Confidentiality: Title 13 of the United States Code authorizes the Census Bureau to conduct censuses and surveys. Section 9 of the same Title requires that any information collected from the public under the authority of Title 13 be maintained as confidential. Section 214 of Title 13 and Sections 3559 and 3571 of Title 18 of the United States Code provide for the imposition of penalties of up to five years in prison and up to $250,000 in fines for wrongful disclosure of confidential census information. The Census Bureau's internal Disclosure Review Board sets the confidentiality rules for all data releases. A checklist approach is used to ensure that all potential risks to the confidentiality of the data are considered and addressed.
Disclosure Statement: A disclosure of data occurs when an individual can use published statistical information to identify either an individual or firm that has provided information under a pledge of confidentiality. Disclosure limitation is the process used to protect the confidentiality of the survey data provided by an individual or firm. Using disclosure limitation procedures, the Census Bureau modifies or removes the characteristics that put confidential information at risk for disclosure. Although it may appear that a table shows information about a specific individual or business, the Census Bureau has taken steps to disguise or suppress the original data while making sure the results are still useful. The techniques used by the Census Bureau to protect confidentiality in tabulations vary, depending on the type of data.
Sampling Frame: Companies, parts of companies (defined by Employer Identification Numbers, or EINs), and single-unit establishments (also defined by EINs) that are located in the United States, have paid employees, and are classified in wholesale trade as defined by the 2002 NAICS. This includes wholesalers that take title to the goods they sell such as jobbers, industrial distributors, exporters, importers, and Manufacturers' Sales Branches and Offices (MSBOs), as well as companies that do not take title of the goods they sell such as agents, merchandise or commodity brokers, commission merchants, and electronic business-to-business markets. The EIN is the identifier employer businesses use to report Social Security payroll withholdings to the Federal government. Read more [PDF] about the AWTS sampling frame.
Sample Design and Size: The sample for AWTS consists of three separate samples: (1) a sample of merchant wholesalers, excluding MSBOs, (2) a sample of MSBOs, and (3) a sample of wholesale electronic markets and agents and brokers. AWTS uses a stratified, one-stage design with primary strata defined by industry (e.g., Motor Vehicle and Motor Vehicle Parts, Furniture and Home Furnishings, Grocery, etc.). There are 59 primary strata: 42 from the merchant wholesale sample, 17 from the MSBOs sample, and 2 from the Agents and Brokers sample. The primary strata are substratified into 4, 7, 10, or 13 annual sales size strata. The largest sales size stratum within each industry stratum consists of companies, all of which are selected with certainty (sampling weight equal to one). The other strata are populated by EINs. Sample sizes are computed to meet multiple coefficient of variation constraints on estimated annual sales and end-of-year inventory totals. Constraints are specified at detailed industry levels and at broad levels up to the total wholesale level. Sampling weights range from 1 to 250. Units are selected independently between strata using simple random sampling without replacement within the size substrata. The sample consists of approximately 2,100 certainty companies and 5,900 EINs. Updates to the sample are made on a quarterly basis to account for new businesses, deaths, and other changes to the universe. Read more [PDF] about how the AWTS sample is stratified,selected, and maintained.
Data Collection: Data are collected by mail, fax, Internet, and telephone. Response is mandatory under the authority of an Act of Congress, Title 13, United States Code, Sections 182, 224, and 225. Firms in the AWTS sample are asked to report their data for the year just ending. Two years of data are requested in the year in which a new sample is introduced.
Data Items Requested: Data items requested include annual sales, e-commerce sales, number of establishments covered by the report, value of inventories, inventory by valuation, inventory outside of the United States, total purchases of products, total operating expenses and the ending date of the report period if the data provided are for a period other than the calendar year.
Estimation and Sampling Variance: Total estimates are computed using the Horvitz-Thompson estimator (i.e., as the sum of weighted data (reported or imputed) for all selected sampling units that meet the sample canvass and tabulation criteria). The weight for a given sampling unit is the reciprocal of its probability of selection into the AWTS sample. These estimates are input to a benchmarking procedure, as described below. Variances are estimated using the method of random groups and are used to determine if measured changes are statistically significant. Read more [PDF] about how the AWTS arrives at its estimates and the reliabilty of those estimates.
Response Rates: Economic surveys at the Census Bureau are required to compute two different types of response rates: a unit response rate and weighted item response rates. Read more [PDF] about AWTS response rates (including the 2011 rates).
Benchmarking: A new sample was introduced with the 2011 AWTS. The new sample is designed to produce estimates based on the 2007 North American Industry Classification System (NAICS). In order to maintain the time series for each industry, an operation was performed to link estimates from the prior and new samples. For the linking operation to occur, two years of data were collected (2010 and 2011) from units in the new sample. The linking is done so that the new sample estimates are implicitly benchmarked using results of the 2007 Economic Census.
Before performing the sample linking operation, historical corrections were made to the prior sample estimates for NAICS 42343 (computer and computer peripheral equipment and software merchant wholesalers) to reflect more accurate information about the type of operation of businesses in this industry which had previously been classified as manufacturers’ sales branches and offices. The estimates for NAICS 42343 and corresponding aggregate estimates have been restated back to 2002. Because of these corrections, the 2007 sales estimates for NAICS 42343 and associated aggregate estimates will no longer match the 2007 Economic Census results. Historical corrections were not made to data for any other industries.
Sales estimates from the new sample for reference year 2010 and subsequent years are linked to the prior sample estimates by multiplying the Horvitz-Thompson estimates from the new sample by a ratio. The ratio is calculated as follows:
The resulting sales estimates (call these “modified” sales estimates) are implicitly benchmarked to 2007 Economic Census results via this linking procedure. Note that no revisions are made to sales estimates prior to 2011, except for NAICS 42343 as described above. The following method is used to produce “modified” estimates for the following items: end-of-year inventories, total operating expenses, purchases, e-commerce and EDI. First the sales ratio described above is multiplied by the Horvitz-Thompson estimate for the given item for 2010 and subsequent years. Then the published estimates for 2004 through 2010 from the prior sample are input into the benchmarking program. Using this program, the estimates for 2005 through 2010 for each detailed industry are revised in a manner that:
A similar method is used for foreign inventories, only using the benchmarked estimate for 2005 from the prior sample as a constraint, because foreign inventories was not collected for 2004.
For Wholesale Electronic Markets and Agents and Brokers (NAICS 425), sales estimates (i.e., gross selling value plus sales on own account) from the new sample for 2010 and subsequent years are linked to the prior sample estimates by multiplying the Horvitz-Thompson estimates from the new sample by a ratio. The ratio is calculated as follows:
The following method is used to provide “modified” estimates for the following items: commissions, gross selling value, operating expenses and sales on own account. First the sales ratio described above is multiplied by the Horvitz-Thompson estimates for the given item for 2010 and subsequent years. Then the published estimates for 2004 through 2010 from the prior sample are input into the benchmarking program. Using this program, the estimates for 2005 through 2010 for each detailed industry are revised in a manner that:
To ensure additivity, estimates of gross selling value and sales on own account are raked to the modified sales estimate.
Modified estimates at aggregate industry levels are computed by summing the modified estimates for the appropriate detailed industries comprising the aggregates.