The content on this page and other DCITA document archive pages is provided to assist research and may contain references to activities or policies that have no current application. See the full archive disclaimer.
6. Regression-based decomposition analysis
In order to infer the relative importance of various factors in driving productivity growth, a number of regressions were run using the explanatory variables discussed in chapter 5. This is a fairly technical approach to assess the impact of different factors, but apart from comparing heavy and light ICT user industries (section 5.9), it is difficult to find a simple quantitative method to tackle this complex issue.
In order to make it easier for readers with little or no knowledge of regression theory to follow the discussion, the analysis is segmented into three parts. Section 6.1 presents some regression results in fairly non-technical terms, in order to familiarise all readers with the type of factor decomposition methodology used. Section 6.2 presents further details about the regressions and methodology. The presentation is moderately technical and readers who are not familiar with regressions may wish to move on to section 6.3. Section 6.3 presents summary and conclusions from the decomposition analysis in non-technical terms.
Back to 5.9 Heavy and light ICT users | Table of contents | Forward to 6.1 Some regression results