers in the HTHS study made almost 4,000 fewer calls to cases that received the cash prepayment compared to cases without the cash prepayment, saving twice the value of the prepayment by making fewer calls. The resultant response rate was 3 points higher for cases with the prepayment (CyBulski et al. 2008).

To further encourage web completions and call-ins, which are less costly to the project, sample members will be offered a higher incentive payment for these completions. A total of $30 will be provided if the respondent completes the interview on the web or calls in, and a total of $20 will be provided if Mathematica initiates the call. The post-paid incentive will be issued by check for the incentive amount less the $5 pre-pay. The Evaluation of the Unemployment Compensation Provisions of the American Recovery and Reinvestment Act of 2009 (OMB 1225-0089), being conducted for DOL, recently received OMB approval to use a similar two-tiered incentive approach to encourage web survey completion.

The offer of incentives is critical to efforts to gain cooperation from sample members and increase response rates ensuring the representativeness of the sample and providing data that are complete, valid, reliable, and unbiased. Given the importance of this study for DOL, the data collection must be held to high standards on these criteria, and offering incentives can help achieve that goal. To leverage fully the benefits of offering incentives, the advance letter to the study participants will fully explain the incentive offer and the conditions governing issuance of the balance. The URL and password needed to access the web survey will be included in the advance mailing. Interviewers will also mention the incentive when they establish contact with the participants and attempt to gain their cooperation.

The offer of incentives is also important because of the declining response to telephone surveys and increasing costs associated with achieving high response. The use of incentives has become a more common practice for survey studies (Curtin et al. 2005) in the current response environment. Singer et al (2000) found that offering incentives can help achieve high response rates by increasing the sample members’ propensity to respond. Studies offering incentives show decreased refusal rates and increased contact and cooperation rates. Among sample members who initially refuse to participate, incentives increase refusal-conversion rates. There has also been evidence that incentives given to respondents in advance, such as the proposed $5 pre-pay, further establish the survey’s legitimacy, leading to increased response rates. Millar and Dillman 2011 found that the inclusion of a cash incentive increased their web survey response rate by 17 percentage points, going from 21 percent to 38 percent.

By increasing sample members’ propensity to respond, incentive payments have been found to significantly reduce the number of calls required to resolve a case and to significantly reduce the number of interim refusals. Thus, incentive payments can help contain costs, and pass some of the costs of conducting the survey as a gain to the participant rather than into additional survey operations.

Although incentives help gain cooperation to increase the overall response rate, they also increase the likelihood of participation from subgroups with a lower propensity to cooperate with the survey request, helping to ensure the representativeness of the respondents and the quality of the data being collected. For example, Jäckle and Lynn (2007) find that incentives increase the participation of sample members more likely to be unemployed. Furthermore, paying incentives does not impair the quality of the data obtained (such as item nonresponse or the distribution of responses) from groups who would otherwise be underrepresented in a survey (Singer et al. 2000).

10. Privacy

This section discusses the general procedures that will be followed to protect the data that are part of this clearance request.

a. Procedures to Protect the Privacy of the Data Collected as Part of the Study

All letters, emails, and other respondent materials will include assurances of privacy protection. In addition, introductory scripts to solicit participation in the survey will be standardized to inform sample members that their responses will be kept private to the extent provided by law. Interviewers will be trained in privacy procedures and will be prepared to describe them in full detail, and to answer any related questions raised by participants. For example, the interviewer will explain that the individual’s answers will be combined with those of others and presented in summary form only.

All data items that identify sample members will be kept only by the contractor, Mathematica, for use in assembling records data and in conducting the interviews. While in use, electronic data will be stored in secure folders only accessible by project staff. Any hardcopy data will be stored in locked file cabinets. DOL has requested that a public data file be created after the end of data collection and analysis of the data. Any data provided to DOL by Mathematica will not contain personal identifiers, thus precluding individual identification and will be provided in a format easily accessible using statistical software. After study completion, Mathematica will purge all data processed during the study from all data storage components as instructed by DOL.

