School-to-work in the 1990s: Modelling Transitions with Large-scale Datasets

In this chapter we explore school-to-work transitions by documenting the activities of young people who reached the minimum school-leaving age in the 1990s. Our starting position is that changes in the economy, education and training lead us to suspect that the landscape of social and economic conditions under which young people grew up during the 1990s were sufficiently different from those a decade before to justify exploration. Through the analysis of data from cohorts of young people who reached minimum school-leaving age in the 1990s we evaluate the ‘detraditionalisation’ thesis.


Introduction
In this chapter we explore school to work transitions by documenting the activities of young people who reached the minimum school leaving age in the 1990s.Our starting position is that changes in the economy, education and training lead us to suspect that the landscape of social and economic conditions under which young people grew up during the 1990s were sufficiently different from those a decade before to justify exploration.Through the analysis of data from cohorts of young people who reached minimum school leaving age in the 1990s we evaluate the 'detraditionalisation' thesis.
'Youth transitions' is a key theme that runs through the sociology of youth literature, and has been the subject of numerous post-war studies (see Clarke 1978).Much contemporary research has been bound up with what is colloquially termed as 'the school-to-work transition' (see Ashton and Field 1976;Wallace 1987;Hollands 1990; Wallace and Cross 1990;MacDonald et al.1993;Irwin 1995;Gayle 1998;Muller and Gangl 2003).In the decades following the war the vast majority of young people in the UK left education at the first opportunity.In more recent decades this situation has reversed and official data illustrate that an increasing proportion of young people have remained in education (Department of Employment 1993;FEFC 2000;Social Trends 2006).This shift has been commented upon by a number of authors (especially Paterson and Raffe 1995;Biggart and Furlong 1996;Cregan 2001).Banks et al. (1992) note that there was always a minority of young people who remained in education for long periods before entering the labour market but only a minority made an early transition straight from school-to-work by the late 1980s.
Sociologists of youth are generally in agreement that the background against which young people grew up in the closing decades of the twentieth century was transformed, and is now radically different from earlier decades.We label this the 'changing times consensus'.It is now widely agreed that the 'normal' school-to-work transition that characterised the 'traditional' rite of passage from youth to adult status has been disrupted (Irwin 1995).Sociologists have deployed a series of adjectives such as 'long', 'broken', 'fractured' and 'uneasy', in order to describe the changing pattern of youth transitions (Craine 1997).Within the 'changing times consensus', authors agree that the transformation was driven by a series of interrelated social and economic changes.
The most dramatic of the economic changes was the virtual collapse of the youth labour market in the early 1980s.This key transformation received a great deal of sociological attention (see Ashton et al. 1982;Atkinson and Rees 1982;Raffe 1984Raffe , 1988;;Roberts, 1984Roberts, , 1997;;Brown and Ashton 1987;Furlong 1987;Bynner 1996;Maguire and Maguire 1997).The growing levels of youth unemployment in the 1970s and 1980s are well documented (Casson 1979;Jackson, 1985 Gallie andMarsh 1994).Concurrently, there was a sharp decline in the number of apprenticeships during this period (Maguire and Maguire 1997).The overall economic environment was one in which there was a reduction in the number of jobs that were suitable for young people, especially minimum aged school leavers.This pattern of economic restructuring lead to a number of policy responses, most notably the widespread introduction of 'youth training' provisions (Raffe, 1982(Raffe, , 1983;;Chapman and Tooze 1987;Stoney and Lines 1987;Roberts 1984;Deakin 1996).The introduction of youth training was coupled with a number of reforms to the welfare system that changed young people's entitlement to state benefits (Maclagan 1992;Irwin 1995;Dean 1997).The provision of further education expanded in the 1980s (Smithers and Robinson 2000;Hyland and Merrill 2003).This was followed shortly afterwards by an expansion in higher education provision (Daniel 1993;Dearing 1997;Archer et al. 2003).

