Scottish Government

Programme for International Student Assessment (PISA) 2003: Initial Report on Scotland's Performance in Mathematics, Science and Reading

Footnotes

1. The level of significance refers to the odds that the difference between the scores is due to there being an actual difference, and not to chance. At the 5% level there is a 1 in 20 chance that the difference is not a true difference (i.e. due to chance), while at the 10% level these odds are greater, a 1 in 10 chance that the difference is not a true difference.
2. See OECD (OECD, 2003)
3. See OECD (OECD, 2001)
4. Note that the OECD counts full national entities as "countries". Thus the UK is counted once only in this figure of 41, although England with Wales, Scotland, and Northern Ireland participated as three independent units within the actual study. Similar situations held in other "countries".
5. Appendix A gives more information on which sampling criteria the UK failed to meet.
6. See: {OECD, 2003. The PISA 2003 Assessment Framework.}
7. The 2003 results were used for this. Thus, the scale is not the same as that in the report on PISA 2000.
8. Although The Netherlands also had a higher mean score than Scotland in 'Change and Relationship' in PISA 2003, there was no data on its performance on this in PISA 2000 (The Netherlands failed to obtain the required sample size in PISA 2000).
9. Slight discrepancies between these four figures, and equally between those below, arise from rounding errors and imbalances in student counts between countries.
10. A comprehensive treatment of performance in these three aspects of reading literacy can be found in the International report on that study.
11. These results must be taken as indicative only and not as well established. The PISA 2000 results came from different tests to those used in PISA 2003, and it is uncertain how this alteration would affect pupils at the extremes of attainment in particular. While mean scores for whole countries would be little affected by the change, scores from students at the extremes may be more greatly so, particularly if the selection of items for PISA 2003 was unwittingly biased towards items that were either more difficult or more easy for these low attaining students.
12. Countries are actually sorted by the statistical significance of the differences in scores, not by the differences themselves, hence Canada appears to the left of Austria despite a smaller score difference.