copy the linklink copied!Annex D. OECD methodology for constructing the OURdata Index

The OECD OURdata Index aims to measure government efforts in line with the three main stages of the data value chain. That is, the Index assesses governments regarding their progresses towards (1) higher data availability, (2) efficient data accessibility and (3) greater support for data reuse. In measuring these three different elements, the Index considers the availability of different formal requirements (either applicable for all ministries and agencies or just in some few agencies), implementation gaps and the presence of oversight mechanisms. Data used for the construction of the composite index derives from the OECD Open Government Data Survey 3.0. The survey is composed of 80 questions representing 170 data points (with some data points corresponding to sub-questions). The survey was designed to monitor the implementation of the International Open Data Charter (IODC) adopted in October 2015. The IODC corresponds to a comprehensive international instrument that provides a set of principles on open government data (OGD). The IODC comes to add to the OECD Recommendation for enhanced access and more effective use of Public Sector Information (PSI) (OECD, 2008b). Additionally, and of course, the survey and the composite index are also based on the expertise of the OECD in the field of OGD and on a roadmap developed in 2013 (Ubaldi, 2013).

The construction of the Index follows the guidelines from the OECD/EU Handbook on Constructing Composite Indicators (2008) that are necessary for meaningful construction of composite or synthetic indexes. This approach is in particular useful to address the common problems associated with composite indexes and is thus the most effective solution to summarise the discrete, qualitative information on key aspects of OGD. Thus, four main types of analyses were conducted with the data to ensure the highest standards in terms of reliability and validity of the indicators (OECD-EU, 2008):

• Correlation analysis

• Confirmatory Principal-Component Factor analysis

• Cronbach alpha testing (scale reliability coefficient)

• Sensitivity testing (Monte Carlo Simulation)

The OURdata Index is based on 140 number of data points (meaning that a total number of 30 data points were dropped). Each pillar of the Index (data availability, data accessibility, and government support for data reuse) has three sub-pillars. The score for each pillar corresponds to an unweighted simple average of each sub-pillar. Regarding the sub-pillar level, implicit weighting was avoided since three sub-pillars were systematically retained under each main pillar. Additionally, each sub-pillar has parameters (factors) identified via expert judgement and factor analysis. The score of each sub-pillar is computed as the unweighted simple average of each parameter. There are 9 parameters in Pillar 1, 8 parameters in Pillar 2 and 7 parameters in Pillar 3.

In the case of Latin American and Caribbean countries, the data collection was conducted from March 2019 to June 2019. Respondents were high-level government officials (in many cases the National Chief Information Officer).