Supporting Statement Part B

Supporting Statement – Part B

U.S. Department of Commerce

U.S. Census Bureau

2015 Management and Organizational Practices Survey (MOPS)

OMB Control No. 0607-0963

B. Collection of Information Employing Statistical Methods

1. Description of Universe and Respondent Selection

The sample for the 2015 MOPS will consist of the approximately 50,000 establishments in the 2015 Annual Survey of Manufactures (ASM) mail-out sample. The mail-out sample for the ASM is redesigned at 5-year intervals beginning the second survey year subsequent to the Economic Census. For the 2014 ASM, a new probability sample was selected from a frame of approximately 101,000 manufacturing establishments in the 2012 Economic Census that have paid employees, are located in the United States, and are associated with multi-location companies or large single-establishment companies. On an annual basis, the ASM mail-out sample is supplemented with large, newly active single-establishment companies identified from a list provided by the Internal Revenue Service and new manufacturing establishments of multi-location companies identified from the Census Bureau’s Company Organization Survey.

In 2010, 78% of respondents returned a form, and the Census Bureau anticipates a similar return rate for the 2015 MOPS.

2. Procedures for Collecting Information

a. Description of Reporting Forms

We mail report forms to approximately 50,000 manufacturing establishments.

b. Sampling Methodology

The 2014 ASM sample was selected from a sampling frame that was constructed from the Manufacturing Sector of the 2012 Economic Census. The sample is supplemented annually to include new establishments in the Manufacturing Sector. This sample will be used through reference year 2018. Below is an overview of the sample design.

The 2014 ASM sampling frame was first partitioned into mail and nonmail strata. Within each of the 364 North American Industry Classification System (NAICS) industries, small and medium-sized single-location companies were identified and defined as the nonmail component. Establishments comprising the remaining portion, including all establishments of multi-location companies, were defined as the sampling frame.

On the sampling frame, establishments that met specified criteria were selected in the sample with certainty. An independent sample was selected within each of the 364 NAICS industries. This allows optimization of the probabilities of selection within each industry, which improves the representativeness and reliability of the survey estimates. Within each industry, each establishment was initially assigned multiple probabilities. These probabilities were based on the establishment’s relative importance within the industry that it is classified and the set of product classes that it produces, as well as the target reliability constraints defined by the survey manager. For sample selection purposes, the establishment’s maximum probability was used to ensure that target reliability constraints were satisfied.

c. Estimating Procedures

Weighted estimates will be produced in which an establishment’s tabulation weight will be based on the product of three component weights. The first component weight will be the sample weight for the 2015 ASM mail-out sample. The second component weight will be an industry-level unit nonresponse adjustment factor, which will be based on unweighted establishment counts from the 2015 MOPS and 2015 ASM mail-out sample. This adjustment factor will incorporate random noise for self-representing establishments to prevent data disclosures on the 2015 MOPS public-use micro data file. The third component weight will be an industry-level calibration factor, which will be based on weighted establishment counts from the 2015 MOPS and 2015 ASM mail-out sample.

3. Methods to Maximize Response

a. Follow-up Procedures

We follow up delinquent establishments with the following:

Due date reminder letter – Approximately 1 month after mailing

1^st follow-up letter – Approximately 2 weeks after due date

2^nd follow-up letter – Approximately 1 month after 1^st follow-up. Note – the same letter will be used for both the 1^st and 2^nd follow-ups.

2. Estimating for Missing Data

Unit nonresponse will be handled in estimation by adjusting weighted reported data. While no estimation for missing data is planned, this is subject to change, depending on the results of research using reported data from the 2010 and 2015 MOPS.

c. Reliability

The estimates developed from the survey are apt to differ somewhat from the results of a complete enumeration of the sampling frame conducted under the same conditions as the sample survey. We will provide estimates of the magnitude of the sampling errors -- the differences between the estimates obtained and the results theoretically obtained from a comparable enumeration of the sampling frame -- by the standard errors of the estimates that will be published with the corresponding estimates. Also, information will be provided on the 2015 MOPS public-use micro data file so that data users may calculate estimated measures of sampling variability for weighted estimates produced from the file.

4. Testing of Procedures

The Census Bureau conducted multiple rounds of cognitive testing on the 2015 MOPS collection instrument. First, the Census Bureau tested the form with 18 businesses in Detroit (MI), Houston (TX), and the Washington D.C. metro areas in order to refine the new 2015 MOPS Data and Decision Making and Uncertainty modules.

Further, The Census Bureau intends to conduct a second round of cognitive testing with 20 businesses in Boston (MA) and San Francisco (CA). The Census Bureau wants to review the new instructions that resulted from the first round of cognitive testing. While the Census Bureau does not foresee significant instrument or burden changes resulting from the second round of cognitive testing, the Census Bureau might observe a lack of clarity on a given question that could lead the Census Bureau to apply less weight to a response while the collected data is analyzed.

After the Census Bureau creates the initial electronic MOPS collection instrument, the Census Bureau will conduct Usability testing on the electronic instrument in early 2016 in order to ensure the electronic interface is effective and efficient.

5. Contacts for Statistical Aspects and Data Collection

Mr. Julius Smith, Jr., Assistant Division Chief for the Manufacturing, Mining, and Construction Sectors (Economy-Wide Statistics Division), serves as consultant on the collection, analysis, and the dissemination of data from the MOPS. He can be reached on (301) 763-7662.

Ms. Amy Newman-Smith, Methodology Director for Manufacturing, Investment, and Construction Programs (Economic Statistical Methods Division), serves as consultant on the statistical aspects of the MOPS. She can be reached on (301) 763-6595.

Attachments:

1. Census Bureau Offers First-Ever Large Scale Look at American Management Practices

2. Letter from George Washington University and Census Bureau Response

3. Report form

4. Respondent Letters and Instructions

5. Select Screenshots from Electronic Instrument

3