Social Exclusion and the Transition from School to ... - Science Direct

Journal of Vocational Behavior 60, 289–309 (2002) doi:10.1006/jvbe.2001.1868, available online at http://www.idealibrary.com on

Social Exclusion and the Transition from School to Work: The Case of Young People Not in Education, Employment, or Training (NEET) John Bynner and Samantha Parsons Center for Longitudinal Studies, Institute of Education, London, United Kingdom In the modern labor market what Côté (1996) describes as “identity capital”—comprising educational, social, and psychological resources—is at a premium in entering and maintaining employment. One consequence is the extension of education and training while young people acquire the qualifications and skills that will enhance their employability. In accordance with the perspective of life span developmental psychology, this places particular pressure on those young people growing up in disadvantaged circumstances and lacking support, especially when attempting to negotiate the transition from school to work. A particular policy concern in Britain has been directed at those young people who leave full-time education at the minimum age of 16 and then spend a substantial period not in education, employment, or training (NEET). This article reports the result of analyzing longitudinal data, collected for a subsample of the 1970 British Birth Cohort Study surveyed at age 21, to model the relationship of NEET status to earlier educational achievement and circumstances and to assess the added difficulties NEET poses in relation to the building of adult identity capital. It is concluded that although poor educational achievement is the major factor in entering NEET, inner city living for boys and lack of parental interest in their education for girls are also important. For young men the consequences of NEET lie mainly in subsequent poor labor market experience. For young women, the majority of whom are teenage mothers, the damaging effects of NEET extend to the psychological domain as well. It is concluded that effective counseling targeted at high risk groups, along the lines of the new UK “ConneXions” service, are needed to help young people avoid the damaging effects of NEET and make a successful transition to adult life. C 2002 Elsevier Science (USA) Key Words: transition to work; human capital; social capital; identity capital; training; unemployment; education; labor market; qualifications; teenage motherhood.

It is well established that the social and economic context of youth transitions is critically important in determining their shape and their outcomes for different groups. These effects, operating across the life course and from one generation to the next, draw attention to the need to study interactions between developmental processes and the social context in which they take place. Life span developmental psychology offers a set of perspectives for doing this (Super, 1980; Vondracek, Address correspondence and reprint requests to John Bynner, Centre for Longitudinal Studies, Institute of Education, 20 Bedford Way, London WC1H 0AL, United Kingdom. E-mail: [email protected]. 289 0001-8791/02 $35.00 C

2002 Elsevier Science (USA) All rights reserved.

290

BYNNER AND PARSONS

Lerner, & Schulenberg, 1986; Savickas, 1985; Blustein et al., 1997; Bynner, 1998; Crocket & Silbereisen, 2000; Silbereisen, 1994). Contexts are changing over time so different cohorts of young people will experience their effects differently. Life course theory (Brooks-Gunn, Phelps, & Elder, 1991; Elder, 1974, 1991; Crockett & Silbereisen, 2000) underlines the point that, regardless of social origin, young people in successive cohorts face different sets of obstacles and opportunities when constructing their own life courses. One facet of social change that has been noted by numerous commentators is a prolongation over the past 20 years or so of the transition from school to work and the increased complexity encountered in passing through it (Jones & Wallace, 1992; Banks et al., 1992; Bynner, Chisholm, & Furlong, 1997). Predictability of life course “trajectories” originating in certain locations in the social structure to particular outcomes in the labor market “opportunity structure” available locally (Roberts, 1984) gives way to the individualized life course in which personal agency is of paramount importance in the “negotiation” of the transition that has to be undertaken (Evans & Heinz, 1994; Crocket & Silbereisen, 2000). In what has been described as the “risk society” (Beck, 1986) there is increasing uncertainty about the choices to make and increasing probability that the wrong ones will lead to inferior life chances. But despite the loosening of structural constraints, as some writers have been at pains to stress, much of the old determinacy remains: Individualization is still bounded by class, gender, and ethnicity (Furlong & Cartmel, 1997; Roberts, Clark, & Wallace, 1994; Breen & Goldthorpe, 2001). The concentrations of disadvantage identified with location in the social structure continue to be reproduced from one generation to the next. Under the conditions of the risk society certification and the skills acquired through kinds of employment experience become increasingly important in maintaining a position in the adult labor market. Those who do not have these “human capital” attributes (Becker, 1975), deemed important by employers, face difficulties not only in entering employment but in sustaining any kind of fulfilling career. Categorized in the United States as the “high risk category of non-college bound youth” (Worthington & Juntunen, 1997), such young people often find themselves on the margins of the labor market, moving between various short-term unskilled jobs and unemployment; young women frequently exit early from the labor market to pursue the alternative route of motherhood (Bynner, Ferri, & Shepherd, 1997; Coles, 2000). Such polarization between the “haves” and the “have-nots” in terms of human capital is increasingly characterized as social exclusion for a substantial minority from mainstream adult life. Apart from patchy employment prospects, subsequent consequences may include difficult relationships, lack of social and political participation, poor physical and mental health, drug abuse, and criminality (Robins & Rutter, 1990; Atkinson & Hills, 1997). Although human capital, as embodied in skills and qualifications, serves as some kind of insurance against social exclusion, may not on its own be sufficient to sustain a fulfilling adult life. In addition to the need for social support networks, or “social capital” (Coleman, 1998), and family know-how, or “cultural

