The experience of school in Scotland, 1970s to 1990s

Reforms to secondary schooling in the 20th century are most commonly discussed in relation to structures — the extension of secondary education to all students in the first half of the century, and the ending of selection into different kinds of school after the 1960s. Yet reformers also sought to give students a more satisfactory experience of school. Understanding statistically the changing experience which students had of secondary school following the reforms of the 1960s requires a lengthy time series of survey data collected contemporaneously and the capacity to link that information to evidence on attainment and on demographic factors such as sex and socio-economic status. Data from a unique series of such surveys in Scotland is used to investigate whether secondary schools became more humane in this period, whether students were more engaged with their schooling, and whether they thought that schools prepared them for life after they left. The conclusions are that the long-term aspirations of reformers to make schooling more satisfactory for students were broadly achieved. There were indeed improvements of experience and sentiment of these kinds, and they extended to students at all levels of attainment, to both sexes, and to all levels of socio-economic status.


Introduction
Educational reform in developed countries in the 20th century is most often discussed in terms of structures or attainment.It is not to detract from the fundamental importance of such questions to say that statistical attention to the lived experience of schooling has been less common.The intention of making schooling more humane was as central to the aims of radical campaigners for what was earlier called a 'new education' as was the provision of wider opportunity.These ideas then gradually entered the mainstream of policy by the 1960s to become part of the understanding of a coherent programme of reform.
This article uses a unique series of mostly biennial surveys of school students in Scotland, mainly covering the period from the mid-1970s to the end of the century.Unusually, these surveys contain not only information about attainment and such demographic factors as sex, social class and parental education, but also data on students' attitudes towards schooling.Although few of these questions were asked in every survey year, enough of them were asked in a sufficiently standard form over a attempted by well over twice that proportion, and so further, more fundamental reform was put in place: Standard Grade courses and assessment, which were intended to provide a suitable curriculum for all levels of ability, and thus may be interpreted as a further attempt to adjust schooling to meet the needs of individual students (Croxford, 1994;Gamoran, 1996;Paterson, 2003;Croxford & Howieson, 2015: 143-144).Vocational courses that were developed from 1984 onwards had similar aims (Raffe, 2009).Hartley (1987: 125) argued that a stronger attention to student motivation was intended to counter alienation from school, a challenge that was becoming particularly urgent because the rise in youth unemployment broke the previous direct links between schooling and work.
The Ordinary and Standard Grade courses were taken mainly in the middle years of secondary, and they probably contributed to the steady rise in the rate of staying in full-time school education after the minimum leaving age, which was raised to 16 in 1972-3.Rising staying on generated new work for guidance, as choices had to be made about courses in post-compulsory schooling.Moreover, as the youth labour market collapsed in the 1980s, a new set of challenges emerged for those school leavers who were not going to university.Various training schemes were set up by government, initially of poor quality but steadily improving (Raffe, 1987).Navigating these options, too, required guidance, but questions about motivation and thus also about student behaviour never went away (Munn et al., 2004a).Nevertheless, by the end of the century there was a growing sense that school was indeed for everyone: in the mid-1970s, hostility to school among low-attaining students was widespread (Gow & McPherson, 1980).A couple of decades later, three-quarters of students were happy in school (Wilson et al., 2004: 48).
Nevertheless, there were also more idealistic concerns, a residue of the more radical aspects of the new education.Part of the pedagogical change was a shift to different kinds of learning and assessment-greater use of project work, a gradually reducing role for examinations, a growth of the idea that assessment might promote learning rather than be merely summative (Simpson & Hayward, 1998: 456).There was also a slow growth of education for citizenship, which emphasised that students were citizens in their own right.The main context for this in Scottish secondary schools was the subject of Modern Studies, which had been developed from the late 1950s in part as a way of preparing students for democratic participation (Maitles, 2009).
The purpose of the analysis reported in this paper is to use a lengthy time series of contemporary surveys to track these changes in the nature of students' experience of secondary school.The intention is not to look in detail at attainment or at rates of progression, and nor can the data allow an assessment of the larger democratic aims.The purpose is more modest-to consider what school leavers thought, as a means of providing insights into the experience of school.

