Choice biographies and transitional linearity : re-conceptualising modern youth transitions

Abstract It has been suggested that youth transitions have become increasingly protracted and complex and that routes between school and work, which were once viewed as linear and predictable, have been replaced by a set of movements that are more fragmented. Our aim in this paper is to contextualise these changes in an attempt to capture the degree of complexity characteristic of modern transitions and to explore the implications for patterns of labour market integration. We argue that there has been a tendency to exaggerate processes of de-linearisation and that the modern tendency to regard transitional complexity as symptomatic of «choice biographies» can help mask structures of disadvantage. Key words: youth transitions, choice biographies, youth employment, youth unemployment. Resum. Biografies d'elecció i linealitat de la transició: la reconceptualització de les transicions juvenils modernes S’apunta que les transicions juvenils cada vegada són més llargues i complexes i que els camins entre l’escola i el treball, que abans es consideraven lineals i previsibles, actualment han estat substituïts per un conjunt de moviments que estan més fragmentats. L’objectiu d’aquest article és contextualitzar aquests canvis per tal de captar el grau de complexitat característic de les transicions modernes i explorar les implicacions que tenen en els models d’integració en el mercat laboral. Defensem que s’ha tendit a exagerar els processos de deslinealització i que la tendència moderna de veure la complexitat de les transicions com un fet simptomàtic de les «biografies d’elecció» pot contribuir a ocultar estructures de desavantatges. Paraules clau: transicions juvenils, biografies d’elecció, treball juvenil, atur juvenil. Papers 79, 2006 225-239 Choice biographies and transitional linearity: Re-conceptualising modern youth transitions Andy Furlong University of Glasgow. School of Business and Management Fred Cartmel University of Glasgow. Department of Sociology and Applied Social Science Andy Biggart University of Ulster Summary Introduction Data The complexity of transitional experiences Conceptualising linearity Linearity and the sequencing of experiences Non-linearity and disadvantage Conclusion Acknowledgements References Papers 79 001-326 6/6/06 09:56 Página 225 Introduction Over the last couple of decades it has been suggested that youth transitions have become increasingly protracted and complex resulting in a greater vulnerability to marginalisation and exclusion. Routes between school and work, which were once viewed as linear and predictable, are seen as having been replaced by a set of movements that are less predictable and involve frequent breaks, backtracking and the «blending» of statuses which previously tended to be quite distinct (Chisholm, 1995; Wyn & White, 1997; Du Bois Reymond, 1998; Looker and Dwyer, 1998; EGRIS, 2001). This new consensus on the changing nature of transitions is rarely challenged, however, there has not been any systematic analysis of the extent to which transitions have increased in complexity and lost their linearity nor of the implications for young people of following non-linear as opposed to linear routes. Our aim in this paper is to contextualise these changes in an attempt to capture the degree of complexity characteristic of modern transitions and to explore the implications for patterns of labour market integration. To achieve these aims we first use cluster analysis to describe a set of typologies of transitional experience in the West of Scotland and then use these typologies to study the degree and type of complexity characteristic of modern labour market transitions. We go on to look at the characteristics of those following linear and non-linear transitions and examine the extent to which transitions are linked to patterns of labour market integration. Data The analysis that is presented in this paper is drawn from the project Youth Transitions: Patterns of Vulnerability and Processes of Social Inclusion funded by the Scottish Executive and Scottish Enterprise. The data used is from the West of Scotland Twenty-07 study conducted by the Medical Research Council’s Social and Public Health Sciences Unit. The Twenty-07 Study is a longitudinal study, located in and around Glasgow, which commenced in 1987. A representative sample of 1009 15 year olds and their parents participated in the first round of interviews. The sample were subsequently followed-up at age 16, 18, 21 and 23 (postal surveys at ages 16 and 21 and face-to-face interviews at ages 18 and 23) and were interviewed again in 2001/02 at age 28/29. Discussion here is limited to experiences between the ages of 15 and 23. The response rates have been good with about 65 per cent of those in the initial sample of 15 year-olds responding at age 23. The complexity of transitional experiences