SOCIAL EXCLUSION AND THE TRANSITION

291

capital” (Bordieu & Passeron, 1977), biological and health factors may also play a part, of which low birth weight has been identified in some studies as significant (Wadsworth, 1991; Silva & Stanton, 1996). According to Côté (1996, 1997) there has also been an increasing premium placed by employers on the possession of “identity capital.” This embraces the three forms of capital just described and a range of psychological attributes. Its active form may be seen as manifested in the personal agency that enables individuals to “navigate” their way into and through the modern labor market (Evans & Heinz, 1994; Evans & Furlong, 1997). A lack of such attributes typically originates in a childhood marked by disadvantaged family circumstances and family values that place little emphasis on educational achievement (Bynner, 1998). In the British context, there has been particular concern about young people who have suffered these difficulties and who are consequently described as “Status Zero,” “Generation X,” “Getting Nowhere,” and “Off Register” (Williamson, 1997; Pearce & Hillman, 1998; Bynner, Ferri, & Shepherd, 1997; Bentley & Gurumurthy, 1999). The common theme of all of these categorizations is disengagement, particularly from the labor market and the means of entering it through education or training. In Britain the group who have attracted particular attention from policy-makers are those who, during the critical period of the late teens, spend a substantial amount of time outside any form of education, employment, or training (NEET). A major report from the UK Government’s Social Exclusion Unit was devoted exclusively to the problems of this group and a new policy of counseling and support for these young people was formulated, “ConneXions,” to help them achieve successful transitions to adulthood (Social Exclusion Unit, 1999). Two questions arise about such young people. First, what characterizes those who enter NEET? Are they the group who have simply failed to do well at school and therefore drop out of all organized activity at the first opportunity or are there other things that are distinctive about them which put them on an even weaker opportunity route? Second, is the experience of NEET no more than a temporary staging post on a life course marred by disadvantage and failure or does the experience in itself constitute a disabling condition or identity capital deficit in its own right, making subsequent adjustment to the demands of adult life significantly more difficult? This second, stronger view of NEET is that failure to gain the critical work experience and job training after leaving school is permanently damaging not only with respect to employment, but also in making a satisfactory adjustment to adult life. In the British context, particularly, employers expect young school-leavers to gain experience in occupationally useful ways (Bynner & Roberts, 1991). Failure to do so marks the young person as an employment risk. Identifying a problematic transition is not the same as defining it. For the purposes of understanding both the origins of NEET and its consequences for subsequent adult statuses, it was necessary to have an operational definition that would capture as precisely as possible the attributes of such youth. In this article, we use longitudinal data from the 1970 British Birth Cohort Study (BCS70) to operationalize NEET and to model entry into it and exit from it into statuses in adult life.

292

BYNNER AND PARSONS

More precisely we want to use the data to assess the penalty in identity capital terms attached to NEET status in the teens, over and above the penalty attached to lack of qualifications and other disadvantaging factors in young people’s early lives. METHODS Data Source BCS70 comprises a sample of all individuals born in Britain during the week 5–11 April 1970 (n = 16,761), who have been followed up subsequently to adult life. Information has been collected from a variety of sources, including interviews with parents, teachers, and medical professionals, together with educational tests and self-completion questionnaires. In addition to the initial survey at birth, follow-up surveys have taken place at ages 5, 10, 16, and 26 and most recently at age 30. In 1991, at age 21, a representative 10% sample (n = 1,623) of the original birth cohort was also followed up in a study of basic skills (Ekinsmyth & Bynner, 1994). In addition to assessing cohort members’ competence in literacy and numeracy and establishing their current employment, housing, family life, and health statuses, this 1991 subsample survey also contained occupational history data, comprising a month-by-month record of relevant statuses—education, employment, and training—back to age 16. It therefore provided exactly the data needed for identifying NEET status. BCS70 was also an appropriate dataset with which to investigate NEET for other reasons. In 1986, when the cohort reached age 16 and were able to leave school, about 50% of the 16-year-old population were leaving full-time education. This compares with 70% leaving school at age 16 in 1976. By the end of the 1980s, the proportion of school children leaving at the minimum age had reduced to about one-third. In consequence, in the 1970 cohort, we see a group of young men and women whose opportunities when leaving school at age 16 have become symptomatic of the situation for young people in Britain ever since. Instead of work, they encountered youth training as embodied in the government’s national scheme (YTS) or unemployment (Dolton et al., 1999). In 1986, YTS lasted 2 years; by 1988 all benefits for unemployed young people between the ages of 16 and 18 were removed to “encourage” them to engage in youth training or stay on in education; but unemployment was still a preferred option for some. The cohort’s experience therefore exemplified the new world, in which young people found themselves, of increasing pressure to enter training, go back to education, or take on any kind of job. Labor market inactivity—including that connected with teenage motherhood—was becoming increasingly stigmatized as an unacceptable option. Variables The variables for inclusion in the models reflected the postulated origins of NEET in terms of the different elements of capital formation and its identity capital outcomes. Table 1a lists the variables representing antecedent influences hypothesized as leading to NEET, including highest qualification achieved by age 16. Table 1b then lists the postulated outcomes of NEET.


293

TABLE 1a Antecedents of NEET Status between Ages 16 and 18 Variable information CM birth weight low (birth) Measured in ounces, converted to grams CM family social class (birth) Registrar General’s Classification (RGSC) based on father’s occupation or mother’s occupation if “no father” or father information missing Parent(s) did not read to CM (age 5) CM parent asked who in the family read to CM Parent(s) had no interest in CM education (age 10) Composite score from information given by CM teacher: (a) if parents had met the teacher or (b) showed interest in CM education CM has no hobbies/interests (age 10) Parent answered whether CM did any of 13 listed spare time activities, coded as follows: often (2), sometimes (1), or never/hardly ever (0). Scores were aggregated and grouped Inner city neighborhood (age 10) Parent selected from a list the best description for the neighborhood where the family lived CM receives free school meals (age 10) Parent reported if CM received free school meals Family receives state benefits excluding pensions and child benefit (age 10) Parent checked all benefits that any member of the immediate family received CM cognitive ability low (age 10) This was measured by performance in two tests. Scores were aggregated and grouped. The Edinburgh Reading Test: a shortened version of this test of word recognition was used after consultation with its authors (Godfrey, Thompson, & Unit, 1978). The shortened test contained 67 items examining vocabulary, syntax, sequencing, comprehension, and retention. Friendly Maths Test: the lack of an appropriate mathematics test for 10-year-olds led to the development of a special test for the BCS70 cohort. It consisted of a total of 72 multiple-choice questions and covered in essence the rules of arithmetic, number skills, fractions, measures in a variety of forms, algebra, geometry, and statistics.

Values

0 = 2515 g or more 1 = under 2515 g 0 = nonmanual or skilled manual 1 = semiskilled or unskilled manual 0 = mother and/or father read to CM 1 = mother and father do not read to CM 0 = very interested 1 = low/no interest 0 = top 3 quartile ranges 1 = bottom quartile range.

0 = rural, village, outskirts of town, other 1 = inner urban, council estate 0 = no 1 = yes 0 = no 1 = yes 0 = top 3 quartile ranges 1 = bottom quartile rang.

294

BYNNER AND PARSONS TABLE 1a—Continued Variable information

No or minimum qualifications at age 16 (age 21) BCS70 were one of the last cohorts to sit the two-tiered examination structure of Ordinary Level (O-Levels) or Certificate of Secondary Education (CSE) qualifications. O-Level grades range from A to E, with A to C being pass grades. CSE grades range from 1 to 6, with grade 1 being equivalent to O-Level grade C and grades 2–5 being lower level passes. CM self-reported all qualifications they held at age 21 and the age they achieved them; these were converted to a scale of “highest qualification achieved”

Values

0 = O-Levels grade A–C or CSE grade 1 1 = CSE grades 2–5 2 = no formal qualifications

Note. CM= cohort member; text in brackets = the survey in which the variable was measured; Code 0 = reference category.

