Letters

Schools' experience of league tables should make doctors think again

BMJ 2000; 321 doi: http://dx.doi.org/10.1136/bmj.321.7274.1467 (Published 09 December 2000) Cite this as: BMJ 2000;321:1467
  1. Peter Tymms, professor,
  2. Andy Wiggins, researcher (andy.wiggins{at}durham.ac.uk)
  1. Mountjoy Research Centre 4, University of Durham, Durham DH1 3UZ

    EDITOR—The BMA discussion paper Clinical Indicators (League Tables) touches on the issue of how performance indicator systems can have dysfunctional behavioural and managerial implications.1 Winston shows how this applies to the Human Fertilisation and Embryology Authority's league tables of in vitro fertilisation clinics.2 Our research looks at the dysfunctional effects of league tables in education and may be relevant to some of the issues surrounding league tables in the health sector.

    We have recently reported a comparison between two systems, one with league tables (English primary schools) and one without (Scottish primary schools). In most other respects the two systems are similar. The research entailed a questionnaire survey of heads and teachers from 54 randomly selected primary schools in England and Scotland during 1999.

    Some of the key findings were:

    • English schools were more likely to report concentrating on meeting their targets, at the expense of other important objectives

    • The target setting and testing process apparently had a narrowing effect on the curriculum in England

    • English schools were more like to say that they had concentrated resources on “borderline children”—those close to reaching the threshold—who would improve their league table position

    • English schools particularly thought that the target setting and testing process had increased the “blame culture”

    These findings were not entirely surprising and had been predicted (for example, by Smith3), but what was surprising was the substantial difference between the two countries. Seven questions were combined to form a dysfunction index. On a scale of 1 (maximum dysfunction) to 5 (no dysfunction), English primary schools scored 2.17 and Scottish 3.19 (P<0.001; effect size 1.68). The figure shows the results.

    Figure1

    Boxplots showing mean (SD) degree of dysfunction for each country (the lower the number the greater the dysfunction)

    As well as the differences, we found several important (though not significant) similarities:

    • Both groups wanted access to performance data for their internal use— that is, they were not against performance indicators as such

    • Schools in both countries seemed to be under similar pressure to meet targets. This suggests that having league tables does not necessarily apply greater pressure than other less public techniques

    • Parents in England were no more likely to make reference to school test results than their Scottish counterparts. This questions one of the fundamental justifications of league tables: giving consumers information that they can use to put pressure on “their” schools

    Although there were difficulties in allowing for other factors that may have an influence on the responses to the questionnaire, the data should give pause for thought. Careful consideration should be given to the unintended consequences of league tables.

    Footnotes

    References

    1. 1.
    2. 2.
    3. 3.
    View Abstract