TECHNICAL NOTE DATA QUALITY
1 Non-sampling errors may arise as a result of errors in the reporting, recording or processing of data. These errors can be introduced through inadequacies in the questionnaire, treatment of non-response, inaccurate reporting by providers, errors in the application of survey procedures, incorrect recording of answers and errors in data capture and processing.
2 The extent to which non-sampling error affects the results are difficult to measure. Every effort is made to minimise non-sampling error by careful design and testing of the questionnaire, the use of efficient operating procedures and systems, and the use of appropriate methodologies.
Reliability of Statistics
3 When interpreting the statistics in this publication, the reliability of the estimates may be affected by the following specific non-sampling errors:
Treatment of non-response
4 Non-responding businesses which reported R&D activity in the previous cycle, had data imputed based on the 2003-04 expected R&D expenditure provided in 2002-03..
- Many businesses provided estimates due to a lack of separately recorded data on R&D activity.
- The OECD standard definition of R&D used in this survey may differ from what businesses regard as R&D activity. This is mainly due to slight differences, from the international standard, in the R&D definition used within the grants for industry R&D schemes (for the allocation of grants) and the AusIndustry administered R&D Tax Concession scheme (for tax deductibility for specific R&D activities). It is possible that some businesses have not reported R&D data as per the OECD definition.
- Data were subjectively classified, by businesses, to research fields and socio-economic objectives at the time of reporting. Some businesses may have experienced difficulty in classifying their R&D programs. The ABS makes every effort to ensure correct and consistent interpretation and reporting of these data by applying consistent processing methodologies.
5 Non-responding businesses which have not previously reported R&D activity, have been assumed to have no R&D activity and do not contribute to estimates.
6 Revisions to previous cycle data occur on discovery of: errors in previously reported data; and newly identified R&D performers in the previous cycle(s). Where newly identified R&D performers indicate that significant levels of R&D have been undertaken in earlier years, details are collected and used to revise previously released estimates.
7 Revisions only occur for significant changes to the data where the impact from individual businesses is of the order of $5m at any classification level, e.g. industry, employment size group, socio-economic objective, research field. Revisions are only applied up to two cycles prior to the current cycle.