Antecedents Variables are labeled to indicate the direction of the postulated influence in precipitating NEET, e.g., no qualifications or family in financial difficulties. They comprise physical characteristics (low birth weight), family circumstances at age 10 (including inner city neighborhood and receipt of state benefits and free school meals), cultural capital of the home (manual social class and parents showed little or no interest in cohort member’s education), educational achievement (combined reading and math score at age 10 in the lowest quartile range, few hobbies of any kind at age 10, and no qualifications at age 16). Outcomes The variables taken to signify identity capital comprise occupational and marital status, self-assessed physical health, mental health (as measured by the Malaise Inventory, Rutter et al., 1970, designed to assess depression), and self-appraisal (fatalism, lack of a sense of control, dissatisfaction with life, life problems). Analytic Approach The analysis was carried out in three stages. Stage 1 comprised the operationalization of NEET based on the BCS70 21year occupational history data. To identify more precisely the distinguishing characteristics of NEET young people we restricted analysis to those who had left school at the minimum age of 16 and were not in full-time education in January 1987 (n = 930, 470 boys and 460 girls). This was to reduce as much possible the confounding effects of educational achievement with NEET; i.e., those young people pursuing the academic route to A levels and higher are by definition


295

TABLE 1b Postulated Outcomes of NEET Status between Ages 16 and 18 Variable Information Employed Whether CM was in full-time or part-time employment at age 21 NEET Whether CM was not employed, in training or education at age 21 Ever married or cohabited by 21 CM reported if they were or ever had lived with a partner or had been married by age 21 General Health poor at 21 CM reported if they had been in excellent, good, fair, or poor health in the 12 months prior to interview Depressed at 21 CM had their psychological well-being assessed by use of the Malaise Inventory (Rutter, et al., 1970). Twenty-four yes/no questions elicited whether feelings of anxiety and depression were currently being experienced. A “depressed” score is assigned if “yes” is answered to 8 or more questions Fatalistic attitude CM opinion on three statements relating to employment and job opportunities:

Values

0 = other 1 = employed 0 = employed, in training or education 1 = other 0 = no 1 = yes 0 = excellent or good 1 = okay or poor 0 = okay 1 = depressed

0 = bottom 3 quartile ranges 1 = top quartile range

Getting a job today is just a matter of chance. Success at work is just a matter of luck. Getting on at work depends on others. Opinion was graded on a 5-point scale ranging from strongly disagree to strongly agree. The average score over the three questions was measured, with a high score representing a fatalistic attitude. Dissatisfaction with Life: does CM get what want out of life? CM had to chose which statement comes closest to their own view Lack of Control: does CM feel they have control over what happens in life? CM had to chose which statement comes closest to their own view

Problems in life: can CM run life as they want to? CM had to chose which statement comes closest to their own view

0 = I usually gets what want out of life 1 = I never really gets what want out of life 0 = I usually have free choice and control over my life 1 = Whatever I do has no real effect on what happens to me 0 = Usually I can run my life more or less as I want to 1 = I usually find life’s problems just too much for me

Note. CM = cohort member; Code 0 = reference category.

296

BYNNER AND PARSONS

engaged in education over the period 16–18 and are therefore by definition non-NEET. Stage 2 used a logistic regression model to assess separately for young men and young women the variables that predicted the status of NEET. The model was built up in two steps; the first with highest qualification at 16 excluded and the second with highest qualification included. The idea was to test whether inclusion of school leaving qualifications eliminated the effects of earlier circumstances and achievement. That is to say we wanted to determine whether the possible influence of these variables on NEET operated entirely through highest qualification achieved or whether as features of social, cultural, and biological capital they continued to have an independent effect on NEET status and its possible consequences for identity capital formation. Stage 3 again used a logistic regression model to assess separately for young men and young women the effect of NEET status on the various outcomes. The model was built up in a number of steps: each of the outcome variables was first predicted from NEET status alone; second, from NEET status plus highest qualification; and finally, from NEET status plus highest qualification plus all the variables used to predict NEET. Under this last condition of maximum statistical control, if NEET status continues to predict the various outcomes we can conclude that the experience of NEET has a distinct effect on the various identity capital outcomes. This influence is over and above that of lack of qualifications and the other potential influences with which the NEET effect might be confounded. The logistic regression model used here always involved the prediction of a binary outcome variable, e.g., “NEET/Not NEET” and “employed/not employed,” in terms of a set of antecedent (or predictor) variables. The results are reported as relative odds or odds ratios for each category of each predictor variable compared with the odds ratio for a reference category, which in this analysis is by definition 1. Odds ratios greater than 1 signify a positive relationship between category membership and the outcome and odds ratio less than 1, a negative relationship. Thus for prediction of NEET status from the three categories of the qualifications variable “O level grade A-C/CSE grade 1 (or higher) qualifications,” “CSE grade 2–5,” and “no qualifications,” with the reference category set at “O level grade A-C/CSE grade 1 (or higher) qualifications,” we might expect the category “no qualifications” to attract an odds ratio substantially greater than 1. As a criterion for establishing the statistical significance of the difference between a given odds ratio and 1 we set p < .05 but also noted odds ratios, which fell just outside this range, i.e., up to p < .10. Missing Value Imputation The longitudinal data on which the modeling was based contained much missing data. To maintain a comparable sample size of 930 cases across all analyses missing values were imputed. The method employed in the SPSS statistical package, MVA, displays and tabulates the patterns of missing data to establish whether the data