Methods
Scotland provides a uniquely long series of surveys of school leavers with a sufficient level of detail for our purposes here, covering the period from the early 1950s to the end of the century.We use 11 surveys which will be referred to by the date at which School experience in Scotland, 1970s-1990s 3 their members turned 16: 1952Scotland, 1970s-1990s 3 their members turned 16: , 1974Scotland, 1970s-1990s 3 their members turned 16: -6, 1976Scotland, 1970s-1990s 3 their members turned 16: -8, 1978Scotland, 1970s-1990s 3 their members turned 16: -80, 1980Scotland, 1970s-1990s 3 their members turned 16: -2, 1984Scotland, 1970s-1990s 3 their members turned 16: , 1986Scotland, 1970s-1990s 3 their members turned 16: , 1988Scotland, 1970s-1990s 3 their members turned 16: , 1990Scotland, 1970s-1990s 3 their members turned 16: , 1996Scotland, 1970s-1990s 3 their members turned 16: and 1998. .These surveys have been used in several of the publications cited above, but not hitherto as a single series.The 1952 survey was a birthcohort study, and was administered in respondents' homes (Macpherson, 1958).The surveys 1974-6 to 1980-2 were leavers' surveys (Gray et al., 1983: 16-23).The 1974-6 survey covered pupils with the full range of attainment only in five regions of Scotland, which included around three-quarters of all pupils; only that part of the survey is used here.The surveys after 1982 were cohort surveys, based on a sample of students in the fourth year of secondary school who were then followed up over subsequent years.Some data for these surveys were obtained from the time series assembled as part of the project 'Education and Youth Transitions' (Croxford et al., 2007).For the surveys in the 1970s and later, the questionnaires were sent to pupils' home addresses and returned by post directly to the survey organisation (the Centre for Educational Sociology until 1990, and ScotCen Social Research after that).Students were assured in the questionnaire that their replies would be treated anonymously, so that, in particular, their school, teachers and parents would not be able to see their answers to the questions.The surveys had response rates at the relevant sweeps ranging from nearly 100% in 1952, through 80% for the leavers' surveys between 1974-6 and 1982, to around 65% for the surveys from 1984 onwards.
Sex is available in all surveys.Parental education is recorded as the age at which the parent left full-time education (summarised into 15 or younger; 16; 17 or older).In every survey it is available for both parents.Social class is the Registrar General social class of the father.Attainment is measured as the number of passes in the mid-secondary examinations that were available in the year in question.These were Lower Grade in the 1952 survey, Ordinary Grade in the surveys from 1976 to 1990, and Standard Grade from 1986 to 1998.Because we are not investigating attainment as such, but rather whether school experience differed according to the level of attainment, this measure is standardised to have mean 0 and standard deviation 1 within each survey year.
For school experience, we use only those measures that were available over quite a long period that includes the late 1990s, at least the early 1980s, and preferably also the 1970s.The only exception was for corporal punishment, which was recorded only up to its abolition (and thus not in surveys after 1982).Only questions that are relevant to a fully representative sample of the target population in the specified years are included.However, for several variables, the question was asked only of people who had stayed on in school beyond age 16.These are converted into variables that describe all respondents by expressing them as a percentage of the whole sample who gave a particular reason for staying on.For example, we then have a variable which shows what percentage of the sample not only enjoyed school but also stayed on because of that.Some questions in some years were asked only of a randomly selected subset of the sample.This does not affect representativeness, only sample size.The sample sizes for each variable in each year are shown in Tables 1-3.
The variables may be grouped under three headings.Each variable is re-coded in the way indicated to give a dichotomy.School experience in Scotland, 1970s-1990s 5

Engagement with school
There are two meaures available for the whole series 1976-1998: • Whether or not the respondent stayed on in school beyond age 16 (which is avail- able also for 1952); this is not analysed in detail here, and is provided only to provide context for the other variables.
• Whether the respondent truanted (even for 'a lesson here and there').

