PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

Ft] 309/ .

Mug E £0819

 

 

 

 

 

 

.161.“ 2° ‘9 £503

 

 

 

 

 

 

 

 

 

1M wm.m.14

YOUTH WORK AND PSYCHOLOGICAL WELL-BEING

By

Deborah J. Safron

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Sociology

1999

ABSTRACT
YOUTH WORK AND PSYCHOLOGICAL WELL-BEING
By

Deborah J. Safron

The present study assesses the psychological consequences of youth work in an
older adolescent sample. Following from Mortimer and colleagues' groundbreaking
research on employment for high school aged adolescents, we hypothesized that poor
work quality would decrease self-efficacy and psychological well-being in an older,
college aged sample. We examined the psychological implications of four aspects of
work quality: noxious work conditions, work/school conﬂict, work complexity, and
work autonomy. We also explored the role of co-worker social support in mitigating the
negative impact of poor work quality on psychological outcomes.

We assessed the above relationships through a closed-ended survey completed by
602 undergraduate students at Michigan State University. Our measures of work quality,
social support and psychological well-being are scales that have been used successfully in
previous research on adolescents and adults. Furthermore, 10 follow-up interviews were
performed to provide a richer context for the survey data.

Results indicate that three out of the four aspects of work quality (noxious
conditions, work/school conﬂict, and low work complexity) are negatively related to
psychological well-being for older adolescent workers. In addition to contemporaneous
effects of work quality, there is some limited support for long term psychological

implications of noxious work conditions. Furthermore, co-worker social support has a

positive impact on psychological well-being. and mitigates the relationship between low
work complexity and psychological well-being. We also explore demographic
differences in the impact of work quality on psychological well-being.

The present study makes an important contribution to two literatures. First, our
results inform the general literatures on work and well-being. Second, our findings have
important implications for the adolescent development literature. Findings are discussed

in terms of theoretical advances as well as practical/policy implications.

ACKNOWLEDGMENTS

I would like to thank Dr. Cliff Broman for his infinite time, energy, and
confidence in this project. His dedication and support will always be appreciated. I also
wish to thank my committee members: Dr. Jan Bokemeier, Dr. Steve Gold. and Dr. Stan
Kaplowitz for providing helpful feedback and support for drafts of this Dissertation. A
special thank you to Lana Hamilton who conducted the qualitative interviews for this
project. I also wish to thank the Survey Research Division of the Institute for Political
and Social Research at Michigan State University for coding the data used in this project.

I would like to thank my wonderful husband, Michael Burke, for his constant
support. devotion and belief throughout this process. I would also like to thank my
parents, Hal Safron & Jacqueline Saltz and Sandra and Jerry Begel for their support
throughout my academic career. A very special thank you to Gordon Weinberg for his

significant role in inspiring my love of research and exploration.

iv

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................ vii
LIST OF FIGURES .......................................................................................................... viii
INTRODUCTION ............................................................................................................... 1
CHAPTER 1:
LITERATURE REVIEW .................................................................................................... 4
A Brief History of Youth Work .......................................................................................... 4
Early vs. Late Adolescence ................................................................................................. 8
Psychological Outcomes ................................................................................................... 12
Self Concept ............................................................................................................. 12
Psychological Distress / Psychological Well Being ............................................... l7
Psychological Outcomes of Youth Work .......................................................................... 22
Work Status ............................................................................................................. 23
Work Hours ............................................................................................................. 29
Quality of Work ................................................................................................................. 32
Salience of Youth Work .......................................................................................... 33
Noxious Work Conditions ....................................................................................... 36
Role Conﬂict ............................................................................................................ 41
Alienated Labor and Occupational Self-Direction .................................................. 47
Demographic Differences in Work Conditions ....................................................... 58
Source of Support .................................................................................................... 64
Type of Support ....................................................................................................... 66
Perceived Support .................................................................................................... 67
Cost of Social Relationships .................................................................................... 67
Long Term Psychological Implications ............................................................................ 69
Importance of the Present Study ....................................................................................... 72

CHAPTER 2:

METHOD .......................................................................................................................... 76
Sample... ........................................................................................................... 76
Survey MethOdOlOgy ......................................................................................................... 78
Measures” ..................................................................................................................... 79
Recall Accuracy of Work Quality ........................................................................... 80
Measures of Work Quality ....................................................................................... 82
Psychological Measures ........................................................................................... 86
Social Support .......................................................................................................... 87
Sociodemographic Variables ................................................................................... 88
Qualitative Interviews ....................................................................................................... 88
Human Subjects ................................................................................................................. 90
CHAPTER 3:
RESULTS .......................................................................................................................... 92
Regression Assumptions ................................................................................................... 92
Employment Status and Work Hours ................................................................................ 93
Psychological Consequences of Work Quality ................................................................. 97
Co-Worker Social Support .............................................................................................. 101
Demographic Differences ................................................................................................ 102
Long Term Implications of Work Quality ...................................................................... 105
CHAPTER 4:
DISCUSSION ................................................................................................................. 108
Noxious Work Conditions ............................................................................................... 108
Work / School Conﬂict ................................................................................................... 1 14
Occupational Self Direction ............................................................................................ 120
Work Complexity .................................................................................................. 121
Work Autonomy .................................................................................................... 125
Emotional Labor .............................................................................................................. 130
Co-Worker Social Support .............................................................................................. 132
Demographic Differences ................................................................................................ 135
Practical Implications ...................................................................................................... 146
Limitations and Directions for Future Research ............................................................. 155
Conclusion... .. ................................................................................................................ 159
APPENDIX A: WORK QUALITY MEASURES .......................................................... 162
APPENDD( B: PSYCHOLOGICAL MEASURES ........................................................ 164
APPENDIX C: SOCIAL SUPPORT SCALE ................................................................. 165
REFERENCES .................................................................................... 166

vi

LIST OF TABLES

Table 1. Means and Standard Deviations for all Variables in Analysis ............................ 80
Table 2. Job Characteristics of Sample ............................................................................. 96
Table 3. Regression of Work Quality Predicting Psychological Distress ........................ 98
Table 4. Regression of Work Quality Predicting ............................................................ 100
Table 5. Regression of Co—Worker Support Predicting Psychological Distress ............ 102
Table 6. Demographic Differences in the Impact of Work Quality on .......................... 105
Table 7. Regression of Psychological Distress on T1 Work Quality ............................. 107

vii

LIST OF FIGURES

Figure 1. Proposed model of work quality, depression and co-worker support ............ 40

Figure 2. Mediating role of self-efﬁcacy ........................................................................... 57

viii

INTRODUCTION

In recent US. history, there has been a vast increase in the number of adolescents
who work at part-time jobs during the school year. Since adolescence is an important
developmental period, researchers, policy makers and parents are concerned about the
short and long term psychological ramifications of youth work. The objective of the
present study is to examine the psychological implications of youth work as well as
explore the mechanisms by which work impacts upon adolescents’ psychological well
being.

To date, literature on the psychological consequences of youth work has been
sparse. In addition, the research that does exist has been conceptually limited in its
measurement of work. Most studies examine the psychological consequences of work
status (employed or not employed) and/or the sheer number of hours worked. Such
research treats youth work as a “black box”, ignoring the substantial variation in the
quality of work conditions. In order to explore the mechanisms by which youth work
impacts upon psychological well being, we must first understand the content or the
experience of work.

Mortimer and colleagues’ groundbreaking research on youth work draws upon
adult literatures on work conditions and psychological outcomes in order to inform the
youth work literature. The present study will replicate and expand upon Mortimer’s work
by testing relationships between various measures of work quality and psychological well
being that have been demonstrated in adult populations. We explore three main aspects

of work quality that have been shown previously to have important psychological

implications for both adults and adolescents: work autonomy, work complexity, role
conﬂict, and noxious work conditions. Work autonomy refers to the degree of decision
making power one has at work while work complexity refers to the variety and
complexity of work tasks. Role conﬂict refers to experiencing contradictory demands
from work and other valued roles. Noxious work conditions refer to negative work
environments such as excessive heat/cold, noise, and time pressures. Our measures of
psychological outcomes follow from both adult literatures as well as research on
adolescent development. Specifically, we examine self-efficacy (the individual’s
assessment of his/her competence) and psychological distress.

We hypothesize that work autonomy and complexity are positively related to
psychological well being and work/school conﬂict and noxious work conditions are
negatively related to psychological well being for adolescents. Furthermore, we explore
the ability of co-worker social support to buffer the negative impact of poor work quality
on adolescent psychological well being. We also examine differences in the impact of
work quality based on gender, class and race. Lastly, we assess the long-term
psychological outcomes of work quality by examining the impact of earlier work
experiences on current psychological well being.

Previous research on youth work (most notably, Mortimer and colleagues) has
focused almost exclusively on early adolescence, neglecting the implications of work
experience for adolescents who are older than high school age. The present study
expandsupon Mortimer’s research by bridging the gap between studies focusing on early
adolescence and adulthood. Speciﬁcally, we will explore the life cycle stage of late

adolescence. As we shall see, the stage of “late

adolescence” has often elluded definition in the literature. For purposes of the present
study, we deﬁne college students as individuals in the stage of late adolescence, since
such individuals are in a transitory stage between early adolescence and adulthood.

The present study assesses the above relationships through the use of a survey
methodology. We distributed closed-ended surveys to 602 undergraduate students at
Michigan State University. Our measures of work quality, social support and
psychological well being are scales that have been used successfully in previous research
on adolescents and adults. In addition, 10 follow-up interviews were performed to
provide a richer context for the survey data.

The present study will make a significant contribution to academic literatures as
well as suggest important practical/policy implications. First we will add to the general
literature on work and psychological well being by clarifying this relationship within a
particular age group (adolescence). Furthermore, this study will contribute to the
adolescent development literature by exploring a largely neglected socialization context
(work) for adolescent psychological well being.

Second, since having a part-time job during adolescence has become almost
normative in the US, it is important that we fully understand its implications for
adolescent development and well being. The present study will clarify which aspects of
work are most harmful to psychological well being and suggest factors (e. g. social
support) that may mitigate such effects on adolescents’ psychological outcomes. It is
only through understanding the mechanisms through which work impacts upon

psychological well being that we can hope to improve the lives of youth workers.

