5342.0 - Balance of Payments Statistics, Information Paper on Quality , 1996  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 20/02/1996   
   Page tools: Print Print Page Print all pages in this productPrint All  
Contents >> Chapter 1. Data quality concepts >> Accuracy and reliability

1.11. In considering the quality of balance of payments statistics, it is useful to make a clear distinction between the concepts of accuracy and reliability.

1.12. They can be defined as follows:

      • accuracy is the proximity of an estimate (especially the final, or fully-revised, estimate) to some notional true value; and
      • reliability is the proximity of initial and intermediate estimates for a particular period to the final estimate for that period.
    1.13. Thus, in the extreme, an estimate may be completely inaccurate but entirely reliable. That is, it bears no resemblance to the true value but, because of the absence of better information, the initial estimate is never revised. The opposite extreme case may also occur. That is, it may be possible to produce highly accurate estimates at long time lags while early estimates are poor and subject to considerable revision.

    1.14. In practice accuracy and reliability tend to be interwoven and reinforcing. Ideally, as the estimate for a particular period passes through a sequence of revisions, the size of revisions gets smaller (the statistic becomes more reliable) and the estimate moves closer to the true value (the statistic becomes more accurate). In balance of payments statistics this is generally the case, but not always.

    Previous PageNext Page