297

are missing at random. Data can be categorical or quantitative for each variable. The program then estimates means, standard deviation, covariances, and correlations using multiple-regression or expectation-maximization (EM) methods. We adopted the latter. This assumes that the pattern of missing values conforms to that of the observed data, i.e., is nonrandom. To obtain the MVA data, a data set was constructed that contained all variables from birth to age 10 that discriminated between NEET and non-NEET. The complete list is displayed in the Appendix, Table A1. DEFINING NEET As NEET status reflects the dynamics of young people’s lives it has to be defined longitudinally, i.e., it must represent a minimum period of time outside education, training, and employment as opposed to being in one or more of them over the same period. However, the precise boundaries for this experience are not obvious. Many young people leave full-time education at the end of the summer term (June/July)—having passed the minimum leaving age of 16—to work over the summer period and then return to education in the autumn. Others do not make up their minds until they have had a term away from school, returning in the following January. Some move between education training and short-term jobs interspersed with unemployment. Another complication is part-time employment, which many young people mix with education or unemployment. In the case of young women with children (teenage mothers), part-time work is often mixed with child-care. Child-care itself is obviously an occupation, but because it involves, for some, complete exit from the labor market, it may also be seen as aligned to NEET. Accordingly the study focused on the education, employment, and training activities of the 1970 cohort during the 24 months from January 1987 to December 1988 inclusive, i.e., January 1987 was taken as the start date instead of September 1986 to allow for a “settling down” period. Numerous exploratory analyses were carried out on the data to determine which cutoff points produced the strongest discrimination between those young people categorized as NEET as opposed to non-NEET. The final decision was to define NEET as “6 months or more during the ages 16–18 outside education, employment, or training.” This contrasts with the category in education employment or training for all of the 24 months between ages 16 and 18 and leaves a missing period of 18–24 months where the status is unclear. For the purposes of the analysis that follows, the latter two categories were combined as “non-NEET.” Two versions of the classification were tested, one with part-time jobs classified as employment and one with part-time jobs classified as unemployment. Table 2 shows the proportions of young men and women classified by their NEET status with and without inclusion of part-time work in NEET. Eleven percent of the total sample experienced 6 months or more with no education, employment, or training over the ages 16–18, comprising 7% young men and roughly twice as many, 14%, young women. The higher proportion of young women partly reflects the status of those who were out of the labor market through having children and

298

BYNNER AND PARSONS

TABLE 2 Grouped Distribution by Number of Months in Education, Full-Time or Part-Time Employment or Training between January 1987 and December 1988 With part-time work excluded from NEET All % NEET: 0 to 17 months education, employment or training 18–24 months education, employment or training EET: 24 months, education, employment or training n

Males

With part-time work included in NEET

Females

All

Males

Females

10

7

14

15

10

19

12

12

11

14

14

13

73

81

75

72

76

68

930

470

460

930

470

460

not actively seeking work, i.e., some of these NEET young women were pursuing the alternative full-time career of motherhood. The first of these two classifications produced the stronger discrimination, i.e., part-time work is best not treated as a feature of NEET. Thus the teenage mothers placed in the NEET category were not engaged in any form of full-time or part-time employment for at least 6 months of the designated period. Notably our specification of NEET corresponds quite closely to the definition used in the pioneering study of “Status Zero youth,” which also used a period of 6 months out of the labor market as indicating lack of engagement (Istance, Rees, & Williamson, 1994; Williamson, 1997). Predicting NEET Status between 16 and 18 Tables 3 and 4 give the odds ratios for young men and young women and the sample as a whole in the prediction of NEET. The figures in the “All” column give the overall profile of NEET status young people. They were likely to have low birth weight and to have grown up in inner city public housing estates in homes marked by poverty (free school meals and state benefits) and lacking cultural capital (parents not reading to the children and lacking interest in their children’s education). Although cognitive ability at age 10 did not appear to be involved, when highest qualification at age 16 was taken into account, a strong effect of educational achievement was evident. Young people with no qualifications were six times as likely to be in NEET status as those with “O level” or above qualifications. The separate analysis for young men and young women showed similarities and some striking differences in the odds ratios. Thus although the low birth weight effect was evident for both sexes parents’ reading at age 5 only featured for the young men and parents’ interest in education at age 10 only for young women. The particularly notable sex difference was for two features of material disadvantage:


299

TABLE 3 Predicting NEET: Odds Ratios Predictors

All

Young men

Young women

Part 1: Without highest qualification at 16 RGSC IV or V = 0 1.30 1.45 Low birthweight = 0 2.50 2.71∗ Parents did not read to child = 5 1.68 2.56 Free School Meals or State Benefits = 10 1.89 1.00 Inner City or Council Estate = 10 2.01 3.84 Low cognitive ability = 10 1.11 1.18 Few hobbies or interests = 10 1.08 0.52 Little parental interest = 10 1.61 0.98

1.00 2.39∗ 1.31 2.55 1.47 1.13 1.46 2.28

Part 2: With highest qualification at 16 RGSC IV or V = 0 1.32 1.25 Low birthweight = 0 2.45 2.95∗ Parents did not read to child = 5 1.52 2.55 Free School Meals or State Benefits = 10 1.59 0.79 Inner City or Council Estate = 10 2.03 4.03 Low cognitive ability = 10 0.83 1.10 Few hobbies or interests = 10 1.10 0.54 Little parental interest = 10 1.26 0.70 Highest qualification: CSE = 16 1.82 0.96 Highest qualification: none = 16 5.84 9.32

1.16 2.15 1.17 2.20 1.48 0.72 1.44 1.75∗ 2.72 6.21

Note. Age at which data were collected is indicated at the end of each variable description. Bold types signifies statistical significance at p < .05; an asterisk signifies statistical significance at p < .10.

inner city housing and family poverty. For boys inner city housing had a large effect (odds ratio = 3.84), whereas for girls family poverty appeared to matter more (odds ratio = 2.55). When highest qualification at 16 was brought into the model the odds ratios for girls were reduced and in the case of low birth weight and lack of parental interest in children’s education reduced to statistical insignificance. For boys the reductions were generally smaller or actually increased. In the case of inner city living the odds ratio rose from 3.84 to 4.03, showing the centrality of geographical location to boys’ experience of NEET. For both sexes highest qualification again had the highest odds ratio of all the predictors, 9.32 for boys and 6.21 for girls, showing the dominance of educational achievement in young people’s life chances. But notably many other factors in early life experience also remained significant independently of qualifications, suggesting that family circumstances are also an important influence. The lack of effects for manual social class and for low cognitive ability were unexpected, but almost certainly reflect the relative homogeneity of school leavers with respect to these characteristics compared with the others that the analysis embraced. Table 4 summarizes the results of modeling the impact of NEET status on the identity capital outcomes. The table shows separately for young men and young women the NEET status odds ratio for each outcome; first for NEET alone, second

300

BYNNER AND PARSONS TABLE 4 Predicting the outcomes of NEET: Odds Ratios Predictors Young men

Outcomes at 21 NEET21 Full-Time or Part-Time Employment Married/cohabiting Poor general heath Malaise Fatalistic attitude Dissatisfaction with life Lack of control over life Problems with life

Young women

NEET with NEET with NEET with CSE or no quals NEET with CSE or no quals CSE or controls + early CSE or controls + early no quals experience no quals experience NEET controls controls NEET controls controls 4.46 .24