CHAPTER 1:
LITERATURE REVIEW

A BRIEF HISTORY OF YOUTH WORK

Before we examine the implications of adolescent employment in the present day,
it is important that we understand the phenomena of youth work within a broader
historical framework. Throughout US. history, the decision of young persons to seek
employment has not simply reﬂected individual choice. Rather, the nature and extent of
youth work has changed over time in conjunction with broader economic and familial
conditions. In fact, as we shall see, the ebb and ﬂow of “youth work” with the historical
tide has also been signiﬁcantly associated with the creation of (and changing status of)
the life stage that we now know as “adolescence.”

In colonial America, child labor was normative for all but those from wealthy
families (who were fortunate enough to attend school). During this time period, a stage
of life distinct from childhood and adulthood did not exist. Rather, children were seen as
“miniature adults” who were primarily prized only for their economic value (Demos
1977). By the time a child reached the teenage years, s/he was capable of covering
his/her own expenses through signiﬁcant contributions to the work of the household, the
family farm, or the family enterprise (Vulcan 1968).

The creation of “adolescence” as a distinct stage of life is strongly linked to the
rise of the manufacturing age in the nineteenth century. With the advent of machines that
displaced large numbers of workers, keen competition arose between adult workers and
child laborers (who were willing to work for low wages). In response, adult

workingmen’s organizations sponsored child labor and

compulsory education laws (Vulcan 1968). As a result, child labor declined substantially
during the ﬁrst half of the twentieth century (Wrigley 1986). Between the years of 1900
and 1950, labor rates for 14-19 year old males declined from 70% to 40%, while the
graduation rate from high school increased from 10% to 50% (Vulcan 1968). Thus, the
20‘h century saw the birth of “adolescence” as a distinct life stage. For the ﬁrst time in
US. history, large numbers of young persons were attending school rather than
contributing substantially to the economy. For the ﬁrst time, many young persons
(particularly from the middle classes) were protected from adult financial responsibilities
and thus were able to develop a separate “youth culture” (Demos and Demos 1973).

Since WWII, the United States has experienced yet another massive economic
change: the transition from a manufacturing to a service economy. Since the 19405,
there has been a rapid growth of the service and retail sectors of the economy at the
expense of production sectors. For instance, between 1940 and 1976, the US. created 9.3
million new jobs in the service sector and 6.6 million jobs in the retail sector (Ginzberg
1977). Furthermore, during the 19803, for every thousand people of working-age, the
US. created 27 clerical, sales and service jobs and lost 16 production, transportation, and
laborer jobs (Wilson 1997 p. 27). These “new jobs” are not comparable, however, to the
previous manufacturing jobs in terms of money, beneﬁts and ﬂexibility. Instead, most
service and retail jobs are characterized by low wages, irregular shifts, low job security,
low beneﬁts, and minimal potential for promotion (Ginzberg 1977; Keithly and Deseran
1995).

Research indicates that countries with the earliest and greatest development of

service economies experience a parallel rise in youth

work. For instance, the United States and Canada, with their large service economies,
also have the largest proportion of student workers among industrialized nations
(Reubens, Harrison, and Rupp 1981). But while young people in the colonial period
worked wad of going to school, recent decades have seen the rise of the “student
worker.” For instance, in the United States, between 1947 and 1980, there was a 65%
increase in part-time work for school enrolled 16-17 year-old boys and a 240% increase
for girls (Greenberger 1988). Recent estimates indicate that 61% of tenth graders, 90%
of eleventh and twelfth graders, and 51% of college student’s work during the school
year (Manning 1990; National Association of Student Employment Administrators). In
addition, over half of employed high school seniors and a quarter of all employed
sophomores work more than 20 hours per/week (Steinberg and Dombusch 1991) while
the average employed college student works 16 hours per week (National Association for
Student Employment Administrators).

But m would the rise of a service economy cause an increase in youth work?
First, the jobs being created in the service sector are particularly well suited to
adolescents. Job attributes such as low wages, low job security and irregular work hours
are likely to be intolerable to an adult worker who needs to support and spend time with
his/her family. These same “bad” job characteristics, however, are typically acceptable
for adolescent employees who require off-hour employment (evenings and weekends),
are not particularly concerned with long term job security, and do not require high wages
since they are typically supported ﬁnancially by their parents (Greenberger and Steinberg
1986). The work provided by adolescents is, in turn, attractive to employers since youth

workers provide a source of cheap and temporary labor

that is necessary in a service economy. Thus, the transition to a service economy in the
US. has produced a new demand for youth workers.

In addition to the impact of the service economy on adolescent employment,
recent changes in family structure have also impacted upon youth work. For reasons that
we shall outline, adolescents are more likely today to return after school to an empty
home. For such youth, getting a paid job may be a more attractive alternative to spending
large amounts of time by him/herself (Greenberger and Steinberg 1986). There are two
main reasons for this change in the ecology of the household.

First, in recent decades, there has been an increase in the likelihood that teenagers
will have employed mothers. The transition to a service economy not only increased
youth work, but also saw unprecedented numbers of women entering the job market. In
the beginning of the 1960s, only 18% of married women with children under age six
worked outside the home, whereas by 1991, 60% of these women were in the labor force
(US. Census Bureau 1992). Thus, teenagers today, (born in the 19803), are the most
likely group to have a mom in the labor force (Greenberger and Steinberg 1986). As a
result, adolescents returning home after school are likely to ﬁnd an empty house since
mom is working. Instead of spending large amounts of time at home alone, adolescents
often follow mom’s example by seeking employment. This serves the dual purpose of
acquiring extra spending money as well as occupying the adolescents’ after school hours
with a viable activity.

Second, in recent decades, there has been a decrease in the number of family
members in any one household. There has been a vast increase in the number of single

parent households and a decrease in the number of

children per family. For instance, one study estimates that 59% of children born in the
early 19803 will live with only one parent for at least a year (Glick 1989) and the
majority of children in mother only families will continue in this situation for the
remainder of their childhood (Bumpass and Sweet 1989). Furthermore, the average
number of children per family is just a fraction more than two (Greenberger and
Steinberg 1986). While adolescents in the past had the potential to interact with many
different family members, today’s youth are simply statistically less likely to have their
schedule mesh with other family members. While an adolescent from a larger family is
no more likely to spend time with any o_n§ family member, s/he is more likely to interact
with various family members at different times. Thus, adolescents from smaller families
(who are likely to spend a signiﬁcant amount of time home alone), may view working as
an attractive alternative activity (Greenberger and Steinberg 1986).

In sum, economic and family changes have had an important impact on youth
work. Although a full analysis of such broad changes is beyond the scope of this brief
historical view of youth work, we have seen that factors such as the rise of the service
economy as well as smaller household structures have increased the number of
adolescents involved in part-time employment. More importantly, the changing nature of
youth work has inﬂuenced our very deﬁnition of “adolescence” as a life stage between

childhood and adulthood.

EARLY VS. LATE ADOLESCENCE
Thus far, our deﬁnition of adolescence has contained an element of ambiguity in

terms of the boundaries of this life stage. Specifically,

when does childhood end and adolescence begin? And when does an adolescent become
a full-ﬂedged adult? Piaget, a prominent early developmental psychologist,
conceptualized adolescence as encompassing the teen years (age 12 to 18). Although
Piaget examined emotional and behavioral components of adolescent development, he
primarily focused on intellectual development. Based on this focus, he broke
adolescence into two periods of mental development: early formal operations (12-15) and
late formal operations (15-18) (Cowan 1978). Following Piaget’s lead, contemporary
scholars generally deﬁne early adolescence as occurring between the ages of 12 and 15
(Brooks-Gunn and Peterson 1984; Brooks-Gunn, Rock and Warren 1989).

The question of concern for our present purposes, however, is the deﬁnition of
“late adolescence.” If Piaget deﬁned this period as occurring between the ages of 15 and
18, the question arises as to whether an individual has made the transition to adulthood by
age eighteen. How have contemporary scholars conceptualized the status transition from
adolescence to adulthood? When does adolescence end and adulthood begin? This
question is not easily answered.

Scholars have proposed many different markers of adulthood such as marriage,
moving out of parents’ home, and full-time job status (Amett and Taber 1994; White
1994). Yet these transition markers do not necessarily occur simultaneously. For
instance, an individual might have a full-time job but still live with his/her parents, or a
person might be married while still attending school. How do we classify such
individuals who appear to be a hybrid of adult and adolescent roles?

Further complicating the picture is that the age at which these transitions occur

has increased in recent decades. First, the age of ﬁrst

marriage has increased substantially in recent decades. In 1960, the median age of ﬁrst
marriage was 22.8 for men 20.3 for women whereas in 1993 the median age was 26.5 for
men and 24.5 for women (U .8. Bureau of the Census 1993).

Second, the status of full-time worker has been delayed in recent decades due to
an increase in higher education. In 1920, only 30% of 14-17 year olds were enrolled in
high school and only 8% of 18-21 year olds were enrolled in college whereas in 1985, the
rates for enrollment in high school and college were 95% and 57% respectively (Elder
1987; Horowitz 1987; US. Department of Education 1988). In the case of college
students, the majority are not ﬁnancially independent. Research indicates that 48% of
undergraduates are ﬁnancially dependent on their parents and 43% received some form of
ﬁnancial aid (National Center for Education Statistics 1993). Thus, the status of full-time
worker (and the ﬁnancial independence that accompanies it) has certainly been delayed
for many young people in recent decades.

Third, the issue of leaving the parents’ home is murky. Even after young people
leave, they typically have not fully established a separate residence since many return for
brief time frames. Studies indicate that about half of children return home for at least a
brief period after their initial leaving (Goldscheider and DaVanzo 1986; Kerckhoff and
Macrae 1992). In addition, although the majority (72%) of college students do n_ot live
with their parents (National Center for Education Statistics 1993), many are still under
the authority and supervision of college dormitory staff. Scholars have adopted the term
“semi-autonomy” to describe this gray area where young people are living in semi-
structured environments and thus are neither dependent nor independent (Goldscheider

and DaVanzo 1989; Young 1987).

