IOP quality and revisions

Details of quality and revisions for the Index of Production.

Results, particularly for the latest Index of Production (IOP) quarter, are provisional and subject to revision as more up to date information becomes available. It is normal practice for IOP estimates to be revised.

Revisions to IOP estimates

Revisions will occur primarily in light of:

  • late responses to surveys and administrative sources
  • where cross survey congruence checks provide more accurate data - for example, data can be revised back and finalised once comparisons have been made with the current year Annual Business Inquiry (ABI) or manufacturing sales and exports returns
  • 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.

The Index of Production was published for the first time on a Standard Industrial Classification 2007 (SIC07) basis in October 2011. This has an impact on the type of businesses classified as production sector.  For example, publishing activities move from the production to the services sector and sewerage and waste disposal moves from services to the production sector.  

Within the production sector there are some new groupings of businesses. The previous IOP data were collected on a SIC03 basis and it has been converted to approximate the SIC07 coverage.

Revisions triangles

Historically, the IOP bulletin had contained information on revisions triangles. These triangles presented a summary of the differences between the first estimates of growth published and those published three years later for the same reference period.

A statistical test was applied to the average revisions to growth to find out if it was statistically significantly different from zero. From Q2 2011 IOP has been published on a SIC07 basis, the coverage of the Index (and the broad industry groupings) is different between SIC03 and SIC07, therefore, it is not possible to compare the results provided in this bulletin with those published prior to and including Q1 2011.

The data for Q1 2011 and earlier have been converted to approximate the SIC07 coverage and are therefore not strictly revisions. As a result no revisions triangles are available. Further information on the construction of SIC07 revisions triangles will be announced in due course.