Part B

Page 2 of 2

B. Collections of Information Employing Statistical Methods



1. Universe and Respondent Selection

Each month from the universe of all newly constructed buildings with five or more housing units the Census Bureau selects a sample for its Survey of Construction (SOC) data collection process. For the SOC, the United States is first divided into primary sampling units (PSUs), which are stratified based on population and building permits. The PSUs to be used for the survey are then randomly selected from each stratum. Next, a sample of geographic locations that issue permits is chosen within each of the selected PSUs. Finally, all newly constructed buildings with five or more units within sampled places as well as a sub sample of buildings with one to four units are selected for the SOC sample.

Each quarter, a random sample of newly completed buildings with five or more housing units in the SOC sample, that were reported as completed during that quarter, are selected as the Survey of Market Absorption of Apartments (SOMA) sample. If there are more than 1,000 buildings completed in a month, a sub sample of no more than 1,000 newly constructed buildings with five or more housing units is randomly selected for the SOMA survey.

2. Procedures for Collecting Information,

Information on the proportion of units rented or sold (absorbed) 3, 6, 9, and 12 months after completion is obtained for units in each of the buildings selected in a given quarter for each of the next four quarters. For example, each building in the sample which is completed in any given month is surveyed at the end of each of the next four quarters or until all the units in the building have been absorbed, which ever comes first. Questionnaires for the selected properties are prepared and provided to regular Census Bureau interviewers for information collection.

Unbiased quarterly estimates are formed by multiplying the counts for each building by its base weight (the inverse of its probability of selection) and then summing over all buildings. The final estimate is then obtained by multiplying the unbiased estimate by a ratio estimate factor. The desired degree of accuracy is that the estimated apartment absorption rate should have a 90% confidence interval of 5 percentage-points or less.

3. Methods to Maximize Response

The response rate over the past several years in which the SOMA survey has been conducted has been above 95 percent and is expected to remain at that level. Due to the high response rate, only normal data collection procedures are followed for non response cases, i.e., supervisory personnel contact reluctant respondents to obtain the data.

4. Testing of Procedures

Discussions have been held about changing the method of data collection from paper questionnaires to computer-assisted survey information collection (CASIC). If such a change occurs, testing of the new data collection method will take place.

5. Contacts for Statistical Aspects and Data Collection

The individuals to contact are:

Henry Tow

Housing and Health Surveys Branch

Field Division

Bureau of the Census

Phone: 301-763-4879

Howard A. Savage

Financial & Market Characteristics Branch

Housing and Household Economic Statistics Division

Bureau of the Census

Phone: 301-763-3199

Dennis J. Schwanz

Longitudinal Surveys Branch

Demographic Statistical Methods Division

Bureau of the Census

Phone: 301-763-1984 

OMB 83-I 10/95

(4266901) -- Created by Preferred User – on 9/12/07