10

Thus, instead of a clear-cut transition from adolescence to adulthood, we find a
multitude of transitions, mixing and matching with one another in terms of time frame;
occurring at later and later stages in the life cycle. How then, are we to deﬁne “late
adolescence” when there is no agreed upon deﬁnition; when there is no one transition that
clearly marks adulthood?

For the purposes of the present study, I will choose a deﬁnition of “late
adolescence” that is relevant to the issues concerning youth work that I wish to address.
As discussed, the combination of school and work roles has been historically relevant in
the deﬁnition of adolescence and thus the “student worker” has emerged as the dominant
adolescent form. As we have seen, college students continue to balance student and work
roles long after early adolescence. Yet the research to date has generally only examined
high schoolers, while neglecting this late stage of adolescence. Since the present study is
interested in the psychological impact of youth work over time, we have chosen a
population (college students) that are in a prolonged state of adolescence. It is a group
that is more mature than young adolescents, yet has not fully reached adulthood since, as
discussed, most are “semi-autonomous”, ﬁnancially dependent, and still combing work
and school roles. This life stage that exists between early adolescence and adulthood has
been neglected in previous research on the psychological impact of work. In the
following pages, we will review the literature on work and psychological well being for
early adolescents and adults. Such previous research will serve as a backdrop for the

present study on the psychological consequences of youth work during late adolescence.

11

W

Before immersing ourselves in the literature concerning psychological outcomes
of youth work, let us ﬁrst explore the general wing of such psychological constructs.
If we were to divide the psychological literature into two broad categories, we would ﬁnd
two main psychological variables: self-concept and psychological distress. Let us

explore each in turn.

Self Concept

The self-concept has received much attention as an important psychological
variable in both psychological and sociological research. The self-concept is deﬁned as
an individual’s conception of him/herself that is shaped through social interactions with
others (Cooley 1902; Gecas 1982; Mead 1934; Rosenberg 1981). Researchers
differentiate among various aspects of the self-concept such as identity and self-
evaluative components. Identities include meanings associated with the self and the
content and organization of self-conceptions. Self-evaluation or self-esteem, on the other
hand, refers to the individual’s overall self-evaluation of him/herself. Recently,
researchers have differentiated among various aspects of self-esteem such as self worth,
(an individual’s sense of moral worth) and self-efﬁcacy (an individual’s assessment of
his/her personal competency and effectiveness) (Gecas 1982).

The present study concentrates on the self-efﬁcacy component of the self-concept.
As we will discuss later in more detail, this aspect of self-concept has been demonstrated
to be an important outcome of work experience (our independent variable of interest).

Self-efficacy refers to an individual’s assessment of

12

his/her effectiveness, competence and causal agency (Gecas 1989). In other words, it is
the individual’s sense that he/she has the capacity to have an impact on the world.
Despite this general deﬁnition, self-efﬁcacy has appeared within psychological and
sociological disciplines subsumed under many different labels.

Researchers generally divide efﬁcacy into two categories: internal and external.
An internal sense of efﬁcacy refers to the individual’s belief that outcomes are contingent
upon their own actions. There are many terms in the literature that refer to the sense of
internal efﬁcacy such as self-efﬁcacy (Bandura 1982; Gecas 1989), mastery (Pearlin
1983), internal locus of control (Rotter 1966), competence (Mortimer and Lorence 1979),
self-conﬁdence (Rosenberg 1979), self-directedness (Kohn and Schooler 1983),
instrumentalism (W heaton 1980) and self-reliance (Greenberger and Steinberg 1986). In
contrast, an external sense of efﬁcacy refers to the tendency to attribute outcomes to
forces outside of oneself; such that outcomes occur independently of the individual's
actions. There are many concepts referring to the external sense of efficacy such as
learned helplessness (Seligman 1975), external locus of control (Rotter 1966), alienation
(Marx [1844] 1964), fatalism (Wheaton 1980) and powerlessness (Seeman 1967).

Although the above terms have much in common, each has a slightly different
ﬂavor based on the speciﬁc discipline’s perspective and research interests. For instance,
behavioral psychologists tend to focus on the behavior that results from exposure to
uncontrollable negative stimuli. For instance, Bandura (1982) focuses on the
compromised performances that result from low conﬁdence in one’s abilities. Cognitive
psychologists, on the other hand, focus more on the pm or perception that one has

causal agency, rather than on performance or behavior

13

per se. Sociologists, on the other hand, tend to focus on the shaping of self-efficacy
based upon an individual’s location in the social structure.

The present study takes the latter sociological focus since we are exploring the
psychological impact of the youth work role (which is tied to larger structural
conditions). As such, the present study will use the term self-efficacy, which is widely
used in literatures taking a sociological focus. Furthermore, self-efﬁcacy has been used
extensively by those exploring the psychological impact of work for adolescents
(Mortimer et a1. 1996). In my review of youth work, however, the reader will see various
terms (as noted above) used interchangeably. Certainly there is some difﬁculty in
comparing results across studies since such terms do not have precisely the same
meaning. However, as Mirowsky and Ross (1989) note, such terms are “roughly

interchangeable.”

Adolescence and Self Efﬁcacy

Although self-efﬁcacy is an important psychological construct at all life stages, it
is particularly important for adolescent development. The adolescent period in the life
cycle represents the transition from childhood to adulthood roles. As such, it is a critical
period for forming personality and acquiring new social statuses (Erikson 1959;
Greenberger 1988). Scholars contend that self efﬁcacy is central in the successful
transition to adulthood. In order to become an independent and psychological healthy
adult, the adolescent must attain a sense of causal agency in his/her world (Bandura 1982;
Dombusch 1989; Finch et a1. 1991; Hauser and Levine 1994; Josselson 1980; Mortimer

et a1. 1996). As the adolescent begins to make the

14

transition to work, spousal or parental responsibilities, s/he must feel efﬁcacious in order
to take on these adult roles successfully. The adolescent can no longer view him/herself
as a dependent child, but rather begin see him/herself as a separate, effective, and
competent adult (Josselson 1980).

But how does the adolescent develop a sense of self-efﬁcacy? Much of the
research looking at the predictors of psychological development for adolescents focuses
on the socialization agents of family, school, media and peer groups. Largely ignored has
been the potentially important socialization context of work experience on adolescent
psychological development. This omission is surprising since adolescents spend a
signiﬁcant portion of their time working in part-time jobs outside of the home or school.
Furthermore, the work role is a signiﬁcant marker of the transition to adulthood. For the
ﬁrst time, the adolescent is given the opportunity to master work tasks and thus to
establish efﬁcacy in a worker role‘. Thus, youth work experiences are implicated in the
development of self-efﬁcacy (Mortimer et al. 1996).

Note the present study is interested in exploring the general self-efﬁcacy that
results from work experiences, rather than concentrating on speciﬁc work-related self-
efﬁcacy. This brings to mind a burgeoning literature on global vs. speciﬁc self-esteem.
For instance, Bandura (1982) contends that an individual’s conﬁdence that s/he can
perform well in a speciﬁed task is more likely to lead to performance outcomes than

general feelings of self-esteem. Recently, a number of studies empirically demonstrate

 

' We will discuss alternative arguments in a later section (i.e. the argument that since youth work is
temporary, that it has little impact on self efﬁcacy).

15

that researchers have inappropriately used general self-esteem measures, when speciﬁc
self-esteem measures would be better predictors of behaviors (Rosenberg et al. 1995).

That being said, it is important to keep in mind the present study’s focus on
adolescents’ development of general self-efﬁcacy. Since the majority of youth work is
dissimilar to later adult jobs, we are not particularly interested in adolescents’ sense of
competency in their work tasks per se. More important for adolescent development is
exploring the broader implications of work quality in terms of its impact on adolescents’
overall sense of competency. As we will discuss in more detail later, Kohn’s
generalization contends that individuals learn or generalize the “lessons of the job to
outside-the-job realities (Kohn 1981: p.290). Thus, workers who experience high levels
of occupational self-direction are likely to have a more self-directed orientation in other
aspects of life. Thus, given the present study’s focus, we are not as interested in
discovering speciﬁc work-related competencies as we are uncovering the impact of youth
work on global assessments of competency in adolescent development.

Furthermore, the present study uses a general measure of self-efﬁcacy because of
our overall focus on psychological well-being. Studies indicate that while speciﬁc self-
esteem is more relevant for behaviors, global self-esteem is most relevant to
psychological well-being (Rosenberg et al. 1995). Speciﬁcally, research indicates that an
individual’s sense of causal agency has important implications for his/her psychological
well being. Studies indicate that external attributions are positively associated with
distress (Mirowsky and Ross 1984; Wheaton 1980), whereas internal attributions are
associated with decreased depression and distress (Benassi et al 1988; Kohn and Schooler

1982; Mirowsky and Ross 1990; Ross 1990). In

16

addition to studies on adults, there is preliminary support for the relationship between
self-efﬁcacy and distress for adolescents (McFarlane, Bellissimo and Norman 1995).
Thus, it is those individuals who feel they do not have the capacity to act that are also
more likely to suffer from psychological distress. Since self efﬁcacy has such important
psychological implications, the present study will also examine psychological distress as

an important psychological outcome variable.

Psychologg' al Distress / Psycholom' al Well Being

Current research and conceptualization of psychological distress can be divided
into two main categories: psychiatric epidemiology and social stress research.
Psychiatric epiderniologists generally use the term “mental health” to refer to speciﬁc and
discrete psychiatric disorders such as depression, bipolar disorder, alcohol/drug
dependence, paranoia, schizophrenia and the like. Large scale national surveys, such as
the Epidemiological Catchment Area program (ECA) (Regier et al. 1984) and the
National Comorbidity Survey (NCS) (Kessler et al. 1994) measure the prevalence of such
disorders in the general population. Such surveys use the Diagnostic Interview Schedule
(DIS) (APA 1980, 1987) and the Composite International Diagnostic Interview (CIDI)
(WHO 1990) respectively to assess the presence, duration and severity of psychological
symptoms. According to predeﬁned criteria, the DIS instrument (and its updated and
revised version, the CIDI) group such symptoms into patterns indicating the presence or
absence of discrete psychiatric disorders as designated in various editions of the

Diagnostic and Statistical Manual (American Psychiatric Association 1980, 1987).

