close
Share with your friends

2019 Change Readiness Index Methodology

2019 Change Readiness Index Methodology

The 2019 CRI combines data from 25 primary survey questions, gathered from 1,400 specialists, with a secondary dataset of more than 125 variables.

1000

Related content

Farmer sowing seeds on land

The Change Readiness Index (CRI) covers 140 countries including the 134 covered in the 2017 CRI plus a combination of additional developed and developing nations.

Country selection

The countries are distributed across a range of income levels. Countries included in this index were selected based on our ability to obtain sufficient or comparable primary and secondary data. The selection of countries provides a useful comparison between the change readiness of advanced and developing nations. Due to poor data availability, previously included countries Syria and Venezuela were not included in the 2019 CRI.

Country income group Economies according to GNI per capita Number of countries (140) Share in index sample (%)
High >US$12,055 40 28.6
Upper middle US$3,896 - $12,054 37 26.4
Lower middle US$996 - $3,895 37 26.4
Low <US$995 26 18.6

Scoring methodology

The 2019 CRI is structured around three pillars (enterprise capability, government capability and people and civil society capability), with subindices for each pillar, and primary survey question responses and secondary data variables feeding each subindex score. The composite/overall change readiness score is comprised of equally weighted pillar scores, which are derived from equally weighted standardized subindex scores. Subindex scores are derived from standardized primary survey question responses and secondary data, with equal weighting given per variable, whether it is a primary survey question or secondary data indicator.

Enterprise capability Government capability People & civil society capability
Relates to broad capability of private and state-owned enterprises Relates to capability of governmental and public regulatory institutions Relates to individual, societal and cultural determinants of capability
Labor markets Macroeconomic framework Human capital
Economic diversification Public administration and state business relations Entrepreneurship
Economic openness Regulation Civil society
Innovation and R&D Fiscal and budgeting Safety nets
Business environment Rule of law Technology use
Financial sector Government strategic planning and horizon scanning Gender
Transport and utilities infrastructure Environment and sustainability Inclusiveness of growth
Enterprise sustainability Food and energy security Demographics
Informal sector Land rights Access to information
Technology infrastructure Security Health

In addition to the secondary data, between December 2018 and February 2019, Oxford Economics conducted a survey of 1,400 country specialists, with 25 survey questions, with a minimum of 10 specialists per country.

Primary survey experts

For the 2019 CRI, we surveyed individuals with at least 7 years of experience analyzing, studying or living in their reporting country. The individual had to have a good knowledge of economic policymaking, social structures and governance institutions in that country and was not currently employed directly by a government department in the country that directly influences and/or enforces policy making. A minimum of a tertiary-level educational qualification from an accredited university or vocational college degree program was required. Country experts came from a range of industries and sectors, where possible, including senior managers within the private sector, academia and trade unions.

Weighting

The CRI is a weighted average of primary (survey) and secondary (published) data. Overall the 2019 Index is weighted 21:79 for primary versus secondary data (25 primary questions and 126 secondary data series). The Index is a weighted average of 3 pillars, 30 sub-indexes, and 151 data points as follows:

  • The CRI is aggregated from the three equally weighted pillar scores. It is in an average of the pillar scores.
  • Each of the three pillars is an average of the 10 sub-indexes within the pillar, all of which have equal weighting.
  • The sub-indexes are, where data is complete, an average of the data points contained within them. The number of individual indicators is different between the different sub-indexes - some have more data points in them than others.
  • Where secondary data is incomplete, sub-indexes are calculated with a methodology which caps the weighting of the primary data, as set out in detail in Box 1.

Box 1; Dealing with Data Gaps

Whilst the primary data is available for every country, secondary data is missing in some cases, with a large number of missing data in certain countries. At the sub pillar level we calculate the average score of the primary data, and the average score of the secondary data, and weight these two averages using the weighting that would be accorded to primary and secondary data in countries with a full data set. So for example if the pillar was meant to have 2 primary survey questions and 8 secondary data points, we would calculate the index with a 20% weighting for the primary questions average, and an 80% weighting for the secondary data average, which could be based on fewer than eight available data points.

Each of the primary survey and secondary data values were transformed to a standardized scale between zero and one, facilitating the aggregation of the variables into sub-index, pillar, and composite scores. Standardization involves converting data to a common scale which can be aggregated. Standardization is calculated as set out in Box 2.

Box 2:  Transformation of data into an Index.

This is achieved by:

  • subtracting the average value for all the countries from the individual country score, and dividing the result by the standard deviation across the data set for all countries.
  • removing extreme values, or “trimming”. The sign of the standardized score indicates whether the raw value for a given country is above or below the sample mean. Outlier values are ‘trimmed’ to avoid distorting effects on the whole scale from countries with extreme values, using 3 standard deviations as a threshold.  In other words this value is restricted to a range which cannot exceed 3 at the upper end and -3 at the lower end.
  • transformation to a normalized range from zero to one.  This is calculated by taking the difference between the country value and the minimum value from all countries, and dividing that by the range from the country with the maximum country value and the country minimum value.  This means that the country with the highest score would have a value of one, and the country with the lowest score would have a value of zero.

*All uses of the word ‘country’ or ‘countries’ in the 2019 Change Readiness Index (CRI) refers to both countries and jurisdictions. The term country is used for brevity.

Connect with us

 

Want to do business with KPMG?

 

Request for proposal