Changes in education and the labour market are often viewed as having resulted in a situation where few individuals follow identical sets of transitional routes and most are faced with the need for reflexive interpretation and constant 226 Papers 79, 2006 Andy Furlong; Fred Cartmel; Andy Biggart Papers 79 001-326 6/6/06 09:56 Página 226 negotiation in an increasingly unpredictable social world. For some, these trends are seen as associated with the emergence of «choice biographies» (Du Bois Reymond, 1995) while for others they are linked to new patterns of vulnerability and to perceptions of risk and uncertainty (Furlong and Cartmel, 1997; Wyn and White, 1997). However the trends are interpreted, few researchers would dissent from the view that modern transitions are characterised by their complexity. Citing evidence from Australia and Canada, for example, Looker and Dwyer (1998) argue that young people tend not to regard their transitions as linear but «report them as complex and interconnected, as involving false starts and redefined possibilities, as requiring negotiation and redefinitions» (1998: 17). While we accept the proposition that modern transitions are frequently complex, two qualifications must be made. First, new evidence (Goodwin and O’Connor, 2002) has led us to accept the view that earlier British research on transitions (e.g. Ashton and Field, 1976; Carter, 1962) had tended to understate the level of complexity which characterised youth transitions in the 1960s and 1970s. At the time researchers tended to concentrate on macro-level analysis, highlighting class (and occasionally gender) specific patterns of reproduction while neglecting individual-level complexities and processes of negotiation. A notable exception to this being Willis (1977) who managed to illuminate some of the ways in which the everyday lives of working class boys involved forms of action, negotiation, penetration and resistance which were central to processes of social reproduction. Our second qualification relates to the levels of complexity characteristic of modern transitions. Here we suggest that the case for the de-linearisation of transitions may have been over-stated. Whereas members of the earlier generation of youth researchers may have focused on structure at the expense of agency, in the rush to embrace late modern or post-modern perspectives, contemporary researchers have perhaps over-stated the significance of processes of reflexivity and life management. It is also important to recognise that the existence of complexity does not necessarily result in the emergence of «choice biographies», as du Bois Reymond (1995) seems to suggest: levels of complexity may actually signify a lack of choice and a vulnerability to the adverse effects of flexible labour markets. We need to distinguish between movement that is a consequence of labour market precarity from movement that suggests flexibility. Before directly addressing the issue of complexity, it is necessary to outline the main sets of routes that young people follow between school and work. A variety of typologies have been used by researchers with distinctions tending to be made between routes involving relatively direct and indirect entry to the labour market, routes involving extended participation in post-compulsory education and training and patterns of movement punctuated by unemployment or training schemes (e.g. Raffe et al., 1998; Anisef et al., 2000; Walther et al., 2002). In this paper cluster analysis is used to determine the main routes followed by young people between the ages of 15 and 23. Choice biographiel and transitional linearity Papers 79, 2006 227 Papers 79 001-326 6/6/06 09:56 Página 227 In the Twenty-07 surveys information was collected on the main status of individuals for each month covered by the survey. As such, between the ages of 15 and 23 status «diaries» cover 96 time-points and provide details on whether young people were employed full or part time, were unemployed, in education or training, were looking after a family or were out of the labour force due to ill-health or a disability. Using cluster analysis, eight transition clusters were constructed using cumulative details of status derived from the monthly diaries collected between the ages of 16 and 23 and reflecting the total time (months) that young people spent engaged in each main activity. We begin by briefly describing these clusters and the characteristics of those making these different transitions. The mean number of months spent in each status by members of each cluster are described in table 1. I. Higher Education Groups The first two clusters (Long Higher Education (Long HE) and Short Higher Education (Short HE) both involve participation in Higher Education (HE). The average duration of HE participation in the first cluster (Long HE) (27 per cent of the sample) is just over 4 years1therefore in the main these are people who had completed full degree courses (74% had degrees at age 23 and a further 11% had higher diplomas2or equivalents). The second cluster (Short HE) (12 per cent of the sample) is comprised of individuals who had a shorter average period of HE participation: just under two years. Included here are those who had taken short courses, those who had dropped out of longer courses and those who had entered late or returned to education after a period away 228 Papers 79, 2006 Andy Furlong; Fred Cartmel; Andy Biggart 1. In Scotland a typical honours degree takes four years to complete. 