17

Social stress researchers, in contrast, do not concentrate on clinical disorders, but
rather on the general psychological health of normal populations. Such scholars
generally use the term “psychological distress” (instead of mental health) to refer to
feelings of depression (feeling sad or hopeless) and anxiety (being tense or worried)
(Mirowsky and Ross). Alternatively, psychological well being simply refers to the
opposite side of the coin (psychological health). Irnportantly, social stress researchers
critique psychiatric epidemiology in terms of its categorization of psychological problems

as distinct entities. Specifically, they argue that the Diagnostic Statistical Manual

 

assumes that there are important cut-off points (based on the number, frequency and
duration of symptoms) that determine the absence or presence of psychiatric disorder.
Under this schema, an individual either has a psychiatric disorder or does not (based on
diagnostic criteria). Social stress researchers argue that in life, things are rarely so black
and white. Rather, psychological problems occur on a continuum, in varying Megs of
severity and duration (MiroWsky and Ross 1989).

Consistent with their conceptualization of psychological problems as occurring

along a continuum, social stress researchers use indexes to measure psychological well

 

being. The widely used Center for Epidemiological Studies’ Depression Scale (CES-D)
contains items pertaining to feelings of fear, depression, and loneliness. Scores on the
CES-D are not used to diagnose speciﬁc psychiatric cases, but rather are summed to
represent each respondents’ distress level. Scores range from high psychological distress
to complete psychological well being. Thus, whereas diagnostic categories include or
exclude cases based on certain criteria, indexes rank respondents based on their le_ve_l of

psychological distress (or psychological well being).

18

Stress researchers do not always agree, however, as to the origins of
“psychological distress.” Some scholars contend that distress is an outcome of role
stressors that are rooted in social statuses (Pearlin 1981). Research from this perspective
has focused on the psychological impact of role stressors related to socioeconomic status
(Kessler et al. 1994; Robins et al. 1981), work conditions (Kohn et al. 1990), race
(Robins et al. 1984; Neff 1984), gender (Kessler et al. 1994; Robins and Regier 1991),
marriage (Gove 1972, 1978), parenthood (Ross and Mirowsky 1988), and disability
(Mirowsky 1994).

Other stress researchers, however, do not view psychological distress as an
outcome of stress, but rather as a perception of stress. Such scholars argue that stressors
are not objective entities, but rather occur within a subjective context that determines
their meaning. Lazarus and colleagues (Lazarus 1991; Lazarus and Folkman 1984, 1987)
argue that whether or not a stressor will be experienced as psychological distress depends
upon the individual’s assessment of the situation. These subjective evaluations will be
determined by individual differences in such things as personal dispositions, early life
experiences, recent events, and ongoing social situations (Dohrenwend et al. 1990;
Farmer and Ferraro 1997; Wheaton 1994). According to this perspective, stress will only
be deﬁned as distress when the individual appraises the stressor as exceeding his/her
personal psychosocial resources. Stressors are not uniformly experienced as distress and
thus subjective individual differences must be taken into account when evaluating the
psychological impact of stress (Lazarus 1991).

Although I acknowledge the importance of individual variation in the meaning of

stressors, the present study will use the previous

19

formulation of stress as causing psychological distress. My interest is in examining the
social patterning of psychological well being. Speciﬁcally with regards to work
conditions, Lazarus (1991) contends that it is not especially useful to identify stressors
that have a negative impact on most workers because stress is ultimately an individual
phenomena (Brief and George 1991). I disagree. Although stress is experienced at the
individual level, it is not an individual phenomena. Rather, as discussed previously,
stressors are rooted in one’s social position and social circumstance. In the present study,
it is useful to discover speciﬁc facets of youth work that have a negative psychological
impact on the majority of adolescents. Since research on the psychological implications
of youth work is still in its infancy stages, it is important to ﬁrst explore large scale
patterns in psychological distress rather than individual variations in response. Let us not
forget that a concentration on the social patterning of outcomes is a central task of

sociology.

Adolescence and Psychological Distress

Psychological distress does not appear spontaneously in adulthood, but rather is
rooted in the individual’s psychological well being during earlier life cycle periods.
Speciﬁcally, reSearch demonstrates that many adult psychological disorders ﬁrst appear
during adolescence (Fleming and Offord 1990; Peterson et al. 1993). Furthermore, there
is recent evidence that psychological distress in adolescence constitutes an important risk
factor for adult psychological distress and disorders (Fleming and Offord 1990; Peterson
et al. 1993). Since psychological well being during adolescence has potentially serious

long term implications for psychological outcomes in

20

adulthood, it is an important area of inquiry for researchers and policy makers interested
in fully understanding the development and course of psychological disorders.

Estimating the rates of psychological distress in adolescence is marred by the
same split in the adult literature between psychiatric epidemiology and social stress
perspectives. Psychiatric epidemiologists tend to use clinical, discrete measures based on
DSM-III criteria (APA 1980). Such researchers generally ﬁnd relatively low rates of
psychological distress with estimates ranging between 0.5% and 2.5% of adolescents
experiencing a major depressive disorder (Anderson et al. 1987; Kashani et al. 1987;
McGee et al. 1990; Whitaker et al. 1990).

In contrast, social stress researchers use indexes to assess adolescents’
psychological distress. Similar to adult populations, the CES-D is a commonly used
index to assess adolescent psychological distress. This scale has been demonstrated to be
reliable and valid for high school and college aged youth coming from diverse
ethnic/racial backgrounds (Roberts et al. 1990; 1991; Schoenbach et al. 1982; Wells et al.
1987). Studies using the CES-D and other index measures ﬁnd that psychological
distress is quite common among adolescents (Emnslie et al. 1990; Fleming and Offord
1990; Gore, Aseltine and Colton 1992; Lewinsohn et al. 1993; Manson et al. 1990; Siegel
et al. 1996). For instance, one study found that over 40% of adolescents showed
“substantial feelings of misery” (Rutter et al. 1981), while another study reported that
48% of an adolescent sample displayed “appreciable misery or depression” (Kaplan,
Hong, and Weinhold 1984). Furthermore, Kutcher and Marion’s (1989) summary of the
literature found that between one-ﬁfth and one-half of the adolescent population displays

some symptoms of depression at any given point in

21

time (p. 897). Thus, in contrast to clinical depression measures, studies employing
general index measures ﬁnd that psychological distress is common among adolescents.

In sum, the present study will focus on psychological distress since it is crucial for
adolescent development and ultimately, adult psychological well being. Since distress
during adolescence is so common, it is important to look at all socialization contexts,
including youth work that may be important in the development of psychological well
being. Since I am interested in the general psychological functioning of the normal
adolescent population, I will use a continuous index measure of psychological distress
rather than a clinical measure. Furthermore, since I am interested in social patterns of
psychological well being, I conceptualize of psychological distress as being an outcome

variable that is reactive to (rather than a deﬁnition of) objective youth work experiences.

PSYCHOLOGICAL OUTCOMES OF YOUTH WORK

Although much research has demonstrated that work conditions have important
implications for gig]; psychological well being, the literature is sparse regarding the
outcomes of work conditions on adolescent psychological well being. The purpose of
the present study is to ﬁll this gap by exploring work as a potentially important
socialization context for adolescents’ psychological outcomes. We will begin by
critically reviewing the literature on youth work and psychological outcomes (self
efﬁcacy and psychological distress). Since this literature is dominated by a focus on
those in early adolescence, I will also speculate on differences for those in late
adolescence. Next, we shall explore how the adult literature on work and psychological

well being can greatly inform the youth work literature

22

and provide a model for exciting new research on youth work and psychological

OUtCOIDCS.

Work Status

There are many methods of measuring youth work experience. Some researchers
use ‘work status’ (working vs. not working) to determine the effects of adolescent work
on psychological outcomes. Are there differences between adolescents who are
employed vs. those who do not work outside the home? Much of the conventional
wisdom assumes that employment is beneﬁcial for youth in terms of increasing self-
efﬁcacy. For instance, William Stephens, a sociologist and author of “Our Children
Should be Working” (1979) asserts that work teaches young people to be self—reliant.
Furthermore, research indicates that both mothers and fathers strongly approve of youth
work. Parents describe their own youth work experience as well as their child’s
employment as being quite positive, with the beneﬁts (such as self efﬁcacy and self
reliance) far outweighing the costs (Aronson, Mortimer, Zierman and Hacker 1996;
Phillips and Sandstrom 1990). Thus, the general public perception in the United States is
that youth work is quite beneﬁcial for adolescent psychological development.

This rhetoric is at least partially based upon one of the earliest and most
inﬂuential empirical studies on the psychological implications of youth work: Elder’s
(1974) study on children of the Great Depression. This study did, in fact, provide support
for the conventional wisdom that work leads to increased self efﬁcacy. Elder examined
the consequences of children contributing to the family’s ﬁnances during the economic

hardship of the Great Depression. Elder found that

23

children who made a ﬁnancial contribution to the family attained a higher sense of
efﬁcacy and a proactive achievement orientation that carried on into adulthood. (Elder
1974)

Although inﬂuential, Elder’s study has some important limitations. First, Elder’s
sample was relatively small (N=l67) and was obtained through non-probability methods.
As a result, it is difﬁcult to generalize results to the general adolescent population. A
second limitation is the inability to assess selection vs. causation effects of work and self-
efﬁcacy. Many researchers interpret Elder’s ﬁndings to indicate that adolescents who
were able to actively improve their families’ economic situation felt a high sense of
control over the world as a result. Another possibility, however, is that adolescents with
higher self-efﬁcacy originally felt more conﬁdent in their ability to help their families
and consequently were more likely to choose to work than adolescents with low self-
efﬁcacy. Since Elder did not measure previous levels of self-efﬁcacy, it is impossible to
test for selection effects (self-efﬁcacy causing work) vs. causation effects (work causing
self-efﬁcacy).

Despite the above limitations, Elder’s study was inﬂuential in providing
preliminm evidence for the positive impact of youth work on self efﬁcacy.
Contemporary researchers must keep in mind, however, that the motivation for youth
work is different today than at the time of Elder’s research. During the Great Depression,
a viable strategy for family survival was to rely on adolescent members of the household
for supplemental income. In fact, at this time period, child labor laws were temporarily
suspended to allow adolescents to contribute to the family income (Angell 1936; Elder

1974).