3.59 .32

3.32 .34

7.76 .13

5.83 .17

5.32 .19

.92 1.73 3.23 2.50 2.34

.85 1.55 2.12 1.95 1.92∗

.76 1.45 2.20 1.85 1.66

4.00 1.38 1.81∗ 2.25 3.51

3.23 1.08 1.76∗ 1.70 2.93

3.09 1.00 1.69 1.56 2.96

2.65

1.77

1.41

4.20

3.36

3.47

1.52

.87

.81

4.13

3.18

3.79

Note. Bold signifies statistical significance, p < .05; an asterisk signifies statistical significance at p < .10.

with qualification level added as a control, and third with qualification level and the set of early experience variables used as controls to predict NEET. The full results giving the odds ratios for all the variables in the models are supplied in the Appendix, Tables A1 and A2. The results support the hypothesis that NEET status has a negative effect on the adult outcomes associated with identity capital formation, particularly for young women. For young men the effects of NEET status in the late teens could be seen mainly through poor labor market performance, especially though the continuation of NEET status itself at age 21. These effects were sustained at a slightly reduced level when controls for qualifications were included in the model and were reduced again (marginally) when the wider set of early experience variables were added in as well. Thus young men who had experienced NEET were over three times as likely as those who had avoided NEET to not be in education, employment, or training at age 21, taking account of qualifications and early life experiences. The odds ratio for NEET was even higher for the young women remaining at 5.3 when all the controls were applied. But this may well be because the young women in the NEET group with one or two children had particular difficulties in reentering the labor market, returning to education, or in undertaking training. Further adverse consequences of NEET for young men were restricted to lack of full-time and part-time employment. Although such other hypothesized outcomes as depression and fatalistic attitudes, dissatisfaction with life, lack of a sense of control, and experiencing problems in life all had significant odds ratios in the model


301

without controls, when controls were applied these odds ratios reduced in size, failing to maintain statistical significance in the model with maximum controls. Notably the variables that replaced them as significant predictors of NEET were lack of qualifications and inner city residence and childhood poverty (Appendix, Table A1). For young women the picture was rather different, with NEET’s effects not only sustained in relation to labor market outcomes, but also extending to early marriage or cohabiting, feelings of dissatisfaction with life, lack of a sense of control, and experiencing problems in life. NEET maintained statistically significant odds ratios for all of these outcomes even in the model with maximum controls. It seemed that these young women were suffering particular problems to which their earlier NEET experience had contributed directly. As for boys, the other factors implicated included lack of qualifications inner city residence and childhood poverty (Appendix, Table A2). DISCUSSION AND CONCLUSIONS The analysis enables us to identify the key characteristics that separate young people with NEET status from others. For young people who leave education at the minimum age of 16, capital in the home, as reflected in parent’s not reading to child (boys at age 5) and lack of parental interest in child’s education (girls at age 10) predict NEET. For boys, living in the inner-city is also significant; whereas for girls, family poverty (e.g. free school meals) matters. Notably these effects persist even when highest qualification achieved at 16 is taken into account, suggesting that the components of identity capital derived from family circumstances and experience add to, rather than operate through, educational achievement in driving some young people toward NEET. The role of inner city housing estate residence for boys gives particularly striking endorsement to the problematic nature of this experience for boys’ life chances (e.g., Power & Tunstall, 1994). For girls the significance of educational interest in the home (or rather lack of it) appears to push them along a path which, for many in the NEET category, is identified with early motherhood (cf. Griffin, 1985; Wallace, 1987). The difficulty in assessing NEET as a distinct category for girls needs to be acknowledged. Numbers were too small to separate NEET girls into two groups: those who were looking after a baby or babies at home and those who had yet to become parents. The latter are clearly closest to the boys in relation to labor market status, but were too few in number to investigate separately in this study. Further work on a larger longitudinal data set will be needed to investigate the differences between them. However, the centrality of child bearing in the construction of female careers (Wallace, 1987; Hakim, 1996; Evans & Heinz, 1994) suggests that young women’s “dropping out” through pregnancy has a certain functional equivalence to young men’s disengagement from education employment and training. Young women who drop out without becoming pregnant are probably very similar to the young mothers in most other respects. The consequences of NEET status in early adulthood point again to the differences in the lives of men and women and the paths they take to social exclusion.

302

BYNNER AND PARSONS

The dominance of poor labor market experience as the main outcome associated with NEET for young men does query whether NEET status does damage their identity capital formation in Côté’s broad sense of the term (1996) rather than just its human capital component. The experience of NEET simply compounds a history of educational failure, reducing prospects of employment or for acquiring human capital through education or training even further. In this sense NEET experience, unaccompanied by other factors, may well be not much more than a staging post on the downward path to the bottom of the labor market and social exclusion. For young women the NEET experience appears to impact on other facets of identity as well. The association of NEET with negative psychological states, including (self-reported) lack of a sense of control over life and problems and dissatisfaction with life, points perhaps to more fundamental damage occurring. And this is at a time when, in terms of educational achievement and progress in the middle to higher echelons of the labor market, women’s prospects have never been better (Hakim, 1996; Bynner and Parsons, 1997; Arnot et al., 1999). Perhaps it is their powerlessness to take advantages of these opportunities that underpins these NEET young women’s negative feelings about themselves. On the other hand, again we need to qualify such a conclusion on methodological grounds. Women are more willing to express their feelings about themselves openly than men, so young NEET men’s lack of acknowledgment of psychological difficulties does not rule out their existence. We set the task at the beginning of this analysis of attempting to identify the category of experience over the ages 16–18, which was both characterized by lack of education, employment, and training and predictive of later social exclusion outcomes at 21. We settled on a category of NEET experience of 6 months or more over the ages 16–18 not in education, employment, or training. When part-time employment was also excluded from the definition of this status we had identified a category of young people whose subsequent lives were clearly marked by difficulty. These signs of social exclusion included poor labor market experience, depression, early parenting, and poor housing. In the case of men, engagement in the labor market was likely to be marginal with much experience of unemployment. In the case of women an early career at home looking after children was more likely. The results underline the importance of taking the social context and changes in it into account in studies of the transition from school to work and vocational choice, as argued in the “Life Span” and “Life Course” perspectives in developmental psychology (Super, 1980; Vondracek, Learner, & Schulenberg, 1986; Savickas, 1985; Blustein et al., 1997; Bynner, 1999; Silbereisen, 1994; Elder, 1991; Brooks-Gunn, Phelps, & Elder, 1991; Elder, 1974; Crockett & Silbereisen, 2000). The cohort born in Great Britain in 1970 faced exceptional difficulties in making a successful transition to work. In the context of a disappearing youth labor market, and considerable uncertainty about the means of maximizing job opportunities in the future, these young people faced the choice whether to staying on in education or leave, and if they left, whether to take any job or training scheme on offer or wait for something better to turn up. Some were lucky with the choices they made;