It is Mathematica’s policy to efficiently protect private information and data in whatever medium they exist, in accordance with applicable federal and state laws and contractual requirements. In conjunction with this policy, all Mathematica staff shall:

1. Comply with a Mathematica pledge that is signed by all Mathematica full-time, part-time, and hourly employees, and with the Mathematica Security Manual procedures to prevent the improper disclosure, use, or alteration of private information. Staff may be subjected to disciplinary or civil or criminal actions or both for knowingly and willfully allowing the improper disclosure or unauthorized use of private information.

2. Access private and proprietary information only in performance of assigned duties.

3. Notify their supervisor, the project director, and the Mathematica security officer if private information has been disclosed to an unauthorized individual, used in an improper manner, or altered in an improper manner. All attempts to contact Mathematica staff about any study or evaluation by individuals who are not authorized access to the private information will be reported immediately to both the cognizant Mathematica project director and the Mathematica security officer.

To allow external verification and replication of the study findings, as well as additional research, public use data files containing key analysis variables created for the LS-UI will be produced at the end of the study and formatted to data.gov specifications. These public use files will follow the current relevant OMB checklist to ensure that they can be distributed to the general public for analysis without restrictions. Steps will be taken to ensure that sample members cannot be identified in indirect ways. For example, categories of a variable will be combined to remove the possibility of identification due to a respondent being one of a small group of people with a specific attribute. Variables that will be carefully scrutinized include age, race and ethnicity, household composition, dates pertaining to employment, household income, household assets, and others as appropriate. Variables will also be combined in order to provide summary measures to mask what otherwise would be identifiable information. Although it cannot be predicted which variables will have too few respondents in a category, the study researchers will not report on categories or responses that are based on cell sizes of less than five. If necessary, statistical methods will be used to add random variation within variables that would be otherwise impossible to mask. Finally, variables that could be linked to identifiers by secondary users will be removed or masked.

1. Systems Security

Mathematica’s computer facilities include state-of-the-art hardware and software. The hardware and software configurations have been designed to facilitate the secure processing and management of both small- and large-scale data sets.

Facility. The doors to Mathematica’s office space and Survey Operations Center (SOC) are always locked, and all SOC staff are required to display current photo identification while on the premises. Visitors are required to sign in and out and must wear temporary ID badges while on the premises. Any network server containing private data is located in a controlled, limited-access area. All authorized external access is through a server under strict password control.

Network. Sensitive data are stored in secure folders that reside on a Windows 2008 Server volume using NT File System (NTFS). BitLocker encryption software, configured to use a 256-bit AES key, encrypts data on the volume as they are stored. The encryption persists for the life of the volume. NTFS/BitLocker makes the data accessible only to users with authorized access, and makes data inaccessible to software that circumvents normal access control, in case the media are stolen. NTFS/BitLocker stores user data in an encrypted format on the volume, but it works transparently with most applications and backup utilities. All the rules of file system trustee assignments, trustee rights, ownership, sharing, visibility, locking, transactions, and space restrictions remain the same on the encrypted volume. Data on the “Secure_Data” folders are backed up using ArcServe 11.5, which encrypts the contents using the 3DES algorithm. These separate backups are overwritten every two months by backups of newer secure data, a process that enables compliance with secure data destruction requirements.

Access to all network features, such as software, files, printers, Internet, email, and peripherals, is controlled by userid and password. Mathematica staff are required to change their password for computer access no less than every month, and passwords must adhere to the following standards: be at least eight characters long, contain at least one letter (upper or lower case), and contain at least one numeric or special character. All userids, passwords, and network access privileges are revoked within one working day for departing staff and immediately for terminated staff. All staff are required to log off the network before leaving for the day.

Printers. Printer access is granted to all staff with a valid userid and password. The physical hard disks on which the printer queues reside are subject to the same security and crash procedures that apply to the file servers. Printer queues are confined to write-access to all staff. No staff have read-access to the printer queues; that is, they cannot browse the contents of the printer queues. Printer stations are appropriately monitored according to the sensitivity of the printed output produced. No private or proprietary data or information can be directed to a printer outside Mathematica’s offices.