24

Today, however, the majority of adolescents do not work to support their families.
Instead, most adolescents spend their money on discretionary items such as cars, stereos,
“extra” clothing, and concert tickets. (Greenberger 1988; Lewin-Epstein 1981; Safyer,
Heahy and Colan 1995; Mortimer et al. 1990). Even in the case of college students,
although many do spend their money on college expenses, the majority (48%) are still
ﬁnancially dependent on their parents and many receive ﬁnancial aid (43 %) (National
Center for Education Statistics 1993).

Research ﬁndings from contemporary studies are mixed. Some support Elder’s
ﬁndings that a positive relationship exists between adolescent employment and self-
reliance (Greenberger 1988, Greenberger and Steinberg 1986; Ortman 1988). However,
many of these studies rely on self-report data, in which adolescents claim that working
caused them to gain greater self-reliance. It is not clear whether these self—reports
actually reﬂected a causal relationship between work status and self-reliance, or whether
these adolescents simply internalized the conventional wisdom that working leads to
greater levels of self-efﬁcacy and self-reliance. Other research, using longitudinal data,
ﬁnds the opposite effect, such that adolescent employment actually causes a decrease in
mastery over time (we will discuss this ﬁnding in more detail later) (Finch, Mortimer,
Shanahan and Ryu 1991). Still other research ﬁnds no difference in self efﬁcacy between
working vs. non-working adolescents (Mortimer and Finch 1996). Thus, the research
evidence is still inconclusive regarding the relationship between work status and self-

efﬁcacy.

Demographic Differences in Work Status

25

Work status is heavily inﬂuenced by class and race. First, families today that are
most in need of ﬁnancial resources are the least likely to have an adolescent family
member in the labor force. Research indicates that lower household income and parental
unemployment actually decreases the odds of adolescent participation in the labor force
(Keithly and Deseran 1995). Although the rates of seeking employment are similar by
class and race (Greenberger and Steinberg 1986), research indicates that white, middle-
class suburban adolescents are more likely to actually work at a part-time job than black,
Hispanic or poor urban youth. Thus, although adolescents from all backgrounds seek
employment, it is the youth from families with the most resources that tend to attain part-
time jobs.

There are several explanations for this inequality in work experience by class and
race. First, proponents of the “spatial mismatch” hypothesis point out that the majority of
low-skilled job growth during the last two decades has been in the suburbs (Kasarda
1985). Since poor and minority youth tend to live in the inner city, it is difﬁcult for them
to travel to jobs located some distance from their homes (Fordham 1996; Giordano 1993;
MacLeod 1995; Wilson 1987). Although some empirical work ﬁnds no effect of
residence on youth work (Ellwood 1986; Leonard 1986), a number of studies provide
evidence for the spatial mismatch hypothesis. For instance, Ihlanfeldt and Sjoquist
(1991) found that as the length of the commute increases, the probability of the
adolescent having a job declines. Furthermore, Rosenbaum and Popkin (1991) found that
low income blacks who moved into subsidized housing in the suburbs were more likely
to subsequently have a job than those moved to housing in the city, even when

controlling for background variables. Thus, one reason

26

that low income and minority adolescents are less likely to have part-time jobs is because
of the mismatch between their residence and “job rich” suburban areas.

Another mechanism explaining the lower rates of employment for low SES and
minority youth is a lack of personal contacts to jobs. Research indicates that white
middle class youth are more likely to ﬁnd employment through family or friend networks
than lower SES or minority youth (Borman 1991; Holzer 1987; Wilson 1987). For
instance, Osterrnan (1980) found that white youth attain 57% of their jobs through
personal contacts whereas blacks only found 33% of their jobs in this manner.
Furthermore, many employers use employee referrals extensively when hiring new
workers; a contact base to which many lower income black youth do not have access
(Korenman and Turner 1996; Moss and Tilly 1992; Turner, Fix and Struyk
1991;Waldinger 1993).

The majority of the research on class/race differences in youth employment status,
however, has focused exclusively on those in early adolescence. The situation is likely to
be different for college students. First, regardless of class or racial status, most college
students (72%) live on/near a college campus rather than at home with their parents
(National Center for Education Statistics 1993). Although such youth still contend with
racial discrimination in hiring practices, the effects of “spatial mismatch” are minimized
or eliminated. In contrast to studies of early adolescence, studies on college students
indicate that minority students are actually more likely to work more than white students.
For instance, one study conducted at Michigan State University found that black and
Hispanic students were more likely to be employed than whites students (Michigan State

University 1993). Thus, living on/near a college

27

campus (rather than in a job deprived inner city environment) is likely to level the
playing ﬁeld in terms of attaining employment for disadvantaged youth.2

Although there is much research on class/race differences in employment status,
there is a paucity of research examining differences in the psychological impact of youth
employment status for minority and lower class adolescents. In order to fully understand
the impact of youth work, it is vital that future research compares the psychological
implications of youth employment status on members of various class/race groupings.
The present study will do a preliminary investigation of this issue by comparing the
psychological impact of employment status on adolescents of different racial and SES
backgrounds.

The literature on youth work has a slightly better record, however, in examining
gender differences in employment status. Although recent studies ﬁnd few differences in
youth employment status by gender (Mortimer et al. 1990; US. Department of Labor
1987), there is some evidence that the psychological impact of employment status differs
for adolescent boys and girls. Studies indicate that employment status increases self
efﬁcacy for adolescent girls, but has no impact on adolescent boys (Greenberger 1984,
1988; Mortimer, Finch, Shanahan and Ryu 1992; Steinberg, Fegley, and Dombusch
1991). One plausible explanation for this difference is that conventional expectations for
adolescent girls have not (until recently) included attaining a part-time job. As a result,
entering the labor force at an early age may represent an act of independence and self
assertion, leading to feelings of self efﬁcacy among adolescent girls (Greenberger 1988;

Steinberg et al. 1991). The present study will do an initial test of such gender

 

2 The playing ﬁeld is not leveled, however, in terms of job quality. We will return to this issue later.

28

differences, exploring whether employment continues to lead to self-efﬁcacy for girls in

late adolescence.

Work Hours

Recently, researchers have proposed that the critical variable is not whether or not
an adolescent works, but the number of hours worked (Bachman, Bare and Frankie 1986;
Mortimer and Finch 1986; Steinberg and Dombusch 1991). The importance of _w_ori(
_ho_ui§ has come to the attention of researchers and policy makers as a result of a
signiﬁcant increase in the average number of hours worked by adolescents over the last
few decades in the United States.

Currently, over half of all employed high school seniors and a quarter of all
employed sophomores work more than 20 hours/week (Steinberg and Dombusch 1991), a
small but sizable minority of high school seniors (10%) work 35 hours/week or more
during the school year (Greenberger 1988), and the average college student works 16
hours per week (National Association for Student Employment Administrators). Thus,
many high school and college students spend a signiﬁcant amount of time at work per
week, even during the school year.

Research evidence on the psychological impact of work hours is mixed. Some
studies indicate that long work hours (over 20 hours/week) are associated with
psychological distress (Steinberg and Dombusch 1991). In terms of self efﬁcacy, some
studies indicate that work hours are not related to self-reliance (Steinberg and Dombusch

1991), mastery (Finch et al 1991), or locus of control (Bachman, Bare and Frankie 1986;

29

see Finch et al 1991).3 Other research, however, ﬁnds a positive relationship between
adolescent work hours and self-reliance (Steinberg and Dombusch 1991; Greenberger
1988). How can we explain this discrepancy?

One possibility is that studies differ in their measurement of self-reliance. Studies
ﬁnding no effect of work on self-reliance use measures that tap the adolescents’ feeling
of control over his/her life in general, such as the sense of mastery or locus of control. In
contrast, studies indicating a positive relationship between work and self-reliance, assess
the extent to which adolescents (rather than their parents) have control over speciﬁc
money management decisions. For instance, adolescents who work long hours report
more latitude in deciding how to spend money on leisure and dating activities
(Greenberger 1988). In large part, this greater money management power is related to the
adolescents simply having more disposable income to spend on consumer and leisure
goods.

Such limited money management, however, does not necessarily imply a sense of
control over one’s world. After all, even if an adolescent decides to spend his/her money
on a new CD or on a movie, this does not imply that he/she feels a sense of overall
control, since his/her parents are likely to make ﬁnancial decisions that have a much
greater impact on the adolescents’ life and long term goals. Most adolescents (working
or not) are still largely dependent upon parents for the basic necessities of life such as
food, shelter, and college expenses. For instance, one study indicates that over 80% of
high school seniors who work do not save their earnings for college expenses or other

long range goals (Johnston, Bachman and O’Mally 1982). Furthermore, almost half of

 

3 Recall from our previous discussion that these terms are all roughly interchangeable with self efﬁcacy

30

college students are ﬁnancially dependent on their parents (National Center for Education
Statistics 1993). Thus, since adolescents are not truly ﬁnancially autonomous, money
management seems to be an inadequate measure of autonomy for adolescents. Instead,
more general measures of autonomy are more likely to tap the adolescents’ overall sense

of control over his/her world.

Demographic Differences in Work Hours

Previously we saw that minority and lower SES youth were less likely to be
employed than white middle class adolescents. When such adolescents do obtain
employment, however, they tend to work much longer hours than their counterparts.
Studies indicate that black and lower SES students work longer hours than white middle
class students in both high school (Schill et al. 1985) and college (Michigan State
University 1993). For instance, one study found that among undergraduate students
majoring in engineering, that black students worked nearly twice the hours of whites
(more than 15 hours per week) (Michigan State University 1988). This intense work
schedule for disadvantaged youth is likely to be the result of economic necessity (for
those fortunate enough to ﬁnd employment). Similar to research on employment status,
however, researchers have not yet explored race/class differences in the impact of work
hours on psychological outcomes for adolescents.

