2021 PAP Satisfaction Survey - Part A and B 9-20-2021 (clean)

OMB Package, EBSA Participant Assistance Program Customer Surveys

OMB Control Number: 1210-0161

Expiration Date: 06/30/2023

Part A:

Supporting Statement

EBSA Participant Assistance Program Customer Survey

Justification

Supporting Statement

EBSA Participant Assistance Program Customer Survey

A. Justification

1. Necessity of the information collection

The purpose of this data collection effort is to solicit inquirers’ feedback and compile reports on the applicability and utility of EBSA’s Participant Assistance Program (PAP). PAP is an integral part of EBSA’s overall mission to: deter and correct violations of the relevant statutes through strong administrative, civil and criminal enforcement efforts to ensure workers receive promised benefits; develop policies and regulations that encourage the growth of employment-based benefits; assist workers in getting the information they need to exercise their benefit rights; assist plan officials to understand the requirements of the relevant statutes in order to meet their legal responsibilities.

This survey will collect customer satisfaction data for a sample of private citizens who call into the participant assistance program to ask about their private sector employer provided benefits such as pensions, retirement savings, and health benefits. Three types of callers will be queried: 1. Those who need benefit claim assistance 2. Those who have a valid benefit claim and 3. Those who have an invalid benefit claim. The results of the survey will be analyzed to provide actionable data that could be used to improve program performance. Examples of improved performance include, but are not limited to:

• Being more attuned to inquirers’ needs- Benefits Advisors should be more adept at identifying issues that lead to benefits recoveries and enforcement leads

• Survey data will enable National and Regional management to identify potential training needs

• Satisfaction scores will guide EBSA leadership to determine which Regions need assistance improving customer service

• Scores on individual BAs will reveal high performers and allow the agency to use those BAs’ techniques as best practices for program-wide improvement.

The study will include data from regional offices in Atlanta, Boston, Chicago, Cincinnati, Dallas, Kansas City, Los Angeles, New York, Philadelphia and San Francisco and District offices in Miami, Seattle and Washington.

2021 Questions Revision

The survey’s demographic questions have been updated in response to Executive Order 13985—Advancing Racial Equity and Support for Underserved Communities Through the Federal Government.

2. Purpose of the information collection

EBSA will conduct the surveys to evaluate the Participant Assistance services provided by the benefits advisor staff nationwide to understand how well the Agency is meeting service delivery goals by; (1) assessing EBSA’s customers’ perception of the services provided by the EBSA benefit advisors, and (2) determining what actions the performance data indicate could enable each regional office to provide the best possible participant assistance service; (3) establishing a current baseline for EBSA’s (Government Performance and Accountability Act GPRA) measurement. EBSA will use the data from the survey to track the agency’s progress on accomplishing it’s GPRA measurement goal. The agency will use the data as a basis to improve individual, office, and Agency level program performance. The study will provide periodic reports (quarterly, midyear, end of year) of enough detail to inform managerial action toward improving performance; and generate informative data, [e.g., regression charts, etc.] and recommendations on which EBSA managers may base decisions and take actions to improve the program’s performance. Such managerial decisions as a result of the PAP study, may result in staff training and development, workload allocation, and financial and human resource allocation.

The contractor will retain control over the information and safeguard it from improper access, modification, and destruction.  The information collection is designed to yield data that meet all applicable information quality guidelines. Information collected in this study is not expected to be disseminated publicly and therefore, will not result in scientific, management, technical or general informational publications.

2021 Questions Revision

In response to Executive Order 13985—Advancing Racial Equity and Support for Underserved Communities Through the Federal Government—the survey information will also be used to provide additional training to EBSA benefits advisors in order to better serve the underserved populations that we assist.

3. The use of automation, electronic, mechanical, or other technological collection techniques

The contractor will use its state-of-the-art Computer-Assisted Telephone Interviewing (CATI) system to conduct the interviews.

4. Efforts to identify duplication

This survey is a direct measure of the services provided by the PAP, there are no other studies currently being conducted to gather data on inquirer’s feedback on services.

5. Impact on small businesses

As the information collection will go to individuals, no impact on small businesses, organizations, or government bodies is expected.

