Second International Handbook of Mathematics Education by Alan J. Bishop

Author:Alan J. Bishop
Language: eng
Format: epub
Publisher: Springer-Verlag Wien 2012
Published: 2015-01-26T16:00:00+00:00

TIMSS

In a globalising world, international comparative assessments make sense. They provide benchmarks for both internal and external comparisons. Such arguments have been made both by the key organisational hub for the Third International Mathematics and Science Study (Plomp, 1998) as well as wider afield (Nebres, 1999). As it re-entered the world in 1994, South Africa found participating in TIMSS in 1995 an attractive option. Here was a possibility for setting up a benchmark against which progress by the post-apartheid government could be mapped and judged (Howie, 1998). The results of TIMSS are now well known and need no rehearsal here. The question we pose is the broader ethical one that drives this whole chapter: Did TIMSS get the description right?

Keitel and Kilpatrick (1999) provide an extensive critique of TIMSS. In their discussion of the promises and perils of international comparisons in mathematics education, they open up numerous ethical issues for the research endeavour. First, they highlight how the direction of the study has been overdetermined by psychometric expertise. In handling the data once collected, and treating possible problems, they argue that “problems of methodological validity, reliability and quality have been resolved purely from a formal point of view. Questions of content — in all its aspects — have usually been seen as secondary” (p. 245).

The second issue they raise is that financial support for the study influences the goals and the extent to which they are politically determined or research oriented, not to mention which countries are able to participate. Few developing countries participated. An interesting and disturbing question is why countries were ranked as in a league table. Whose interests are served by this ranking, and what kind of description is it?

This question points to the third issue they raise: control over the framing and dissemination of results. The power of TIMSS publicly is the sense that numbers do not lie and that the results obtained through the careful data collection and analysis processes were somehow objective. We only need to pause and reflect for a moment on Cooper and Dunne’s (2000) extensive research on performance on ‘realistic’ test items in the United Kingdom. They show convincingly how more complex forms of assessment, like contextualised questions or questions requiring extended elaborated responses, can produce false negatives and tend to do so in ways that disadvantage working-class learners. When Cooper and Dunne interviewed learners who presented or selected wrong answers to such items, they found on probing that the answers selected masked learners’ mathematical competence. And there were significant differences here between working- and service-class learners. Cooper and Dunne’s research raises important questions about test validity and, we would add, about the ethics of using such results for determining the position of individual learners, let alone countries on a ladder of achievement. A number of the TIMSS items were embedded in realistic contexts. From this perspective, TIMSS could not get the description right. Although the data and analysis revealed important information about how learners across countries performed on a particular

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.