Electronic communication. Each of Mathematica’s locations has a site-specific local area network. A combination of T1 and ethernet private line (EPL) lines links the site-specific local area networks into a wide area network (WAN) and supports cross-office communications. Traffic on the Mathematica internal network, which is not encrypted, is secured by these links, all of which are private, point-to-point communication lines dedicated to Mathematica traffic and completely contained within Mathematica’s firewalls. As each office is connected to other offices solely by these private point-to-point lines and not through the Internet, all WAN traffic is contained and protected within Mathematica’s firewalls; no WAN traffic is routed through the Internet.

2. Treatment of Data with Personal Identifying Information

All data containing personal identifying information (PII)—including Social Security number (SSN), name, home address, date of birth, and telephone number—are considered to be sensitive, or private, data. The LS-UI is in compliance with the aforementioned company security policies. Listed below are specific details regarding the handling and processing of private information in this evaluation.

Access. Electronic files with private data are stored in restricted-access network directories. Access to restricted directories is limited through access control permissions, on a need-to-know basis to staff who have been assigned to and are currently working on the project. When temporarily away from their work area, project staff are instructed to close files and applications and to lock their workstations using the CTRL-ALT-DEL command. Workstations automatically lock within a set number of minutes, and a password must be used to regain access through the protected screen saver.

Electronic communication. For internal emails, staff are forbidden to transmit sensitive study information as a regular file attachment; they are instructed instead to use the “insert hyperlink” feature in Outlook to include a shortcut to the file. This allows the receiver to go to the file directly but will not allow access to unauthorized individuals. In addition, staff are instructed to avoid including sample member names or other PII in internal emails, so that there is no potential for these to be viewed by others.

Emails sent outside Mathematica are not automatically encrypted, and therefore neither the text nor attachments are secure. Before sending an email containing sensitive information, the sender is obligated to ensure that the recipient is approved to receive such data. When files must be sent as attachments outside Mathematica, staff are instructed to use SecureZip, a FIPS 140-2 compliant encryption tool to password-protect the file and transmit the password to the recipient using a separate form of communication, preferably via phone. When a sample member’s name and contact information are sent outside Mathematica, the information is included in a secure attachment rather than in the text of the email.

LS-UI databases. Project databases containing private information are password-protected and accessible only to staff currently working on the project. To access the project’s database, users must first log on to their workstations and then, upon starting the database, log in again using a separate prompt. Project databases will be removed from the company servers and securely destroyed at the end of the data-processing period.

Telephone interviewing. Telephone interviewers for the LS-UI will be seated in a common supervised area. As part of the process to verify sample member identity, interviewers will have access to respondents’ names and birth dates, as well as the last four digits of their SSN. Birth date and the last four SSN digits will be displayed on the computer screen only temporarily, at the beginning of the survey, so that the interviewer can verify the sample member’s identity. Interviewing staff for this project receive training that includes general SOC security and privacy procedures, as well as project-specific training that includes explanation of the highly private nature of this information, instructions to not share it or any PII with anyone not on the project team, and warnings about the consequences of any violations. Telephone interviews are recorded for monitoring purposes to allow project staff to make improvements to question flow and wording, and for educational and training purposes, to aid SOC staff in improving their interviewing skills. All recordings will be securely destroyed by the end of the data collection.

Locating. Staff that work on updating sample member contact information when the original contact is not successful must have access to key identifying information for short periods. These staff members receive training that includes general SOC security and privacy procedures, as well as project-specific training that includes clear instructions on what data and databases can be accessed and what data are required and can be recorded.

Locators may talk to a sample member’s family, relatives, or other references to obtain updated contact information. To protect the sample member, locators are given scripts that describe what they can and cannot say when using these sources to obtain information. For example, interviewers will be instructed not to tell anyone that the sample member has been selected to participate in a study of the unemployed. Rather, they will indicate that Mathematica is trying to reach the sample member for an important study sponsored by DOL.

