ABS undermines strong labour data


From the ABS on today’s Labour Force figures (h/t The Lorax):

The incoming rotation group for February 2014 had a higher proportion of employed persons and persons in the labour force (i.e. less persons not in the labour force) than the sample it replaced. This incoming rotation group contributed, in original terms, 37% of the increase in total employment and 29% of the decrease in persons not in the labour force in February 2014. The trend estimates provide a better measure of the underlying level and direction of the series especially when there are significant rotation group effects.


6 Responses to “ “ABS undermines strong labour data”

  1. The Lorax says:

    One should always read the sampling error note as well:


    The estimates in this publication are based on a sample survey. Published estimates and the movements derived from them are subject to sampling variability. Standard errors give a measure of sampling variability (see Standard Errors section). The interval bounded by two standard errors is the 95% confidence interval, which provides a way of looking at the variability inherent in estimates. There is a 95% chance that the true value of the estimate lies within that interval.

    At the 95% Confidence interval the change in total employment was anything from -9,900 to +104,500.

    So yeah, if a public figure is drawing conclusions from a single seasonally adjusted number, consider again whether you should be taking this person seriously.

  2. Wing Nut says:

    Never let the fine print get in the way of a good headline.

  3. Lef-tee says:

    Yup. I would have thought that such an explosive gain out of the blue in the absence of data supporting the notion that the economy was suddenly booming would have raised some suspicion.

  4. ronfire says:

    No kidding

  5. Mining Bogan says:

    Does this mean I wasted my day in going off to play golf?