6. Consequences of less frequent collection

EBSA plans on using the information from this study to track progress towards its GPRA goals and to use it to improve the services provided to the public. As a result, having frequent and current information is important for helping the agency identify and prioritize actions.

7. Explain any special circumstances

There are no special circumstances that require the collection to be conducted in a manner inconsistent with the guidelines in 5 CFR 1320.5.

8. Public comment Federal Register notice and consultation with outside representatives

The Department published a Federal Register Notice on July 21, 2021 (86 FR 38500), providing the public with 60-days to comment about this information collection as required by 5 CFR 1320.8(d).  No comments were received. 

9. Payment or gifts to respondents other than remuneration to contractors or grantees

Respondents will not receive any payment or gifts for completion of surveys.

10. Assurances of confidentiality provided to respondents and the basis

Individuals contacted will be assured that the survey is anonymous and that all sensitive information will be protected to the greatest extent allowed by law.

11. Questions of a sensitive nature

The survey will include questions about the inquirer’s interaction with PAP benefits advisor and the address areas of overall satisfaction and feedback on perceptions of the quality of the service received, along with some general demographic questions about the inquirer.

12. Burden hour estimates and annualized costs to respondents for the hour burdens

Study Year	Type of Respondent	No. of Respondents	No. Responses per Respondent	Average response time (in minutes)	Total Burden Hours
FY20	EBSA Inquirers	11,200	1	10	1,867
FY21	EBSA Inquirers	11,200	1	10	1,867
FY22	EBSA Inquirers	11,200	1	10	1,867
FY23	EBSA Inquirers	11,200	1	10	1,867
FY24	EBSA Inquirers	11,200	1	10	1,867

13. Total annual cost burden to respondents or record-keepers resulting from the collection

There are no start-up or annual operation and maintenance costs incurred by respondents.

14. Annualized cost to the Federal government estimates

Based on the cost for research contractor, contract oversight and IT contractor for support the annualized cost is $696,084.50.

15. Reasons for program changes or adjustments reported in Items 13 or 14 of the OMB 83-I

The survey’s demographic questions have been updated in response to Executive Order 13985—Advancing Racial Equity and Support for Underserved Communities Through the Federal Government. Each survey should take approximately an additional two minutes to complete. The cost to the government increased from $624,984.56 to $696,084.50.

16. Plans for publication

There are no plans to publish the findings from this study.

17. If seeking approval not to display the expiration date for OMB approval, explain reason(s)

EBSA is not requesting approval not to display the expiration date for OMB approval.

18. Exception to the certification statement identified in Item 19 of the OMB form 83-I

EBSA is not requesting an exception to the certification requirements.

2. SURVEYS AND EVALUATIONS EMPLOYING STATISTICAL METHODS

B.1. Describe (including a numerical estimate) the potential respondent universe and any sampling or other respondent selection methods to be used. Data on the number of entities (e.g., establishments, State and local government units, households, or persons) in the universe covered by the collection and in the corresponding sample are to be provided in tabular form for the universe as a whole and for each of the strata in the proposed sample. Indicate expected response rates for the collection as a whole. If the collection had been conducted previously, include the actual response rate achieved during the last collection.

The goal of this study is to evaluate EBSA’s Participant Assistance Program (PAP). The universe consists of participant inquiries (individuals who had contacted EBSA for assistance) handled by the 10 Regional offices (Atlanta, Boston, Chicago, Cincinnati, Dallas, Kansas City, Los Angeles, New York, Philadelphia, San Francisco) and the 3 District offices (Miami, Seattle and Washington DC). The total number of inquiries across all 13 offices is expected to be between 170,000 and 195,000 over the course of each fiscal year.

Every two weeks, EBSA will send data on all of the participant inquiries identified as newly “closed” during the previous two weeks. The newly closed cases consist of all inquiries with a determination or closure analysis status in the Technical Assistance Information System (TAIS). Closure types include Benefit Claim- Assistance, Benefit Claim-Valid Recovery, Benefit Claim-Not Valid, Benefit Claim-Enforcement Lead, Benefit Claim-Secondary Enforcement Lead, or Benefit Claim – Referral as Abandoned Plan. Gallup will select samples of participant inquiry records on a bi-weekly basis using a simple stratified sample design. Based on information from the past administration of this study conducted by Gallup for fiscal year 2014, we anticipate an average response rate of around 30.0% for each two-week period across all offices. However, it differs across offices ranging from 20.7% for New York to 34.7% for the Seattle office. In order to examine the issue of non-response bias, a non-response bias analysis will be conducted. The non-response bias analysis plan is described in Section B3.

