MORE EQUAL AND LESS EQUAL AT THE SAME TIME? MEASURING INEQUALITY IN EDUCATIONAL ACHIEVEMENT OF 15-YEAR OLDS IN 37 COUNTRIES

  • The authors:
    Gwyther Rees
    Anna Gromada
    Yekaterina Chzhen
  • Issue: July 24-26th, 2019
  • Pages: 666-673
  • Section: EDUCATIONAL MOBILITY AND SOCIAL INEQUALITY
  • URL: http://conference-ifl.rudn.ru/666-673/
  • DOI:
    10.22363/09669-2019-666-673

Abstract. The article aims to explore how different approaches to the
conceptualisation and measurement of educational inequalities affect the
conclusions that are drawn from comparisons between countries and over
time. It contains new analysis based on the most recent waves of the
Programme for International Student Assessment (PISA). It shows that
different measures of inequality lead to diverging, and sometimes even
contradictory, conclusions about which countries are faring better and worse
in terms of educational equality compared to other countries, and over time.
We focus on countries that are a member of the European Union (EU)
and/or OECD. The study produces nationally representative and crossnationally comparable data on schoolchildren’s skills and knowledge in
reading.
According the Pearson correlations between each pair of inequality
measures across 37 countries, using data from PISA 2015 all measures of
inequality of outcome are strongly related to one another – with correlations
ranging from 0.92 to 1.00. Measures of inequality of opportunity show
somewhat weaker relationships, with correlations ranging from 0.56 to 0.83.
The limited range (P90-P10) is a good ‘representative’ measure of
inequality of outcome for international comparisons because it is easy to
communicate to non-specialist audiences (see UNICEF Office of Research
2018). It shows perfect correlation with standard deviation and very high
correlation (0.93***) with the Gini Index. P90-P10 will be subsequently
compared with three measures of inequality of opportunity.
The complex relationship between the inequality of outcome and inequality
of opportunity will be discussed by comparing Iceland and Hungary. The
two countries find themselves on the two extremes of the inequality of
opportunity spectrum. Iceland is the single most “equal” country using
ESCS R-squared and the third most equal using the gradient. As such, it
666
DOI: 10.22363/09669-2019-666-673
emerges as a country with the single weakest influence of economic, social
and cultural status on children’s reading skills. Yet, Iceland’s ranking goes
down radically to the 24
th
rank when we opt for the R-squared which takes
into account factors outside of children’s control that are not directly linked
to their parents (i.e. school location, migration, language and gender).
By contrast, Hungary emerges as one of the three most “unequal” countries
using all three measurements. It is the single most unequal using F-G and
ESCS R-squared and the third most unequal using the ESCS gradient. The
record 35% of reading score variance can be attributed to factors outside
children’s control, including 22 percentage points attributable solely to
ESCS status – making for an especially strong link between the family
status and educational outcomes. However, the most “equal” Iceland and the
most “unequal” Hungary turn out to have almost identical inequality of
outcome regardless of the measurement method. They show the standard
deviation of 99 and 97 respectively, Gini of 0.10-0.11 and P90-P10 of 255-256 points. Furthermore, these outcomes place them in the bottom half of
the equality of outcome ranking among the analysed countries.
Keywords: inequality of opportunity, inequality of outcome, educational
achievement, reading, PISA

Gwyther Rees¹, Anna Gromada², Yekaterina Chzhen³
¹The UNICEF Office of Research – Innocenti, Florence, Italy,
e-mail: grees@unicef.org.
²The UNICEF Office of Research – Innocenti, Florence, Italy and Polish
Academy of Sciences, Warsaw, Poland, e-mail: agromada@unicef.org
ORCID ID: 0000-0001-8135-1424
³The UNICEF Office of Research – Innocenti, Florence, Italy,
e-mail; ychzhen@unicef.org

Ferreira, F.H.G. and Gignoux, J. 2014. ‘The Measurement of Educational
Inequality: Achievement and Opportunity’, World Bank Economic Review,
28 (2): 210–246.
Põder, K., Lauri, T. and Veski, A. 2017. Does School Admission by
Zoning Affect Educational Inequality? A Study of Family Background
Effect in Estonia, Finland, and Sweden, Scandinavian Journal of
Educational Research, 61(6): 668–688.