303

others without qualifications or work experience faced an increasingly uncertain future. Those boys growing up in impoverished inner city areas, often lacking good schools and housing, and those girls growing up in families without educational commitment, more frequently than others drifted into the NEET status, which disadvantaged their prospects even further. Given that ever-more complex forms of identity capital are likely to define employability in the future (Côté 1996, 1997), the social context in which such capital is acquired becomes increasingly important. Clearly the ages 16–18 represent a critical stage in the lives of such young people which underlines the importance of professional intervention to move their careers off the exclusion path toward fulfilling occupations. If, through the help of career counselors and others, they can land secure jobs with training and prospects of career progression, then a secure future for them may still be assured (Worthington & Juntunen, 1997). However, lack of qualifications will continually be a problem if the jobs NEET young people enter terminate. For those who spend a substantial part of the period not engaged in employment and who do not or cannot take the opportunity to engage in education or training either, the future may be bleak. The lack of physical amenities, educational resources, and employment opportunities that also frequently characteries the inner city neighborhoods, in which many of these young people grow up, exacerbates their difficulties even further (cf. Dolton et al., 1999). Clearly, these young people have not been excluded from training and employment altogether. In fact, under the definition we used, up to 18 months of the period had been spent in doing one of these. But probably what characterized these experiences was the lack of a genuine base for employability. This makes the case for investment in an education and training infrastructure that will keep opportunities open. It also underlines the need for much stronger commitment on the part of employers to ensuring that first jobs, as well as training experiences, are all seen as part of an educational progression into proper adult work. The British system, with its variety of routes to skilled employment, offers, through training, a pale shadow of the much stronger systems of vocational preparation in evidence in other European countries (Heinz, 1990; Rose, 1991; Bynner & Roberts, 1991; Evans, 2000). In the United States staying on to age 18 to graduate 12th Grade is the norm rather than the exception, though concerns in the United States with dropouts and the need for new models of vocational preparation have a striking resonance with these results (Hamilton & Hamilton, 1999). The modern apprenticeship now promoted by the British government and targeted at all early school leavers goes some of the way toward the German apprenticeship model, combining work-based training with off-site vocational education (DES, 1991). But despite high hopes for modern apprenticeship, it is unlikely that it will embrace all young people over the ages 16–18 and already there are signs that many young people who embark on it do not complete it: only 40% get a vocational qualification from their apprenticeship. There is clearly a long way to go still in terms of the strengthening of early labor market experience in employability directions than the current systems in many

304

BYNNER AND PARSONS

countries allow. Effective counseling services and educational investment become ever more vital as the means of bridging the gap. APPENDIX A Variables Used in the Imputation of Missing Values At birth CM birthweight CM mother smoking habit during pregnancy Age CM mother at her first birth Age CM mother and father left full-time education Marital status of CM parents Social class (based on father’s occupation or mother’s occupation if “no father” or father information missing) Age 5 CM living with natural/adopted mother CM living with natural/adopted father Parents’ divorced/separated Number of family moves since birth Housing tenure: home ownership, rented accommodation, tied property, etc. Overcrowded living accommodation: ratio of number people in house to number of rooms (excluding kitchen and bathroom) CM ever been in Local Authority care CM ever separated from mother for 1 month or more Presence of a long-term illness in the household CM father experienced unemployment in the last 12 months CM attended preschool or nursery CM parents read to CM CM father figure helped mother with domestic duties and childcare responsibilities Behavior adjustment [measured on the Rutter (home) scale, a modified version of the Rutter “A” Scale (Rutter et al., 1970)] Cognitive development [measured by CM performance in the Copying Designs Test (a test to obtain some assessment of the child’s perceptuomotor ability) and the Human Figure Drawing Test (a modified version of the Draw-a-Man test originally devised by Goodenough, 1926, and later developed by Harris, 1963) Age 10 CM family in receipt of state benefits CM has free school meals CM ever been in care Presence of a long-term family illness in the household CM living with natural/adopted father CM Parents divorced/separated Overcrowded living accommodation: ratio of number people in house to number of rooms (excluding kitchen and bathroom) Housing tenure: home ownership, rented accommodation, tied property, etc. Number of family moves since birth Description of neighborhood CM family lived in (inner city, suburbs, rural, etc.) Parent(s) interest in CM education Parent(s) education aspirations for CM: wanted them to leave ft education at 16, pursue post-16 education, etc.

305


Number of CM’s hobbies and interests Behavior adjustment [measured on the Rutter (home) scale, a modified version of the Rutter “A” Scale (Rutter et al., 1970)] Cognitive development [measured by CM performance in The Edinburgh Reading Test (a shortened version of this test of word recognition was used after consultation with its authors, Godfrey, Thompson, Unit, 1978, which examined vocabulary, syntax, sequencing, comprehension, and retention) and the Friendly Maths Test (a special mathematics test developed for the BCS70 cohort covering the rules of arithmetic, number skills, fractions, measures in a variety of forms, algebra, geometry, and statistics)] Age 16 (from age 21 data set) Highest qualification: each CM listed all qualifications they held at age 21 and the age they attained the qualification. A “highest qualification” variable was then derived from all the public examinations a CM had passed at age 16. As the BCS70 cohort was one of the last to experience the two-tiered examination structure of Ordinary Level examinations (O-Levels) and the less academic Certificate of Secondary Education (CSE) examinations. Some CMs also sat General Certificate of Secondary Education (GCSE) examinations, the examination system which was to replace O-Levels and CSE examinations. Note. CM = cohort member. TABLE A1 Prediction of Adult Outcomes from NEET—Odds Ratios for Young Men Not get No Not run FT or Married/ Poor Fatalistic what want control life as PT emp Neet 21 cohab health Malaise attitude out of life over life want to NEET NEET Hq16-cse Hq16-none

0.24 0.32 0.72 0.28

4.46 3.59 1.45 2.78

0.92 0.85 0.93 1.30

1.73 1.55 1.07 1.50

3.23 2.12 1.57 5.01

2.50 1.95 2.15 3.35

2.34 1.92 1.38 2.28

2.65 1.77 4.25 9.14

1.52 0.87 1.72 6.14

NEET Hq16-cse Hq16-none RGSC IV or V 0 Low Birth weight 0 Parents read to child 5 FSM or State Benefits 10 Inner City or Council 10 Cognitive ability 10 Few hobbies 10 Little parental interest 10

0.34 0.81 0.36 0.87

3.32 1.31 2.17∗ 1.33

0.76 0.79 1.00 1.02

1.45 0.96 1.42 1.04

2.20 1.62 5.99 0.53

1.85 2.05 2.72 1.10

1.66 1.40 2.16 1.03

1.41 4.70 9.94 1.06

0.81 1.99 6.77 0.76

0.73

1.76

0.44

0.58

0.57

1.09

2.14∗

4.78

3.36∗

1.03

1.08

1.24

1.01

1.66

1.11

1.22

1.24

0.84

0.62∗

1.67∗

1.42

0.88

0.74

1.02

0.77

0.79

0.41∗

0.79

1.16

1.27

1.19

0.90

1.11

1.49∗

1.66

1.35

0.86

1.06

1.29

1.75

1.23

0.86

0.69

0.72

0.62

0.77 0.85

1.24 1.08

0.82 1.15

0.83 0.75

0.62 0.90

0.66 1.92

1.03 1.59∗

1.25 1.20

0.64 2.11∗

Note. Bold type signifies statistical significance, p < .05; figures marked with an asterisk signify statistical significance, p < .10.