For the purpose of sampling, the universe of all inquiries will be first stratified by the 13 offices listed above. Each of those offices will be further sub-stratified by BA (Benefit Advisor) level inquiries. To achieve reporting at the BA level for each full-time experienced BA (defined as having been taking calls for at least the full prior quarter and closed at least 100 cases each quarter), at least during the mid-year and year-end reporting cycles, a minimum of 30 completed interviews for the reporting period (so a minimum of 60 per year per BA) is planned. There will be roughly 100 BAs meeting the criteria for full time. Sampling, as noted above, will be done once for each two-week data collection period. For one full fiscal year, there will be 24 two-week data collection periods and each quarter will consist of six such periods.

Table 1 below presents the volume of inquiry (universe counts) for each office based on FY 2019 reports. Within an office, each of the sub-strata corresponds to a full time BA while the residual substratum ((non- full time and non-BAs) will include the rest of the inquiries. It may be noted that the stratification scheme will be based on this approach but the exact number of sub-strata in any particular quarter will depend on the actual number of inquiries received during the previous quarter at the office/BA level. At the beginning of a specific quarter, the determination of the sub-strata will be carried out based on the volume of inquiries received in the previous quarter.

Table 1: Number of Inquiries by Strata (Office) FY2019

Participant Assistance Program (PAP) Office	Total Inquiries in a fiscal year	Sample Size (Targeted Number of Completed Surveys in a fiscal year)
Atlanta	17,824	800
Boston	12,534	800
Chicago	10,085	800
Cincinnati	19,250	800
Dallas	18,681	800
Kansas City	13,824	800
Los Angeles	13,092	800
Miami	14,772	800
New York	12,700	800
Philadelphia	10,114	800
San Francisco	18,410	800
Seattle*		800
Washington D.C.	4,525	800
Total	165,811	10,400
[*The number of inquiries for Seattle office are combined with those for San Francisco office. At the time of sampling, it will be possible to identify inquiries separately for these two offices (San Francisco and Seattle). Samples will be drawn independently from each office and so these two offices will be treated separately for the purpose of sampling and reporting.]

For each office, as shown in Table 1 above, we plan to complete around 400 interviews for a six-month reporting period i.e. about 800 interviews over a fiscal year as part of the regular sample. That will require completing a total of 10,400 (=13 X 800) surveys for the entire year for all 13 offices. At the BA level, the average number of interviews per BA is likely to be around 100 for the entire fiscal year (based on a total sample of 10,400 interviews distributed across 100 BAs across all 13 offices). Gallup, however, understands that there is significant variation in the volume of inquiries handled by different BAs and differing numbers of BAs in each office. If sampling is done at the office level (and not at the BA level), a simple random sample will include proportional representation of cases (inquiries) from each BA within that office. An office-level simple random sampling strategy may therefore lead to significant variation in the number of inquiries sampled for BAs within that office. BAs dealing with relatively smaller number of inquiries will not be adequately represented in the sample. To address this issue, Gallup plans to stratify by office and then sub-stratify by full time BA within each office. There will be a residual sub-stratum containing all non-BA and non-full time BA staff in each office. The goal will be to achieve at least 30 interviews per BA for mid-year and year-end reporting (a similar approach to that used in previous administrations). The average expected number (100) of completed interviews per year at the BA level will produce a margin of error + 9.8% at 95% level of confidence. About 15 interviews for each BA will be targeted quarterly, with BAs with a proportionally higher level of closed cases having additional completed interviews. Inquiries handled by non-full time BAs or other non-BA staff (e.g. SBA and interns) will also be sampled (although at a lower sampling rate) from the residual sub-strata to ensure an accurate and full picture of the inquirer experience in each office. In general, the goal will be to draw a proportional sample across the BAs within an office and to ensure to the extent possible a minimum of 15 interviews for each full-time BA (sub-stratum) per quarter.