L. Paterson
The remaining variables are recorded only in the specified years: • Whether or not school was worthwhile (1980, 1990, 1996, 1998).Typical wording (from 1980) was: 'on the whole, do you feel your last year at school was worthwhile?Yes/No'.An approximately similar question (about being happy in school) was asked in 1952.
• Stayed on to study specific subjects (1980, 1986, 1988, 1990, 1996, 1998).Typical preamble (from 1980) was 'why did you start a fifth year?', with an option 'I planned to do subjects for Highers' (the colloquial name for the main school courses at that stage).
• Stayed on because always assumed would stay on (1980, 1986, 1988, 1990, 1996,   1998).Typical preamble as for the previous variable, with option 'I had always assumed would start a fifth year'.
Typical wording (from 1980): 'my teachers helped me to do my best: true/untrue'.
Preparation for life after school • Teachers taught student things useful in a job (1976, 1996, 1998).Typical word- ing (1976): 'how much did teachers help you to learn things that would be useful to you in a job' (grouping 'a lot' and 'quite a lot').
Although constructing the variables in each group into a continuous index might have allowed the analysis to be more succinct, there was not enough overlap of years School experience in Scotland, 1970s-1990s 7 to allow this to track the same length of time as can be provided by the individual measures.For example, in the 'school environment' group, the first four measures are available together only in 1996 and 1998, and so a composite index could not track change.Moreover, the correlations of the measures was not strong enough to justify reducing them to single indexes.Applying principal componments analysis to the available attitudinal variables (that is, omitting the variables recording truancy, staying on and corporal punishment, and using pairwise deletion of missing values for years where variables were not available), the first component explained 58% of the variance for engagement, 38% for environment and 48% for the future.None of these reaches the threshold of 75% that would conventionally justify retaining only one component (Krzanowski, 1988: 67).
We model the variables by linear regression using the package 'svyglm' in R.This allowed the clustering of students into schools to be taken into account, and also allowed the analysis to be weighted.For all the surveys from 1974-6 to 1998, there were post-stratification weights to compensate for varying rates of response; the weighting categories were sex by attainment (e.g.Gray et al., 1983;Croxford et al., 2007: 7).For several surveys, the sampling fraction varied by attainment and by region, and so the weights take account of this design as well; the sampling fraction also varied by year, and so the weights were standardised to have the same sum in each year (which affects the predicted proportions but not the effective sample sizes for the analysis of deviance).We report the results as analysis of deviance and as predicted proportions attaining the specified threshold.Analysis of deviance tables are shown using Type II tests (with the 'Anova' function from the 'car' package in R), which are the results of dropping each term in turn from the model shown in the table.

Analysis
Tables 1-3 describe the changes over time in students' reports of their experience of school.Table 1 shows clear evidence of growing engagement.On the one hand, increasing proportions of students chose to be in school, and to be there for longer.The proportion who truanted decline from 64% in 1976 to 43% in 1998.Serious truanting of staying away for days or weeks at a time also fell, from 15% to 6% (not shown in the table).The choice to stay on beyond age 16 rose from 14% in the early 1950s (when the minimum leaving age was 15), through a third in the 1970s (by when it had been raised to 16), and reached a half in the late 1980s and two-thirds at the end of the century.
The reasons which people gave for staying on are consistent with a growing attachment to school.The proportion who stayed on because they enjoyed school rose from 14% in 1980 to nearly three times that in the late 1990s.That was partly because a similarly large minority had come to think that staying on was the norm, in the sense that they could not imagine doing otherwise.
Although these variables indicate improvement, the one variable in Table 1 which shows no change-judging school to be worthwhile-is stable at a level of widespread approval of school.Generally positive attitudes to school were also found 30 years earlier: in 1952, 91% were happy (54% being 'fairly happy' and 37% 'very happy').