Growing Up in the 1990s
We argue that the transformations in the structural social and economic conditions that are identified above as key changes in the climate in which young people grew up largely took place in the 1980s.By contrast, the 1990s was a decade of employment growth in the UK (DfEE 2000).It is plausible that in the 1990s young people may have benefited from a more buoyant economy.We conjecture that changes in the economy, as well as in education and training demark the 1990s.Therefore the opportunities available to young people and the choices that they made may have differed from those made by their counterparts a decade before.
The Education Reform Act 1988, is sometimes regarded as the most important single piece of post-war education legislation.The result was that the early 1990s saw rapid changes in the curriculum, organisation, management and financing of schools (Spence 1993).A major change for pupils was the introduction of the General Certificate of Secondary Education (GCSE) (Department of Education 1985; Mobley et al. 1986;North 1987).GCSEs differed from the established Ordinary Level General Certificate of Education (GCE O'Level), Certificate of Secondary Education (CSE) and 16+ examinations which they replaced.A new grading scheme was established and all pupils were to be entered for a common set of examinations.There were also changes in the nature and format of examinations and assessment by coursework was introduced (Ashford, Gray and Tranmer 1993).It is conceivable that, coupled with the reorganisation of schools, changes in the system of examinations and qualifications could have had an impact on the choices that pupils made when they reached the end of compulsory schooling.
In addition to the changes in academic education, a new apprenticeship initiative called 'Modern Apprenticeships' was established in order to enhance the technical and vocational skills of young workers (Saunders et al. 1997;Ainley and Rainbird 1999).Young people were now eligible for new, nationally recognised, vocational qualifications (Smithers 1999).These opportunities had the potential to influence the decisions that young people made as they approached the end of compulsory education, although they were not exclusively targeted at minimum age school leavers.
In 1997 New Labour came to power with a distinctive education policy agenda under the more general umbrella of tackling social exclusion.Williamson (2005) comments that it is virtually impossible to present a full catalogue of the measures which have now been established to address the challenge of social exclusion.Because New Labour came to power late on in the 1990s we suspect that these policy changes were only likely to affect the choices made by those young people reaching the end of compulsory education towards the end of the 1990s.Hodgson and Spours (1999) argue that New Labour's education and training policies were largely dominated by responses to the Conservative legacy.Smithers (2001) notes that what is remarkable about all the apparent changes brought in by New Labour is how little they differed at root from the educational policies of the preceding Conservative administration.Hodgson and Spours (1999) highlight a difference of approach towards compulsory and post-compulsory education.They argue that changes in compulsory education were prioritised and changes in postcompulsory education were positioned lower in the policy hierarchy.This was because of the more complex interrelationship between post-compulsory education, training and the labour market.
Under the New Labour administration minimum age school leavers continued to be excluded from the unemployment benefits available to older workers (CPAG 1998;Mizen 2004).However, a notable example of an early New Labour policy initiative in the area of training was the New Deal for Young People (NDYP).This initiative resonated within the wider 'welfare to work' agenda (Riley and Young 2001;Brewer et al. 2002;Fraser 2004).Introduced in 1998 NDYP was aimed at older young people.The scheme aimed to provide opportunities to work, gain new skills, and get work experience for 18-24 year olds (Wilkinson 2003).Participation was mandatory for young people claiming unemployment benefits (i.e.Jobseekers Allowance) continuously for six months (IER 1999).We contend that the introduction of the New Deal for Young People would not have directly affected minimum age school leavers.However, it clearly signalled how the government aimed to treat unemployed young people and this may, albeit indirectly, have affected pupil's choices.
In the same period New Labour also introduced the minimum wage.The Low Pay Commission was established as a result of the national minimum wage legislation in 1998.As a result, from 1 st April 1999 workers aged 18-21 were entitled to a minimum wage at the development rate (i.e. a lower level than the adult rate).This legislation was introduced explicitly to target poverty and social exclusion and, more recently, has been extended to include workers aged 16 and 17.Whilst the timing of the introduction of the national wage and its focus on older young people was unlikely to directly influence the choices and activities of pupils as they reached the end of compulsory education, it is at least plausible that it may affect the pay and conditions of more recent groups of minimum age school leavers.

