Abstract
Over the years, different policies and infrastructural interventions for sustainability have been implemented by various countries. Yet, despite these interventions, the economic growth that has translated to an improved wellbeing of citizens has been a challenge for the countries, which underlines the need for more research and strategies for investigating the effectiveness of the interventions. In this regard, there have been calls for integrating sustainability with economic planning as a potential solution. However, the dearth of research on examining sustainable development indicators (SDIs) together with the economic development (ED) adds more complexity to solution seeking processes. In order to address these challenges, this research sets out to evaluate the importance and impact of SDIs on ED. To this end, the study adopts the use of the random forest algorithm, which is a widely used machine learning technique, to model ED using 408 SDIs. In this regard, the study, conducted in the African context, initially identified the most important predictors and then, examined the direction of the relationships to determine if the impact of the identified indicators on ED was positive or negative. During these analyses, ED was proxied using the human development index, which is a more encompassing measure of the wellbeing of a society as it includes economic measures and social and health indices. The obtained findings revealed that some indicators (e.g. GNI, CO2 emission and mortality rate) are very related to the ED of the six examined countries while some indicators (e.g. forest area, school enrolment, external debt stock) were either important to some countries or remained entirely unimportant (e.g. Bird species threatened) to the ED of the examined countries. Accordingly, the findings highlight that, since countries have idiosyncratic characteristics, each country should first examine its unique environment, growth attainment and available resources before adopting or implementing any ED strategy. To do this, the research suggests that countries should recognise the varying influence of SDI on their Ed and then, use this model as the basis for their policy interventions towards ED. Thus, by using the data-driven approach, the research holds potential to help policymakers in countries improve their policy formulation and implementation strategies.