The three-day week of 1974 and measurement error in the FES and NCDS data sets
暂无分享,去创建一个
In the first three months of 1974, a drastic energy policy on the part of the British government limited the workweek to three days. Workers responding to income surveys during this period may report a “three-day week” wage rather than their usual weekly (five-day) income. The National Child Development Study (NCDS) and the Family Expenditure Survey (FES) questionnaires do not make clear which figure the respondents should have given. This study uses responses during and after the three-day week policy period to determine the magnitude of the measurement error in these two national surveys. It appears that very few respondents gave an “incorrect” three-day figure. In the FES, the estimated fraction of misreports is no larger than 3.2%; in the NCDS, the best estimate is zero.
[1] R. Blundell,et al. Intergenerational Mobility in Britain , 1995 .
[2] Mark B. Stewart,et al. On Least Squares Estimation when the Dependent Variable is Grouped , 1982 .