Locating and calling contact sheets. Project team members keep only the minimum amount of printed private information needed to perform assigned duties. Hard-copy materials (such as locating or calling contact sheets) containing data with any individual identifiers (e.g., name, street address) are stored in a locked cabinet or desk when not being used. When in use, such materials are carefully monitored by a project supervisor and are never left unattended. At the conclusion of the project, a final disposition of all remaining samples will be made, and contact sheets and other associated materials will be destroyed.

Hard-copy printouts. Sensitive temporary work files, used to create hard-copy printouts and stored in temporary work files on local hard drives, are deleted on a periodic basis. Hard-copy output with private information is shredded or stored securely once no longer needed. Test printouts of data records containing personal identifiers that are generated during file construction are shredded.

Data files. When possible, electronic files for everyday use are created without personal identifiers. Data and sample files that must contain sensitive data are stored and analyzed on one of Mathematica’s “Secure_Data” drives. Specifically, staff working on this project will be instructed to maintain all files with private data in project-specific, encrypted folders on the Mathematica network. Access control lists restrict access on a need-to-know basis and only to project staff who are specifically authorized to view the sample data (as designated by the project director or survey director) to select and process the sample or to process the data files. Sensitive data that are no longer needed in the performance of the project will be magnetically erased or overwritten using Hard Disk Scrubber or equivalent software, or otherwise destroyed.

11. Questions of a Sensitive Nature

The LS-UI will include questions about current employment, employment characteristics related to the separating employer, earnings, job search, job offers, reemployment, participation in government assistance programs, income sources and amounts, financial hardships, and satisfaction with the UI program. Any of these topics could be perceived as sensitive by sample members, depending on their individual circumstances. Obtaining information about these potentially sensitive topics is, however, integral to addressing the research questions posed by the study and to describing the characteristics of UI recipients. Mathematica has extensive experience in developing these types of sensitive questions.

All survey questions (see Appendixes B-D) have been worded to show the highest level of objectivity and sensitivity. Interviewers will also be trained to show sensitivity to respondents while remaining impartial. In addition, many questions in the current survey have been included without modification from other surveys of similar populations, such as the TAA Evaluation, the Accelerated Benefits Demonstration, the COBRA Subsidy Study, and the National Longitudinal Survey of Youth, and have been used extensively with no evidence of harm. All questions in the current survey, including those deemed potentially sensitive, have been thoroughly pretested.

Further, as described in Section A.10, all respondents will be assured of privacy at the outset of the interview and will be reminded of this, as appropriate. Respondents can refuse to answer any question in the survey.

12. Hour Burden of the Collection of Information

The hour burden for each round of the LS-UI is estimated at 25 minutes per completed survey. Since this is a longitudinal study—our goal is to interview sample members three times within an approximately nine month period—we will follow only those sample members who complete a survey at the first round. Because we will be contacting sample members early in their UI experience, we expect contact information to be good and survey topic salience to be high. Based on these factors, coupled with the combination of techniques discussed in Section A.9 to maximize response, we expect to interview 1,089 respondents (80 percent of the sample) in each area at round one; at round two we estimate that we will complete interviews with 85 percent of the round one completers, for a total of 926 interviews; and will complete interviews with 90 percent of round 2 completers at round 3 for a total of 833 completed interviews at round three and 2,848 interviews overall per area. In the Americans’ Changing Lives (ACL) panel survey, Lepkowski and Couper (2002) found that prior round survey experience was strongly correlated with cooperation--gaining the cooperation of 86.5 percent of located Wave 1 participants at Wave 2. Rapport established at each round and maintaining contact with respondents between rounds will enhance our ability to achieve targeted response rates at rounds two and three. The total estimated time burden is 2,373 hours for the projected 5,695 respondents across all three rounds and both areas (5,695*(25/60). These estimates are presented in Table A.3. The total monetized hours cost of conducting this survey is estimated to be $52,223. This cost represents the time to complete three rounds of the survey, multiplied by the number of projected respondents at an estimated average hourly wage of $22.01 per hour.⁹

