Part B. Collection of Information Employing Statistical Methods
1. Universe and Respondent Selection
Approximately 4,500 firms, representing about 3.6 million retail and food service establishments, are included in this voluntary survey. These firms are selected as a subsample of the Census Bureau’s larger monthly sample of approximately 13,000 firms, used to derive preliminary and final estimates of retail sales for more detailed kind-of business categories. The following chart provides response rates. Dollar volume response represents the percent of total sales accounted for by response data. The unit response is represented by both the total number of cases providing response data as a percent of the total sample and the total number of cases providing response data as a percent of the total mailed.
Data Month Dollar Volume Unit Response Unit Response
(% of Total Sales) (% of Total (% of Total
Sample) Mailed)
April ‘06 60.0 % 58.0 % 76.4 %
January ‘06 63.7 % 60.6 % 79.0 %
October ‘05 63.4 % 61.4 % 78.9 %
July ‘05 59.6 % 59.2 % 75.1 %
2. Procedures for Collecting Information
The advance sample is selected by stratifying the companies in the larger monthly sample by dollar size of sales within the respective major kind-of-business group and randomly selecting on a probability basis, the desired number of cases in each stratum. As a result, small as well as large retail firms are represented in the sample. Some firms, because of their relatively large effect on the sales for the kind-of-business category in which they are classified, are selected with a weight of one, to represent only themselves, in order to reduce the variance associated with the estimates. Weights are assigned to each selected firm to produce the appropriate representation in the estimation procedure.
The sales estimates are developed by applying a ratio of current month to previous month sales (from the advance sample) to the preliminary estimate for the previous month (from the larger sample) by kind of business and summing to obtain the summary estimates.
The median standard error for estimates of month-to-month change for Total Retail Trade and Food Services for May 2006 is 0.8 percent.
There are no unusual problems requiring specialized sampling procedures.
On a monthly basis, report forms are mailed to respondents five working days before the end of the month. For respondents who have a fax number listed, a form is sent via fax to them on the last work day of the month (other than on Fridays). The sales estimates are collected by the National Processing Center in Jeffersonville, Indiana by the end of the sixth working day following the data month. The data are tabulated, edited, analyzed, and reviewed on the sixth, seventh, eighth, and sometimes ninth working days.
Approximately every two and one-half years, the sample used for this survey is revised. Births are not added to this sample and, therefore, as firms go out of business, refuse to respond, etc., the sample deteriorates and is less representative. By re-selecting the sample, it better represents current business conditions and, in addition, small-size and medium-size firms are relieved of the reporting burden. We are currently in the process of revising our sample.
3. Methods to Maximize Response
A laser printer facsimile machine connected to an “800" telephone line permits facsimile reporting to our collection facility on a 24-hour basis. The Census Bureau also provides an “800" toll-free telephone number to permit respondents to call in data or ask questions without any extra expense. The National Processing Center in Jeffersonville, Indiana performs the telephone follow-up for all firms that have not responded by the due date, and those firms that have not completely filled out the report forms or have reported questionable data. Additionally, firms that refuse to respond to the survey are called or visited in an attempt to convey to them the importance of the survey.
We have instituted an earlier beginning of our telephone follow-up of nonresponse cases. The National Processing Center calls these cases beginning the afternoon of the second work day of the month, rather than the morning of the third. This became necessary because the larger sample size meant a larger number of nonresponses to followup.
Special telephone follow-up is conducted each month for a limited number of respondents to get an extra week of sales data from companies that were excluded from the totals because the sales ending dates were unacceptable. Unacceptable ending dates are those that fall too early or too late in the month and, therefore, don’t represent the calendar month.
4. Testing of Procedures
We continuously edit the reported data and monitor procedures and methods for data collection in an effort to reduce reporting burden and improve data quality.
5. Contacts for Statistical Aspect of Data Collection
Questions regarding the sample design and statistical methodology used for this survey should be directed to Ruth Detlefsen, Assistant Division Chief for Research and Methodology, Service Sector Statistics Division, (301) 763-5171. Planning and implementation of this survey are under the direction of Scott Scheleur, Chief, Retail Indicators Branch, (301) 763-2713.
Attachments
1. Copy of Form SM-44(06)A
2. Copy of Form SM-44(06)AE
3. Copy of Form SM-44(06)AS
4. Copy of Form SM-72(06)A
5. Copy of Form SM-44(00)FA
6. Copy of Form SM-44(00)FAE
7. Copy of Form SM-44(00)FAS
8. Copy of Form SM-72(00)FA
9. Copy of Introductory Letter MARTS-L1
10. Copy of Introductory Letter MARTS-L2
11. Copy of Letter of Support from BEA
| File Type | application/octet-stream |
| File Modified | 0000-00-00 |
| File Created | 0000-00-00 |