Detraditionalised Transitions?
Throughout the 1980s and 1990s numerous writers have described how young people, differentiated by social class, gender and ethnicity, follow different paths during the teenage years after they leave school (MacDonald 1999).Empirical projects, for example Banks et al. (1992) and Bynner and Roberts (1991) provided useful statistical evidence and contributed to conceptual frameworks that centred around notions of 'career trajectories'.These trajectories were based on broadly similar routes to employment and had their origins in education and family background and led towards the predictability of ultimate destinations in the labour market (Bynner and Roberts 1991).Adjectives such as 'pathways', 'trajectories', 'navigations', and 'niches', have been deployed as metaphors to describe youth transitions (Evans and Furlong 1996).Within these conceptual frameworks the emphasis was on the importance of social class, gender and ethnicity and the influence of economic realities such as labour markets and unemployment rates (Evans and Rudd 1998).
More recently, youth researchers have been keen to argue that we have moved, or at least are moving towards, a postmodern era (Furlong and Cartmel 1997).The sociological ideas of mainstream theorists, notably Giddens (1990;1991) and Beck (1992), have been influential within youth sociology.The central argument being that social life has undergone a profound change, although without undergoing a complete epochal break with the modern period.In essence, contemporary (often termed as 'risk') societies are typified by greater opportunities for individual action and decision-making, however this involves increasing risks.We are aware that the individualisation thesis is more subtly nuanced.Some theorists are frequently cautious and, for example, suggest a 'movement towards' rather than a total shift in societal conditions.Indeed Bauman (2000) highlights the 'fluid' nature of this development.
Central to the individualisation thesis is the concept of 'detraditionalization'.In essence the idea that structural factors such as social class, gender and ethnicity cease to be determinants for the individual who is pursuing the imperative of living a life of one's own (Beck and Beck-Gernsheim 2002).A consequence of detraditionalisation is life as a 'planning project' (Beck-Gernsheim 2002).The individual biographies carved out in late modern society are considered as being 'self-reflexive' and 'selfdetermined' (Giddens 1991;Beck 1992).Therefore 'biographies of choice' replace 'standard biographies', and the individual's life becomes less predictable because previous standard lifecourse sequences cannot be taken for granted.Inevitably a life of one's own, under conditions of late modernity, is a reflexive life (Beck and Beck-Gernsheim 2002).
As youth researchers Brannen and Nilsen declare 'intrinsic to the theory is the thesis that the process sweeps away structural divisions of gender, social class, and age.Individual choices become all the more important, and the choice biography takes over from the standard biography' (2002 p.515).Similarly, Arnot et al. (1999) suggest that the process of individualisation is considered to transform individuals' relationships with their class and ethnic status, family connections and traditional loyalties.Following from this position is the view that young people are now required to adopt calculative, strategic and reflexive personalised strategies towards the new risks and opportunities of the post-school world, rather than following the traditional (and now often obsolete) pathways associated with their collective social class, gender and ethnic identities (MacDonald and Marsh 2005).
The theorised experience of youth transitions under detraditionalised conditions is neatly summarized: 'In the place of these collective guides and traditional institutions are much more individualized identities and biographies where individuals have a greater scope beyond traditional markers of class, race and gender to create complex subjective lifestyles' (Cieslik and Pollock 2002, p.3).
The idea that the conditions in which young people grew up underwent a process of change in the closing decades of the 20 th Century is not in contention.In this chapter our overall analytical aim is to investigate whether, or not, evidence of detraditionalisation can be detected through the analysis of cohorts of nationally representative survey data.If a process of detraditionalisation is taking place we would expect that the influence that individual level factors (e.g.social class, ethnicity and gender) have on young people's transitions (i.e.participation in education, employment and training) would be in decline.

Survey Data
Researchers in the UK have enjoyed access to youth data through the birth cohort studies.No new birth cohort data was collected between The British Cohort Study in 1970 and the Millennium Cohort.A gap in data provision exists with no birth cohort data for young people growing up in the 1990s.In this analysis we concentrate on data from the Youth Cohort Study of England and Wales (YCS), which partially fills this gap (see Gayle 2005).
The YCS is a major longitudinal study which began in the mid-1980s.The study is designed to monitor the behaviour of young people as they reach the minimum school leaving age and either stay on in education, enter the labour market or undertake training.The sample is representative of all Year 11 pupils in England & Wales (that is all pupils reaching the minimum school leaving age).The data are therefore nationally representative and have mainly been collected by postal questionnaires.The survey mostly collects individual level information.A small amount of information is collected from the young people on their parents and family life.In particular, detailed information on experiences in education (especially qualifications), training and employment is collected.The cohorts are tracked for 3 (sometimes 4) waves (sweeps) of data collection.The first sweep of data collection takes place in the spring (this is usually near the Easter vacation) after the pupils complete Year 11 (e.g.pupils remaining at school will be in Year 12, historically this was known as the lower sixth form).
In the present analyses we focus on five YCS cohorts which span the 1990s.The cohorts comprise young people who reached the minimum school leaving age in 1990, 1993, 1995, 1997 and 1999.The design of the questionnaire, and question content have changed over the life of the YCS, partially due to changes in substantive interests and alternative policy concerns.In the present analyses we have harmonised a number of measures in order to compare across cohorts.