Table A.3. Burden Estimates for LS-UI Respondents

Number of Respondents	2,178a
Number of Responses per Respondent	2.615
Hourly Cost of Burden (Dollars)	$22.01
Hours per Response	25/60
Total Monetized Hours Cost (Dollars)	$52,223b

^aApproximately 1,666 respondents are expected to complete all 3 interviews, 186 respondents are expected to complete 2 interviews, and 326 respondents are expected to complete 1 interview.

^bCalculations vary due to rounding.

13. Estimated Total Cost Burden to Respondents and Record Keepers

No financial costs will be borne by respondents.

14. Estimated Annualized Cost to the Federal Government

The total cost of the study to the federal government is $1,596,308. Over the 30-month study period, this translates to an annualized cost to the federal government of $638,523. These costs include the major expense categories required to conduct this study that are shown in Table A.4.

Table A.4. Cost to the Federal Government

Activity	Cost (dollars)
Kickoff Meeting	16,707
Develop Work Plan, Evaluation, Design & Analysis Plans	41,309
Sample Frame	25,901
Questionnaire Development	131,695
OMB Clearance	57,685
Obtain UI Administrative Dataa	94,139
Data Collection/Survey Fieldingb	573,755
Survey Operations—Locating	35,589
Survey Operations—Coding	27,776
Survey IS Programming	295,472
Technical Work Group Meetings	41,843
Data Analysis	37,298
Report Writing	145,755
Client Briefings	20,498
Public Data File	22,331
Project Management	28,555
Total Cost to the Government	$1,596,308

^aIncludes reimbursable costs to states and contractor costs.

^bRepresents the costs of administering all survey rounds, including costs for all mailings, contact attempts, and incentives.

15. Changes in Burden

This is a new information collection. The survey data collection efforts for this study will count as 2,848 hours toward DOL’s information collection burden.

16. Publication Plans and Project Schedule

The study will use tabulations (univariate analyses) and multivariate analyses to describe how the experiences of UI recipients change over time during and after they collect UI benefits. This section describes the tabulations in the study, which will be used to examine each of the main topics identified in Section A.1. The multivariate analyses are discussed in Section B.2 of this package.

Across all topics, the study will always present tabulations of measures separately for the two areas to reflect their different UI program features and local economic conditions.¹⁰ For any subgroups for which statistical tests are conducted, the study will report minimum detectable differences for key measures at a 5 percent significance level based on actual sample sizes and variances. T-tests will be used to identify statistically significant differences in means of variables between subgroups that are explored in the analysis, and chi-squared tests will be used to identify statistically significant differences in distributions of variables. The analysis plans for each topic are described in more detail below.

Financial hardship. Measures of financial hardship will be constructed based on family income, savings amounts, whether the household reports it did not make a payment such as for their mortgage or rent, and whether UI recipients make use of other government assistance programs. The analysis will calculate the percentages of UI recipients in each study area who experienced financial hardship at any of the three survey rounds and at each survey round. The rates of financial hardship will be compared for individuals whose pre-UI wages were above or below the median wage in their study area (high- or low-wage earners), and for individuals who are single or part of a household couple. High- and low-wage earners will be defined relative to the median over the study area rather than over the survey sample because it is a more stable measure of the local economy and less likely to reflect mass layoffs or other shocks that might affect UI recipients in the survey. This information will be obtained from the May 2013 Occupational Employment and Wage Estimates for all civilian workers in the study areas. Among exhaustees, the analysis will compare the rates of financial hardship before and after UI benefits were exhausted. Finally, the average total amount of UI benefits received will be calculated for UI recipients overall and for each of the subgroups of focus.

