Transparency of Gender Indexes
All gender indexes should be accompanied by complete documentation of their design and construction. These guidelines will help governments and other relevant stakeholders better monitor progress in WEE-related outcomes. A technical report or equivalent documentation should include the following:
|A conceptual framework that identifies the ultimate outcome monitored (for example, women’s economic empowerment, gender equality, statistical capacity, governance) as well as the outcomes represented by the index’s dimensions and sub-dimensions.|
|A measurement model that identifies the indicators used to measure each sub- dimension and dimension and the criteria used to select the indicators.|
|The data sources and the years for which the data for each indicator are available.|
|All adjustments made to the raw data (for example, recoding, normalization. imputation of missing values, suppression of extreme values, censoring of values exceeding certain limits).|
|Description of any analytical methods used to construct the index (such as correlation, regression, principal components, factor analysis, item response theory, structural equation modeling) .|
|Methods used to combine the indicators into sub-dimensions, dimensions and the overall index, including the formulas and weights used at each level of aggregation (for example, unweighted arithmetic means, weighted geometric means).|
|Description of any methods used to evaluate the internal or external validity of the index (for example, criterion validity, sensitivity analysis, comparison of country rankings to those of other indexes).|
Full transparency requires that the index and the data used to construct it should be readily available in a convenient form that facilitates its analysis.
|The downloadable data set should include the raw data prior to any adjustments.
|The data should be in the format of a widely used statistical package, such as R, SPSS, Stata, or SAS, and when feasible in CSV format, so that interested users can analyze the data without having to invest a lot of time converting, recoding and re-labeling.
|All indicators should be clearly labeled, with labels no longer than necessary to uniquely identify the indicator.
|The “country” variable should include standard UN country names when available and ISO 2-alpha or ISO 3-alpha country codes (to facilitate comparisons with external data sets).
Once equipped with the data and the technical documentation, an interested user should be able to reproduce the outcome measure and country rankings.
|ABOUT THESE GUIDELINES
These guidelines were developed by Mayra Buvinic (CGD and Data2X), James C. Knowles (Data2X consultant), Megan O’Donnell (CGD), and Eric Swanson (Open Data Watch). The guidelines are informed by an evaluation of indexes reflecting countries’ progress in WEE-related outcomes and technical audits of selected gender indexes.
The full report on Measuring Women’s Economic Empowerment: A Compendium of Selected Tools is available here.
Technical audits of four tools measuring WEE at the country level by James C. Knowles is available here.