- Mar. 20, 2019
- Mar. 19, 2019
- Mar. 18, 2019
Home > Statistics > Outline of Statistics and Statistical Release Schedule > Notices of Changes and Corrections 2012 > Addition of Sample Enterprises and Financial Institutions to the Tankan
February 21, 2012
Bank of Japan
Research and Statistics Department
The Bank of Japan has decided that, from the March 2012 Tankan (Short-Term Economic Survey of Enterprises in Japan), 102 enterprises and 5 financial institutions will be added to the sample of the survey, and the total number of sample enterprises and financial institutions will be 10,894 and 202 respectively.
The statistical accuracy of the Tankan is examined once a year, to prevent a decline in the statistical accuracy due to a decrease in the number of the sample caused by bankruptcies, mergers, and other factors. New sample enterprises and financial institutions are added to the Tankan when the examination reveals that statistical accuracy has declined (see "Sample Design and Sample Maintenance of TANKAN," June 7, 2004, for details).
The Bank of Japan has decided to add 102 enterprises and 5 financial institutions to the Tankan (see the Appendix) after the recent examination of the statistical accuracy of the latest Tankan (December 2011). As a result, the total number of sample enterprises and financial institutions for the March 2012 Tankan will be 10,894 and 202 respectively.
Details of the examination are as follows.
Population enterprises are classified into 397 strata by industry, scale based on capital size, and number of regular employees to calculate the population estimates.
For each stratum subject to the sample design1, whether the distribution of the sample enterprises properly reflected that of the population enterprises was examined. The result showed that there was no stratum in which the distribution of the sample deviated from that of the population2. However, there were some strata for which the sample distribution indicated a high likelihood of deviation from the distribution of the population with a further decline in the number of sample enterprises. To maintain statistical accuracy, sample enterprises were added to these strata.
For all six main divisions by industry and scale (manufacturing/nonmanufacturing and large/medium-sized/small, see below), the error ratios of the population estimates3 measured by sales data remained within the target range of 3 percent for manufacturing and 5 percent for nonmanufacturing after the addition of 102 sample enterprises.
E-mail : firstname.lastname@example.org