Part B MBDA Global Commerce Survey 052610

SUPPORTING STATEMENT
U.S. Department of Commerce
Minority Business Development Agency
Survey of Minority-Owned Business Participation in Global Commerce
OMB Control No. 0640-XXXX

B. COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL
METHODS
(If this collection does not employ statistical methodology, state this information. Delete
the following questions.)
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 and the corresponding sample are to be provided in tabular
form. The tabulation must also include expected response rates for the collection as
a whole. If the collection has been conducted before, provide the actual response
rate achieved.
The universe of firms includes 16 million firms from the Dunn & Bradstreet (D&B)
business database, of which approximately 500,000 are Minority Business Enterprises
(MBEs). From that universe a sample was generated randomly stratifying for minority
businesses and for company size. The sample includes 200,000 MBE and non-MBE firms
in the Dunn & Bradstreet (D&B) business database. D&B constructed their database by
using customers who apply for a DUNS number, through their own due diligence
searching through court and state filings, and through the business registry of new
businesses from all 50 states. As a result, the D&B database contains the most complete
source of observations. The 200,000 firms will be split between 60,000 MBEs (4,000 of
which are exporters) and 140,000 non-MBEs (35,000 of which are exporters). The D&B
database provides a marker for minority business as well as exporting vs. non-exporting
firms.
To ensure that the sample is representative of the population of all U.S. businesses,
Minimax will run descriptive statistics and frequency analyses on the responses and
based on these results, Minimax will check if the features of the sample (i.e., in terms of
receipts, SIC codes, employer vs. non-employer status, MBE vs. non-MBE, etc.) match
those of the known universe of firms. The population distribution of those features will
be obtained from publicly available census data.
If the distribution of the sample data does not match that of the population, poststratification weights will be applied to the data. Since more than one variable will be
taken into account for the stratification (as mentioned in the previous paragraph), a raking
or iterative proportional fitting procedure will be used to create the weights. These
procedures assign a weight to each of the observations in a manner that the frequency
distribution of the relevant variables in the weighted sample is equal to the frequency
1

distribution of the population. In this way, a larger weight will be given to strata that
might happen to be under-represented in the sample, so that the weighted sample data
will match the distribution of the population.

2. Describe the procedures for the collection, including: the statistical methodology
for stratification and sample selection; the estimation procedure; the degree of
accuracy needed for the purpose described in the justification; any unusual
problems requiring specialized sampling procedures; and any use of periodic (less
frequent than annual) data collection cycles to reduce burden.
Minimax will use the D&B database to contact the sample of MBE and non-MBE firms.
Minimax does not expect that any specialized sampling procedures will be needed, and
the survey will be conducted only one time (so there will be no periodic data collection
cycles to take into account). The survey will be a web-based instrument and the
participants will be contacted to notify them about the survey prior to sending the
participation package. Follow ups will be conducted immediately. If the contact
information in the D&B database is outdated for some of the potential participants so that
they cannot be contacted, those participants will be assumed to be non-responses.
The target degree of accuracy required to address the objectives of this paper will be such
that the bounds of 95% confidence intervals computed around the sample means are no
greater than 110% or 90% of the sample mean. For nominal/count data, Minimax will
target a 5% margin of error at 95% confidence. It is expected that even with a 1-2%
response rate, the final sample will be large enough to attain an adequate margin of error.

3. Describe the methods used to maximize response rates and to deal with
nonresponse. 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.
It is significant to note the extreme unlikelihood of not gathering enough data, and that
the projected reliability figures are well in excess of 95%, compared to the standard
reliability cutoff of 70%. Non-responses will not be included in the final analysis and the
response rate will be presented. Moreover, as described in question B1, frequency
analyses will be conducted to determine whether the distribution of certain variables in
the dataset (such as receipts, SIC codes, MBE vs. non-MBE, etc.) matches the known
distribution from the population of U.S. firms. If they do not match (for example, because
the response rate in some stratum was significantly higher or lower than in another one),
weights will be applied to correct for strata that might be under- or over-represented in
the sample. The variables that will be taken in account to weight the observations are:
MBE vs. non-MBE, receipts, employer vs. non-employer, 2-digit NAICS code and
exporter vs. non-exporter status.

2

4. Describe any tests of procedures or methods to be undertaken. Tests are
encouraged as effective means to refine collections, but if ten or more test
respondents are involved OMB must give prior approval.
No pilot testing will be conducted. The data collection procedure will be straightforward,
and the survey items will be concise and objective, so we will not require feedback from
test respondents to adjust the survey.

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.
MBDA is overseeing the development and administration of the survey for the
production of the research study. The contact at MBDA is Ivonne Cunarro, Chief
Knowledge Officer, 202-482-2157. The contractor, Minimax Consulting, will be
collecting the data. Minimax Consulting Clark Guo, Director of Government Contracts
can be reached at 401-331-6360.

3