Teacher Literacy and Numeracy Skills: International Evidence from PIAAC and ALL
Teacher Literacy and Numeracy Skills: International Evidence from PIAAC and ALL
Bart H. H. Golsteyn 0 1
Stan Vermeulen 0 1
Inge de Wolf 0 1
JEL Classification I 0 1
0 Inspectorate of Education and Department of Economics, Maastricht University , P.O. Box 616, 6200 MD Maastricht , The Netherlands
1 Department of Economics, Maastricht University , P.O. Box 616, 6200 MD Maastricht , The Netherlands
Using the OECD-studies PIAAC and ALL, this paper shows that teachers on average have better literacy and numeracy skills than other respondents in almost all of the 15 countries in the samples. In most countries, teachers outperform others in the bottom percentiles, while in some countries they perform better than others throughout the skill distribution. These results imply that the scope to improve teachers' skills varies between countries and that policy makers should take the shape of the skills distribution into account when designing interventions in order to most efficiently raise teachers' skills.
Teachers; Skills; Human capital
1 Introduction
Teachers are essential for the development of human capital in society. Their skills
are formed in teacher training programs, but are also highly influenced by the type
and overall quality of the students who enter these programs and become teachers.
Understanding which segment of the population is part of the teacher corps is important
in order to determine the focus of interventions which can improve the quality of
teachers.
This paper compares the dispersions of literacy and numeracy skills of primary and
secondary school teachers relative to those of other respondents. We use international
data of 15 different countries from the Programme for the International Assessment
of Adult Competencies (PIAAC) and the Adult Literacy and Lifeskills Survey (ALL),
both conducted by the OECD. These data sets are representative samples of the adult
population in various countries. They include reading and math test scores, and
contain detailed information about occupations. For each country, we compare average
math and literacy skills between teachers and other respondents, and we investigate
differences at the 10th and 90th percentiles of the distributions.
In virtually all countries, both primary and secondary school teachers score on
average higher on literacy and numeracy tests than the country average. Secondary
school teachers score higher than primary school teachers on both skill measures. Our
analyses of the differences in skill distributions between teachers and others show
that the lowest scoring teachers significantly outperform the lowest scoring other
respondents both on literacy and numeracy tests. At the top of the distributions, we
find that the best secondary school teachers are not strongly outperformed by the best
other respondents, and that the best primary school teachers score only slightly lower
than the best other respondents.
The extent to which low scoring teachers outperform other low scoring respondents
differs substantially across countries. For instance, in the Netherlands, primary school
teachers at the 10th percentile perform much better than other respondents at the
10th percentile, while in Denmark, primary school teachers at the 10th percentile do
not outperform other respondents at the 10th percentile that strongly. In Denmark, it
might therefore be an effective policy to focus on the bottom of the distribution (e.g.,
by raising barriers to enter into teaching, or focusing training on the worst teachers),
while in the Netherlands little can be gained in becoming more restrictive at the lower
end.
The main message of the paper is that it is important to investigate the shape of
the relative skill distributions in addition to the differences in average skills when
designing policy to improve the teacher corps. Our results persist when restricting the
comparison to the tertiary educated subsample. The results are not driven by age or
gender, and are not sensitive to the inclusion of measures for the frequency of skill
use.
This paper contributes to the literature on teacher characteristics and teacher quality.
Teacher quality has been recognized as one of the most important determinants of
educational productivity
(Hanushek 2011; Barber and Mourshed 2007)
. Hanushek
(1992) finds that being taught by a high quality teacher results on average in 1.5 years’
worth of progress in one academic year, while being taught by a low quality teacher
results on average in 0.5 years’ worth of progress. Although the exact characteristics
of teacher quality are not well-defined
(Hanushek and Rivkin 2006)
, teachers’ skills
as measured by scores on achievement tests have been found to be associated with
educational productivity
(Metzler and Woessmann 2012; Eide et al. 2004)
. Hanushek
et al. (2014) furthermore show that teachers’ cognitive skills are an important factor
in explaining international differences in student performance.
While such (...truncated)