Automated data pipeline (FlowKit) in Ghana

Partnership

MNO (Vodafone Ghana); Ghana Statistical Services (GSS); Flowminder Foundation.

Background

In response to the global drive for National Statistical Agencies to harness the use of big data systems, Ghana Statistical Service (GSS) collaborated with Vodafone Ghana and Flowminder Foundation to explore the use of anonymised and aggregated mobility data to produce official statistics. This partnership has been supported financially by Vodafone Foundation and the William and Flora Hewlett Foundation. As early as 2018, this partnership commenced the establishment of an automated CDR data pipeline through the installation of FlowKit software. Since then, the project has been working to strengthen the capacity of GSS staff in the analysis and use of CDR data. In the early days of the SARS-CoV-2 pandemic, the pre-existing collaboration framework and CDR data pipeline in Ghana enabled the project team to rapidly deliver mobility insights to support and inform the government's intervention strategies. Four days after the government's proclamation of mobility restrictions, the first mobility report was published, highlighting changes in mobility patterns based on the analysis of aggregated CDR data. A further report was produced the following month and additional analyses have been undertaken since (Li et al 2021).

In addition, the use of the CDR aggregates by other third-party data requesters has been made possible because of a pre-established legal and partnership framework, which included the setup of a Steering Committee led by the Statistician General with representation from GSS, Vodafone Ghana and other stakeholders including civil society representatives. Using criteria stipulated in advance, and agreed to by members of the Steering Committee, this body was engaged in providing approval for further sharing of aggregated CDR data to organisations such as the World Bank and London School of Hygiene and Tropical Medicine who are using the data in their disease transmission models.

Insights on this approach

  • Leveraging the use of pre-existing collaboration and data pipeline provided timely mobility insight for informed decision on intervention to support COVID-19 strategies
  • Data governance has been strengthened through the collective and transparent decision-making process engendered by the Steering Committee

Key steps taken for developing the institutional framework and analytical pipeline

  • Legal agreements and partnership frameworks established upfront
  • Installation of FlowKit software and subsequent ongoing troubleshooting to maintain the pipeline

Areas of improvement and challenges

  • Significant challenges have been faced in relation to establishing the necessary connection remotely to the API, which serves the aggregated CDR data, for GSS staff
  • There is need to provide further capacity strengthening to GSS on the use and analysis of the CDR aggregates

Links for further information



  • No labels