In addition to the ongoing interviewing, one office will be oversampled to better identify areas for improvement in struggling offices and to provide reports for each BA in the selected office (with 30 completed interviews each in each quarter). Based on this goal and looking at the likely number of interviews which will fall out naturally for BAs across the 13 offices, it may be necessary to conduct up to an additional 800 interviews per year for a specified office. The criterion for selection of offices to be oversampled will be based on performance (satisfaction rating) at the overall office level. The actual selection of the office will be made in consultation and with the approval of EBSA. Once the office to be oversampled is selected, the exact sample size for oversampling will be determined depending on the number and size (volume of inquiries) of BAs from that office. As mentioned before, it may be necessary to complete up to an additional 800 interviews as part of the oversample.

The population parameter of primary interest will be the proportion of customers in specific categories. For example, the “proportion of customers in the population who are satisfied with EBSA overall” or the “proportion of customers who think EBSA always delivers on what they promise” will have to be estimated based on survey data. The corresponding sample estimates will be computed based on responses to the survey. On a satisfaction question such as : “How satisfied are you with EBSA overall?”, the proportion of satisfied customers may be estimated based on the proportion selecting one of the top two boxes on a 5-point likert scale. Customers will also be asked to indicate their level of agreement with statements like “EBSA treats me like a valued customer” or “EBSA always delivers on what they promise” on a 5-point scale. The proportion of customers who select one of the top two boxes will provide an estimate of the corresponding population proportion. The sample based estimate (p) of the parameter representing an unknown population proportion (P) can be expressed as:

p = ,

where Yi = 1 if the i^th sampled respondent belongs to the category of interest (satisfied, for example) and 0 otherwise; W_i is the sample weight attached to the i^th respondent and ‘n’ is the number of completed surveys.

These parameters (proportions or means) may have to be estimated at the overall EBSA level, for each of the offices separately and possibly for other domains of interest within each stratum. For example, it may be of interest to generate similar estimates by Closure Types. The bulk of the calls (about 80 to 90% on an average) will belong to the first Closure type (those who need benefit claim assistance) and hence the number of completed surveys for this subgroup will be large enough to generate estimates of acceptable precision. Similar estimates for the other two Closure types can also be generated but the sample size within individual offices may be low and hence the estimates for these two Closure types may have to be generated at the overall EBSA level.

B.2. Describe the procedures for the collection of information including:

• Statistical methodology for stratification and sample selection – For the purpose of sampling, the universe of all inquiries will be first stratified by the 13 offices listed above. Each of those offices will be further sub-stratified by BA (Benefit Advisor) level inquiries. Sampling will be done once for each two-week data collection period. In total for the entire year, there will be 24 two-week data collection periods and each quarter will consist of six such periods. As mentioned before, inquiries handled by each full-time BA within an office will constitute one sub-stratum. All smaller BAs (not full time) within an office will be pulled together to form one combined sub-stratum. An office with 8 full time and 3 non-full time BAs will, for example, have 9 sub-strata – one for each full-time BA and an additional sub-stratum consisting of all three non-full time BAs. For any particular quarter, the determination of full-time or non-full time status for any BA will be determined based on the number of inquiries handled by that BA in the previous quarter. Sampling will be done independently within each sub-stratum (office by BA level) following a simple stratified sample design and the process will be repeated in each two-week data collection period. Within a sub-stratum, the sample size for each period will be based on the targeted number of completes and the expected response rate. In each period, the targeted number of completes will be derived by taking into account the requirement of a minimum number of 15 completes for full-time BAs and also the need for oversampling for selected offices. Once the sample sizes are determined, simple random samples of specified sizes will be drawn from the sub-strata once every two weeks.