In terms of gender, research indicates that adolescent boys work longer hours than
adolescent girls in high school (Lewin-Epstein 1981) and college (Michigan State
University 1993). Some tentative research suggests that long work hours (more than nine

hours per week) decreases psychological well being for

31

girls, but has little impact on adolescent boys (Cherlin and Furstenberg 1986a, 1986b;
Yamoor and Mortimer 1990). Researchers have speculated that since there is a higher
expectation that boys will work in adulthood, that long work hours conﬁrm their image of
masculinity and thus does not have as deleterious psychological effects as for adolescent
girls (Shanahan, Finch, Mortimer and Ryu 1991). The present study will perform an
initial test of racial and gender differences in the psychological impact of work hours for
those in the stage of late adolescence.
QUALITY OF WORK

Thus far, we have discussed youth work in terms of work status and work hours.
The ﬁndings are mixed regarding the impact of work on adolescents’ psychological
outcomes. How can we explain these inconsistent results? One possibility is that the
psychological ramiﬁcations of youth work vary based on the content of the job itself. In
other words, it is not employment per se, or even the number of hours worked that
predicts psychological well being, but rather it is the quality of the job that is pertinent.

Let us begin with an illustrative example. Suppose two adolescents, Jack and Jill
each work at their respective jobs for 20 hours/week. Jack is a fast food cook whose
tasks are routine and involve extreme time pressure. Jill, in contrast, is a swim instructor
whose job affords challenge and variety as well as a more leisurely pace. Although Jack
and Jill each spend 20 hours per week at work, the quality of the activities they engage in
are quite different. From this anecdotal example, it seems plausible because of the many
stressors that Jack experiences at work, his psychological well being is more likely to
suffer than Jill’s. Yet a study measuring only ‘work status’ and/or ‘work hours’ would

rank Jack and Jill’s jobs as exactly equal (since they

32

are both employed and both work for 20 hours/week). Such a study, then, would not
account for the variation in work quality when considering psychological implications.

Thus, it is important that we construct variables that adequately capture the
experience of work. Since we are exploring work as a potentially important socialization
context for adolescents, it is crucial that we fully understand what this context contains.
For how can we understand work’s effects if we do not understand work itself? What
sorts of activities/tasks does the adolescent engage in while at work? What are the
supervisor’s expectations? What sorts of stressors is the adolescent exposed to while at
work? These and other questions are important in understanding why, how, and when
work affects psychological well being.

Fortunately, we have much to guide us in this quest. Although the youth literature
is limited, the effects of work conditions on adult_psychologieal well being has a long
history, with roots in Adam Smith’s conception of work as “toil and trouble”, Marx’s
work on capitalism and worker alienation, Weber’s “Protestant Ethic,” and Durkheim’s
work on social structure and anornie (Dupre and Gagnier 1996; see Mortimer, Lorence
and Kumka 1986, p.10). This existing literature on adults provides a fertile starting point
for studies on youth work.

Salience a Youth Work

Some argue, however, that the adult work literature is unlikely to be applicable to
adolescents. Since the jobs that individuals have during adolescence are often quite
different from those attained later in life (particularly for those attending college), youth
may not incorporate their current worker role into their identity. In other words, the

worker role may not be salient (or important) to the

33

adolescent’s long term sense of self. There is research evidence that the psychological

 

impact of experiences depends upon the salience of the role in which such experiences
occur (see Brown et al. 1987; Simon 1995; Thoits 1992, 1994). Since many youth do
not expect to stay at such jobs over the long term, their work experiences may not be
salient and thus will not have deleterious psychological outcomes.

For instance, suppose that Jack, college student majoring in engineering, has a
part time job at McDonalds. Since Jack plans on obtaining a professional engineering job
after graduation, he is unlikely to consider “fast food worker” to be a salient part of his
long term identity. Although Jack’s job involves repetitive and exhausting tasks, the
knowledge that performing such tasks is temporary and his ultimate worker role will be
quite different, allows Jack to distance himself from his “fast food worker” role and thus
not experience a decrease in self efﬁcacy or psychological well being.

Although this is a plausible argument, there are important reasons to suppose that the
youth work role is, in fact, highly salient to adolescents and thus has an important impact
on psychological outcomes. First, adolescence marks the ﬁrst entry into the worker role
for most individuals. Evidence suggests that individuals are most sensitive to
environmental experiences at the point of ﬁrst acquisition of a role (Van Maanen and
Schein 1979; Mortimer, Finch and Maruyama 1988). Thus, adolescents may be highly
responsive to work conditions simply because it is their ﬁrst introduction to the world of
work (Shanahan et al. 1991).

Second, adolescence is widely recognized as a critical period for the acquisition of a
work identity and the formation of attitudes and orientations towards work (Erikson

1963). Since the work role is an important part of

34

adult identity, the adolescent is likely to imagine his/her future “possible self" in terms of
the adult worker role. Although jobs during adolescence may have a different character
than many adult jobs, the very agt of working is likely to be salient in terms of its
perceived connection to later adult work experiences (Clausen 1993; Mortimer and
Johnson 1997). For instance, the employed adolescent may ponder how his/her current
job will differ from adult working experiences. Or the adolescent may ask him/herself
what job conditions are important in his/her imagined future adult work role. For those in
late adolescence (such as college students), such questions are likely to be even more
paramount since such individuals are closer to attaining an adult worker role (Super
1990). Thus, the very experience of employment is likely to be salient for adolescents
since this life cycle period constitutes an important time for contemplating future adult
worker roles.

Third, adolescents spend a signiﬁcant amount of time working at part time jobs.
As we shall see shortly, many facets of such jobs have been demonstrated to be harmful
for adult psychological functioning. Even if the adolescent is certain that s/he will not
encounter such conditions in his/her later adult work role, s/he is still likely to experience
psychological discomfort during the time period in which s/he is working. For instance,
suppose that Jack’s job (cook at McDonalds) involves exhausting and dirty work under
very hot conditions. The fact that in 3 years or so (after graduation) he will no longer
encounter such conditions is unlikely to be of much comfort wlﬂe Jack is experiencing
noxious work conditions. Since depression is so common in adolescence and

furthermore is a critical risk factor for later adult psychological functioning, any

35

experiences that may cause psychological distress (even temporarily) are of great
concern.

Thus, it is important to study the psychological implications of work experiences
for adolescents. Although the experiences and developmental tasks of adolescence and
adulthood are not identical, a modified form of adult research is applicable to
adolescents. The objective of the present study is to draw from and integrate the adult
literature into an exploration of youth work and psychological well being. We shall now
turn to a review of the adult literature on work conditions and psychological well being as
well as review the few existing studies on adolescence that draw upon this adult
literature.

Noxious Work Conditions

One important dimension of work quality in the adult literature is the various
conditions experienced during the course of a job. There is considerable evidence that
noxious work conditions are negatively related to adult workers’ psychological well
being. First, work conditions that involve uncomfortable physical tasks, such as dirty
work (e. g. garbage collecting) lifting heavy objects (e. g. construction work), and
excessive heat/cold conditions (e. g. winter maintenance) are related to distress and
depression for both men (Kohn and Schooler 1983) and women (Miller, Kohn and
Schooler 1986).

Other job conditions involve a time dimension such as work overload (too much
work to perform in the time allotted), work underload (not enough work to perform),
machine paced work, and time pressure. There is considerable evidence indicating that

time related negative work conditions are positively

36

related to depression (Broadbent and Gath 1981; Bromet, Dew, Parkinson and Shulberg
1988; French, Caplan, and Van Harrison 1982; Houben 1991; House et al. 1986;
Johansson 1981; Karaseck and Theorell 1990; Landsbergis 1988; Margolis and Farran
1984; Sutherland and Davidson 1993).

Other researchers, however, contend that noxious work conditions can have a
925% (or at least neutral) impact on psychological well being. Such scholars argue that
previous research has ignored individual agency in the psychological reactions to
stressors (including noxious work conditions). According to this view, individuals are
motivated to protect and enhance their well being (Thoits 1994b). Reactions to stressful
conditions involve both emotion-focused coping (changing the affective meaning of the
stressor) and problem-focused coping (removing the source of the stress). Empirical
evidence conﬁrms that individuals actively solve many of the problems that confront
them (Kessler Turner and House 1989; Riessman 1990; Brown, Lemyre and Bifulco
1992; Silver, Boon and Stones 1983; Thoits 1994). As a result of such coping efforts, the
argument goes, stressors have little impact on psychological well being (Turner and
Avison 1992).

Some scholars take this argument one step further. They contend that not only do
individual coping efforts mitigate the impact of stressors on psychological well being, but
that coping with stress can actually have a my; impact on psychological well being.
Following Hans Seyle’s biological model of the stress process (Seyle 1976), some social
stress researchers argue that although stress can sometimes produce distress, that more
often, it creates eustress, or an increase in resistance to such stressors. Actively coping

with stressors leads to positive changes such as an

37

increased stress capacity and a strengthening of coping resources. Furthermore, when the
individual perceives him/herself to be successful in overcoming stressors, there is likely
to be a heightened sense of self efﬁcacy and psychological well being (Shanahan and
Mortimer 1996; Wheaton 1994).

The above argument is limited, however, in that it ignores social structural
conditions that constrain individual coping efforts. Stressors and coping resources are
not randomly distributed, but rather vary based on one’s social position (Menaghan and
Merves 1984). As Pearlin puts it, “Certain kinds of life exigencies seem to be
particularly resistant to individual coping efforts...There are situations in which ‘problem
solving’ is not a realistic option” (Pearlin 1991 p.267).

