The Scottish Government

« Previous | Contents |

Listen

Footnotes

1. In 2005 the reporting authorities were Aberdeen City, Angus, East Ayrshire, Edinburgh City, East Dunbartonshire, East Renfrewshire, Highland, Inverclyde, North Ayrshire, North Lanarkshire, Perth and Kinross, Renfrewshire, South Ayrshire, South Lanarkshire, Stirling, West Lothian. In 2006 the other 16 authorities will become reporting authorities, and the 16 in focus in 2005 will become 'non-reporting'.
2. The banding criteria, while in principle arbitrary, were based on the judgment of subject specialists. They were first used for reporting reading attainment in the 2001 AAP English Language survey, and were then used in all subsequent AAP surveys in other subject fields.
3. It is worth noting that, contrary to a commonly held misconception, the precision of an estimated proportion depends virtually exclusively on the size of the sample, the size of the population being irrelevant unless extremely small.
4. The fact that population size, unless extremely small, has no bearing on the precision of estimated proportions means that reporting authorities of very different sizes nevertheless require the same pupil sample sizes to achieve the same degree of estimation precision.
5. In practice, since the SSA pupil samples had to support separate attainment reporting by authority in more than one curriculum area, viz. reading and numeracy, the 450 pupils were divided between the two types of assessment, with the consequence that margins of error were closer to seven percentage points than to five (see Appendices C and E).
6. Since pupil samples can only comprise integer numbers of pupils, the proportions of pupils taken from each school at a particular stage within a reporting authority varied slightly around this nominal figure, and the target of 445 pupils per stage was sometimes slightly over-reached and at other times slightly under-reached (see Table B.2).
7. School and pupil sample numbers were computed by Assessment Europe. School sampling was also carried out by Assessment Europe, while pupil sampling was conducted by SEED, on the basis of ScotXed school and pupil records.
8. Determined on the basis of pupils' home data zones, each zone having a ' SIMD rank' indicating its position within a national rank order based on a complex deprivation measure: for full information, see Scottish Index of Multiple Deprivation 2004. Summary Technical Report (Edinburgh: Scottish Executive, 2004), or access www.scotland.gov.uk/SIMD2004Mapping and www.scotland.gov.uk/SIMD2004Data.
9. An important difference in the application of these criterion cut-off scores in reading and mathematical literacy on the one hand and numeracy on the other is that in reading all the items contributing to a pupil's '% correct' score were included in a single multi-item task whereas in numeracy the items concerned were spread over two different test booklets.

« Previous | Contents |