• Estimation procedure – Sample data will be weighted to generate estimates for the target population subgroups. Within each sampling stratum (office by BA), weighting will be carried out to adjust for (i) probability of selection in the sample and (ii) non-response. Once the sampling weights are generated, weighted estimates will be produced for different unknown population parameters (means, proportions etc.) for the target population and also for population subgroups of interest. For the purpose of illustration, let us assume that we receive a total of 100 inquiries in a sub-stratum (office by BA) in a particular two-week period and we select a random sample of size 50 from those 100 inquiries. Also, assume that 15 of those 50 sampled cases actually respond i.e. we get 15 completed surveys from that sub-stratum. The weight assigned to each of those 15 completed surveys will consist of two weighting factors: (i) selection probability weight (100/50) and (ii) non-response weight (50/15). The first weighting factor is the inverse of the selection probability while the second factor is the ratio of the sample size and the number of completed surveys. The final weight will be the product of these two factors. In this specific example, the final weight assigned to each of those 15 completed cases will be (100/50) * (50/15) = 100/15. The sum of the weights of these 15 cases will add up to 100, the total number of inquiries for that two-week period for that particular sub-stratum.

Based on our previous experience in conducting this survey, we do expect some “ineligible” cases in the sample. The weighting procedure can be easily adjusted to account for ineligible cases. If, for example, 5 out of the 50 sampled cases turn out to be ineligible, the non-response weight factor will be equal to 45/15 and then the final weight assigned to each of the 15 completed surveys will be (100/50)*(45/15). The sum of the weights for all 15 cases will then equal 15*(100/50)*(45/15) = 90, the estimated number of eligible inquiries in that sub-stratum during that particular data collection period.

In terms of mathematical symbols, the weighting steps can be described as follows. Let and denote the population size (total number of inquiries received) and the corresponding sample size (number of inquiries sampled) for any particular sub-stratum for a specific two-week data collection period j (j=1,2,…,24). Also, let denote the number of responding units in the sample in i^th sub-stratum and j^th data collection period. Then, the base-weight or the probability weight factor ( ) assigned to k^th sampled unit (k=1, 2, …, ) will be derived as:

= N_ij/n_ij… (1)

At the next step, the non-response adjustment factor ( will be derived as:

= … (2)

if the k^th unit (k=1, 2, …, n_ij) is a responding unit and 0 otherwise;

e_ijm = 1 if the m^th unit in the sample is eligible and 0 otherwise; d_ijm = 1 if the m^th unit is eligible and responds to the survey and 0 otherwise.

In the right-hand side of equation (2) above, note that the summation in the numerator is over all sampled eligible cases whereas the summation in the denominator is over all selected eligible persons who actually respond to the survey.

The final weight (W_ijl) assigned to all r_ij responding units (in i^th sub-stratum in j^th data collection period) will be the product of the two weighting factors:

… (3) (l = 1, 2, …, r_ij).

Construction of non-response adjustment cells based on Closure types may be considered. However, the bulk of the calls (80 to 90%) are likely to belong to the first Closure type category (those who need benefit claim assistance) only and the number of calls belonging to the other types will be small. Collapsing of non-response adjustment cells, therefore, will be necessary. We anticipate using rules based on the (i) size of the cell and (ii) value of the non-response adjustment factor. For this study, given that bulk of the calls will belong to the first category only, we anticipate using the entire stratum as the non-response adjustment cell for most strata. In some strata, it may be possible to use two cells (those who need benefit claim assistance and Others).

• Degree of accuracy needed for the purpose described in the justification – For any particular sub-stratum (office by BA), the total number of completed interviews over the 24 data collection periods will depend on the number of full-time BAs in that office. However, a minimum of about 800 surveys is expected to be completed annually within each of the 13 offices. For offices with relatively higher number of BAs and also for those undergoing oversampling, the number of completed surveys will be more then 800. For estimation of any unknown population proportion (P), for example, this will result in a margin of error (MOE) of about 3.5 percent at the 95% level of significance (ignoring any design effect) for the yearly sample of size 800. For half-yearly sample size of about 400, the MOE at each office level will be about 5%. Following the most conservative approach, these numbers are estimated assuming the unknown population proportion to be around 50%. If that value is different from 50%, the actual MOE values are likely to be even smaller.

The margin of error (MOE) for estimating the unknown population proportion ‘P’ at the 95% confidence level can be derived based on the following formula:

MOE = 1.96 * where ‘n’ is the sample size (i.e. the number of completed surveys).

