Status of Education System: Interactions of Input and Output Composite Indicators
Dovilė Stumbrienė
Vilnius University
Audronė Jakaitienė
Vilnius University
Rimantas Želvys
Vilnius University
Published 2017-12-20


principal component analysis
composite indicator
education system pagrindinių komponenčių analizė
sudėtinis indeksas
švietimo sistema

How to Cite

Stumbrienė, D., Jakaitienė, A. and Želvys, R. (trans.) (2017) “Status of Education System: Interactions of Input and Output Composite Indicators”, Lithuanian Journal of Statistics, 56(1), pp. 31–40. doi:10.15388/LJS.2017.13669.


Education is one of keystones that guarantees well-being of a country, therefore the understanding about the educationsystem status might be crucial. It motivates to measure the state of the education system, to understand its determinants and tomonitor changes over time that would allow the implementation of evidence-based education policy. Measurement and assessment ofthe state of the education system is a complex task, as the analysis of individual indicators of the educational system is insufficient tomonitor and evaluate education as a multidimensional phenomenon. To achieve a comprehensive and generalized assessment of theeducation system, we have chosen to calculate the composite indicators, namely, indicators of resources and outcomes. Using thelatter indicators we evaluate state of resources and output of the educational system, understand the factors, determining the state, andcompare it over time and in the context of other countries. Indices were calculated for the Baltic countries and three “old” EUmember states: UK representing the Anglo-Saxon liberal model, Germany for the Continental corporatist model and Finland as anexample of the Scandinavian model. For the analysis we used 2002-2014 annual publicly available data from EUROSTAT, OECD,and IEA databases. We have employed a simple weighted additive method with equal weights and principal components analysis forthe construction of indices. We have found that the differences between composite indicators, constructed by the simple weightedadditive method with equal and principal components analysis weights, are limited. The increase in the number of sub-indicators byalmost two-thirds does not affect dynamics of the output indices over time. We have established that inertia of the education systemis different for the countries: the impact of the output on the results is observed with 2–4 year lag for the Baltic States, as there is notime lag or there is a one year lag for Germany and the United Kingdom. Finland's results are different as compared with the othercountries examined. The dynamics of the Baltic indices is similar and possibly constitutes a separate group.