Results
The proportion of young people remaining in education after the minimum school leaving age increased in the 1980s and 1990s and in the later cohorts nearly three quarters of young people remained in education (see Table 1).Eight per cent of the young people in the 1984 cohort were unemployed in the October following their completion of compulsory education.We envisage that youth unemployment was at its zenith in the early 1980s and unemployment was halved in the 1990s cohorts.
Overall, the increased participation in education is a result of lower participation in employment and training.Around a quarter of the age cohort participated in training in the 1980s cohorts but this fell to less than ten percent for the later 1990s cohorts.Around 45% of young people left education and entered employment or training in the 1980s and the 1990 cohorts.This proportion fell sharply in the 1993, 1995, 1997 and 1999 16,208 14,116 14,511 18,021 15,899 14,662 13,698 Note: Weighted percentages. 6 In general, the picture for the 1980s was one of increasing participation in postcompulsory education.This trend has stabilised in the more recent cohorts however.The majority of young people growing up in the 1990s remained in education after the minimum school leaving age.Theorists have made the claim that an increasing number of pathways have become available to young people in contemporary society.This may be the case, but in reality increasing proportions are choosing the path of continuing in education.
In the 1990s cohorts, young people were thinly distributed across the pathways outside of education.This poses the empirical question, which factors are associated with early school-to-work transitions?In the remainder of this chapter we attempt to identify how gender, ethnicity and social class relate to post-compulsory educational participation.From the outset we conjecture that we may no longer identify the key structural effects of gender, ethnicity and social class under conditions of increasing detraditionalisation.

Qualifications
Qualifications remain a powerful dynamo, and in general young people with high levels of educational attainment tend to remain in education.Across the 1990s YCS cohorts there was a steady improvement in year 11 GCSE attainment, measured by the familiar official measure 5+ passes at grades A* -C (see Table 2).Year 11 GCSE attainment is a strong predictor of participation in post compulsory education.We note that only a small minority of young people with 5+ passes at grades A* -C did not remain in education (see Table 3).More interestingly, over half of the young people who failed to obtain 5+ passes in 1990 left education, but in subsequent cohorts there was a small increase in the portion of this group who remained in education after they reached the minimum school leaving age.Logistic regression coefficients (with comparison intervals) In Education Table 4 reports the results of a logistic regression model which models the interaction between cohort and year 11 GCSE attainment.Figure 1 depicts that whilst the effects of having less than five passes is significantly different for the cohorts after 1990 there is no clear trend and overall these cohorts are not significantly different from each other1 .Similarly, passing at least five GCSEs in year 11 has a different effect for some cohorts but there is no clear trend.There is a changing distribution of participation in post-compulsory education and a changing distribution of year 11 GCSE attainment.We believe that explicitly modelling this interaction is useful and will provide increased statistical control when modelling this data, which pools five cohorts of survey data, in order to better explore trends over time.

Gender
During the 1970s and 1980s the primary focus of research on gender in the field of education was on girls (Warrington and Younger 2000).The overall message was that expectations, aspirations and choices were structured along traditional gender lines to the disadvantage of young women (see for example Sharpe 1976;Deem 1980;Griffin 1985).In recent years the situation is reversed as there is now growing concern about the lack of participation, and the under-achievement of boys (Young and Warrington 2005).In the present analysis there is a persistent gender effect with a higher proportion of girls remaining in education in each of the five cohorts (see Table 5).