Under the most conservative assumption (P=0.5), the MOE for a half-yearly sample size of 400 will be 1.96* = 4.9%. After accounting for design effect, the average sampling error is likely to be around 5% for estimates at the individual office level. It may be noted that, for any given office, the allocation of sample across all 24 data collection periods is not strictly proportional because a fixed number of completed surveys will be targeted in each two-week period. However, for the same office, the total number of cases (inquiries) is not expected to vary significantly over time and hence its impact on design effect for estimates based on data for several data collection periods should be minimal.

At the overall agency level, the total sample size will be around 10,400 (plus any oversampling that may be carried out) for the entire year and hence the MOE is expected to be around 1.96* = 1.0% under the assumption of no design effect. For a half-year period, the MOE is likely to be around 1.4%. However, at the agency level, the disproportional sample allocation across different strata will contribute to the design effect. The design effect was defined formally by Kish (1965) , Section 8.2, p. 258) as “the ratio of the actual variance of a sample to the variance of a simple random sample of the same number of elements.” Based on Kish’s approximate formula

{design effect= (sample size)*(sum of squared weights)/ (square of the sum of weights)},

the design effect at the overall agency level based on past data is not expected to exceed 2.5. After accounting for design effect, the margin of error for estimates of population proportions at the overall agency level based on a sample size of 10,400 for the entire year will be about 1.5%. For a half-year period, the MOE will be about 2.1%.

The formulas for two-sample proportion test are:

[1-tailed test]

[2-tailed test]

In these expressions,

is the sample size required to achieve the desired statistical power;

are the normal abscissas that correspond to the respective probabilities;

are the two observed sample proportions in the two-sample test;

is the simple average of and and .

For comparison between two offices, for example, a difference of 8% in proportions (minimal detectable effect) can be detected based on this sample size with α=0.05 and 80 percent power ignoring design effect when the proportion is about 50% in one quarter and 42 percent, for example, in another. Based on the assumption of a design effect of 1.25, a difference of 9% in proportions (minimal detectable effect) can be detected based on this sample size with α=0.05 and 80 percent power ignoring design effect when the proportion is about 50% in one quarter and 41 percent, for example, in another.

• Unusual problems requiring specialized sampling procedures and – We don’t foresee any unusual problems requiring specialized sampling procedures.

• Any use of periodic (less frequently than annual) data collection cycles to reduce burden – For this study, Gallup will sample once every two weeks and the data collection for every bi-weekly sample will be completed within the following two weeks. The data collection will therefore be a continuous process throughout the year but every respondent will be contacted/interviewed within two weeks after his/her inquiry is closed. This will be done to minimize the recall error and thereby increase the overall accuracy of the survey data.

B.3. Describe methods to maximize response rates and to deal with issues of non‑response. The accuracy and reliability of information collected must be shown to be adequate for intended uses. For collections based on sampling, a special justification must be provided for any collection that will not yield "reliable" data that can be generalized to the universe studied.

Methods to maximize response rates – In order to maximize response rates, Gallup will utilize a comprehensive plan that focuses on (1) a call design that will ensure call attempts are made at different times of day and different days of the week to maximize contact rates, (2) conducting an extensive interviewer briefing prior to the field period that educates them about the content of the survey as well as how to handle reluctance and refusals, (3) having strong supervision that will ensure that high quality data are collected throughout the field period, and (4) utilizing troubleshooting teams to attack specific data collection problems that may occur during the field period. Gallup will use a 5+5 call design i.e. a maximum of five calls will be made on the phone number to reach the specific person that we are attempting to contact and another up to five calls will be made to complete the interview with that selected person.

Issues of Non-Response: Survey based estimates for this study will be weighted to minimize any potential bias including any bias that may be associated with unit level non-response. The non-response weighting procedure described on page 14 for the purpose of illustration assumes that information will be missing completely at random (MCAR). However, that is a strong assumption and may not be realistic. A more realistic assumption will be missing at random (MAR) that assumes that the sample can be partitioned into classes (or non-response adjustment cells) such that the missing observations from a class can be assumed to be a random subsample of all cases (respondents and non-respondents) coming from that class. Based on our findings from the non-response bias analysis, the goal will be to form such non-response adjustment cells. If found necessary, methods based on response propensity stratification may be used to form such non-response adjustment cells. The bi-weekly files (sampling frames) to be received from DOL will contain some useful information (like Closure types) for all cases including the non-respondents. Non-response adjustment cells may be formed based on these variable and then ratio type adjustments will be carried out to correct for non-response. This will make the non-response weighting procedure quite effective in terms of minimizing non-response bias, if any.

