It is normal practice for Quarterly Employment Survey estimates to be revised. All quarters back to the latest published Census of Employment (currently Q4 2015 onwards) are provisional and subject to revision in each subsequent first release. These revisions will occur primarily in light of (a) late responses to surveys and administrative sources, (b) where cross survey congruence checks provide more accurate data and (c) where there are revisions to seasonal adjustment factors which are re-estimated every quarter.
Very few statistical revisions arise as a result of ‘errors’ in the popular sense of the word. All estimates, by definition, are subject to statistical ‘error’ but in this context the word refers to the uncertainty inherent in any process or calculation that uses sampling, estimation or modelling. Most revisions reflect either the adoption of new statistical techniques, or the incorporation of new information, which allows the statistical error of previous statements to be reduced. Only rarely are there avoidable ‘errors’ such as human or system failures, and such mistakes are made quite clear when they do occur.
In addition the QES has begun publishing “Revisions Triangles”, these are summary tables which indicate the size and direction of revisions. The table presents a summary of the differences between the first estimates published and those published 3 years later for the same reference period. A statistical test has been applied to the average revision to find out if it is statistically significantly different from zero.