Ethnicity
It is well observed that there are differing levels of participation in post-compulsory education across ethnic groups (see Drew et al. 1992;Drew 1995;Bhattacharyyal et al. 2003).We observe higher participation from non-white young people even after year 11 GCSE attainment, cohort and gender have been accounted for (see Table 6).Biggart and Furlong (1996) suggest that one explanation as to why pupils from minority ethnic families remain in education for longer periods is that they might be sheltering from (either actual or perceived) discrimination in the labour market.Whilst all of the minority ethnic groups in the five cohorts have higher participation than their white counterparts no single group has a significantly higher level of participation overall (see Figure 2).At the current time there is emerging concern regarding the educational participation of white boys from poorer families (see Cassen and Kingdon 2007).This is a point that we will return to in the next section.

Family Social Class
In this section we explore the effects of family social class through measures of parental occupation.We use what is commonly termed as a 'dominance approach', namely the social class of the family is measured by the occupational position of the higher of the two parents (or step parents) (see Erikson 1984).We consider that a dominance approach logically resonates with the idea of an overall 'family' social class, and is a more empirically suitable measure for the present analysis.Given the rising levels of divorce, separation and lone parenting this is much more appealing than adopting father's social class as a measure.

Logistic regression coefficients (with comparison intervals)
In Education  We have calculated four recognizable measures of social class.Registrar General's Social Class (see Leete and Fox 1977), NS-SEC (see Rose and Pevalin 2003), CASMIN (see Goldthorpe 1987) and CAMSIS (see Stewart et al. 1980).We are aware that established conceptions of social stratification and social class analysis have recently drawn sharp criticism in mainstream sociology (Bergman et al. 2002).
Our motivation in choosing a number of different measures of social classification is that they measure social stratification through alternative dimensions (Rose and Pevalin 2003).
There is an ordinal relationship between the family's Registrar General's social class and the young person's participation in post-compulsory education (see Table 7).The picture is very similar when social class is measured by NS-SEC and CASMIN (see Table 8 & Table 9).Young people from families with parents in less advantaged occupations have lower levels of participation in post-compulsory education.
Similarly, Table 10 reports that young people who do not remain in education have parents with lower mean (male) CAMSIS scores.
The message is unambiguous.Net of year 11 GCSE attainment, cohort, gender and ethnicity there is a clear negative relationship between family social class measured by the Registrar General's classification and participation in post compulsory education (see Table 11).Young people from families with parents in less advantaged occupations have greater chances of making an early school-to-work transition.Figure 3 illustrates that there are significant differences in the levels of participations in postcompulsory education for young people from professional, intermediate and skilled non-manual backgrounds.Whilst those young people from skilled manual, partly skilled and unskilled families are less likely than counterparts from the three more advantaged social classes to remain in education, they are not significantly different to each other.This is an important finding and signals that there is a genuine cleft between young people from non-manual and manual backgrounds.
The picture is very similar when family social class is measured by NS-SEC (see Table 12).Young people from families with parents in less advantaged occupations have greater chances of making an early school-to-work transition.The modelling results for NS-SEC and CASMIN are almost identical, therefore we have only reported one full set of model results.Figure 4, however, depicts the high similarity of these two alternative measures of family social class.We are also keen to note that the effect of the family's Registrar General's social class and participation in postcompulsory education varies little across each of the five cohorts.This pattern is identical when family social class is measured by either NS-SEC or CASMIN and overall modelling results bear remarkable similarities using any of these social class measures.
The overall substantive conclusions are comparable when family social class is measured by the (male) CAMSIS score (Table 13).Young people from families with parents in less advantaged occupations have greater chances of making an early school-to-work transition.We note that CAMSIS has additional appeal in this type of analysis as it leads to a better fitting statistical model that is more parsimonious because it estimates fewer parameters.
As Drew et al. (1992) suggested over a decade ago, modelling approaches are attractive in understanding the multivariate nature of post-compulsory educational participation.Returning the Table 11, we estimate that in the 1990 cohort a young white male with professional parents (Registrar General's Social Class I), and with fewer than 5 year 11 GCSE passes (at grades A* -C) has a probability of remaining in education of 61%.By contrast, if he had attained 5+ GCSE passes (at grades A* -C) in year 11, his probability of staying on in education would be 77%.This hopefully demonstrates the incredibly strong effect of year 11 GCSE attainment.
We estimate that in the 1999 cohort a similar young white male with professional parents (Registrar General's Social Class I), and with fewer than 5 GCSE passes (at grades A* -C) in year 11, would have a probability of staying on in education of 73% (compared with the 61% probability for a counterpart in the 1990 cohort).This illustrates the slight increase in participation by those with lower GCSE attainment in the later 1990s cohorts.
We estimate that a young white male with professional parents (Registrar General's Social Class I), and with 5+ GCSE passes (at grades A* -C) in year 11 would have at least a 95% chance of remaining in education in the 1993, 1995, 1997 and 1999 cohorts.We suggest that this illustrates a potential plateau, or ceiling effect.Participation in education immediately after year 11 may have reached an upper limit for those with good levels of year 11 GCSE attainment.
The effects of family social class can similarly be illustrated within the multivariate explanation of post-compulsory participation.We have estimated that in the 1990 cohort a young white male with professional parents (Registrar General's Social Class I), and with fewer than 5 GCSE passes (at grades A* -C) in year 11 has a probability of staying on in education of 61%.In contrast if he had unskilled parents (Registrar General's Social Class V) his probability of participating in post-compulsory education would be reduced to 29%.Even when a young person from the less advantaged social class backgrounds has obtained 5+ GCSE passes (at grades A* -C) in year 11, on average, they have about a 12% lower chance of remaining in education.Taken together, the negative effects of being from a lower social class background, being male and being white lends some support to the emerging concerns regarding boys from poor white families.