The data collection mode for this study is telephone and so the item non-response rate is expected to be less compared to data collected using self-administered modes. The respondent will however have the opportunity to refuse to answer a specific question. The item missing rate is not expected to be significant and so imputation of missing item level data was not used in the past and is not recommended for the 2020 study.

As described above in Section B2, the sampling error associated with estimates of proportions at the individual office level is expected to be around 5% and that for the overall agency level is likely to be around 1.0% at the 95% level of confidence. For any other subgroup of interest (based on closure types, for example) the sampling error will depend on the sample size. Also, all estimates will be weighted to reduce bias. It will be possible to calculate the sampling error associated with any subgroup estimate in order to ensure that the accuracy and reliability is adequate for intended uses of any such estimate

Non-response Bias Analysis: A non-response bias analysis will be conducted to identify potential source of non-response bias. Non-response bias associated with estimates consists of two factors - the amount of nonresponse and the difference in the estimate between the groups of respondents and non-respondents. The bias of an estimate can be expressed mathematically as follows:

Bias (y_r) = (1 – r) {E (y_r – y_n )}

where y_r is the estimated characteristic based on survey respondents only, ‘r’ is the response rate and so (1 – r) is the nonresponse rate, y_n is the estimated characteristic based on the non-respondents only, and E is the expectation for averaging over all possible samples.

Bias may therefore be caused by lower response rate and/or by significant difference in estimates between respondents and non-respondents. As described earlier in this section (B3), necessary steps will be taken to maximize response rates and thereby minimize any non-response bias that may be caused by low response rates. Also, non-response weighting adjustments (refer to “Issues of non-response” above) will be carried out to minimize potential non-response bias. However, despite all these attempts, non-response bias can still persist in estimates. The goal of the non-response bias analysis will be to identify potential sources of nonresponse bias on estimates and to identify potentially biased estimates.

The non-response bias analysis will compare the “Early” respondents to “Late” respondents on selected key variables of primary interest. The basic assumption in such an approach is that later respondents to a survey are more similar to non-respondents than are earlier respondents. In this study, data collection will be conducted using telephone and a respondent can receive anywhere between 1 and 10 calls to complete an interview. Respondents will be divided into two groups (Early and Late respondents) based on the number of calls received. The exact definition of these two groups will be finalized after examining the distribution of the ‘number of calls’ needed to complete an interview for this study. Comparison of estimates (proportions or means of selected key variables like proportion of satisfied customers, for example) between these two groups will be carried out by testing the hypothesis of equality of proportions (or means). This process will help identify estimates that may be subject to non-response bias.

The key variables (or survey questions) for the comparison of “Early” and “Late” respondents will include the following ten questions that are expected to be strong predictors of overall customer satisfaction. (Each of these ten questions uses a five-point scale, where 5 means strongly agree and 1 means strongly disagree and the respondent is asked to tell how much he or she agrees or disagrees with each statement as it applies to EBSA).

1. EBSA treats me like a valued customer

2. EBSA is willing to work with me to make sure my needs are met

3. EBSA acts in a timely fashion

4. EBSA does what it says it will do

5. EBSA is easy to reach

6. The information I receive from EBSA is clear and easy to understand

7. EBSA does its best to help me out

8. EBSA thoroughly answers all of my questions

9. EBSA is proactive in addressing my question or issue

10. If you had a need to work with EBSA again in the future, you would want to interact with this same benefits advisor

For each of these selected variables, the mean of the two groups (‘early’ and ’late’ respondents) will be compared based on a t- test using software SUDAAN so that the sample design and the resulting sample weights can be taken into consideration.