Labor market experiences and job search strategies. The key measure of job search will be the mean number of hours that UI recipients reported searching for work during a week. This measure will be calculated at each survey round for all respondents in a study area, for high- and low-wage earners, and for individuals who are single or are part of a household couple. The analysis will also tabulate the types of methods that UI recipients used to look for work, whether they used services provided at an American Jobs Center, and the types of services that they received. For those who are not employed and report not looking for work at the time of each survey round, the analysis will tabulate their reasons for not looking for work.

Statistics about reemployment expectations will include the percentage of survey respondents who expect to be recalled to their former jobs and their expected time to reemployment. Data about the quality of the desired job also will be tabulated. For example, the analysis will calculate the percentage of survey respondents who expect to need to change their industry or occupation from the industry and occupation of their pre-UI job. Statistics also will be calculated to show the minimum earnings and fringe benefits that the respondent requires before he or she would accept a job. The study will tabulate these measures at each survey round for all survey respondents in a study area, for high- and low-wage earners, and for individuals who are single or part of a household couple. For exhaustees, the analysis will compare these measures before and after UI benefits were exhausted.

Reemployment outcomes. The study will calculate the percentage of individuals who are reemployed at each survey round. Among reemployed individuals, tabulations will be used to examine the percentages of individuals in each study area whose new job is in the same industry or occupation as their pre-UI job. The study will also compare the wages and fringe benefits of their new job and their pre-UI job. These measures will be tabulated over all UI recipients, as well as important subgroups (such as high- and low-wage earners) for whom the analysis is feasible. Cross-tabulations over all survey respondents in a survey round will compare reemployment expectations about industry and occupation, wages, and fringe benefits with realized reemployment outcomes.

Satisfaction with UI program operations. The measures of satisfaction with UI program operations will be based on survey questions about the clarity of instructions about UI filing, staff helpfulness in answering questions, the amount of time required to file a claim, challenges filing initial and subsequent claims, the speed of payment, and overall program satisfaction. The analysis will compare means and distributions of these measures for important subgroups for whom the analysis is feasible.

The study will present findings in a final report. The tentative schedule for the fielding of the data collection efforts, the delivery of the report, and the provision of a public use data file and documentation is provided in Table A.5.

Table A.5. Schedule for Project Tasks

Tasks	Schedule (pending OMB approval)
Fielding of the LS-UI Survey	5/5/2014 - 1/29/2015
Final Report	5/25/2015
Public Use Data File and Documentation	6/18/2015

17. Reasons for Not Displaying Expiration Date of OMB Approval

The OMB control number and expiration date issued for this data collection will be included on all materials sent to sample members.

18. Exception to the Certification Statement

Exception to the certification statement is not requested for this data collection.

REFERENCES

Addison, John T., and McKinley L. Blackburn. “The Effects of Unemployment Insurance on Postunemployment Earnings.” Labour Economics, vol. 7, no. 1, 2000, pp. 21–53.

Card, David, Raj Chetty, and Andrea Weber. “The Spike at Benefit Exhaustion: Leaving the Unemployment System or Starting a New Job?” Working paper #12893. Cambridge, MA: National Bureau of Economic Research, 2007.

Congressional Budget Office. “Unemployment Insurance Benefits and Family Income of the Unemployed.” Washington, DC: Congressional Budget Office, 2010. Available at [http://www.cbo.gov/sites/default/files/cbofiles/ftpdocs/119xx/doc11960/11-17-unemploymentinsurance.pdf]. Accessed April 17, 2013.

Curtin, R., S. Presser, and Eleanor Singer. “Changes in Telephone Survey Nonresponse over the Past Quarter Century.” Public Opinion Quarterly, vol. 69, no. 1, spring 2005, pp. 87–98.

CyBulski, Karen, Tom Barton, and Barbara Lepidus Carlson. “Prepays, Promises, and Postpays: Additional Evidence on What Helps Response Rates.” Proceedings of the American Association for Public Opinion Research. New Orleans, Louisiana: American Association for Public Opinion Research, 2008.