L. Paterson
The stability of broad satisfaction despite very extensive social and educational change in this period is striking.Leaving school at the first opportunity in the earlier years was perhaps more to do with economic necessity than with attitudes to school.We return to this point about stability below.
The pattern for staying on to study specific subjects may confirm this.It fell and then rose, from a third in 1980 to a sixth in the mid-1980s, then reaching over a half in the late-1990s, after the new, broad curriculum had been fully developed.The pattern of this variable may suggest that, when staying on rose from the mid-1980s, the reasons were initially more diffuse than they had been.In the 1970s, for the minority who stayed on, the main reason to do so had been the intention to study specific subjects.In the late 1980s, when the rate of staying on rose again, there was a wider range of reasons, for example to gain qualifications for employment (as we will see in connection with Table 3).Only the development of courses for that wider range of students could bring the attention back to specifically educational reasons in the 1990s.Again we return to this below.
One reason for the generally growing attachment to school may have been a strengthening sense that school valued students and that students valued school (Table 2).The proportion who said that teachers gave them confidence doubled to around 70% between the mid 1970s and the end of the century.The proportion who were in social circles that took school seriously rose to over a half, probably one reason why, in column 6 of Table 1, there was a rise in staying on because that was the norm.
As with the persistently high proportion who judged school to be worthwhile, there was no change in the percentage reporting that teachers helped them do their best, remaining at the very high level of three-quarters from the mid-1970s to the late 1990s.So the more positive evaluation of the school ethos in Table 2, and the greater engagement with school noted in Table 1, may be due to greater attention by teachers to students as whole people-giving confidence to be independent (an all-round quality) rather than just helping with study.Although, in Table 2, there was a growing perception of unacceptable disruption from other pupils, respondents seem to have been able to distinguish between the educational practices of the school as a source of teaching and potential learning, and the social context in which these activities took place.This sense of being valued personally by the school was no doubt helped by the fall and then the abolition of corporal punishment (not in the table).The proportion who reported ever having been punished in this way fell from 80% in 1976 to 61% in 1982, on the eve of the abolition.Yet having received corporal punishment was quite consistent with positive evaluation of school.In 1978, for example, 74% of people who had received corporal punishment reported that teachers helped them to do their best, not much lower than the 84% among people who had never received corporal punishment.
A growing proportion of students also felt that school was preparing them for life, despite policy makers' recurrent concern that this was not happening (Munn et al., 2004b).In Table 3, the proportion who reported that teachers taught things that    School experience in Scotland, 1970s-1990s 11 Contrary to many beliefs at the time (summarised by Paterson & Raffe, 1995), the rise in staying on was not due to the rise of youth unemployment or to the inadequacy of the training schemes which government put in place to try to deal with that: fewer than 1 in 20 gave that as a reason, even in the period of high youth unemployment in the 1980s.For a growing minority of students, school gave them unprecedented choice over their own futures, enabling them to postpone decisions by staying on: this rose from 18% in 1980 to around a third by the late 1990s.That there was a rise in the early 1980s in all the reasons for staying on except to study specific subjects does tend to confirm further that a diffuse change was taking place in students' attitudes to postcompulsory school.As well as being for pursuing further study, school was also becoming a route into the labour market, an opportunity to reflect on future life and an experience to be enjoyed with friends who felt similarly about it.
The next questions are whether this improving experience was shared equally, first with respect to attainment and second in relation to sex and social class.Summaries of the analysis of the statistical effects of attainment are given in The models assess whether attainment is associated with each dependent variable, and also whether any such association varied over time.The models are not intended to imply any causal impact either way; presumably there are influences in both direction, which would require longitudinal data at the level of the individual to be investigated.The purpose is descriptive.The results are illustrated in Figures 1-3 by predictions at three schematic levels of attainment: in each year, the mean, and the mean plus or minus half of a standard deviation.
In Table 4, all the measures of engagement with school show strong evidence of an association with attainment and of change in that over time.Most of the criteria improve over time at all levels of attainment.Where the level was already very high at high attainment, the level at mean and at low attainment caught up.This is illustrated in Figure 1 for the belief that school was worthwhile (including being 'very happy' or 'fairly happy' at school in 1952): the gap between low and high attainment was only 3 points for the different measure in 1952, and then fell from 10 percentage points in 1980 to 7 in 1998.Notably, though, over three-quarters of even quite low-attaining students already in 1980 saw school as worthwhile.
A corollary of these points is that, where satisfaction was not uniformly high, improvement at low attainment rose at the same rate as improvement at mean or higher attainment.This is illustrated in Figure 2, for staying on because of enjoying school.At low attainment, the percentage was merely 8% in 1980, still below 20% up to the mid-1980s, but then rose to around 37% in the late 1990s, more than a fourfold increase.At high attainment, the increase was from 23% to 53%.There was a similar pattern for staying on for the reason that doing so was always assumed.On staying on to study specific subjects, the dip in the mid-1980s (in Table 1) did not happen with students of low attainment.The one exception to these patterns was on truancy.Although there was a sharp fall at all levels of attainment, it was steeper at high attainment than at low.Thus the difference increased from 10 points to 14.
On school environment, summarised in Table 5, there is again strong evidence of a relationship with attainment and of change in the relationship.Improvement at low attainment was greater than at high levels.For example, for reports that teachers helped the respondent to do their best, the percentage at low attainment rose from 69% in 1974-6 to 78% in 1998, while remaining stable at 82-83% at high attainment.A similar pattern was evident for the proportion who reported that their friends took school seriously, which rose from 55% in 1980 to 62% in 1998 at high attainment, but more steeply from 40% to 54% at low attainment.The pattern for reporting that teachers gave the respondent confidence was only weakly related to attainment.For disruption in class, the attainment differences narrowed because reported conditions worsened most at high attainment (27% in 1980 to 42% in 1998 at high attainment, but 37% to 48% at low attainment).
In the analysis of opinions about how school related to students' future life, summarised in Table 6, there is again evidence of an association with attainment, and of a change over time.Staying on to improve qualifications became more important at all levels of attainment: at low attainment, the proportion rose from 11% in 1980 to 60% in 1998; at high attainment, the change was from 21% to 88%.Not yet having School experience in Scotland, 1970s-1990s 13 decided about the future grew much more at low attainment (10% to 34%) than at high (29% to 48%), reinforcing the point that a greater range of students was being given the opportunity to choose a future life.++ 20 d.f. for column 1; 45 d.f. for columns 2 and 3; 25 d.f. for columns 4-6.

