Reform Scotland

Cultural resources and school attainment – Lindsay Paterson

The importance of children’s homes in their educational progression is widely recognised  – and was made very apparent during the home-schooling caused by Covid. But do all the potentially relevant aspects of home life matter equally? There are two kinds of home resource. There is the technology of learning, ranging from the most basic – having a quiet room in which to study – to the most advanced, such as having access to laptops and wifi. On the other hand, there are the cultural resources of the home, what the U.S. writer E. D. Hirsch has called cultural literacy. For each of these two, there is also the questions of how they relate to more economically fundamental aspects of a child’s home environment, and of how schools might compensate for poverty. Although laptops might not be affordable by families with few economic resources, cultural literacy costs less. So a key question is whether rich cultural resources overcome the educational effects of economically deprived living conditions.

Scotland and England provide a quasi-experiment on this question in the past decade because of their different paths of curricular reform. On the one hand, the dominant strand of English reform was shaped by Michael Gove when he became Secretary of State for education in 2010, and taken further by Nick Gibb as schools minister until he was dropped by Boris Johnson in autumn 2021. Inspired by Hirsch and others, these reforms tried to ensure that schools would provide the same kind of cultural literacy for all students. The changes shifted the attention of the curriculum to a core of knowledge, away from skills and from any direct concern with pastoral matters such as emotional well-being. The principle was that well-being and skills would develop best from a sound understanding of a powerful intellectual inheritance. In some respects, Mr Gove and Mr Gibb were merely intensifying a shift in policy that had already started in England during David Blunkett’s tenure at Education in the first years of the Blair government.

Scotland’s Curriculum for Excellence has moved in a quite different direction, eschewing what the Scottish debate tends to disparagingly refer to as academic knowledge, and towards not only a greater variety of skills – such as those that might be required in the workplace – but also towards well-being as a deliberate aim. This policy, too, has commanded cross-party support, even though that has implicitly brought the Britain-wide parties to a different position in Scotland from that taken by their counterparts in England.

Both the English and the Scottish policies have claimed to be able to overcome social inequalities of learning. In England, this is on the basis that cultural literacy can be provided by schools to compensate for its lack at home. Scottish policy is founded on the principle that students’ self-confidence, aspirations and belief in their own capacities is an effective way of overcoming the depressing effects of poverty

PISA data on educational resources at home
Is there any evidence that might allow a comparison of the effects of these two broad approaches to the curriculum? Unfortunately, the best kind of comparison is not available because of the loss in Scotland of good-quality series of statistics (as Reform Scotland’s Commission on School Reform has frequently pointed out). There is nothing in Scotland that would allow a comparison with the publicly accessible data from England’s National Pupil Database. However, the three-yearly data from the Programme for International Student Assessment (the PISA studies) do give some limited opportunity to compare. In each survey since 2006, the samples have been large enough to give reliable evidence – on average about 2,800 in Scotland and 4,700 in England. The assessments in the PISA surveys are of reading, mathematics and natural science. Because they are done in the same way in every country, using them to compare Scotland and England is more valid than other kinds of comparison (such as trying to compare Highers and A-levels).

Home cultural resources

In the most recent survey (in 2018), the students were asked whether they had any of five key indicators of cultural resources at home:

  • works of classic literature (in the British surveys giving the example of Shakespeare);
  • books of poetry;
  • works of art;
  • books on art, music, or design;
  • musical instruments (for example guitar or piano).

These are used here to construct a scale of cultural possessions, with summary values ‘one or none’, ‘two’, ‘three’ and ‘four or five’. (The distribution is shown in Table 1 below.)

This scale is not ideal for our purposes. In particular, it would clearly be desirable to have some information about scientific resources at home. Nevertheless, in the absence of anything better, the intended interpretation of this scale is not so much for its overt content as for its suggestion of a general ethos. A home with all these resources is likely to be one which is educationally stimulating in all respects, not only in the explicit ways referred to in these measures.

Home educational equipment
A scale of educational resources at home was similarly constructed from questions about whether, at home, the student had:

  • a desk to study at;
  • a quiet place to study;
  • a computer available for school work;
  • educational software;
  • books to help with school work;
  • technical reference books;
  • a dictionary.

