Supporting Statement-Part-B_07

SUPPORTING STATEMENT - PART B

2007 ECONOMIC CENSUS COVERING THE CONSTRUCTION SECTOR

Collections of Information Employing Statistical Methods

1. Universe and Respondent Selection

Approximately 762,500 establishments with payroll have construction activity as their primary type of business. These establishments are stratified into two groups: establishments of multi-establishment firms and single-establishment firms. The expected number of establishments for each of these strata and the corresponding expected sample for the 2007 construction sector are given in the table below. Approximately 130,000 establishments will receive the report forms and instruction sheets as shown in Attachments C and D respectively. Response rates shown reflect response rates from this information collection in 2002.

Stratum	Universe	Sample	Response Rate
Establishments of multi-establishment firms	12,500	12,500	.81
Single-establishment firms	750,000	117,500	.77
Total	762,500	130,000	.78

2. Procedures for Collecting Information

a. Mail Selection Procedures

The census will select establishments with paid employees for its mail canvass from a frame obtained from the U.S. Census Bureau’s Business Register. To be eligible for selection, an establishment will be required to satisfy the following conditions: (i) it must be classified in the construction industry sector; (ii) it must be an active operating establishment of a multi-establishment firm, or it must be a single-establishment firm with payroll; and (iii) it must be located in one of the 50 states or the District of Columbia. Mail selection procedures will distinguish the following groups of establishments:

(1) Establishments of Multi-Establishment Firms

Selection procedures will assign all active construction establishment firms to the mail component of the potential respondent universe. We estimate that the census mail canvass for the 2007 construction sector will include approximately 12,500 construction establishment firms in this category.

(2) Single-Establishment Firms With Payroll

We will use a stratified random sample of single-establishment firms. The population of eligible single-establishment firms will be partitioned into state by NAICS groupings. Within each group, we will further subdivide the population into a minimum of two strata based on the annualized payroll of the firms within these groupings. Independent random samples will be selected for each stratum. The total sample size for each state by NAICS cell will be determined based on the reliability constraint for the cell and any additional U.S. level reliability constraints for the industry. The impact of the multi-establishment firms in producing the estimates will be factored into determining the sample sizes for the single-establishment firms. We estimate that the census mail canvass for 2007 will include approximately 117,500 establishments in this category.

b. Estimation procedures

Within sampled strata, estimates are formed by multiplying the census data from each sampled case by the inverse of the probability of selection and summing for all sampled cases.

Data for non-responding cases are estimated using the response patterns of the responding cases.

For the strata of establishments without payroll, data from the Internal Revenue Service and Social Security Administration are reviewed and tabulated. The latter tabulations are very limited in scope. Data are shown only for number of establishments, proprietors and working partners, and business receipts for the major construction groups.

c. Required Accuracy

(1) Sampling Error

Due to the large size of the national sample, estimates at the national level are reliable and accomplish the purposes listed in Section 2. The degree of accuracy is a consideration only for NAICS estimates at the state level. In order to make these estimates more accurate, the probabilities of selecting small, single-unit establishments with payroll (those establishments not taken with certainty) are increased so more are selected.

(2) Nonsampling Error

The accuracy of all census data is influenced by nonsampling errors, such as those affecting coverage, administrative records, questionnaire design, reporting, processing, and tabulation. Although we make no direct measurement of nonsampling errors, we take precautionary steps in all phases of planning, report form development, data collection, processing, and tabulation to minimize their influence.

d. Problems Requiring Specialized Sampling Procedures

There are no known problems that will require specialized sampling procedures for the 2007 census.

e. Use of Periodic Data Collection to Reduce Burden

The census uses periodic (5-year) data collection, as required by Title 13 United States Code, Section 131.

3. Methods to Maximize Response

This information collection will maximize response through the following means: (i) mailing materials that emphasize the mandatory and confidential nature of census reports, as provided by Title 13 United States Code; (ii) effective census questionnaires and instructions that try to simplify reporting and minimize response burden; (iii) toll-free assistance for any business that has questions about completing its census report; and (iv) systematic mail follow-up for nonresponse, supplemented by telephone follow-up for selected firms. We expect to maintain or improve the level of response obtained by the 2002 census. This level of response will yield accuracy and reliability that are adequate for intended uses of economic census data.

4. Tests of Procedures or Methods

This information collection will use procedures that are based on a considerable body of experience with the economic census and surveys. Previous economic censuses also have been the subject of evaluation studies that have examined methodology, conceptual issues, and related statistical questions. As a result, the procedures used by the 2007 Economic Census are very well tested.

5. Contacts

Statistical methodology is developed under the direction of Paul L. Hsen; Assistant Division Chief for Research and Methodology; Manufacturing and Construction Division; U.S. Census Bureau, Washington, DC 20233 (301-763-4586).

The 2007 Economic Census covering the construction sector is conducted under the direction of Mendel D. Gayle; Assistant Division Chief for Census and Related Programs; Manufacturing and Construction Division; U.S. Census Bureau; Washington, DC 20233 (301-763-4586).

Data collection and analysis are managed by Mary Susan Bucci, Chief; Construction and Minerals Branch; Manufacturing and Construction Division; U.S. Census Bureau, Washington, DC 20233 (301-763-4680).

Attachments

A. Consultations with Trade Associations and Federal Agencies: Organizations Contacted

B. Changes in Form Content from 2002 to 2007

C. Draft of Construction Census Forms CC-23601, CC-23701, CC-23702, CC-23801,

CC-23802, CC-23803, and CC-23804

D. Construction Census Form Instructions

10