306

BYNNER AND PARSONS TABLE A2 Prediction of Adult Outcomes from NEET—Odds Ratios for Young Women Not get what No Not run FT or Married/ Poor Fatalistic want out control life as PT emp Neet 21 cohab health Malaise attitude of life over life want to

NEET NEET Hq16-cse Hq16-none

0.13 0.17 0.49 0.26

7.76 5.83 2.18 3.89

4.00 3.23 1.34 2.77

1.38 1.08 1.40 2.54

1.81∗ 1.76∗ 1.00 1.15

2.25 1.70 1.41 2.91

3.51 2.93 1.22 2.20

4.20 3.36 2.38 2.10

4.13 3.18 1.87 2.65∗

NEET Hq16-cse Hq16-none RGSC IV or V 0 Low birth weight 0 Parents read to child 5 FSM or State Benefits 10 Inner City or Council 10 Cognitive ability 10 Few hobbies 10 Little parental interest 10

0.19 0.62∗ 0.39 0.80

5.32 1.74 2.64 1.44

3.09 1.30 2.58 1.00

1.00 1.22 2.12 1.13

1.69 0.98 1.20 1.03

1.56 1.21 2.69 0.99

2.96 1.11 2.09∗ 1.37

3.47 2.58 2.45 0.93

3.79 1.85 3.13 1.17

1.14

0.97

1.08

0.54

1.23

0.81

0.99

1.23

1.13

0.95

1.06

1.03

1.12

1.10

1.08

1.23

0.60

1.62

0.60

1.53∗

1.56

1.58

1.29

1.59∗

0.98

1.07

0.76

0.89

1.25

0.81

1.12

2.14

2.36

1.37

1.52

1.10

0.56

1.73

1.04

1.27

0.97

1.54

1.31

0.89

1.15

0.76 0.98

1.32 0.98

0.72 1.15

1.24 0.92

0.72 0.59

0.87 0.56∗

1.64 0.54

0.64 0.83

1.29 0.34

Note. Bold type signifies statistical significance, p < .05; figures marked with an asterisk signify statistical significance, p < .10.

APPENDIX B Comparison Categories for Variables in Logistic Regressions Personal Attributes 1. 2. 3. 4.

Started nursery/school after age 4 vs Started nursery/school by age 4 More than 5 days absent from school at age 10 vs No days absent from school at age 10 No qualifications at 16 vs O-Level/CSE grade 1 or NVQ2 qualifications at 16 Low grade qualifications at 16 (CSE grades 2 to 5/NVQ1) vs O-Level/CSE grade 1 or NVQ2 qualifications at 16

The BCS70 cohort was the last to sit the two-tiered system of O-Level (Ordinary Level) and CSE (Certificate of Secondary Education) examinations at age 16. O-Level exams are graded A–E, with grade C being the lowest pass. CSE exams are graded 1–5, with grade 1 being deemed equivalent to an O-Level grade C pass. The National Vocational Qualification (NVQ) level system attempt place all academic and vocational qualifications within one system. NVQ levels range from NVQ1 to NVQ6, with NVQ5 being equivalent to degree level qualifications.


307

Socioeconomic Characteristics of Family 1. Father in manual occupation at birth/no father vs father in nonmanual occupation at birth 2. Overcrowded accommodation at age 10 (more than 1 person per room) vs Accommodation at age 10 with up to 1 person per room 3. Living in an inner city environment at age 10 vs Living in outskirts of town or a rural environment at age 10 4. Few household goods in the home at 5 vs Average of above number of household goods in the home at 5 Of a list of household possessions such as television, telephone, washing machine, and so on, the variable was split into cohort members with less than average and those with the average and above number of goods. Scores ranged from 0 to 7, with 5 being the average number. 5. Father not in regular paid work at 10 vs Father in regular paid work at 10 6. Gross family income less than £100 per week vs Gross family income £100 or more per week

Education of Mother/Educational Aspirations of Parents for Cohort Member (CM) 1. Mother not staying at school past end of compulsory education (controlling for changes to the end of compulsory education by accounting for age of mother) vs Mother experienced some form of extended education 2. Mother not having any qualifications vs Mother with some qualifications 3. Mother having little interest in CM education at 10 vs Mother with interest in CM education 4. Parents were unsure/did not want CM to continue training after they left school at 16 5. Parents wanted CM to continue training after they left school at 16.

REFERENCES Arnot, M., David, M., & Weiner, G. (1999). Closing the gender gap. Cambridge, MA: Polity Press. Atkinson, A. B., & Hills, J. (Eds.) (1997). Exclusion, employment and opportunity. London: Center for the Analysis of Social Exclusion, London School of Economics and Political Science. Banks, M., Bates, I., Breakwell, G., Bynner, J., Emler, N., Jamieson, L., & Roberts, K. (1992). Careers and identities. Buckingham: Open University Press. Becker, G. S. (1975). Human capital. Washing DC: National Bureau of Economic Research. Beck, U. (1986). Risk society. London: Sage. Bentley, T., & Gurumurthy, R. (1999). Destination unknown: Engaging with the problems of marginalised youth. London: Demos. Blustein, D. L., Phillips, S. D., Jobin-Davis, K., Finkelberg, S. L., & Roarke, A. E. (1997). A theory building investigation of the School-for-work transition. The Counselling Psychologist, 25, 364– 402. Bourdieu, P., & Passeron, J.-C. (1977). Reproduction in education, society and culture. Beverley Hills: Sage. Breen, R., & Goldthorpe, J. H. (2001). Class, mobility and merit: The experience of two British birth cohorts. European Sociological Review, 17(2), 1–27. Brooks-Gunn, J., Phelps, E., & Elder, G. H. (1991). Studying lives through time: Secondary data analyses in developmental psychology. Developmental Psychology, 27, 899–910. Bynner, J. (1998). Education and family components of identity in the transition from school to work. International Journal of Behavioral Development, 22, 29–54. Bynner, J. (1999). New routes to employment: integration and exclusion. In W. R. Heinz (Ed.), From education to work: Cross national perspectives. Cambridge, UK: Cambridge Univ. Press.

