8(A) Web Supporting Statement Part B 3-10-2011

B. Collections of Information Employing Statistical Methods

1. Describe (including a numerical estimate) the potential respondent universe and any sampling or other respondent selection method to be used. Data on the number of entities (e.g. establishments, State and local governmental 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 potential sampling universe is approximately 12,800 Participant firms. The firms are those that were certified in the 8(a) BD program between 1998 and 2009. Econometrica will randomly draw a sample of 1,000 Participant firms for this survey.¹ To plan for contingencies, such as firms that could not be contacted or were inappropriately classified, Econometrica will initially oversample. The universe includes Participant firms currently in the program, as well as firms previously in the program that are no longer participating because they have graduated. The sample will not include companies owned by Alaskan Native Corporations, Native Hawaiian Organizations, tribally owned concerns and Community Development Corporations, since the purpose of the 8(a) BD program for organizationally-owned firms is different than that for individually-owned firms. The goal of the program for individually-owned firms is to provide business development so that such firms can compete in the American economy while the purpose of the program for organizationally-owned firms is to provide business development so that the firms can give back to the communities of which they are a part. The sample also will not include firms currently in the program subject to investigations regarding their participation in and activities related to the 8(a) BD program since participation in such a survey might impede the investigations. The table below provides data on the number of entities covered by the collection for the universe as a whole and for each of the strata in the proposed sample.

Status of Participant Firms	Universe Covered by the Collection	Corresponding Random Sample	Oversample
Total number of Participant firms	12,811	1,000	300
Active	7,276
Completed the program	2,296
Terminated	1,964
Withdrawn	1,147
Graduated	36
Not classified	92

SBA estimates the overall response rate for this proposed Web survey will be at least 30 percent, based on Econometrica’s previous experience with surveys of this type on behalf of various Federal Government agencies.

2. Describe the procedures for the collection, including:

*Statistical methodology for stratification and sample selection;

* Estimation procedure;

* Degree of accuracy needed for the purpose described in the justification;

*Unusual problems requiring specialized sampling procedures; and

*Any use of periodic (less frequent than annual) data collection cycles to reduce burden.

The sample will be drawn randomly; each Participant firm in the potential sampling universe has an equal chance of being selected. SBA compiled a list of Participant firms in the 8(a) BD program certified between 1998 and 2009, including the case number, the year of entry into the program, and the name of each firm’s SBA District Office. The list contains approximately 12,811 Participant firms. A simple random sample will be selected without replacement, in which each individual firm can be selected only once. All firms in the universe have an equal chance of being selected for the sample. Simple random sampling will guarantee that the sample chosen is representative of the population, ensuring that the statistical conclusions will be valid. Assuming a population of 12,800, a sample of 1,000 firms and an estimated response rate of 30 percent, if 267 firms respond there is 90-percent confidence that the results would be between +/- 5 percent.

This is a one-time collection. The responses will be submitted via the Web, but Participant firms can download a PDF file to prepare responses prior to entering and submitting their responses to the survey. SBA has not previously conducted any surveys of this type. As noted in Section A, question number 4 above, there are some existing SBA data on Participant firms. However, to the extent that data is incomplete, outdated or otherwise unreliable, the Web survey will facilitate verification that the data are relevant, valid, and reliable.

3. Describe the methods used to maximize response rates and to deal with non-response. The accuracy and reliability of the information collected must be shown to be adequate for the intended uses. For collections based on sampling, a special justification must be provided if they will not yield “reliable” data that can be generalized to the universe studied.

The accuracy and reliability of the collected data depend upon response rate. To achieve a high response rate, Econometrica will implement the following procedures that are consistent with contemporary strategies in the field of survey research.

• Potential respondents will receive an e-mail invitation (see the proposed text of the e-mail invitation below), which will serve as the formal launch of the survey. The invitation provides a clear and direct explanation of the survey, including the reasons for the data collection, and urges the Participant firm to participate in the survey. The e-mail invitation also encourages response by explaining that the sampling process is random and responses will be kept confidential. Offering confidentiality will also contribute to the accuracy and truthfulness of the responses.

• The survey is respondent-friendly. It is divided into sections to provide visual relief from a continuous listing of questions. The content of the questions is designed to ask for types of information that Participant firms are familiar with and that relate directly to their operations. The terms used in the survey are widely known among Participant firm respondents and are defined in technical materials issued by SBA and the District Offices.

