LEMMA II - Structures for building, learning, applying and computing statistical models

Principal Investigator: Professor Fiona Steele

LEMMA II (Structures for building, learning, applying and computing statistical models) focused on the development of new multilevel modelling methodology to address important social science research questions and capacity building in quantitative social science.

Existing multilevel modelling methodology cannot adequately model dependencies among schools and dependencies between schools and neighbourhoods, and the effects these dependencies have on pupils' learning progress. LEMMA II extended existing methodology to represent these dependencies better which allows more realistic modelling of the effects that school competition and parental selection processes have on pupil’s educational progress.

LEMMA II had three main substantive research areas: school completion, family dynamics and social segregation. Their main areas of statistical methodology development were correlated classifications and missing data in multilevel models. Researchers at LEMMA II implemented the new methodology in a robust user-friendly software which together with an online learning environment and other learning materials and activities formed a programme to build capacity in quantitative social science and facilitate cutting edge social science research

LEMMA II was based at the University of Bristol.

For further information please see LEMMA II website.