Making Sense Programme: final report

Annex G : Schools and local authorities’ data

Pupils with Dyslexia: Pupil Census

This paper is set out as follows:

• Introduction: this describes the background to the task and explains how the various statistical analyses were carried out.
• Results: this summarises the main results of the statistical analyses.
• Conclusions: here the results are discussed

1. Introduction

Data considered

2.1 Data from the annual pupil census on pupils in local authority funded schools with dyslexia recorded as a reason for having an additional support need (ASN) is used in these analyses.

2.2 This is not a direct analogue of the number of pupils in these schools who have been diagnosed with dyslexia. The pupil census figures may include pupils who do not have a formal diagnosis of dyslexia but who receive support as they exhibit dyslexic characteristics. It may also exclude pupils with dyslexia but who the school and/or local authority do not consider to have an additional support need resulting from this diagnosis. However, the figures are likely to be reasonably close to the number of pupils diagnosed with dyslexia so their use in these analyses is appropriate.

2.3 The core datasets are the number of pupils with dyslexia recorded as a reason for having an ASN in 2014 and 2018 in:

(a) primary schools in each local authority,

(b) secondary schools in each local authority.

2.4 Equivalent data for the intervening years were also considered and these data showed an increase in the number of pupils identified with dyslexia over the period 2014-2018. Therefore, comparing the 2014 and 2018 data will provide the clearest indicator of change.

2.5 The data for pupils with dyslexia in special schools were examined but these data comprise relatively small numbers of pupils. Therefore, for the purposes of this exercise the data from special schools were not included in the statistical analyses.

2.6 Data for pupils in independent schools was not available as they are not included in the pupil census. They are therefore also excluded from the statistical analyses.

Determining progress

2.7 As noted above, The Making Sense report asked local authorities to ‘improve the quality and use of data regarding the number of children and young people identified as having dyslexia.’

2.8 The data from the pupil census does not, by itself, shed any light on the quality and use of data. However, examination of the data can provide information on trends in the recording of dyslexia as a reason for ASN in local authority funded primary and secondary schools.

2.9 Two assumptions are be made to frame this analysis of the census data. Firstly, that there was an underreporting of pupils with dyslexia as a reason for ASN in 2014. Secondly, that variations between local authorities in the recording of dyslexia as a reason for ASN in 2014 was the result of incorrect and/or inconsistent reporting.

2.10 Using these assumptions, two indicators of progress which can be examined using the pupil census data can be formulated. These are whether there is:

(a) an increase in the proportion of pupils being recorded as having dyslexia as a reason for having an ASN from 2014 to 2018; and

(b) a reduction in the variability of numbers of pupils with dyslexia as a reason for ASN across local authorities (signalling greater consistency in reporting).

Statistics calculated

2.13 Given the data available in the pupil census the following statistical calculations, in Table 1, were carried out to provide evidence about 2.9 (a) and (b) above.

Table 1: Description of statistical analyses carried out.

Calculation	What information does it provide?
The percentage of pupils in each local authority recorded as having dyslexia in (a) primary schools and (b) secondary schools.	This is the core data set which enables comparisons to be made across local authorities and across primary and secondary schools. All other calculations are carried out on this data set.
Range This is the difference between the highest and lowest values in the data set.	The Making Sense report states (page 38) that ‘There is significant variability in levels (of recorded data in the pupil census) across different education authorities.’ This range provides one estimate of the variability in the data set. The larger the range the greater the variability.
Interquartile Range (IQR) The IQR describes the middle 50% of values when ordered from lowest to highest.	While the range provides an estimate of variability it can be heavily influenced by just one very atypical piece of data. The IQR is a more robust measure of variability because it is not influenced by extreme values. The IQR can be used to interrogate the data from 2014 to 2018 to determine if the IQR has decreased over this period which we would expect if there is a reduction in the variability of the data recorded.
Variance Variance is a measurement of the spread between numbers in a data set. It measures how far each number is from the mean.	Variance provides another measure of the variability of data. It has been considered here along with range and IQR as a check to determine if all three measures of variability are providing consistent results.

