The use of indicators to assess innovation development: A panel data analysis

This research brought some important contributions to the field of innovation studies in terms of use of indicators to assess innovation performance over time. Through the three panel regression analyses accomplished here, some important findings were reached, which deserve to be highlighted in this final part of the article.

Even though the innovation outputs had the best predictive power in the first regression, iinnovation is not something that appears in isolation; there is a lot of institutional support, human capital, and adequate infrastructure that enable its development. However, when evaluating only the variance of innovation over the years, the outputs are considered the best variables to explain such variability. In this way, increases in creative and knowledge and technology outputs will lead to an increase in the innovation score of the GII; lower scores of creative and knowledge outputs, in turn, will imply in a decrease in the overall innovation score.

In addition, another important finding was the importance given to business sophistication in both regression analyses that had the innovation outputs as dependent variables. This may be because the variables covered by the business sophistication pillar are the ones that encompass indicators related to intangible assets, especially the sub-pillar innovation linkages, which involves university-industry R&D collaboration, state of cluster development, gross expenditure on R&D financed from abroad, joint venture and strategic alliances, and patent families. This is also supported by literature, as seen in the previous section, which establishes a positive and significant relationship between innovation outputs and collaboration.

Moreover, this research supports that innovation has a direct impact on the sustainability of a country’s economic growth, which is why it is crucial to adopt strong policies that encourage more research and development.

This article presents a few limitations. This study only comprehended the years between 2013 and 2021. Future studies can look at a longer span, thus enabling the identification of other patterns in the data. Another limitation is the fact that this study focused only on the GII, thus neglecting other possible indicators at the country-level that could be considered in order to determine and assess innovation performance.