8158.0.55.003 - Research Paper: Experimental Estimates of a Multi-Year Innovation Rate, Australia, 2017  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 26/10/2017  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All


The Organisation for Economic Co-operation and Development (OECD) publishes innovation indicators for a number of member countries on a biennial basis, offering opportunity to compare rates of innovation across nations. In the 2015 Innovation Indicators publication, Australia had the 5th highest proportion of innovation-active businesses of the countries included (OECD, 2015). However, there are some methodological differences (including survey scope, design and reference period) that obfuscate direct comparison of results between some countries.

This research paper outlines key differences in survey scope and methodology between Australia’s Business Characteristics Survey (BCS) and Europe’s Community Innovation Survey (CIS) to enhance understanding of how Australian innovation rates compare internationally.

An experimental methodology has been developed by the Australian Bureau of Statistics (ABS) that uses BCS innovation data collected for a single reference year to estimate of the proportion of Australian businesses that were innovation-active over a three year reference period. This is referred to as the Multi-Year Innovation Rate (MYIR). The MYIR methodology is then applied to a sample scoped to match business industry and employment classifications used by the CIS. Applying this experimental methodology aligns the BCS survey methodology as closely as possible to the survey methodology used in the CIS, although some remaining differences are not controlled for.

The experimental estimates presented in this paper indicate that the proportion of innovation-active businesses in Australia is greater when both industry and employment scope and reference periods are aligned with those used by the CIS. These results indicate that the ranking of 5th highest proportion of innovation-active businesses of the countries included in the OECD’s 2015 Innovation Indicators publication would be higher when these methodological differences are addressed using the experimental methods.

Despite a level of uncertainty in the experimental estimates due to unaddressed methodological differences and provider recall, this research adds to the body of knowledge related to innovation survey methodologies and their impacts on estimates of business innovation. The methodology in this paper enhances understanding of how Australian innovation rates compare internationally, without making changes to the current innovation survey methodology used.