Czajka, John. “Appendix A: SIPP Data Quality.” Reengineering the Survey of Income and Program Participation. Washington, DC: The National Academies Press, 2009.

Ehrenberg, Ronald G., and Ronald L. Oaxaca. “Unemployment Insurance, Duration of Unemployment, and Subsequent Wage Gain.” American Economic Review, vol. 66, no. 5, 1976, pp. 754–766.

Gruber, Jonathan. “The Consumption Smoothing Benefits of Unemployment Insurance.” American Economic Review, vol. 87, no. 1, March 1997, pp. 192–205.

Herkenhoff, Kyle F., and Lee E. Ohanian. “Foreclosure Delay and U.S. Unemployment.” Working paper 2012-017A. St. Louis, MO: Federal Reserve Bank of St. Louis, 2012.

Jäckle, Annette, and Peter Lynn. “Respondent Incentives in a Multi-Mode Panel Survey: Cumulative Effects on Nonresponse and Bias.” Working paper presented to the Institute for Social and Economic Research, University of Essex, Colchester, United Kingdom, 2007.

Lepkowski, James M., and Mick P Couper. 2002. “Nonresponse in the Second Wave of Longitudinal Household Surveys” in Survey Nonresponse, edited by Robert M Groves, Don A. Dillman, John L. Eltinge, and Roderick J.A. Little. Hoboken, NJ: John Wiley & Sons, Ltd, pp. 259–272.

Katz, Lawrence, and Bruce Meyer. “The Impact of Potential Duration of Unemployment Benefits on the Duration of Unemployment Outcomes.” Journal of Public Economics, vol. 41, no. 1, 1990, p`p. 45–72.

Kay, Ward R. “The Use of Targeted Incentives to Reluctant Respondents on Response Rates and Data Quality.” Proceedings of the American Association for Public Research. Montreal, Canada: American Association for Public Opinion Research, 2001.

Kingston, Jerry L., and Paul L. Burgess. “The Adequacy of Unemployment Insurance Benefits: An Analysis of Adjustments Undertaken Through Thirteen and Twenty-Five Weeks of Unemployment.” Employment and Training Administration, U.S. Department of Labor. Washington, DC: U.S. Department of Labor, 1978.

Krueger, Alan B., and Andreas Mueller. “Job Search and Unemployment Insurance: New Evidence from Time Use Data.” Working Paper # 175. Princeton, NJ: Princeton Center for Economic Studies, 2008. [Also published in the Journal of Public Economics, vol. 94, nos. 3–4, 2010, pp. 298–307.]

Krueger, Alan B., and Andreas Mueller. “Time Use, Emotional Well-Being and Unemployment: Evidence from Longitudinal Data.” American Economic Review, vol. 102, no. 3, 2012, pp. 594–599.

Marcus, Steven S., and Joseph W. Frees. “U.S. Department of Labor Unemployment Insurance Claimant Satisfaction Study.” OWSOP 99-2. Employment and Training Administration, U.S. Department of Labor. Washington, DC: U.S. Department of Labor, September 1998. Available at [http://wdr.doleta.gov/research/FullText_Documents/op_02-99.pdf]. Accessed April 9, 2013.

Martinez-Schiferl, Michael, Sheila Zedlewski, and Linda Giannarelli. “Reports of Unemployment Compensation in the American Community Survey: A Data Note.” Washington, DC: Urban Institute, 2011. Available at [http://www.urban.org/UploadedPDF/412323-Reports-of-Unemployment-Compensation.pdf]. Accessed April 17, 2013.

Millar, M., and Don Dillman. “Improving Response to Web and Mixed-Mode Surveys.” Public Opinion Quarterly, vol. 75, no. 2, summer 2011, pp. 249–269.

Moffitt, Robert. “Unemployment Insurance and the Distribution of Unemployment Spells.” Journal of Econometrics, vol. 28, no. 1, 1985, pp. 85–101.

Singer, Eleanor, John Van Hoewyk, and Mary P. Maher. “Experiments with Incentives in Telephone Surveys.” Public Opinion Quarterly, vol. 64, no. 2, summer 2000, pp. 171–188.

