Data Scandal in the 2018 Labor Law Reform Debate in Japan

- A Review of Literature on a Series of Governmental Surveys on Working Hours
TANAKA Sigeto <http://tsigeto.info>
(Graduate School of Arts and Letters, Tohoku University)
Annual Reports of Graduate School of Arts and Letters, Tohoku University. 68:68-30 (2019-03-07)

[Published online version] [Journal site]

[Manuscript PDF] [Manuscript PDF on OSF] [Manuscript PDF on Academia.edu]

[Communication] [Links]



URI: http://tsigeto.info/19a
OSF project: https://osf.io/phu6x/

Abstract

This paper reports the findings of a review of literature on a series of surveys on working hours in Japanese workplaces, conducted by the Ministry of Health, Labor and Welfare. The surveys have attracted public attention since Prime Minister Abe Shinzo deployed flawed data from the latest survey, conducted in 2013, to build up his argument for a labor reform bill in the Diet in January 2018. This scandal led to a partial retraction of the bill. An inspection team of the Ministry subsequently explored how and why flawed data were generated by the survey. However, the inspection team's report was inadequate: It only focused on how the 2013 survey carelessly replicated previous surveys. It left unresolved questions of how the past surveys were conducted, and how they affected labor policies. I reviewed primary and secondary literature on the series of surveys, which started from 1986 and deeply impacted the work of the deliberative council on labor policy, to uncover defects in the structuring of questionnaire s, sampling schemes, control of inspectors, statistical analysis, and result reporting. I observed symptoms of imprecise and incomprehensible explanations of various elements of the surveys by the Ministry in these 30 years, with no criticism leveled at the surveys before the scandal in the Diet in 2018.

[Contents in the Japanese page]

[Full-text files]


Acknowledgement

This work was supported by JSPS KAKENHI Grant Number JP17K02069.


Additional information

After finished this paper, I found the following article:

This article in 1986 included a table of the sample composition by industry and the size of workplace. I wrote in my paper that the sample composition was known only for the 1990 and 2005 survey, but it was also opened for the 1986 survey.


Communication

Questions/comments are welcome.


Related sites and pages

[Published online version] [Manuscript PDF] [Manuscript PDF on OSF] [Manuscript PDF on Academia.edu]


Tohoku Univ / School of Arts and Letters / TANAKA Sigeto


History of this page:


Generated 2024-03-11 12:01 +0900 with Plain2.

Copyright (c) 2019,2020 TANAKA Sigeto