L. Paterson
Staying on because of a lack of immediate employment prospects was rare at all levels of attainment: at low attainment, this reason was given by 3% in 1980 and 11% in 1998; the proportions at high attainment were 7% and 13%.There was only weak association of attainment with reporting that school had taught things relevant to jobs.
So, in most respects the improvements were at all levels of attainment.Our final body of evidence then relates to whether this was also true for both sexes and at all levels of socio-economic status (SES).The socio-economic measures changed over time in familiar ways.For example, the proportion of students who had at least one parent educated to age 17 or older was 4% in 1952, 11% in 1980 and 34% in 1998.The proportion with a father who worked in classes I or II (professional or semi-professional jobs) was 12% in 1952, 21% in 1980 and 30% in 1998.Summaries of the models are in Tables 7-9.The models are illustrated by predictions at mean attainment.To avoid excessive complexity, we combine class and parental education, for this purpose of presentation, into three levels of SES: high, medium or low, defined to be people in classes I, III or V whose parents had the modal level of education for that class in that year.
Table 7 shows that the statistical effects of attainment and change in these effects are far stronger than the independent effects of sex or SES or any interactive effect of these with attainment.The quite small remaining effect of sex and SES after controlling for attainment is illustrated for truancy in Figure 3.There is a slight tendency for the fall to be greater at high SES, corresponding to the statistically significant interactive effects of year by class and by parental education in Table 7, but generally the pattern is common at all three levels of SES and for both male and female.
The predicted proportions who thought that school was worthwhile (or in 1952 were happy at school) were mostly similar to those in Figure 1, with SES in place of attainment.The main difference was that, in 1952 and 1980 (but not later), the satisfaction among female students was about 6 percentage points higher than among male students.In particular, among low-SES students, the level reporting that school was worthwhile (or, in 1952, that they were happy at school) was always at least around 80% or higher.
The proportion who stayed on because they enjoyed school was similar to those in Figure 2, again with SES in place of attainment, and again with a sex difference of about 6 points which here persisted to 1998.The same was true for staying on because that was always assumed to be appropriate.The proportion of low-SES males who stayed on to study a specific subject rose from 27% in 1980 to 54% in 1998.Low-SES females had a similar trajectory, but about 4 points higher.At high SES, the rise was 59% to 80% for males and 63% to 83% for females.School experience in Scotland, 1970s-1990s 17 Table 8 summarises the models for perceptions of the school environment.The statistical effects of attainment and the change in these effects are again greater than any of the effects of sex or SES.There is some evidence here of a widening in the 1990s of the differences with respect to SES. Figure 4 shows this for reports that friends took school seriously, where, between 1980 and 1998, there was a rise to a high plateau for high-SES students, but a less certain rise at low and medium SES.For no category, though, were levels in 1998 below those in 1980.For reports of too much disruption, the overall worsening in the late 1990s (seen in Table 2) was in fact due to reports from high-SES students: low-SES and medium-SES students reported little change in that period.Moreover, whatever the problems from fellow students, all groups reported high levels of support from teachers: by the late 1990s, around 78-80% in each sex-by-SES group reported that teachers helped them to do their best (much the same as in the 1970s), and around 73-75% reported that teachers encouraged their confidence (about double the proportion in the 1970s).
Table 9 shows the models for experiences relating to life after school.Once more, sex and SES are less important statistically than attainment.The proportion staying on to improve qualifications rose for both sexes and all SES groups, as Figure 5 illustrates.Staying on because the respondent had not decided on their future rose for both sexes and all SES groups, as did the reports of learning material relevant to jobs and (at rates of 16% or lower for all groups) staying on because no suitable job or training place was available.
The main point about sex and SES is then that attainment mattered far more than these in a direct sense.Although, in this period, attainment was itself related to these demographic variables (Paterson, 1997), the statistical explanation of attitudes to the school experience was much more strongly attainment than demography.Differences with respect to sex fell to negligible levels by the late 1990s.However, there is also evidence that SES differences in some aspects of the experience of school widened in the 1990s, having closed in the previous decade.

