CAMCES

Computer-Assisted Measurement and Coding of Educational Qualifications in Surveys

Team: Roberto Briceno-Rosas, Verena Ortmanns
Leader: Dr. Silke Schneider
Scientific unit: Survey Design and Methodology (SDM)

Abstract

Education is an important cause, mediator and outcome of social inequality. The individual’s level of education to a considerable degree influences his/her labour market and wider life chances, e.g. income and health, but also his/her children’s education. Educational attainment is also an influential factor of individuals’ behaviour, values and attitudes. One reason for those relationships is that education increases the individual’s human and cultural capital (specific knowledge, skills, and competences) that have positive outcomes in later life. Another reason is the socialising effect of education. Finally, educational qualifications assume symbolic meaning and signalling power on their own. Therefore, the educational attainment variable is of central importance for many economic, social and education research and policy questions. It is consequently also one of the most used control variables in survey research.

The individual’s educational attainment is thus a core social background variable in standardised surveys. However, the centrality of this variable contrasts starkly with its inadequate measurement, especially in the case of migrants and cross-national surveys. Usually, educational attainment is measured by means of a closed question on the highest educational qualification achieved, providing a limited number of fixed response categories containing the most common qualifications in the country of survey, which are harmonised post-hoc. An inconsistent level of detail of response categories across surveys and countries, increasing differentiation of educational systems as well as education and work-related migration however increasingly complicate the measurement and harmonisation of educational attainment in surveys. Current practice also does not yet take advantage of the technological opportunities offered by computer assisted interviews, which are becoming increasingly common. The aim of the proposed project is to develop a tool for measuring educational qualifications in computer-assisted surveys, based on 1) an international database of educational qualifications, 2) optimised questionnaire instruments and 3) an interface to directly access the database for use in computer-assisted surveys.

The basic workflow would be as follows: When asked about the highest educational qualification achieved, the survey interviewer/respondent would enter the response into a text field and submit it to a search algorithm. It would then be matched with the educational qualifications in the database, produce a list of likely response options, and the most suitable response would be selected (potentially following a prompt or probing by the intereviewer). In addition, a search tree (like a sequence of nested showcards) could be provided to help respondents who cannot spontaneously name their highest qualification. The tool would thereby allow interviewers/respondents to select the ‘best match’ for the given textual information.

This new service could be integrated into a large number of surveys. The new tool would allow respondents to report their educational attainment in their own words and relative to the educational system of the country where they completed their education, rather than forcing them to choose from a limited number of response categories or guess the closest equivalent in the country where the survey is conducted. It would also provide instant harmonisation. As a GESIS product, it is planned that the service will be free to use for academic and non-profit surveys. It will be comprehensively documented for users on a dedicated web page.

Runtime

01.06.2013-31.05.2017

Sponsored by

 (Leibniz Competition)

Partner

Partner

  • Amsterdam Institute for Advanced labour Studies (AIAS), Universität Amsterdam
  • German Socio-Economic Panel Study (SOEP), German Institute for Economic Research (DIW), Berlin
  • TNS Infratest, Munich