2. Higher diplomas are two year advanced courses at a sub-degree level. Table 1. Transitional cluster, by mean months spent in each status between the age of 16 and 23. School FE HE Unemp Scheme FT Job PT Job Domest Other Long HE 23.76 2.2 50.6 3.25 1.29 15.48 0.53 0.005 0.69 Short-HE 22 1.45 23.21 4.33 2.36 43.18 0.65 0.02 0.77 Enhanced Ed 19.52 3.85 1.62 1.46 1.14 67.23 1.9 0.85 0.28 Direct job 8.14 0.28 0 1.43 1.01 85.7 0.61 0.08 0.7 Assisted 7.22 0.2 0.15 5.64 16.48 65.95 0.25 1.36 0.49 Unemployment 7.78 0.76 4.63 44.07 11.81 25.31 0.63 2.15 0.73 Domestic 7.8 0.95 0.52 3.8 8.8 24.04 2.19 49.04 0.47 Other 14 0 3.28 5.57 11.14 16.42 0.85 1.42 45.28 Note: Definitions of Long HE and Short HE (Higher Education) appear below. Papers 79 001-326 6/6/06 09:56 Página 228 and who were still at college at age 23. At age 23, 45 per cent of this latter group had higher diplomas or equivalent qualifications while a further 24 per cent had degrees. Both of the higher education clusters contain similar proportions of males and females and, in general, these two clusters are comprised of relatively advantaged young people. II. Employment transitions The second two clusters describe those who experienced relatively direct school to work transitions. Within the first cluster (14 per cent of the sample), young people typically remained in education until the age of 18: the average period spent in school beyond the minimum leaving age being 19 months with some having followed Further Education (FE) based routes (these are described as the enhanced education group). Following educational completion, entry into work was fairly direct: the average time spent unemployed between the ages of 16 and 23 being a month and a half and with an average of a month being spent on government training schemes. With members of this group having made a fairly direct entry into the labour market, by age 23 they had been in full-time employment for an average of five and a half years. The second employment cluster (17 per cent of the sample) tended to have left school at an early stage (described as the direct job group) having spent an average of eight months in school beyond the minimum leaving age. Average time spent unemployed and on schemes between the ages of 16 and 23 was similar to the above group (a month and a half and a month respectively). Having made early and relatively direct transitions, members of this group had the most extensive experience of full-time employment in the sample (average of 7 years). In terms of characteristics, these employment clusters contained more females than males. Those who had followed enhanced education routes tended to be slightly more advantaged than those who entered the labour market earlier: they were less likely to live in a deprived neighbourhood or to have parents in the lowest social classes. III. Assisted and unemployment based clusters Some of the more difficult transitions are contained within the assisted and unemployment based clusters (accounting for 20 per cent and 6 per cent of the sample respectively). Many of those following assisted transitions had spent time on government sponsored training schemes (average duration of 16.5 months) and their average duration of unemployment over the time span was 5.6 months. These young people tended to have left school at a relatively early stage. However, their average duration of full-time employment is very similar to those within the enhanced education cluster (66 months compared to 67 months). Choice biographiel and transitional linearity Papers 79, 2006 229 Papers 79 001-326 6/6/06 09:56 Página 229 The labour market histories of those within the unemployment based cluster were dominated by unemployment and members had spent an average of 44 months unemployed between the ages of 16 and 23. Many had also spent time on government training schemes (average of 11.8 months) and some had spent short periods in Higher Education. Over the whole period they had spent an average of just 25 months in full-time employment. At age 16, the educational qualifications of these two groups was poorer than any of the groups previously described, with the performance of the unemployment based group being particularly low. Nearly three in ten (29%) of the assisted employment group had no qualifications while almost seven in ten (68%) of the unemployment based group lacked