Conclusion
The aims of educational reform in the 20th century were not confined to raising attainment or reducing structural barriers to opportunity.The purposes also came to include making schools more congenial and more encouraging of effort and achievement.From the campaigning of the advocates of a so-called 'new education' in the first decades of the century, these ideas became part of the shift to comprehensive secondary schooling between the mid-1960s and the late 1970s.The purpose of the present analysis has been to investigate whether the students who went to school in the last three decades of the century noticed any effect.
The strength of the analysis is that it is based on an unusually long time series of survey data relating to experiences of school.While series of data on attainment and its relationship to such demographic factors as sex and socio-economic status are quite widely available, having similar evidence about attitudes is much less common.Even though we confined attention to variables that were available in the late 1990s and at least as far back as 1980, and for about half of our measures back to the mid-1970s, we were still able to analyse in detail 13 indicators of students' experience.
There are, however, weaknesses in the data used here.Although the variables cover 25-30 years of schooling, they are mostly intermittent, with quite long gaps for some of them.The variables also try to capture in a small number of categories (usually just two) the complexity of students' experience of school.A fuller picture would require a range of other measures, whether from qualitative interviews or from ethnographic studies of schooling.Examples of such non-statistical studies from this period of Scottish education are Gow and McPherson (1980) on the 1970s and Duffield et al. (2000), Wilson et al. (2004) and Macbeath (2006) for the turn of the new century.The present study pays the price of loss of subtlety for the capacity to compare many thousands of students over a long period of time.
The findings generally confirm that the aims of treating students with greater respect and individual attention were achieved, although in many respects not till some three decades after the initial move to a comprehensive system in the mid-1960s.The proportion of students who were committed to school gradually rose, whether measured by behavioural indicators such as staying on beyond age 16 or not truanting, or by measures of attitudes such as enjoying school.There was evidence School experience in Scotland, 1970s-1990s 19 that policy played a role, for example by reforming the curriculum in ways that encouraged increasing proportions to stay on in order to study specific subjects.But perhaps the most extensive influence was from professional practice, seen in the growing appreciation by students that teachers provided both encouragement to attainment and also pastoral support.In several ways, schools were already providing levels of support to students in the 1970s that were widely appreciated: large majorities were already finding school to be worthwhile at that time, and were also appreciating teachers' help.It must also always be borne in mind that dissatisfaction among students was probably never as high as the reformers claimed: overall happiness at school was very high in 1952.
The beneficial changes extended to all levels of attainment, to both sexes and to all levels of SES.The rate of improvement was greater for students with low attainment and students with low SES.Some of this convergence of experience might be attributable to ceiling effects in the sense that the proportions in one category might have reached as high a level as could be expected, so that any further change would be bound to lead to some reduction of inequality.This is analogous to the concept of maximally maintained inequality (Raftery & Hout, 1993).For example, the proportion reporting positive attitudes in the high-SES or high-attainment group was at around 90% or greater in the late 1990s for school being worthwhile, for teachers helping the respondent to do their best and for staying on to improve qualifications.But that is only 3 of the 13 measures.As we have illustrated with several of the graphs, for most measures there was steady improvement in all SES groups, and so the high-attaining or high-SES groups had not reached any apparent limit.
Although there was some evidence that the gains by low-SES students were not advancing further at the end of the century, in no respect was there a fall back to levels that were common in the late 1970s.This growth in most measures of satisfaction among low-SES students shows that the overall improvement was not due to the changing social composition of the student population.Because most measures have always been more positive among high-SES students than among medium-or low-SES students, the growth of student satisfaction might have been due to the growth in the relative size of the higher social classes and in the average education level of parents.But that compositional change was not the main explanation.Moreover, that the levels of satisfaction have not deteriorated in the high-SES groups shows that the greater heterogeneity of these groups at the end of the century has not had any general effect on their attitudes to schooling.
In short, the evidence reported here indicates that Scottish secondary schools became more humane in the last quarter of the 20th century.For the advocates of universal secondary education in the 1920s and later, this outcome would have been as welcome as the increase in attainment or the rising rates of entry to higher education.The ethos of schools might be used, these reformers thought, to shape the character of citizens in a democracy.If being treated with respect, and returning that appreciation, are signs that these aims were being met, then we may conclude that Scottish schools at the end of the century were fulfilling some of these pioneers' most ambitious hopes.