Logistic regression coefficients (with comparison intervals)
In Education

Logistic regression coefficients (with comparison intervals)
In Education

Conclusion
The 'pathways' or traditional routes that young people followed in the immediate post-war decades have altered.The structural changes in the climate in which young people grew up in the 1980s are not in question.Over the final two decades of the twentieth century increasing proportions of young people remained in education after they reached minimum school leaving age.The upward trend away from early transitions reached a plateau by the mid 1990s.In the cohorts since then only a minority of young people made the early transition out of education.This is consistent with commentaries within the sociology of youth that have deployed various adjectives that describe the lengthening of contemporary school-to-work transitions.
We reiterate that the cohorts of data which were analysed above are nationally representative.Educational attainment, measured by qualifications gained at the end of compulsory school, is very important.The 1990s saw overall improvements in performance at GCSE.Like Ashford, Gray and Tranmer (1993) we are reluctant to label this as a 'GCSE effect' however.Compared with the 1990 cohort, in the more recent cohorts an increasing proportion of young people with lower GCSE attainment in year 11 remained in education after the minimum school leaving age.We see no reason to expect that attainment will not continue to be a main factor influencing participation in post-compulsory education.Therefore if the government wishes to further increase participation in post-compulsory education, then arguably more pupils with lower levels of years 11 GCSE attainment must be encouraged to continue in education.2 Net of qualifications, and changes to the effects of qualifications across the cohorts, differences in participation across gender lines clearly exist.More young women remain in education and it is their male counterparts who are more likely to make an early transition out of education.This pattern is in sharp contrast to earlier decades.Marked differences in educational participation by different ethnic groups remain.White young people are most likely to make a transition out of education directly after compulsory education.
Family social class, measured by parental occupation has an effect on postcompulsory educational participation.Family social class effects persists net of qualifications, gender and ethnicity, regardless of which social class scheme is deployed.Young people from more advantaged social class backgrounds have very high levels of participation in post-compulsory education.Over the 1990s there has been some improvement in the proportions of young people from less advantaged social class backgrounds participating in post-compulsory education.Young people from the least advantaged families still lag behind their counterparts from more advantaged social class positions however.
We envisage that cohorts of young people reaching the minimum school leaving age after the millennium will continue to exhibit similarly high levels of participation in post-compulsory education.The Labour administration is committed to minimising the proportion of young people not in education, employment or training (NEET).We do not imagine that the government will make major changes allowing 16 and 17 year olds greater access to welfare benefits, therefore we suspect that levels of unemployment for minimum age school leavers will remain low.We envisage that the introduction of the minimum wage for younger workers and the introduction of the policy initiatives (e.g.Modern Apprenticeships, which have more recently been repackaged as Apprenticeships) may, albeit indirectly, influence the decisions and choices that young people make as they reached the end of compulsory education.Data from more contemporaneous cohorts of young people is required however to quantify the effects of the raft of policies directed towards combating social exclusion that have been introduced by the New Labour administration.
It is undisputable that transitions to work have lengthened, because fewer young people make an early transition from education.Nevertheless, our overall theoretical conclusion is that there is little empirical evidence supporting 'detraditionalisation'.The 1990s data on early transitions, analyzed above, show clear patterns of structure along gender, ethnicity and social class lines.There appears to be little loosening of these social bonds and our evidence refutes the core claims of the 'detraditionalisation' thesis.
