Growing Up in Scotland: changes in language ability over the primary school years

3 Examining the Gap in Language Ability

3.1. Introduction

This chapter examines the gap in expressive language ability between children from different socio-economic backgrounds at the time they were in Primary 6. The gap is considered according to measures of household income, area deprivation and parental education.

3.2. Expressive language ability by social background - Primary 6

Figures 3-1 to 3-3 display the standardised vocabulary scores of children in Primary 6 by household income, area deprivation and parental education. The distribution of vocabulary scores is shown using box plots (Figures 3-1, 3-2 and 3-3). These are explained below. The numbers informing the box plots are provided in Tables B1 to B3 in Appendix B.

As noted in section 2.2, the scores shown for each socio-economic group are relative to those recorded for all children. A score of 0 represents the average vocabulary score recorded for all children, irrespective of social background; a score of -1 represents a score which is one standard deviation below the average for all children and a score of 1 represents a score which is one standard deviation above the average for all children. Children scoring between -1 and 1 make up roughly 70% of all children, while those with scores above 1 and those with scores below -1 make up approximately 15%, respectively.

In each chart, the average (median^[6]) score for each socio-economic group is represented by the horizontal line that divides the box into two parts. Half the scores are greater than or equal to this value and half are less. For example, in Figure 3-1, the average vocabulary score for children in the lowest income quintile was -0.25. The box for each socio-economic group represents the middle 50% of scores for that group. The boxes and median lines allow us to compare average vocabulary ability across different socio-economic groups.

The lines extending above and below the boxes - the upper and lower 'whiskers' -represent the range of scores outside the middle 50%. That is, the highest point of the top whisker for each group represents the highest score for children in that particular group while the lowest point of the bottom whisker represents the lowest score for children in that particular group (no outliers were removed). This allows us to consider variations in the full range of ability within each socio-economic group and not just differences in average scores.

Regardless of the characteristic considered, the graphs show a clear difference in vocabulary ability by social background. In 2014/15, among children in Primary 6, those in higher income households, in less deprived areas, and those whose parents had higher levels of educational qualifications had, on average, better vocabulary than those in lower income households, those in more deprived areas and those whose parent(s) had lower levels of educational qualifications.

The largest differences are visible by parental education. As shown in Figure 3-3, at the time they were in Primary 6, children whose parent(s) had lower Standard Grade qualifications or below had an average vocabulary score of -0.49, compared with 0.25 for children whose parent(s) had a degree. The smallest differences are seen in relation to area deprivation (Figure 3-2), where the average vocabulary score among children in the most deprived areas was -0.29 compared with 0.11 for children in the least deprived areas. Clear differences were also visible by household income (Figure 3-1). Here, the average vocabulary score among children in households in the lowest income quintile was -0.25, compared with 0.18 for children in households in the highest income quintile.

Thus, the charts illustrate a clear gap in language ability by social background when considering average scores for each socio-economic group. Nonetheless, they also illustrate substantial variation within socio-economic groups. As noted above, this variation is illustrated by the boxes which mark the middle 50% of scores for each group but also, in particular, by the whiskers at the top and bottom of each box which illustrate the range of ability within each group.

For each of the three social characteristics considered, the charts clearly show that not all children in disadvantaged circumstances did poorly - with significant proportions returning scores above average - and not all children in advantaged circumstances did well - with many having scores below average. It is clear that many children in the least advantaged groups had vocabulary ability as high as or higher than their more advantaged peers whilst some children in the most advantaged groups had poorer ability than some of their less advantaged peers. For example, as shown in Figure 3-1, some children in the bottom income quintile had scores as high as 3.21, higher than any score recorded for children in the highest income quintile.

Figure 3‑1 Standardised vocabulary ability score by household income - Primary 6

Base size (unweighted/weighted) = 2553/2519. See also Table B-1 in Appendix B.

Figure 3‑2 Standardised vocabulary ability score by SIMD - Primary 6

Base size (unweighted/weighted) = 2726/2698. See also Table B-2 in Appendix B.

Figure 3‑3 Standardised vocabulary ability score by highest parental level of education - Primary 6

Base size (unweighted/weighted) = 2722/2694. See also Table B-3 in Appendix B.

3.3. Summary

In summary, the analysis showed a gap in expressive language ability between children in the most and least advantaged circumstances towards the end of primary school (in Primary 6). These findings are in line with existing research which has demonstrated a strong correlation between a pupil's socio-economic status and their educational attainment in Scotland, with pupils from disadvantaged backgrounds having a higher risk of not succeeding in school (Scottish Government, 2017a).

Similar to an earlier GUS report which examined the gap in cognitive ability over the pre-school period (Bradshaw, 2011), the largest gap was seen in relation to parental level of education, with smaller but still substantial gaps according to household income and level of area deprivation.

Notably, though, although there were clear differences in average language performance between children in different socio-economic groups, there was also significant variation within these groups. Despite the poorer average performance of children from the most disadvantaged background, some of these children were performing well. Conversely, despite a better average performance, some children in the most advantaged groups were performing less well than some children in the most disadvantaged groups. These variations indicate that social background, whilst an important factor, is not the only factor that influences language ability. Although being from a disadvantaged social background increases the risk of poorer language skills, it does not equate to poorer language skills for all children from disadvantaged backgrounds.