Children's development at the start of school in Scotland and the progress made during their first school year: An analysis of PIPS baseline and follow-up assessment data

6. Trends over time for 2012/13, 2013/14 and 2014/15

This section investigates trends over time for three academic years.

6.1 Age on entry

The chart below shows the distribution of age on entry for the three cohorts.

Figure 28: Comparison of age of entry by cohort

The distributions across the three years are very similar and as reported in Figure 6 the mean age remained constant at 5.0 years. There remained a higher proportion of older children starting school than would be expected, reflecting those whose parents had opted to defer entry for one year as discussed previously.

6.2 Start of Primary 1

6.2.1 Cognitive development

Figure 29 shows the change, in months, in total PIPS scores, early reading and early mathematics at the start of Primary 1 over the three academic years.

Figure 29: Months difference in start of year scores between academic years

	2013/14 compared with 2012/13	2014/15 compared with 2013/14
Total	-1.21**	-1.72**
Early Reading	-0.99**	-2.96**
Early Mathematics	-0.99**	-1.28**

^** Significant at the 1% level ^* significant at the 5% level

At the start of year, the scores for cognitive development declined between 2012/13 and 2013/14, and again between 2013/15 and 2014/15; this was statistically significant.

Between 2012/13 and 2013/14, the total score declined by 1.21 months of cognitive development and by 1.72 between 2013/14 and 2014/15. Looking more closely at early reading and early mathematics, the declines were similar for 2012/13 to 2013/14 but for 2013/14 to 2014/15, there was a larger decline for early reading than early mathematics. Over these same years, the demographics have changed a little in Scotland with an increase from 3.2% to 4.7% of children for whom English was an additional language. It is possible that this change is associated with the slight decline in PIPS scores however this is a hypothesis and it would need further exploration to establish a causal relationship. We were not able to identify children as having English as an additional language and so were not able to investigate the issue further.

Other research (http://www.gov.scot/Resource/0048/00486755.pdf) found an increase in children's vocabulary acquisition at age three years between 2004 and 2010 and a similar trend of increasing vocabulary might be expected within the current study. However, as said we found a drop in vocabulary over the three year period but it was very slight. For more detail, see Appendix D.

6.2.2 Cognitive measures with age, gender and deprivation

Figure 30 shows correlations of the cognitive measures over three years with the three key variables: age at the start of school, sex and deprivation level (SIMD).

Figure 30: Correlations between cognitive measures, age, sex and SIMD at the start of

Primary 1 by academic year

	2012/13	2013/14	2014/15
Picture vocabulary
Age	0.12*	0.10*	0.12*
Sex	0.10*	0.11*	0.14*
SIMD	0.18*	0.16*	0.19*
Early mathematics
Age	0.19*	0.19*	0.20*
Sex	0.02	0.02	0.04*
SIMD	0.23*	0.20*	0.21*
Phonics
Age	0.14*	0.13*	0.13*
Sex	0.10*	0.09*	0.12*
SIMD	0.20*	0.17*	0.20*
Early reading
Age	0.15*	0.12*	0.15*
Sex	0.06*	0.06*	0.09*
SIMD	0.18*	0.16*	0.18*

^** Significant at the 1% level ^* significant at the 5% level

There was some slight variation but the correlations can be considered to have remained stable over time^[13]. This means that age, sex and deprivation level did not differentially impact on cognitive development over time.

6.3 Progress during Primary 1

The differences over the years at the end of Primary 1 can be found in Appendix D, Figure D-3. This section focuses on progress during Primary 1, and the months of difference between academic years is shown in Figure 31.

Figure 31: Months difference in progress between academic years

	2013/14 compared with 2012/13	2014/15 compared with 2013/14
Total	0.68**	0.69**
Early Reading	0.39	2.28**
Early Mathematics	1.06**	1.47**

^** Significant at the 1% level ^* significant at the 5% level

Pupils' progress during Primary 1 in early reading and mathematics increased slightly over the three year period. These gains meant that there were no significant drops over the three years for mathematics levels at the end of Primary 1. For reading, a drop was still apparent although it was very small. It appears that the decline in children's cognitive development at the start of school over time was being compensated for by their time in school.

6.4 Personal and social development

6.4.1 Start of Primary 1

Although the number of respondents decreased noticeably in 2014 the scores at the beginning of Primary 1 increased very slightly across all areas as the two figures below show. The mean was the average score of the five-point scale and the standard deviation (SD) indicates the spread of scores. For example, the Adjustment comfortable mean score was 3.6 on the five-point scale at the start of 2012/13. The standard deviation was 1.08, which implies that 68% of children had scores between 2.52 and 4.68 (3.6 plus or minus 1.08). The Adjustment comfortable mean score rose slightly to 3.7 in 2013/14 and then again to 3.9 in 2014/15. Over the three years, the mean score rose by 0.3 points on the five-point scale. Figure 32 gives details of the other PSD measures.