The summary values here are ‘four or fewer’, ‘five’, ‘six’ and ‘seven’ (with distribution in Table 2 below). Again, this scale is not ideal, though it is probably a better indicator of a range of educational resources than is the cultural scale for culture.

Home economic circumstances
The PISA survey does not collect financial data about students’ homes, but it does include a measure of the occupational class of parents. This socio-economic index has been developed over many decades by researchers mainly in the Netherlands specifically for the purpose of making international comparisons. Here, we use the higher of the father’s and mother’s value of the index (or of a sole parent where information on only one was available), and divide the resulting scale into four quarters. Typical occupations in each quarter are:

class 1       sales staff, waiter, hairdresser, personal carer, cleaner, construction worker, vehicle repairer, lorry driver;
class 2:      retail manager, technician, secretary, nursing associate, construction manager, ambulance worker, unpromoted police officer;
class 3:      nurse, primary school teacher, social worker, software developer, marketing manager;
class 4:      scientist, engineer, architect, accountant, doctor, secondary teacher, university teacher, lawyer.

The main question is whether and how the measure of home cultural resources relates to students’ attainment. The results summarised here were similar for male and female students, and so they have been presented without any differentiation by sex.

Cultural resources and attainment
England has a higher proportion of students in the highest category of cultural resources than Scotland, a difference that is seen in each of the occupational classes. This is shown in Table 1. Overall, 39% of students in England are in the highest category of cultural possessions, whereas in Scotland the proportion is 26%. In the highest-status occupational class, these proportions are 53% and 39%.

Figure 1 shows the average performance in the reading tests in Scotland and England classified on two dimensions: by the index of occupational class, and by the index of cultural possessions. (In each year, in each subject, the scale of attainment is defined to have an average of about 500 across all the economically developed countries.)

For example, in the left-most graphs, there are students from families in the lowest-status classes. Within each of these graphs, the left-hand bar is for students in these classes whose households have the fewest cultural possessions. So these students are disadvantaged both economically (through the occupations of their parents) and culturally. At the other end of Figure 1 are students who are highly advantaged in both respects. The narrow grey lines at the top of each bar give some assessment of random-sampling variability – indicating, for each category of class and culture, the range in which the true value probably lies (with 95% probability).



There are very large differences associated with class: the blocks of bars rise from left to right across the panels of the Figure, similarly in England and Scotland. There are also large differences associated with culture: the bars also tend to rise from left to right within each panel. But there is a difference in this cultural respect between Scotland and England. In Scotland, there is a gradient across cultural categories in each class block. In England, that cultural gradient is clear only in relation to the highest cultural category. In the lowest-status class (at the left) there is hardly any cultural gradient in England, in contrast to Scotland.

The most pronounced difference between specific bars in the two countries is for students in the higher-status classes, but who have few cultural possessions at home: these are the dark-blue bars of the panels labelled Class 3 and Class 4. The bars for England there are higher than the Scottish bars. Indeed, a student from the highest class in England who has low cultural resources from home does at least as well as all but the highest-culture Scottish students: that is, the dark-blue and bright-blue bars for the highest class in England are at least as high as all but the brown bars in the other classes in Scotland.

The only case where the Scottish bar is higher than the corresponding English one is for people in the lowest-status class who have many cultural possessions at home (the brown bars in the left-hand panels). In fact, these Scottish students have attainment almost as high as students with less culture in all but the highest-status class: the brown bar in the left-most Scottish panel is as high as all but the brown bars in the middle two panels for either country.

In short, home culture in England has less of an effect on reading attainment than it does in Scotland. Scottish students in the lowest-status class who have rich cultural possessions overcome the general educational effect of being in that class; but (from Table 1) only 16% of Scottish students in that class have that advantage.

The patterns for mathematics and science were, if anything, even stronger. The mathematics graphs looked very like the reading ones, but with no combination of class and culture in which students in Scotland performed better than corresponding students in England. Figure 2 shows science. Here, attainment at the lowest level of home culture (the dark-blue bars) is higher in England than in Scotland at all levels of class. As with mathematics, there is no reliable evidence that any category of Scottish students had higher average attainment than the corresponding English students. This relationship of the cultural scale to science attainment (despite there being no explicit reference to science in the components of the scale) tends to vindicate our interpretation of the scale as reflecting something about the general ethos of the home.