Figure 1 .
Figure 1.Predicted proportion who thought school was worthwhile, by attainment.Standard error of each point is 0.01.Source: Model 1 in Table 4

Figure 2 .
Figure 2. Predicted proportion who stayed on because enjoyed school, by attainment.Standard error of each point is 0.01.Source: Model 5 in Table 4

Figure 4 .
Figure 4. Predicted proportion reporting that friends took school seriously, by sex and socioeconomic status

Figure 5 .
Figure 5. Predicted proportion staying on to improve qualifications, by sex and socio-economic status

Table 1 .
Distribution of variables recording engagement with school

Table 2 .
Distribution of variables recording school environment

Table 3 .
Distribution of experiences relating to life after school

Table 4 .
Models of engagement with school (1975Models of engagement with school ( -6 to 1998)), by attainment

Table 6 .
Models of experiences relating to life after school(1975-6 to 1998), by attainment would be relevant to jobs doubled from 35% in 1976 to 76% in 1998.The proportion who stayed on beyond age 16 to improve qualifications for employment or further study rose from under a half in the 1980s to nearly two-thirds in the late 1990s.

Table 7 .
Models of engagement with school (1975Models of engagement with school ( -6 to 1998)), by attainment, sex, class and parental education

Table 8 .
Models of school environment (1975Models of school environment ( -6 to 1998)), by attainment, sex, class and parental education

Table 9 .
Models of experiences relating to life after school(1975-6 to 1998), by attainment, sex, class and parental education