Resources for working with MICS Foundational Learning Skills data

New AFLEARN resources for working with MICS Foundational Learning Skills data

The UNICEF Multiple Indicator Cluster Surveys (MICS) are an incredibly rich source of data for exploring foundational learning across countries. Because the Foundational Learning Skills (FLS) module is administered through a household survey, the data allow researchers to look beyond headline learning indicators — examining the full distribution of skills and how learning varies with schooling, household characteristics, wealth, disability, language and other factors.

But getting the MICS FLS microdata research-ready is not always straightforward.

AFLEARN has developed a set of practical resources to make it easier for researchers to work with the MICS6 FLS data. These include:

• guidance on accessing data through UNICEF MICS and IPUMS MICS
• tools for organising and merging the original UNICEF files
• harmonised variables for cross-country analysis
• guidance on survey weights, sampling design and standard errors
• worked examples in R and Stata
• detailed documentation of harmonisation and country-specific issues.

We have also developed research-ready reading assessment variables from the original UNICEF microdata. This is particularly important for researchers using IPUMS MICS: in countries where MICS administered more than one reading passage, the IPUMS harmonised reading variables retain only one passage, meaning that children assessed using other passages can be systematically missing. Our reading resources retain the different passages and assessment languages and can be merged with an IPUMS extract, allowing researchers to combine IPUMS's very useful harmonised background variables with the complete MICS reading data.

The aim is not simply to make the data easier to use, but to make it easier to use them well — understanding the sample, assessment structure, routing, language issues and comparability decisions that sit behind the variables.

Explore the AFLEARN MICS guide and resources

We would be very interested to hear from researchers using MICS FLS data — and particularly about other resources or examples that would make the guide more useful.