Isled – A World-wise Comparative Measure of Level of Education

Author – Harry Ganzeboom.

Abstract :

In comparative social research, the level of education is routinely measured using one of two methods. The qualification method measures the level by highest completed (or most recently attended) program, most often by asking for the actually completed certificate / diploma / degree. Best practice here is to measure the qualifications in country-specific terms and then to post-harmonize these using a common denominator. Alternatively, comparative research measures level of education using its duration, best collected as a question to respondents about the (net) length of their educational careers.

The recent development of the three-digit International Standard Classification of Education 2011 [ISCED-2011] has become a major game-changer, because for the first time a detailed and rigorous harmonization framework has become available for country-specific qualifications, which allows the researcher to scale these to an internationally valid measure with fine-grained values. Schröder & Ganzeboom (2014; see also Schröder (2014)) have proposed two methods to improve the comparative measurement of level of education. First, using data from R1-R4 of the European Social Survey, they developed the International Standard Level of Education [ISLED], which was conceived and constructed as an optimal scaling of educational qualifications in an intergenerational status attainment model, with parental statuses as inputs and respondent’s occupation and spouse’s education as output (see Figure 1). As to date, the ISLED measure is associated to and has been mostly used in combination with the ESS data. In the present paper I generalize the construction of ISLED in a world-wide database, consisting of all ISSP waves that contain information on parental occupation(s). In order to do so, all country-specific information on educational qualifications has been harmonized in retrospect to the ISCED-2011 classification. The database covers some 4000 qualifications in five ISSP waves (1987-2019) in some 45 countries worldwide, with a total N of 131,047 cases.

Second, Schröder & Ganzeboom (2014) have developed a latent variable model to measure the true (‘hidden’) level of education, with optimally scaled qualifications (i.e. ISLED) and duration as parallel reflective indicators. Using (rare) instances of double measurement of respondent’s and spouse’s education, this latent variable measurement model can be recast as a Multiple Trait Multiple Method [MTMM] model, that can distinguish the validity and reliability of each indicator (Figure 2). I use this MTMM model to examine the measurement quality of the newly constructed ISLED, as well as the duration measure.