Method

The project is mainly based on the statistical techniques and procedures for an analysis of aggregated data. Broadly speaking, these procedures to collect and to analyse the data quality of the SMRE and to give own estimates consist of five distinct steps.

1. Data research and data entry

Relevant data sources are identified and added to the metadatabase via the data entry tool either by external scientists and country experts or by the SMRE team.

2. Verification and data cleaning: error sorting out

All available statistics are integrated into the metadatabase. For each of the data sources, the research team clarifies the existency and the origin of the data reported. Due this, it is possible to identify those sources from which the reported figures really originate.The data will be harmonized with the  SMRE-categories of religious denominations and groups. This procedure of standardizing the data structure is called “mapping” (step two in the data entry section). For all available statistics, original or not, a detailed data description will be added (step three in the data entry section). The data description is an important part of the endeavour.

The very important last step for cleaning is the “error sorting out”. By adapting the total survey error theory the SMRE looks for various sorts of errors in the data sets like sampling errors or non-sampling errors. All data sets with errors will be sorted out and are not used for the further SMRE calculations. Verifying and cleaning data leads to a robust data collection within the SMRE for its calculations.

3. Technical analysis

Degree of pluralisation: In the analyses of the available data after verifying, cleaning and error sorting out, all data will be categorized according to three degrees of pluralisation.

Country data quality: Based on datasets without errors and the degree of pluralisation the SMRE evaluates the quality of data for each country). The algorithm enables intersubjective data evaluations. The country data quality is used as an evaluation scheme country by country and reporting period by reporting period.

4. SMRE Estimates

Based on the classification of country data quality, the SMRE proceeds to calculate its final SMRE estimates. The distribution shown as SMRE-estimates originate from different pathways of producing estimates based on the particular level of country data quality.

5. Data output

All available data (with and without errors) including the SMRE estimates are presented in the Data Output. You can retrieve data and you can easily prepare customized tables, diagrams and maps.