Educational resources and attainment
Next we replace home culture with the measure of home educational resources. The distribution of these resources is shown in Table 2. They are more common than the cultural possessions (Table 1): over one third in Scotland – and one half in England – have all seven. The difference between Scotland and England is less than for cultural possessions, especially in the highest-status class.

When average attainment is calculated in relation to these educational resources, there is certainly a gradient at each level of occupational class, but there is no difference between Scotland and England, unlike the cultural scale: see Figure 3 for reading. The same was found for mathematics and science.

Cultural resources and attainment over time
Some further insight into this may be had by tracking these differences over time, including now also the PISA data from 2006, 2009, 2012 and 2015. The definition of family class was broadly the same in each survey, but the survey questions about cultural possessions at home were fewer up to 2012 than subsequently, asking only about the first three topics noted for 2018 above (classic literature, books of poetry, and works of art). So for consistency in this time series, the 2018 and 2015 definition of cultural possessions has been revised to include only that information. Figure 4 shows the trends for reading. To make the patterns clearer, the graphs are restricted to the highest-status and lowest-status classes (the two columns of the array in Figure 4), and to the highest and lowest categories of cultural possessions (the red and blue lines). The patterns for the other classes and cultural categories were intermediate between those shown.

The main theme to emerge from Figure 4 is that the cultural difference is fairly constant in Scotland but diminishes in England. That reduction in England is, in both of the classes, in part because of a rise in the group with the weakest home culture (the blue lines), but in the lowest-status class it is also because of a weakening performance in the group with the strongest home culture (the red line in the top-left graph). The Scottish group that out-performed the corresponding English group in 2018 – the low-status, high-culture students (who are the red line in the bottom-left panel of Figure 4) – did not improve their position over time, and so we can now see that they are ahead of that group in England in 2018 only because of the English decline.

The patterns for mathematics and science were similar to this, as illustrated for science in Figure 5. Again, there was no change in Scotland in the association of attainment with home culture. The association weakened in England. And the attainment of the highest-culture group in England slowly declined. In both mathematics and science, the highest-culture group also declined in Scotland.

The importance of home culture
What conclusions might we draw from this? One is the importance of home culture of the kind that is recorded in the survey questions which we have used here. It seems to mean something different from educational resources as conventionally measured, such as textbooks and laptops. These certainly matter, but they matter in the same way in both Scotland and England. There is a national difference in the way in which cultural resources relate to attainment, in all three domains.

The second conclusion is about policy in England. The Gove-Gibb reforms may be a plausible explanation of the apparently weakening importance of home culture. It is noticeable that the change over time in Figures 4 and 5 became clear in the 2012 or 2015 surveys, which is what we would expect if the reforms were the explanation, because these would have started to have an effect in secondary schools around the middle of the last decade. But the evidence from PISA is that the most striking impact was not on children from the lowest-status social classes but rather on those in affluent families who had few cultural possessions at home. There was also a positive effect on children from low-status classes who had weak home culture. The group that did not move ahead were the high-culture, low-status category, where the downward-sloping red lines in the top-left graphs of Figures 4 and 5 might cause Mr Gove and Mr Gibb some disappointment.

The third conclusion is about Scotland. Even these partial improvements in England were completely absent. There was a cultural gradient in each class, and it did not change over time. This would be consistent with a weaker effect of school culture in Scotland than of home culture.

Overall then we might say that the policy attention to high culture in schools in England in the past decade or more seems to have had some impact in compensating for an absence of that culture at home, even though not as straightforwardly as Mr Gove and Mr Gibb might have hoped. Not building on the cultural resources of those working-class families which value high culture seems a particular failure. But Scottish policy cannot even claim that partial success. If there is any indication of Scottish success here it is by parents, not schools. Those parents in the lowest-status class who value cultural richness have enabled their children to counteract quite a lot of the general educational effects of economic disadvantage. But the consequence, when looked at over time, is that home cultural resources remain as strongly influential on students’ attainment as ever. For the children of all social classes who are fortunate enough to have parents who value culture of this kind, the home provides some educational insurance. For others, Scottish schools are failing to compensate for the cultural disadvantage.

Lindsay Paterson is Professor of Education Policy, School of Social and Political Science, Edinburgh University.