308

BYNNER AND PARSONS

Bynner, J., & Parsons, S. (1997). Getting on with qualifications. In J. Bynner, E., Ferri, & P. Shepherd (Eds.), Twenty something in the 1990s: Getting on, getting by, getting nowhere. Gower, UK; Ashgate. Bynner, J., Chisholm, L., & Furlong, A. (Eds.) (1997). Youth, citizenship and social change. Aldershot, UK: Ashgate. Bynner, J., Ferri, E., & Shepherd, P. (Eds.) (1997). Twenty something in the 1990s: Getting on, getting by, getting nowhere. Aldershot, UK: Ashgate. Bynner, J., & Roberts, K. (Eds.) (1991). Youth and work: Transition to employment in England and Germany. London: Anglo-German Foundation. Coleman, J. (1998). Social capital in the creation of human capital. American Journal of Sociology, 94 (Suppl.), 95–120. Coles, R. (2000). Changing patterns of youth transition: Vulnerable groups, welfare careers and social exclusion. In J. Bynner & R. K. Silbereisen (Eds.), Adversity and challenge in the life course in England and Germany. Basingstoke, UK: Macmillan. Côté, J. E. (1996). Sociological perspective on identity formation: The culture-identity link. Journal of Adolescence, 19, 417–428. Côté, J. E. (1997). An empirical test of the identity capital model. Journal of Adolescence, 20, 577–597. Crocket, L., & Silbereisen, R. (Eds.) (2000). Negotiating adolescence in times of social change. Cambridge, UK: Cambridge Univ. Press. DES (Department of Education and Science). (1991). Education and training for the 21st century. London: H.M Stationery Office. Dolton, P., Makepiece, G., Hutton, S., & Audas, R. (1999). Making the grade: Education, the labour market and young people. York, UK: Joseph Rowntree Foundation. Ekinsmyth, C., & Bynner, J. (1994). The basic skills of young adults. London: ALBSU. Elder, G. H., Jr. (1974). Children of the Great Depression: Social change in life experience. Chicago: Univ. of Chicago Press. Elder, G. H., Jr. (1991). Lives and social change. In W. R. Heinz (Ed.), Theoretical advances in life course research: Volume I. Status passages and the life course (pp. 58–86). Weinheim: Deutscher Studien Verlag. Evans, K. (2000). Educational systems and transitions. In J. Bynner & R. K. Silbereisen (Eds.), Adversity and challenge in the life course in England and Germany. Basingstoke, UK: Macmillan. Evans, K., & Furlong, A. (1997). Metaphors of youth transitions: Niches, pathways, trajectories or navigations. In J. Bynner, L. Chisholm, & A. Furlong (Eds.), Youth, citizenship and social change. Aldershot, UK: Ashgate. Evans, K., & Heinz, W. R. (Eds.), (1994). Becoming adults in the 1990s. London: Anglo German Foundation. Furlong, A., & Cartmel, F. (1997). Young people and social change: Individualisation and risk in late modernity. Buckingham, UK: Open University Press. Griffin, C. (1985). Typical girls: Young women from school to the full-time job market. London: Routledge & Kegan Paul. Hakim, C. (1996). Key issues in women’s work. London: Athlone. Hamilton, S. F., & Hamilton, M. A. (1999). Creating new pathways to adulthood by adapting German apprenticeship to the United States. In W. R. Heinz (Ed.), From education to work: Cross national perspectives. Cambridge, UK: Cambridge Univ. Press. Heinz, W. R. (1990). The transition from school to work in crisis: Coping with threatening unemployment. Journal of Adolescent Research, 2, 127–141. Jones, G., & Wallace, C. (1992). Youth, family and citizenship. Buckingham, UK: Open University Press. Istance, D., Rees, G., & Williamson, H. (1994). Young people not in education, training or employment in south glamorgan. Cardiff, Wales: South Glamorgan Training and Enterprise Council. Pearce, N., & Hillman, J. (1998). Wasted youth: Raising achievement and tackling social exclusion. London: Institute for Public Policy Research.


309

Power, A., & Tunstall, R. (1994). Swimming against the tide. York, UK: Joseph Rountree Foundation. Roberts, K. (1984). School leavers and their prospects. Milton Keynes: Open University Press. Roberts, K., Clark, S., & Wallace, C. (1994). Flexibility and individualization: A comparison of transitions into employment in england and Germany. Sociology, 20, 31–54. Robins, L., & Rutter, M. (1990). Straight and devious pathways from childhood to adulthood. Cambridge, UK: Cambridge Univ. Press. Rose, R. (1991). Prospective evaluation through comparative analysis in time-space perspective. In P. Ryan (Ed.), International comparisons of vocational education and training for intermediate skills. London: Falmer Press. Rutter, M., Tivard, J., & Whitemore, K. (1970). Education, Health and Behavior. London: Longman. Savickas, M. L. (1985). Identity in vocational development, Journal of Vocational Behavior, 27, 329– 327. Silbereisen, R. K. (1994). Differential timing of vocational choices. Presented to the 13th Biennial Meeting of the Society for the Study of Behavioural Development, Amsterdam. Silva, P. A., & Stanton, W. R. (1996). Child to adult: The Dunedin Multidisciplinary Health and Development Study. Oxford, UK: Oxford Univ. Press. Social Exclusion Unit. (1999b). Bridging the gap: New opportunities for 16–18 year-olds not in education employment or training. London: The Stationery Office. Super, D. E. (1980). A life-span, life-space approach to career development. Journal of Vocational Behavior, 16, 282–298. Vondracek, F. W., Lerner, R. M., & Schulenberg, J. E. (1986). Career development: A life-span developmental approach. Hillsdale, NJ: Erlbaum. Wadsworth, M. (1991). The imprint of time: Childhood history and adult life. Oxford, UK: Clarendon Press. Wallace, C. (1987). For richer or poorer: Growing up in and out of Work. London: Tavistock. Williamson, H. (1997). Status zero youth and the ‘underclass’: Some considerations. In R. McDonald (Ed.), Youth, the underclass and social exclusion. London: Routledge. Worthington, R. L., & Juntunen, C. L. (1997). The vocational development of non-college bound youth. The Counselling Psychologist, 25, 323–363. Received: October 1, 2001