The Scottish Survey of Achievement (SSA) aims to find out how well pupils are learning in primary schools and the first two years of secondary schooling in Scotland. It looks in particular at the performance of P3, P5, P7 and S2.
To gather the evidence it needs, SSA uses a sample survey rather than testing every single pupil at these four stages. First it chooses a sample of schools and then a sample of pupils from within these schools.
The sample of schools and pupils is large enough to ensure that the findings of surveys are representative of Scotland as a whole and present a reliable picture of pupils’ performance at each of the four stages. The survey can therefore report confidently on the performance of P5 in science, for example, for Scotland. In contrast, the survey cannot report reliably on the performance of individual schools; that is not what it is designed to do.
When SSA has to report at the level of a local authority, as it sometimes does, then the number of schools in that authority involved in the survey will increase. This is necessary to make sure that there is enough data to report reliably at the level of the authority. This often means that all the schools in the authority will be asked to participate. The number of pupils involved will also be greater.
Schools
Choosing a national sample of schools
Each year SSA draws a nationally representative sample of schools for the survey. Typically around 200 to 250 schools are needed at each of the four stages. The schools are selected at random. However, an attempt is made to ensure that the random selection includes representative samples of small and large schools, rural and urban schools, etc. The aim is to ensure that there are enough schools - and pupils - to provide reliable data for reporting at a national level at P3, P5, P7 and S2. At each of these stages around 2500 pupils are required.
Authority level reporting
When a survey has to report at the level of a local authority, as it sometimes does, then the number of schools in that authority involved in the survey will increase. This is necessary to make sure that there is enough data to report reliably at the level of the authority. This often means that all the schools in the authority will be asked to participate. The numbers of pupils involved will also be considerably greater than for a survey reporting only at national level.
Pupils
The sample of pupils
SSA needs around 2500 pupils to be able to report reliably on the performance of a single stage for Scotland as a whole. So a typical survey requires around 10,000 pupils.
Having drawn a sample of schools, SSA draws this sample of pupils randomly and anonymously from information already available from the school census. Schools sometimes express concern that the pupil sample is not representative of the pupils in the school. This is an understandable concern, but one which should be set aside. The sample in any one school is only a small part of a national sample that is designed to represent Scotland as a whole. The sample is not intended to represent the school.
Using a large sample ensures that the pupils in total represent the full range of performance for the specified stages.
Pupils with additional support needs
The pupil sample is entirely random. Consequently it is likely that the sample in any one school may include pupils with additional support needs. If this does happen, there are two options available to a school.
The school can let the pupil take part in the survey with the support that would normally be provided. So for example a pupil who normally has a scribe, could have a scribe.
The school can withdraw the pupil from the survey, if the school knows that the pupil will be unduly stressed or adversely affected by taking part. This is a judgement only the school staff can make, based on their knowledge of the pupil. However, a school should not withdraw a pupil simply because he/she 'is not very good at mathematics', for example. A pupil should be withdrawn only where the school judges that involvement in the survey would be genuinely stressful for the pupil.
Withdrawing pupils from the survey
Once a school knows which pupils are in the pupil sample, it will be able to judge which, if any, of them might be unduly stressed or adversely affected by taking part in the survey. The school may decide to withdraw such pupils from the survey before the survey starts. See also Pupils with additional support needs.
A choice for parents/carers
Parents/carers, too, can withdraw their child from the survey.
As part of the preparations for the survey, schools are sent a letter that they are asked to send to the parents/carers of every child taking part in the survey.
The letter explains the purpose of the survey, what pupils are expected to do, and how the results will be reported. If parents/carers want to withdraw their child from the survey for any reason, they are asked to contact the school to request the withdrawal. The school should then make sure that the pupil does not take part.