Let the mean (or equivalently the proportion of 1^s for a 0-1 variable) of ‘early’ and ‘late’ respondents for a specific variable (Y) based on survey data be denoted by p₁ and p₂ respectively. Then, p₁ can be written as

p₁ = ∑W_iy_i/∑W_i, where y_i is 1 if the value of variable Y for the i^th respondent is 1 and ‘0’ otherwise; W_i is the weight assigned to the i^th respondent and the summation in both numerator and denominator is over all ‘early’ respondents in the sample. p₂ can be similarly defined. The t-statistic for testing the equality of means for those two groups (H_o: P₁=P₂ vs. H₁:P₁ ≠ P₂ where P₁ and P₂ are the corresponding population means) will be computed as:

t=(p₁ – p₂)/SE (p₁ – p₂) , where SE (p₁ – p₂) is the standard error or the estimated square-root of the variance of (p₁ – p₂).

In order to obtain the value of t-statistic (and the corresponding significance level or p-value), the main SUDAAN commands using the DESCRIPT procedure will be as follows:

PROC DESCRIPT DATA=XXXX FILETYPE=SAS DESIGN=STRWR;

nest strata;

WEIGHT FINALWT;

class early_late;

var Y;

contrast early_late = (1 -1)/name = "early vs. late";

print nsum t_mean p_mean mean;

The variable strata (obtained by crossing the levels of regional offices, data collection periods and BAs) will represent all the strata. The WEIGHT statement specifies the final weight variable. The CLASS statement defines the independent variables as categorical and REFLEVEL specifies the reference level for each of these variables. The early_late variable will contain two distinct values (0-1 for example) to identify the two groups (‘early’ or ‘late’) for each case in the data set. The VAR statement will include the variables for which the mean has to be compared between the two groups. For each selected variable included in the VAR statement, the hypothesis of equality of means will be rejected (or not) based on the p-value (less than 0.05 or not).

Non-response bias analysis may also involve comparison of survey based estimates to known Population Values that may be available. For this study, variables like Subject entry code, Closure types and Days_to_close (of inquiry) will be available for all cases on the sampling frame. It is, therefore, possible to compute the actual value of population parameters that are functions of these variables and compare those with the corresponding sample based (weighted) estimates. For example, the following three population parameters may be used for this comparison.

• P_BCA: Proportion of Benefit Claim Assistance (BCA) calls (derived from the Closure Analysis variable on the sampling frame)

• M_Close: Mean of the variable Days to Close (derived from Days_to_close variable on the sampling frame)

• P_PBS: Proportion of calls falling under a certain subject entry code (for example, the proportion of calls under PBS (Pension Benefits, Social Security Notice). This will be derived from the Subject Entry Code variable on the sampling frame.

The corresponding sample-based weighted estimates for the three population parameters will be generated using the values of those variables from the completed surveys. The comparison will be carried out using a one-sample t-test based on a t-statistic = (p–P)/SE(p) where ”p” is the sample-based estimate of the corresponding population proportion (or mean) ”P,” and SE(p) is the estimated standard error of p. The main SUDAAN statements to be used for the computation of SE(p) for the variable “Days to Close”, for example, will be as follows:

PROC descript data = XXXX filetype = sas design = strwr;

NEST strata;

WEIGHT finalwt;

VAR days_to_close;

PRINT nsum wsum mean semean;

[NEST statement specifies the strata determined by office, BA and period (2-week data collection period) and FINALWT is the final weight variable. The VAR statement will include the variable (M_Close: Days to Close) for which the mean was to be compared between the two groups. The hypothesis of equality of means will be rejected (or not) based on the p-value (less than 0.05 or not).]

Once SE(p) is estimated using SUDAAN, the t-statistic will be calculated using the values of p and P, and the hypothesis of equality (p = P) will be rejected (or not) based on the observed significance level (less than 0.05 or not).

B.4. Describe any tests of procedures or methods to be undertaken. – Cognitive/pilot tests for minimizing burden and improving utility were conducted in the past and the survey questionnaire for this study has now been standardized.

B.5. Provide the name, affiliation (company, agency, or organization) and telephone number of individuals consulted on statistical aspects of the design and the name of the agency unit, contractor(s), grantee(s), or other person(s) who will actually collect and/or analyze the information for the agency.

The following individuals consulted on statistical aspects of the design and will also be primarily responsible for actually collecting and analyzing the data for the agency.

Name	Agency/Company/Organization	Number Telephone
Manas Chattopadhyay	Gallup Organization	202-715-3179
Camille Lloyd	Gallup Organization	202-715-3188

-25-