3.0 Results

3.1 Table 2 below provides a summary of the results of an analysis of the pupil census reports for 2014 and 2018 carried out using the statistical approaches described in Table 1

Table 2: Results of the analysis of the pupil census carried out in 2014 and 2018.

	2014		2018
	Primary	Secondary	Primary	Secondary
Percentage of pupils with an additional support need	19.31	20.82	25.39	31.71
Percentage of pupils with dyslexia as reason for ASN	1.10	4.02	1.36	5.61
Range	2.77	10.03	3.72	10.26
Interquartile range	0.90	2.12	0.93	3.08
Variance	0.42	4.57	0.77	5.31

Primary

3.2 In the primary sector, the percentage of pupils identified and recorded with dyslexia as a reason for ASN has increased from 1.10 to 1.36. This represents a 28% increase in the number of pupils with dyslexia as a reason for ASN. This increase occurred at the same time as an increase in the overall percentage of primary school pupils with an ASN (from 19% to 25%). These increases may therefore have been driven by the same factors.

3.3 The correlation between local authorities’ figures over the two periods is significant (r=0.7). That is, those local authorities recording the highest proportions of primary pupils with dyslexia as a reason for ASN in 2014 also had the highest proportions in 2018 (and vice-versa for those recording lowest proportions).

3.4 On all three measures of variability, the results in Table 2 indicate that variability has increased numerically over the four-year period. However, it is not clear if the increase is statistically significant so it is safer to conclude that variability has not decreased over the 2014-18 period in primary schools.

Secondary

3.5 In the secondary sector, the percentage of pupils with dyslexia as a reason for ASN has increased from 4.02 to 5.61, representing a 40 percent increase in the number of pupils with dyslexia as a reason for ASN. As with the increase in pupils recorded with dyslexia as a reason for ASN in the primary sector, this increase was concurrent with a rise in the overall proportion of secondary pupils with dyslexia as a reason for ASN (from 21% to 32%).

3.6 The correlation between local authorities’ figures over the two periods is significant (r=0.8). Those local authorities recording the highest proportions of secondary pupils with dyslexia as a reason for ASN in 2014 were doing so in 2018 (and vice-versa for those recording lowest proportions).

3.7 As with the primary sector, on all three measures of variability, the results in Table 2 indicate that variability has increased numerically over the four-year period. As with the results for the primary sector, it is not clear if the increase is statistically significant, so it is safer to conclude that variability has not decreased over the 2014-18 period in secondary schools.

Primary and secondary compared

3.8 In 2014, the proportion of pupils with dyslexia as a reason for ASN was 3.6 times higher in secondary schools than in primary schools. In 2018, the comparable figure was 4.1 times. Therefore, the gap between the proportion of pupils recorded as having dyslexia as a reason for ASN in the primary and secondary sectors has widened since 2014.

3.9 The variation in the proportion of pupils recorded as having dyslexia as a reason for ASN across local authorities is also larger in secondary schools than in primary schools. This is true for all of the measures of variance included in Table 2.

4.0 Conclusions

4.1 The analyses of the pupil census results of 2014 and 2018 indicate that the numbers of pupils being recorded with dyslexia as a reason for having an ASN has increased as a proportion of the school population over the four year period in both primary and secondary sectors, but more so in the secondary sector. However, variability across local authorities in terms of those identified with dyslexia as a reason for ASN has not decreased. So the conclusion of the Making Sense report of 2014 (page 39) that ‘Local authorities have a wide variation in the percentage of children and young people identified as needing support for dyslexia’ still holds.

As noted above, the analyses of the results of the pupil census do not provide information on the quality and use of data. The analyses reported here can only describe trends and not the underlying explanations.

End of report and annexes