Figure 32: Personal and social development scores at the start of Primary 1 for three academic years

	2012/13	2013/14	2014/15	Overall
Adjustment comfortable, n	659	625	477	1761
Mean (SD)	3.6 (1.08)	3.7 (1.08)	3.9 (1.05)	3.7 (1.07)
Adjustment independence, n	657	625	477	1759
Mean (SD)	3.6 (1.07)	3.6 (1.08)	3.7 (1.14)	3.6 (1.10)
Personal confidence, n	657	622	477	1756
Mean (SD)	3.5 (0.98)	3.6 (1.00)	3.6 (1.10)	3.5 (1.02)
Personal concentration (self-directed), n	657	623	477	1757
Mean (SD)	3.2 (0.98)	3.3 (0.99)	3.3 (1.09)	3.3 (1.02)
Personal concentration (teacher-directed), n	650	623	477	1750
Mean (SD)	3.2 (0.95)	3.4 (0.96)	3.4 (1.06)	3.3 (0.99)
Personal actions, n	650	623	477	1750
Mean (SD)	3.5 (0.96)	3.5 (0.92)	3.6 (1.03)	3.5 (0.97)
Social relationship (to adults), n	650	623	477	1750
Mean (SD)	3.4 (0.85)	3.4 (0.86)	3.6 (0.96)	3.5 (0.89)
Social relationship (to peers), n	650	623	477	1750
Mean (SD)	3.6 (0.86)	3.6 (0.88)	3.7 (0.98)	3.6 (0.90)
Social rules, n	650	623	477	1750
Mean (SD)	3.6 (0.89)	3.7 (0.85)	3.8 (0.98)	3.7 (0.90)
Social cultural awareness, n	650	623	477	1750
Mean (SD)	2.7 (1.04)	2.8 (0.93)	2.8 (1.10)	2.8 (1.02)
Social communication, n	650	623	477	1750
Mean (SD)	3.6 (1.06)	3.7 (1.02)	3.8 (1.12)	3.7 (1.07)

n=number of pupils; SD=standard deviation

The information in Figure 32 is shown graphically in Figure 33 below, which emphasizes the small degree of increasing scores on the five-point scales over the three years. The vertical axis shows the scores on the five-point scale.

Figure 33: PSD rating in the first term of Primary 1 over three years

6.4.1.1 Age, gender and deprivation

The correlations between age and PSD over time are shown in Figure 34.

Figure 34: Correlations between PSD and age at the start of Primary 1 for three academic years

	2012/13	2013/14	2014/15	Overall
Adjustment comfortable	0.04	0.11**	-0.02	0.05*
Adjustment independence	0.05	0.15**	0.09	0.10**
Personal confidence	0.04	0.15**	0.05	0.08**
Personal concentration (self)	0.06	0.14**	0.04	0.08**
Personal concentration (teacher)	0.04	0.14**	0.03	0.07**
Personal actions	-0.03	0.11**	-0.05	0.01
Social relationship (to adults)	-0.03	0.07	-0.04	0.00
Social relationship (to peers)	-0.06	0.10*	-0.05	0.00
Social rules	-0.02	0.09*	-0.04	0.01
Social cultural awareness	0.08*	0.09*	0.04	0.07**
Social communication	-0.01	0.06	0.04	0.03

^** Significant at the 1% level ^* significant at the 5% level

For each academic year, the correlations are very low and not statistically significant. There was virtually no relationship between the two variables. The low correlations did not change over time.

The correlations between PSD and gender are shown in Figure 35.

Figure 35: Correlations between PSD and gender at the start of Primary 1 for three academic years

	2012/13	2013/14	2014/15	Overall
Adjustment comfortable	0.10*	0.06	0.05	0.07**
Adjustment independence	0.20**	0.13**	0.18**	0.17**
Personal confidence	0.12**	0.05	0.13**	0.10**
Personal concentration (self)	0.29**	0.22**	0.28**	0.26**
Personal concentration (teacher)	0.28**	0.22**	0.24**	0.25**
Personal actions	0.27**	0.20**	0.29**	0.25**
Social relationship (to adults)	0.20**	0.15**	0.19**	0.18**
Social relationship (to peers)	0.15**	0.11**	0.14**	0.13**
Social rules	0.30**	0.24**	0.27	0.27**
Social cultural awareness	0.16**	0.10*	0.13**	0.14**
Social communication	0.13**	0.07	0.17**	0.12**

^** Significant at the 1% level ^* significant at the 5% level

On average girls scored higher than boys at all time points. Whilst the correlations between PSD and gender were quite low, several were statistically significant but the size of the correlations remained broadly stable over time.

Figure 36: Correlations between PSD and SIMD at the start of Primary 1 for three academic years

	2012/13	2013/14	2014/15	Overall
Adjustment comfortable	0.08*	0.00	0.02	0.04
Adjustment independence	0.05	0.05	0.11*	0.07**
Personal confidence	0.09*	0.03	0.06	0.06*
Personal concentration (self)	0.08*	0.08	0.10*	0.08**
Personal concentration (teacher)	0.10*	0.06	0.12*	0.09**
Personal actions	0.05	0.02	0.10*	0.05*
Social relationship (to adults)	0.14**	0.02	0.15**	0.10**
Social relationship (to peers)	0.05	0.00	0.09*	0.05
Social rules	0.10**	0.02	0.02	0.05
Social cultural awareness	0.15**	0.15**	0.14**	0.15**
Social communication	0.09*	0.10*	0.16**	0.11**

^** Significant at the 1% level ^* significant at the 5% level

Figure 36 shows the correlation between PSD and SIMD quintiles and was noted earlier when the data from 2012/13 was examined: the scores of children from the most affluent areas were significantly higher than those of children from the most deprived areas on several of the scales. Although many of the correlations between PSD and deprivation at the start of the academic year are statistically significant they are low, and the size of these correlations remained stable over time.