• The e-mail invitations are personalized. The invitations are addressed to specific Participant firm respondents identified by SBA staff, rather than to a generic recipient or respondent. An SBA official will sign the e-mail invitations.

• Econometrica will send follow-up personalized e-mails. The follow-up will include two e-mail reminders, as needed, to non-respondents, to prompt a response and, if necessary, a follow-up telephone call from Econometrica staff members as a final step to encourage response. The e-mail reminders will be e-mailed at 2-week intervals from the initial survey mailing to those who have not responded. The third follow-up is a telephone call from Econometrica to the respondent to encourage participation in the survey.

Although the sampling methodology yields a “representative” sample of the population as a whole, non-responses may make it likely that some sample units will remain unmeasured. However, Econometrica will measure response bias and use post-survey adjustment to reduce non-response error.

There may be unit non-response bias. Although every effort will be made to reduce the number of non-respondents in order to reduce the bias, Econometrica can evaluate unit non-response bias in several ways, including:

• Compare initial and late respondents on several key measurements. Sample statisticians theorize that those who respond after several attempts have some similarities to non-respondents. Differences would be considered an estimate of non-response bias.

• Compare the respondents and non-respondents on several dimensions. As noted in Section A, question number 4, SBA has information on the characteristics of the Participant firms, such as the type of firm (i.e., NAICS code), status of firm (i.e., whether a firm is in the program or graduated from the program), minority code, firm location, and year of participation. If the comparison results in clear differences, SBA can conclude that these differences indicate there may be non-response bias in the data.

It is not anticipated that item non-response bias will be an issue, because many precautions have been taken to reduce item non-response. Web surveys have the potential to help reduce item non-response. For example, the Web survey does not allow respondents to accidentally skip a question. Several techniques can be used to deal with item non-response, such as imputation and direct estimation. Econometrica will use statistical methods to interpolate missing values (e.g., those contracts excluded from the data collection effort). Econometrica expects few missing values. In some cases, there will be SBA data sources to impute values.

When reporting the results of this survey, Econometrica will include a sufficient caveat regarding the limitations of the survey’s methodological approach.

4. Describe any tests of procedures or methods to be undertaken. Testing is encouraged as an effective means of refining collections of information to minimize burden and improve utility. Tests must be approved if they call for answers to identical questions from 10 or more respondents. A proposed test or set of test may be submitted for approval separately or in combination with the main collection of information.

During OMB’s review, the Web survey instrument will be tested electronically by SBA staff members, Econometrica staff members, and nine Participant firm respondents who are not included in the random sample. The SBA will work with Econometrica to select nine firms from the universe of Participant firms not in the sample. The selection will be based on firms that the SBA considers best prepared or most appropriate to provide the data needed. Following implementation of changes, the instrument will be retested by the team of analysts, as well as by SBA program subject matter experts.

5. Provide the name and telephone number of individuals consulted on the 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 tables below provide the names, affiliation, and contact information for those consulted on the statistical aspects of the design and who will actually collect and/or analyze the information.

Individuals Consulted on Statistical Aspects/Data Collection & Analysis
Name	Affiliation	Contact Information
Dr. Fred Bellemore	Econometrica, Inc.	301-657-9883
Warren J. Prunella	Econometrica, Inc.	301-657-9883
Priscila J. Prunella	Econometrica, Inc.	301-657-9883
Alex Thackeray	Econometrica, Inc.	301-657-9883
Craig Davis	Econometrica, Inc.	301-657-9883
Steve McGonegal	Econometrica, Inc.	301-657-9883
Dr. Richard Hilton	Econometrica, Inc.	301-657-9883

SBA Staff who Advised on the Survey Design
Name	SBA	Contact Information
Darryl Hairston	Office of Business Development	202-205-5852
LeAnn Delaney	Office of Business Development	202-205-6731
Joan Elliston	SBA COTR	202-205-7190
Sheila D. Thomas	Office of Business Development	202-205-6613
LaTanya Wright	Office of Business Development	202-205-7020

1 SBA chose the 1998 through 2009 timeframe because the last major revision to the 8(a) BD program occurred in 1998. Two thousand nine was selected as the end point because any firm that entered the program in 2009 will have had at least one annual review available to assist in providing answers to the research questions posed in the survey.