| InCiSE indicator | RAG rating | Potential routes for future development |
|---|---|---|
| Integrity |  |
Addition to, or replacement of, existing metrics with non-perception based measures. |
| Openness |  |
n/a |
| Capabilities |  |
Identification of data sources with more recent data and/or more regular update frequency. |
| Inclusiveness | |
Additional metrics providing objective measurement of ethnic/religious diversity, and metrics providing objective/subjective measurement of inclusion for other under-represented groups (e.g. disability, age, socio-economic background, LGBT). |
| Policymaking | |
Addition of non-perception based measures, including on themes such as timeliness, accuracy, and use of evidence. |
| Fiscal & financial management | |
n/a |
| Regulation | |
n/a |
| Crisis & risk management | |
Replacement of the data sourced from the UN’s Hyogo Framework for Action monitoring reports as monitoring data from the Sendai Framework becomes available. |
| HR management | |
Identification of data to measure additional themes such as skills gaps/talent deployment, quality of learning and development, and level of satisfaction with HR services. |
| Procurement | |
Additional themes such as value for money and the capabilities of procurement officials. |
| Tax administration | |
Additional themes such as preventing tax evasion. |
| Digital services | |
Identification of non-perception based measures, including average transaction time, up-time of systems, proportion of government services available online. |
| Table 6.3.A in the original 2019 publication | ||
17 Future development
The 2019 index is the second edition of the InCiSE project, following the pilot edition published in 2017. The 2019 edition builds on and strengthens the methodology of the pilot edition. The InCiSE Partners have used a combination of stakeholder feedback, continued engagement with data providers and further desk research to develop the methodology for the 2019 edition of the InCiSE Index.
Given the frequency of data updates and to provide suitable time to reflect on each edition’s results, we propose that future editions of the InCiSE Index are repeated on a biennial timescale. This chapter sets out considerations for future development of the InCiSE methodology.
17.1 Social security administration
The InCiSE framework (described in Section 1.3) identifies social security administration as one of the constituent functions of an effective central civil service, and the 2017 Pilot edition of the InCiSE Index included an indicator for social security administration. The indicator was based on a single metric, which was the administrative costs of social protection as a proportion of total social protection expenditure. Feedback from the pilot edition included a critique of this metric, saying it was unsuitable given the inclusion of state provided healthcare which varies significantly across countries. Furthermore, the data was available solely for European Union member states, so data for non-EU countries was imputed based on correlated perception measures from the Quality of Governance study used elsewhere in the InCiSE model.
Exploration of the source data did not identify an appropriate method to exclude healthcare costs from the calculations. A review of further data sources identified neither alternative metrics that included non-EU countries nor imputation predictors with a closer intellectual or theoretical relationship to the indicator’s conceptual basis.
It was therefore decided that the social security indicator should be removed from the 2019 edition of InCISE. For future editions of the InCiSE Index, we will continue to explore whether there is suitable data to reintroduce a social security indicator.
17.2 Functions and attributes not yet measured
In addition to social security, four of the functions and attributes identified in the InCiSE framework have not been measured in either edition of the index: IT for officials, internal finance, staff engagement, and innovation. No suitable data has been identified since the pilot that would allow for measurement of these four potential indicators. Future editions of the InCiSE index will continue to explore whether suitable data exists to introduce indicators for these four areas.
17.3 Functions and attributes already measured
The 2019 edition of InCiSE has used an additional 46 metrics compared to the 2017 Pilot: six form the new procurement indicator and 40 are distributed across the existing indicators measured in the 2017 Pilot.
While this has strengthened a number of indicators, as Table 2.7 shows only three of the indicators have been given a final ‘RAG’ rating of green (data quality score of 0.75 or more). Table 17.1 below provides some considerations for future improvements of each of the indicators measured in the 2019 edition of InCiSE, with amber or red ‘RAG’ ratings.
- High data quality
- Medium data quality
- Low data quality
17.4 Extending country coverage
While coverage of the InCiSE results has increased from the 31 countries in the 2017 Pilot to 38 in the 2019 edition, the group of countries remains broadly homogeneous, made up of OECD and EU member countries with high or upper-middle incomes. Future editions of the InCiSE Index will continue to use the data quality based approach to country inclusion set out in Section 2.3, however this requires greater data availability for non-OECD/EU countries.
There are a number of potential options, such as creating regional versions of the InCiSE Index using existing multi-country data collections for different regions (but for which either OECD or EU countries are not members). Alternatively, subsets of the existing InCiSE Index could be created as some indicators have wider data coverage than others.
The InCiSE Partners are committed to identifying ways to increase coverage, and have conducted two short studies of how the InCiSE framework applies in Brazil and Nigeria to inform future thinking.
While extending country coverage will generate a greater set of results, careful consideration will be needed on developing alternative versions of the index and how (if at all) to compare between them.