The overall pattern of societal change might be a movement towards greater individualisation.We are keen to assert that with temporally comparative data such theoretical claims should be empirically testable however.It may be that individual young people feel that the choice biography is taking over from the standard biography, or that their individual choices are becoming more important.The strong presence of traditional structural influences (e.g.gender, ethnicity and social class) on early transitions does not especially persuade us that processes of individualisation are taking hold and influencing school-to-work transitions.Some youth researchers, for example Furlong and Cartmel (1997), following Roberts et al. (1984), have suggested a more nuanced theoretical conception and have deployed the label 'structured individualisation'.Under these conditions young people believe that their choices and decisions are individualised, but in reality they are still heavily mediated by social structures.
We can conclude that there is no evidence of detraditionalisation and that gender, social class and ethnicity remain apparently strong.From the data analysed above it is not directly possible to determine whether or not young people increasingly believe that their choices and decisions are becoming ever more individualised.At the current time we believe that the view that young people have changed, or are in a process of changing, how they view (or interpret) their circumstances is empirically unproven.What would be required to convince us is large-scale individual level data collected over time (e.g. through cohorts) showing clear changes in attitudes, aspirations, expectations and preferences, in a way that demonstrated that young people were moving towards understanding their circumstances and choices in more 'individualised' ways.
Finally we caution against uncritically accepting the concept of a process of detraditionalisation. Historical sociological research has demonstrated the paradox that life course career stability can coexist with periods of sustained structural and economic change (Savage 1993).The Annales school of social history argue that social changes occur at an extremely slow pace, far more incrementally than is commonly appreciated by contemporary sociologists (Penn 2006).Other empirical reviews in British economic sociology have reported long terms patterns of social stability within the experiences of employment and social mobility over the 20th Century (Penn 2006;Goldthorpe 2007;Lambert el al. 2007).These points illustrate that at a macro level social change within certain contexts might seem more evident but at a micro, or individual, level the experience of social stability can dominate.
cohorts, with 20% or less leaving education and entering employment or training.The data show a decline in those young people not in education, training or employment (NEET) which concurs with official statistics (see DfES 2005).

Table 3
In Education October Following Year 11 and GCSE Attainment Year 11 by YCS Cohort (%) Note: Weighted percentages.

Table 4
Logistic Regression Model -In Education October Following Year 11 -cohort and year 11 GCSE effects Note: Weighted survey data, survey regression estimated using pseudo maximum likelihood.Goodness of fit measures based on standard regression (i.e.non-weighted data); Pseudo R 2 = .18;Deviancenull model 77439; Deviance current model 61389; Change in deviance 16050 @ 9 d.f.Source: YCS, Cohorts 1990 -1999.Model: Logistic regression predicting young person is in education October after year 11.

Table 5
In Education October Following Year 11 by Gender and YCS Cohort (%) Note: Weighted percentages.

Table 6
Logistic Regression Model -In Education October Following Year 11-cohort and year 11 GCSE attainment, gender and ethnicity effects

Table 8
In Education October Following Year 11 by Family Social Class (NS-SEC 3 Classes) and YCS Cohort (%) Note: Weighted percentages.

Table 9
In Education October Following Year 11 by Family Social Class (CASMIN 3 Categories) and YCS Cohort (%) Table 11 Logistic Regression Model -In Education October Following Year 11-cohort and year 11 GCSE attainment, gender, ethnicity and parental social class (Registrar General) effects Table 13 Logistic Regression Model -In Education October Following Year 11cohort and year 11 GCSE attainment, gender, ethnicity and parental social class (Male CAMSIS) effects