Research conducted on the positive impact of stressors has generally focused on
negative life events such as divorce (Riessman 1990) or unemployment (Kessler, Turner
and House 1989). In these cases, individual coping efforts, such as searching for a new
job or learning to live independently are often successful and lead to increased self
efﬁcacy and psychological well being. Noxious work conditions, in contrast, are more
often chronic conditions (that are part and parcel of the job itself) and thus are more
resistant to individual coping efforts. For example, it is difﬁcult to imagine a garbage
collector changing the dirty and heavy job conditions associated with his/her job or a
waiter/waitress changing the fast paced nature of waiting on tables. Such conditions are
ingrained within the structure of the job itself and individual coping efforts are unlikely to
lead to any real change. Since the present study is interested in the psychological impact

of noxious work conditions, we will follow the lead of most previous literatures by

38

conceptualizing of stressors as having a negative impact on psychological well being.
Adolescence and Noxious Work Conditions

Although there is considerable support for a negative relationship between
noxious work conditions and psychological well being for adults, there is a paucity of
research on work conditions in adolescence. This is surprising since adolescents, who
primarily work in low-skilled service sector jobs, are likely to be exposed to poor
working conditions. Research indicates that the majority of adolescents (54% of males
and 59% of females) are currently working in retail trade jobs such as gas stations and
eating/drinking establishments (Aronson et al. 1996). Such jobs are likely to include
negative conditions such as time pressure and heavy, dirty work. For instance, the main
task of a fast food worker, a cashier, or a waiter/waitress is to perform duties as quickly
as possible, and the tasks of a gas station attendant or a construction worker include
heavy lifting and dirty work.

Another noxious working condition that many adolescents face is the frequent
adjustment to new job conditions. Many adolescents “job jump”; meaning that they have
many different jobs over short periods of time. The frequent acquisition (and quitting) of
jobs carries with it the responsibility of learning new work procedures, establishing work
relations with co-workers and supervisors, and becoming proﬁcient at new work duties.
Adjusting to new job environments on a relatively frequent basis is likely to be more
noxious than having one job over a long period of time. Thus, frequent job jumping is
likely to be a salient work condition for adolescent workers.

There is little empirical research, however, on the impact of noxious work

conditions on psychological outcomes for adolescents.

39

One notable exception (Shanahan et al. 1991) suggests that adolescents exposed to
excessive time pressure at work experience more depressive affect over time. The
present study attempts to add to the literature by examining the impact of noxious work
conditions on adolescent psychological well being. The present study is limited in that it
does not test for all relevant work conditions for adolescents. Since research on work
conditions for adolescents is in its infancy stages, we will follow the lead of previous
literatures on noxious work conditions for adults. We wish to explore the application of
such previous literatures for adolescent workers. Thus, we will examine whether the
negative relationship between noxious work conditions and psychological well being
demonstrated for adults holds for adolescents. Speciﬁcally, we will examine whether
noxious work conditions adolescents encounter at work (time pressure, excessive

heat/cold or noise) have a negative impact on adolescents’ psychological well being.

 
 
 
 

 
  

Work Quality

Noxious Conditions
Work/ School conﬂict
Low work Autonomy
Low Work complexity

     

Depression

 
 

 

 
 

 
 
   
 

Co-wprker Support

Thus, we will test the following hypothesis.
Figure 1. Proposed model of work quality, depression and co-worker support
Hypothesis 1: Noxious work conditions (excessive heat/cold, noise, and time pressures)

are negatively related to psychological well being.

40

Role Can ict

Another important work dimension explored in the adult literature is the conﬂict
experienced between work and other valued role domains. Role conﬂict refers to roles
containing contradictory expectations of what the individual should be doing at a
particular time. For adults, role conﬂict is most likely to occur between work and family
demands. For instance, an individual’s work role may require her to be on time for work,
whereas her parental role may require her to care for a sick child. She cannot perform
both of these roles adequately. Research indicates that the frustrations resulting from
combining contradictory work and family demands lead to psychological distress,
particularly for women who have greater household responsibilities (Aneshensel et al.
1981; Rosenﬁeld 1989; Ross and Mirowsky 1988). Adding children into the picture
further increases the contradictory nature of heavy work and family demands. For
instance, research indicates that employed women with children have more psychological
distress than those without children (Kessler and McRae 1981).

Since the student role is widely regarded as the central “business” of the
adolescent’s life (Shanahan et al. 1991), and because youth work is so prevalent,
adolescents are likely to experience contradictions between work and school demands
(which may subsequently cause psychological distress). Before we examine the
psychological impact of work/school conﬂict, let us ﬁrst examine the ways in which
work and school might contain contradictory demands.

First, daily time is ﬁnite and thus hours spent engaging in work tasks diminish the
time potentially available for schoolwork. Researchers taking this perspective generally

examine the impact of work hours on school related

41

outcomes. Many studies fail to ﬁnd an association, however, between work hours and
school achievement (measured by the student’s grade point average) (Mortimer and
Shanahan 1991, 1994). Researcher generally explain this null ﬁnding by pointing to the
meager amount of time required for homework in the United States. The national
average for time spent on homework during high school has been estimated as less than 4
hours/week (Greenberger and Steinberg, 1986) to as little as one hour per week (Lewin-
Epstein 1981). Furthermore, in the case of college students, although homework time is
more demanding, classroom time decreases signiﬁcantly from about 35 hours/week in
high school to 12-15 hours/week in college. Since this leaves college students with an
20-25 hours per week to study, the total time spent on school related activities is likely to
be similar in high school and college. As a result of modest time requirements for school
related activities, Steinberg and Dombusch (1991) argue that it is unlikely that work
(even intense work) would adversely impact upon school achievement.

Other researchers however, argue that measuring school achievement in terms of
grade point averages has masked the negative impact of long work hours. Students may
compensate for time lost to work by taking less demanding course work (that requires
less effort), curbing participation in outside school activities or copying school work from
classmates in order to maintain their GPA. Empirical studies conﬁrm that long work
hours in high school (usually deﬁned as >20 hours per/week) are associated with more
school-related deviance (e. g. cheating, copying others homework), taking less demanding
course-work, lower participation in extracurricular activities, and lower aspirations and
attainment of post-secondary education (Carr 1996; D’Amico 1984; Greenberger 1988;

Marsh 1991; Mortimer and Finch 1986; Steinberg and

42

Dombusch 1991). Thus, adolescents with overly demanding jobs appear to suffer in
terms of achieving their full potential for school achievement.

Another reason why adolescents might experience work/school conﬂict is because
of the ti_mi_gof work shifts. In other words, w_he_n do adolescents work in relation to
time spent at school? First, the issue of whether adolescents work on school days or
weekends (or both) is important. Since homework assignments and exams are given
during the school week, an individual who works on most school days will have a more
difﬁcult time completing such assignments than an individual working primarily
weekend shifts. It could be argued that an individual working during the school week
could play “catch up” on the weekend or study in advance for the following week. Yet
homework assignments and study guides for tests are often distributed during the week
and study groups usually do not form until a day or two before an exam. Thus,
adolescents working primarily during the school week are at a disadvantage in terms of
taking advantage of academic resources.

Furthermore, the ti_mi_ng of work shifts is likely to differ for high school vs.
college students. Consider that young adolescents have legal restrictions on the number
of hours they can work per week. For instance, the Fair Labor Standards Act allows 14
and 15 year olds to work 3 hours a day when school is in session, and only allows such
youth to work between the hours of 7:00am and 7:00pm (Walker 1990). Although older
high school students do not have such legal restrictions, they are still restricted in the
hours available to work since they attend school approximately 7 hours per day and often

have an evening curfew imposed by their parents. Thus, high school students often have

43

restrictions on the times they can work (early evenings or weekends).

In contrast, college students are only in class for a few hours per day and have no
legal work restrictions. Furthermore, since the majority (72%) of college students do not
live at home (National Center for Education Statistics 1993), they are not as closely
monitored by their parents. As a result, college students are not restricted from working
at any time of the day or night. For those adolescents who choose to work late evening
hours or other inconvenient times, work/school conﬂict is likely to ensue. For instance, a
high school student who works for 3 hours each evening (e. g. 7-10pm) is likely to
experience less interference with school responsibilities than a college student who works
those same 3 hours per day but from 2am to 5 am. Although the adolescents in the above
example are working the same number of hours, the late shift is more likely to interfere
with getting adequate sleep to attend and be alert in class the next day as well as
completing one’s homework assignments.

In addition to school/work conﬂict in terms of the time dimension, a less obvious
reason that work and school can conﬂict is in terms of energy levels. A job that contains
a high degree of work stress (e. g. excessive heat/cold, time demands, etc) is likely to
require a great deal of energy; energy that is not left over for schoolwork. For instance,
suppose Jack and Jill both work for 25 hours per week. Jack’s job involves heavy lifting
and extremely hot conditions whereas Jill’s job involves non-strenuous tasks such as
paper work and answering phones. Even though Jack and Jill technically have the same
amount of 9111; outside of work to engage in school responsibilities, the amount of my
left over from work is quite different. Jill often comes home from work and spends a few

hours on homework, whereas Jack is simply too

exhausted to expend mental energy on homework after his physically demanding job.

Furthermore, work/school conﬂict in terms of energy levels is likely to increase
from high school to college. Recall that school requirements involve more 11% time for
high school students and more homework time for college students. Even if these
activities take the same amount of overall time per week, they are not equivalent in terms
of the energy required. Sitting in class and taking notes is a more passive activity
whereas writing papers, studying for exams, and preparing student presentations are
active activities requiring more thought and energy. Whereas a high school student may
still be able to sit in class even when exhausted from work, a college student may choose
to delay or ignore homework assignments (for which there are no immediate
consequences). Thus, work and school demands are more likely to conﬂict for college
students as a result of the increased energy levels required to perform school related
tasks.

There are also reasons to suppose, however, that work/school conﬂict might

decrease from high school to college. First, juggling work and school demands is a skill

 

that can be learned and developed over time. Since most adolescents begin working in
high school, by the time they reach college, most have had ample experience in handling
conﬂicting demands of work and school. Over time, adolescents are likely to develop
time management skills as well as become more competent (and thus more efﬁcient) in
performing both school and work activities. Thus, although work/school conﬂict cannot
be completely overcome, older adolescents are less likely to experience intense
work/school conﬂict as a result of their increased skill capacity in dealing with such

conﬂict.

45

Another reason why work/school conﬂict is likely to decrease from high school to
college is because college students’ perception of the relationship between work and
school is likely to change. As we have seen, high school students generally use their
earnings for short term needs such as clothing, transportation and entertainment
(Bachman et al. 1986; 1987; Greenberger 1988). In contrast, although almost half of
college students are still dependent on their parents (National Center for Education
Statistics 1993), many still contribute to their college or living expenses (Greenberger
1988). Since working is often essential in order to attend the university, college students
are likely to view their work role as contributing to (rather than conﬂicting with) school
outcomes.