qualifications. Both of these groups were dominated by males (56 per cent and 66 per cent respectively) and the majority lived in deprived neighbourhoods (56 per cent and 71 per cent respectively). IV. Domestic and other clusters The two remaining clusters are both relatively small: the domestic cluster (3 per cent of the sample) and the «other» cluster (1 per cent of the sample). The former had spent an average of 49 months working in the home while the latter mainly comprise of the few with long-term health problems and disabilities. The latter group is too small to discuss in any meaningful sense. While the domestic group is also small, they were all female, tended to come from the lower social classes and more than two in three lived in deprived neighbourhoods. Conceptualising linearity In discussing the concept of individualisation in relation to youth transitions, Roberts (1997) has argued that modern transitions can be viewed as individualised to the extent that few young people can now identify others who have followed identical trajectories. Without implying that our clusters represent homogenous sets of experiences it is important to recognise that although contexts have changed, some young people still follow smooth and linear routes. Here we aim move beyond the clusters introduced in the previous section in order to assess the extent to which complex non-linear transitions have become the typical experience of Scottish youth. To address this question, we first have to determine what range of experiences should be defined as linear or non-linear. In our view, linearity involves a fairly smooth and straightforward transition in which there are no major breaks, divergences or reversals. It is important to recognise here that some experiences which were once relatively uncommon have become normalised and/or institutionalised. Few young people, for example, manage to avoid unemployment altogether and for early entrants into the labour market, government training schemes have become an important mechanism for training 230 Papers 79, 2006 Andy Furlong; Fred Cartmel; Andy Biggart Papers 79 001-326 6/6/06 09:56 Página 230 and occupational socialisation. As such, we suggest that a transition can still be described as linear if the young person making the transition has encountered short periods of unemployment or has been «trained» in the context of a Government programme. Thus someone who has a short period of unemployment between leaving education and gaining a job can still be seen as having made a linear transition. In contrast, non-linear transitions involve breaks, changes of direction and unusual sequences of events. They can involve extended or repeated experiences of unemployment, frequent moves between jobs and returns to education and training after periods in employment. The way in which we operationalised these two broad transitional patterns from the statuses listed in the diary is described in figure 1. Here, by operationalising core theoretical assumptions, non-linear transitions are defined as sequences that do not involve straightforward progressions through education or training to employment or domestic work. Even when overall transitions follow a «traditional» sequence, they are also defined as non-linear if an individual experiences a single period of unemployment lasting for three or more months or have a cumulative experience of unemployment of a year or more or has a high number of status changes over the period (defined here as exceeding the median for a person leaving education at the same stage by two). Using this definition, similar proportions of the sample could be viewed as making linear and non-linear transitions (52 per cent as compared to 48 per cent) with males being more likely than females to follow non-linear routes (55 per cent of males compared to 42 per cent of females). Not only are linear transitions just as characteristic of the modern age as non-linear transitions, we also suggest that if the same criteria were applied to studies conducted in the 1960s or 1970s, there would also be a good mix of transitional types: Goodwin Choice biographiel and transitional linearity Papers 79, 2006 231 Figure 1. Conceptualising linear and non-linear transitions. Linear transitions Non-linear transitions School to employment Any sequence of statuses not conforming to those listed on the left of the table School-training scheme-employment A single period of at least three months unemployment at any stage School-HE-employment Total (cumulative) experience of unemployment exceeding 12 months School-FE-employment A number of status changes which exceeds the median for a person leaving education at the same stage as themselves by two School-FE-HE-employment School-any status-domestic work Papers 79 001-326 6/6/06 09:56 Página 231 and O’Connor (2002), for example, show that in the 1950s some young people moved jobs on a very regular basis while in periods of