U.S. Census Bureau. “Table 2A. Reported Internet Usage for Individuals 3 Years and Older, by Selected Characteristics: 2010.” Current Population Survey, October 2010 (Internet release: July 2012). Washington, DC: U.S. Census Bureau, 2010. Available at [http://www.census.gov/hhes/computer/publications/2010.html]. Accessed September 10, 2012.

1 See “Unemployment Insurance Weekly Claims Report,” available at [ http://www.oui.doleta.gov/press/2014/062614.pdf]. Accessed June 30, 2014. More than 2.4 million people were receiving benefits from some type of unemployment compensation program, which includes the state program and several other programs that are either offered on an ongoing basis or that were enacted in response to the most recent recessionary downturn.

2 See “Comparison of State Unemployment Laws,” available at [http://workforcesecurity.doleta.gov/unemploy/comparison2014.asp]. Accessed June 30, 2014.

3 Krueger and Mueller (2012) used data from a longitudinal survey of UI recipients in New Jersey (“Survey of Unemployed Workers in New Jersey”) that attempted to conduct interviews with UI recipients on a weekly basis for up to 24 weeks. Of the initial sample of about 64,000 recipients, only 10 percent participated in the first wave of the survey. The respondents in the first wave of the survey participated in about 40 percent of the remainder of the waves.

4 The 2012 Agricultural Census provided information on values of agricultural production for Fresno County: [http://www.agcensus.usda.gov/Newsroom/2014/05_02_2014.php], accessed June 30, 2014. A comparison of economic indicators for Fresno County and Los Angeles County is available through California’s Employment Development Department, accessed on June 30, 2014: [http://www.labormarketinfo.edd.ca.gov/cgi/databrowsing/LocalAreaProfileComQSResults.asp?menuChoice=localAreaCom&selectedindex=0&area1=0604000037&countyName=&area2=0604000019&countyName=&area3=0601000000&countyName=].

5 UI claimants in California typically file for benefit payments on a biweekly basis. During this process, they attest to their eligibility for benefits for each particular calendar week for which they are requesting benefits. The precise eligibility criteria are determined by the state, but claimants generally must be able to work and be available for and searching for work. In addition, they must report their earnings for the week, because having employment affects eligibility for benefits.

6 Substate patterns in UI recipiency are not reported by states to DOL. However, given the size of the Los Angeles MSA and the ability to build a geographic area from several Central Valley MSAs, it is expected that only one cohort will be needed to provide an adequate number of sample members. Use of one cohort, rather than two, would be more cost-efficient for the evaluation. However, the study plans allow for the selection and use of a second cohort, if needed.

7 The CPS is collected by the Census Bureau and published by the Bureau of Labor Statistics. Approximately 60,000 households are selected in a probability sample and are in the survey for two sets of four consecutive months, paced eight months apart. A core group of questions about labor force participation during a specific calendar week is asked each month. Supplemental questions about income, health insurance, and other topics are collected annually in the Annual Social and Economic Supplement or less frequently.

8 The ACS is conducted by the Census Bureau and administered to more than 2 million housing units a year. In the ACS, individuals are first asked to report the amounts of income that were received from each of seven specific sources, including employment and Social Security. Then, individuals are asked to report the total income received from all other sources; UI benefits, Veterans’ payments, child support, and alimony are listed as examples of what can be included in this category. Survey respondents also are asked to report the total income that was received from all other sources besides the seven specified ones, but they are not asked to distinguish between the types of other sources. Therefore, it is not directly possible to learn from survey respondents who reported income from other sources how much income was from UI benefits.

9 This hourly wage estimate is the mean hourly wage for all occupations published by the Bureau of Labor Statistics, U.S. Department of Labor, May 2012 National Occupational Employment and Wage Estimates.

10 As discussed in Section A.2, the two MSAs are purposively selected, and statistics about them are not intended to be generalizable to a broader population.