SEED Sponsored Research: Children starting school in Scotland
4 What do children know and what can they do when they start school in Scotland?
4.1 The Cognitive Profile Of Children Starting School In Scotland
Rasch scaling was used to estimate the relative difficulties of the items for the entire PIPS On-entry Baseline assessment and for each section separately (for an explanation of Rasch scaling see Bond and Fox, 2001). The interval scale enables direct comparisons between items and sections to be made. Figure 2 gives a general overview of the stages of development of children on entry to school. Figure 3 gives a more detailed comparison of the reading and phonological awareness, mathematics and vocabulary sections.
Figure 2 Developmental descriptors of children starting school in Scotland.
Each box describes activities which have been selected so that a typical child in the group can do about half of the tasks described. They can generally manage the items described in the box below but not above.
Figure 3 Descriptors of children staring school in specific cognitive areas
4.2 How Does The Cognitive Profile Vary?
How then does this cognitive profile vary? Data on age, gender, home background, pre school experience, first language and special educational needs are collected at the time of the PIPS BLA and the analysis below shows how the cognitive profile varies in relation to these variables.
Table 2 (Sex) indicates the advantage to girls over boys in standard deviation units (Effect Sizes 3) for the four areas being considered.
Table 2 Sex
Advantage to girls
In general, the girls started school in Scotland a little ahead of the boys, although in mathematics they were exactly on a par. The greatest difference between boys and girls was in the reading section where there was an advantage of about a fifth of a standard deviation. There was a similar but slightly smaller advantage in phonics and in the vocabulary there was just 0.06 of a standard deviation, which was statistically significant but educationally not important. It is worth pointing out also that the spread of scores for the girls was significantly less both for vocabulary and for mathematics, although not for phonics or for reading. The implications of the standard deviations differences are that the girls form a more homogeneous group than the boys. Extreme scoring groups, the very highest and the very lowest, will have a greater preponderance of boys amongst them. There will be fewer girls with special needs in both of those categories.
4.2.2 Home Background
Table 3 Home background
Advantage to those without free meals
SD with FSM
SD no FSM
Home background has long been established as being an important variable when looking at children's performance at school, especially when no earlier cognitive measure is available. There are a variety of ways of looking at home background and later on in this report we use post codes linked to deprivation indices, but a straightforward and widely used measure is the entitlement to free school meals which gives a dichotomous outcome. Quite clearly in each of the 4 variables there were highly significant differences between the average scores of pupils from the groups with and without free school meals. This was most noticeable for reading and vocabulary where the difference amounted to nearly 7/10 of a standard deviation and almost as much in vocabulary and rather less in phonics. It might be that it was rather less in phonics because the phonics measure was less reliable. The overall general pattern averages out at about 7/10 of a standard deviation.
4.2.3 Pre-school Experience
Table 4 Pre-school experience
Advantage for each term
Result from England
Very little connection was found between the amount of pre-school that pupils had experienced and their scores on the baseline assessment. The amount of pre-school experience recorded varied from no full-time terms to 6 and it was expected that a strong relationship would be found. On analysis, only a minimal link was found for mathematics (0.03 of a standard deviation unit per term). For a child attending pre-school for six terms the advantage would be 0.18 standard deviation units, which is not of great educational importance. By contrast in England a very clear and uniform relationship was found in the same year group amounting to figures that were typically around 0.1 standard deviation units per term. Figures 4 and 5 below show a very strong and clear relationship between mathematics and the amount of pre-school in England but not for Scotland. The values on the Y-axis of Figures 4 and 5 are mean scores with error bars denoting the 95% confidence interval. The datasets were smaller for Scotland and so one might expect to find a weaker pattern because of the errors on the measurement and it might also be that, because the children were older in Scotland, proportionately fewer pupils had experienced only one or two terms in pre-school. But to find such a weak relationship effect is a little odd and at this stage no clear explanation is apparent.
Figures 4 & 5 Maths and terms in pre-school
4.2.4 Special Educational Needs
Not enough data were recorded to be able to comment.
Table 5 (Age) compares the difference between the youngest children (four and a half years old) and the oldest children (five and a half years old) starting school in Scotland and England (age four and five respectively) for each area of the PIPS BLA. The differences are expressed as Effect Sizes.
Table 5 Age
Gain per year
Gain per year
Clearly the age of children is of considerable importance in their cognitive development. To quantify the relationship between age and measured attainment, regression analyses were carried out to estimate the difference that a year makes. That is essentially the difference between youngest and the oldest children starting school in Scotland. Comparing four-and-a-half year-olds to five-and-a-half year-olds, the difference amounts to about half a standard deviation unit. The difference was slightly higher for mathematics, rising to two thirds of a standard deviation, and a little less for phonics and vocabulary. Generally, the pattern can be translated into a figure of about 0.05 per month. There was a stronger relationship in England. That is to be expected, at least to some extent, since the mean age of children starting school was lower than in Scotland and age is a more important factor for younger children than older.
4.2.6 Summary: What children know and can do when they start school in Scotland
In the first part of this research the PIPS on-entry baseline assessment was used to describe what children know and can do when they start school in Scotland. This was presented in charts derived from Rasch analyses. It showed a very large range of cognitive development. In the top 1% were fluent readers and individuals who were very adept at working with numbers and had extensive vocabularies. They could read passages which include words such as "everyone" and do problems such as "what is 3 fewer than 7?". In the lowest 1% were children whose progress towards literacy had reached the stage of recognising the activities of reading and writing without themselves having skills in those activities. They knew mathematically relevant words such as biggest and smallest but had difficulty counting just a few objects.
The age, sex and home backgrounds of the children showed systematic links to the developmental levels and the chart below summarises the results for mathematics.
Figure 6 Summary chart for mathematics
There was no difference between girls and boys in their average starting points for mathematics but older children and those from more affluent homes had higher starting points. Although it is not shown on the chart, the amount of pre-school experience was positively related to the starting point in mathematics, but to a very small extent.
Important though age and home background were, they were small in comparison to the very large differences between pupils more generally. Similar results were found for vocabulary, reading and phonics.
A very weak relationship was found between the amount of pre-school experience and the PIPS scores at the start of Primary 1.