recession labour market experiences are frequently chequered by unemployment and enforced job changes. Of those making non-linear transitions, some had encountered difficulties soon after leaving education while others had made fairly smooth transitions into the labour market but encountered difficulties some time later. Just over a third (33 per cent of males and 37 per cent of females) of those following non-linear routes encountered their first period of unemployment lasting at least three months directly after leaving education. More than four in ten (43 per cent of males and 48 per cent of females) first encountered unemployment after a job or Youth Training (YT) but within the first three years after leaving education. A smaller number, (24 per cent of male and 15 per cent of females) survived in the labour market for more than three years without a significant period of unemployment3. These findings challenge the view that young people who are unemployed for more than a few weeks are often not «job-ready». Some young people had also clearly experienced more fragmented transitions than others. Given that our definition of linearity rests heavily on unemployment experience, we would expect to find a higher incidence of unemployment among those following non-linear routes. However, it is worth noting that among those who made non-linear transitions, some pathways are chaotic: at the extreme the maximum number of status changes for males is 17 and for females 15. Similarly the maximum amount of time that a male has spent unemployed was seven years (84 months), while for females it was nearly six years (71 months). Non-linear transitions tended to be slightly more common among those from less advantaged class positions. In other words, complex transitions are best regarded as disadvantaged routes, not the «choice biographies» of the affluent. For both males and females, a small majority of those from social class I and II made linear transitions while for males, a majority of those from social class V and VI made non-linear transitions. Young people who had followed nonlinear routes were also more likely to report that their father (but not mother) had experienced a significant period of unemployment (over 6 months) at some stage. Of those who had followed non-linear routes, 24 per cent of males and 32 per cent of females said that their father had experienced protracted unemployment, while the corresponding figures for those who had followed linear routes were 16 per cent and 21 per cent respectively. Overall, linear transitions tended to be most common among those with the highest level of educational attainment (degree or equivalent) and least common among those with no qualifications. However, below degree level, the majority of males tended to make non-linear transitions whereas females with 232 Papers 79, 2006 Andy Furlong; Fred Cartmel; Andy Biggart 3. Gender differences not significant. Papers 79 001-326 6/6/06 09:56 Página 232 lower level qualifications were still more likely to make linear transitions. In other words, educational qualifications were more strongly linked to the linearity of transitions for males than for females. Linearity and the sequencing of experiences While the cluster analysis that we introduced earlier helped us to describe the main transitional paths and the characteristics of young people who followed these routes, within each cluster young people will have had a range of experiences that have occurred in different sequences. With most clusters combining those who have made relatively smooth transitions with those encountering significant turbulence, they allowed us to group key routes and identify typical sets of experiences within them, but do not facilitate a full exploration of patterns of vulnerability. However, as operationalised here, the concept of linearity/non-linearity enables us to explore the extent to which those individuals who have made relatively turbulent transitions are concentrated within key «pathways» as represented by the clusters. In this section we discuss the relationship between clusters of experience and patterns of linearity. The first thing to note is that some clusters are strongly dominated by those who we had defined as having made linear transitions (figure 2). The majority of those experiencing direct job transitions, long HE transitions and enhanced education transitions followed linear routes. Conversely, most shortHE transitions and assisted transitions are non-linear. With unemployment being central to our definition of linearity, all of those following unemployment based transitions had followed non-linear routes. Choice biographiel and transitional linearity Papers 79, 2006 233 Figure 2. Patterns of linearity among transitional clusters. 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% non-linear linear Long HE Short HE Enhanced Ed Direct job Assisted Unemployed Papers 79 001-326 6/6/06 09:56 Página 233