It is important to note, however, that even if an individual perceives work and
school roles as complementary, s/he may still experience actual work/school conﬂict.
For instance, suppose that Jay must work 35 hours/week in order to attend university.
Although he views work as essential to his schooling, he is still likely to have difﬁculty
keeping up with school demands as a result of ﬁnite energy and time resources (that are
heavily devoted to work). Thus, in this instance, it is perceived (not actual) work/school
conﬂict that is likely to decrease for college students.

In sum, in terms of the change in level of work/school conﬂict from high school
to college, we have seen that there are reasons to expect that work/school conﬂict
increases from high school to college (i.e. because of increased energy required for
college work). There are also reasons to suspect, however, that work/school conﬂict
decreases for college students (i.e. because of increased competency in juggling roles or

because work and school roles are viewed as

46

complementary). Thus, although we do not hypothesize the direction of effect, we will

test for changes in work/school conﬂict as adolescents move from high school to college.

Work/School Conﬂict and Psychological Well Being

Now that we have examined the ways in which work and school role demands are
contradictory, let us explore a more relevant question for purposes of this study: what are
the psychological implications of work/school conﬂict? Although there is ample research
on the existence of work/school conﬂict for adolescents, there is a paucity of studies
examining the psychological implications of such conﬂict. The studies that do exist
indicate that adolescents who perceive signiﬁcant levels of work/school conﬂict are more
likely to be depressed (Shanahan et al 1991), whereas those who perceive low levels of
work/school conﬂict experience increased psychological well being (Finch et a1 1991).

The present study will examine the impact of contradictions in work and school
roles for an older adolescent sample. Consistent with previous research on youth work,

we hypothesize the following (see Figure 1):

Hypothesis 2: Perceived work/school conﬂict is negatively related to psychological well

being

Alienated Labor and Qccupational Sell-Direction

Another important dimension of work quality in the adult literature is
occupational self-direction. Before discussing this contemporary term, let us brieﬂy

explore its classical roots in the study of alienated

47

labor. Alienated Labor is deﬁned as a situation in which the worker has little control
over his/her own labor and is doing a job in which his/her skills or capacities are
underutilized (see Greenberg and Grunberg 1995). The origins of alienated labor lie in
the advent of industrialization and increased division of labor in the late 19th and early
20th centuries. Period scholars argued that although industrial capitalism increased
productivity, it also produced extremely negative living and working conditions.
Beginning with Adam Smith’s conceptualization of work as “the real price of everything”
(Smith 1937), scholars have theorized on the impact of capitalist labor on the individual
worker.

Modern writings on alienated labor owe a large debt to the early writings of Karl
Marx (Marx 1964 [1844]). Marx argued that capitalist production increases the value
placed on material goods while it decreases the value placed upon individuals. Marx
believed that industrialized capitalism denied workers’ the right to control their work
activity and the products of their labor. Speciﬁcally, he argued that workers became
separated from four aspects of their work: the products of their labor, the process of
work, the ability to be creative, and the need to be part of a collective group. As a result,
workers become alienated not only from their work, but from the society as a whole. In
other words, “Workers ﬁnd no place, no thing, and no experience that they can truly call
their own (From 1968).

Since WWII, many industrialized nations have moved from an economy based on
manufacturing, to one based on service and technology (Wilson 1997). For instance,
during the 19803, for every thousand people of working-age, the US. created 27 clerical,

sales and service jobs and lost 16 production,

48

transportation and laborer jobs (Wilson 1997; p.27). Contemporary theorists argue that
although the speciﬁc types of jobs may have changed in the post-industrialized economy,
many of today’s occupations contain their own ﬂavor of alienated labor.

Some theorists focus on technological_changes in postindustrial society and its
impact on alienated labor. Robert Blauner, in his groundbreaking book, Alienation and
Freedom (1964), performed case studies on four manufacturing industries (printing,
textiles, automobiles and chemicals) at various stages of technological development.
Blauner’s ﬁndings indicate that although alienated labor increases during the initial
stages of technological development (c. g. assembly-line technologies), alienation
decreases again with more advanced continuous-process technologies.

More recent studies, however, ﬁnd that advanced technologies do not always
reduce alienated labor. Instead, it depends upon how the technology is used within the
job itself. For instance, although clerical data entry workers use advanced technology
(e. g. computers), they still experience a signiﬁcant degree of alienation because of the
repetitiveness of their tasks (e. g. Noble 1984). Thus, the degree of alienated labor
depends not only on the nature of the technology, but also upon the use of such
technologies within various occupations.

Other theorists have focused on alienated labor in relation to the many service
jobs in today’s economy. Scholars argue that in postindustrial economics, the capacity to
deal with people (rather than work with “things”) has become the central task of today’s
service jobs (Bell 1973). Arlie Hochschild (1983) refers to this as “emotional labor”
which requires the worker to “induce or suppress feeling in order to sustain the outward

countenance that produces the proper state of mind in

49

others.” (p. 7). Estimates indicate that between one-half and one third of US. workers
have jobs that require signiﬁcant levels of emotional labor. Hochschild argues that the
constant display of emotion based on capitalist needs causes workers to separate
themselves from their own feelings. Eventually this limits the worker’s capacity to feel
and thus workers become alienated from the services that they provide. Thus, just as
alienated labor occurs in a manufacturing economy, so it does also in today’s service and
technology economy.

The question of interest in the present study is: what is the impact of alienated
labor on psychological functioning? Contemporary pioneers in the ﬁeld, Kohn and
colleagues, have done groundbreaking research examining the effect of alienated labor on
worker’s sense of efﬁcacy and psychological well being. These researchers have
speciﬁed and operationalized the concept of alienated labor, which they refer to as
occupational selfodirection. Kohn and colleagues argue that there are three interrelated
conditions of work that form occupational self-direction (Kohn and Schooler 1983). The
ﬁrst dimension is the substantive complexity of one’s job. Complex work involves tasks
that require independent judgment and initiative. Generally, substantively complex jobs
involve working with “data” (e. g. engineers) or “people” (e.g. consultants) rather than
“things” (e.g. assembly line worker). Certainly this is a general rule containing notable
exceptions such as an artist whose profession involves highly complex work involving
“things” or a secretary who works with “data” despite the low substantive complexity of
the job. Yet regardless of whether one works with data, people, or things, substantively
complex work always involves high levels of thought, creativity, and independent

judgment.

50

A second factor that may limit occupational self-direction is the closeness of
supervision at work. A supervisor who tightly regulates employees’ activities precludes
workers from exercising their own judgment in performing work tasks. For instance,
consider two employees at a survey research center. First, a phone survey interviewer is
closely supervised since he/she receives explicit instructions before each work task and is
often monitored will; he/she makes phone calls. In contrast, a data analyst is not as
closely supervised since he/she spends a signiﬁcant amount of time deciding how to
analyze and present data before his/her work is evaluated by a supervisor.

The third factor determining occupational self-direction is the routinization of
work tasks. Jobs involving self-direction involve tasks that can be performed in a variety
of ways whereas jobs with low self-direction provide only one method of solving work
related problems. For instance, the survey interviewer is given a script to read for each
phone call, with little room for variation. In contrast, each time the analyst receives data,
he/she evaluates a variety of possible methods for analysis.

It is important to note that these three factors are interrelated and work in tandem.
Each factor by itself is a necessary but not sufﬁcient condition for occupational self-
direction. For instance, suppose Barb is a data analyst for a market research company.
She has a good deal of autonomy since her supervisor does not closely monitor her and
she uses a variety of statistical methodologies to analyze data. However, the work itself
is not substantively complex. She usually analyzes customer satisfaction surveys, which
lack a theoretical component. As a result, the work does not afford Barb an opportunity
for creativity and foresight. Thus, although Barb has two out of the three requirements

for occupational self-direction (variety and ﬂexible

51

supervision), the lack of substantive complexity precludes her from having the full
experience of occupational self-direction. In sum, although the dimensions of
occupational self-direction often occur together, this is not necessarily the case. Rather,
the experience of occupational self-direction involves all of the above mentioned factors
(Kohn and Schooler 1983).

Kohn and colleagues argue that occupational self-direction is intimately tied to
self-efﬁcacy and psychological well being. Specifically, Kohn and Schooler (1983)
argue that experiencing opportunities for self—direction at work, fosters a self-directed
personality. According to Kohn’s generalization theory, individuals learn or generalize
the “lessons of the job to outside-the-job realities (Kohn 1981: p.290). Thus, workers
who experience high levels of occupational self-direction are likely to have a more self-
directed orientation in other aspects of life.

There is considerable evidence to support the position that occupational self-
direction has a negative impact on self-efﬁcacy, or the individual’s assessment of his/her
competence in the world. Research indicates that all elements of occupational self
direction (low substantive complexity, close supervision, and highly routinized work
tasks) are related to a self-directed orientation, intellectual ﬂexibility, and self-efficacy
(Gecas and Seff 1989; Kohn, Naoi, Schoenbach, Schooler and Slomczynski 1990; Link,
Lennon and Dohrenwend 1993; Mortimer et al 1986; Kohn and Schooler 1983; Spenner
and Otto, 1985). Such ﬁndings have been replicated even in cultures that do not stress
individual autonomy and self-reliance. For instance, research indicates that the

relationship between occupational self-direction and selfoefﬁcacy holds for Polish and

52

Japanese men (Kohn et a], 1990) and Japanese women (N aoi and Schooler 1990;
Schoooler and N aoi 1988).

In turn, low self-efﬁcacy and a reduced assesssment of effectiveness is associated
with psychological distress. There is considerable evidence for a negative relationship
between self-efﬁcacy and psychological distress for adults (Link et al. 1993; Mirowsky
and Ross 1989; Pearlin et al. 1981; Stets 1995) and adolescents (McFarlane et al. 1995).
Furthermore, studies indicate that the negative relationship between occupational self-
direction and psychological well being is mediated by self-efﬁcacy (Kohn and Schooler
1983; Kohn et al. 1990). Thus, self-efﬁcacy is an important vehicle through which
occupational self-direction impacts upon psychological well being.

Although the above argument is plausible, an alternative explanation is that social
selection processes play a role in the relationship between occupational self-direction and
psychological well being. Speciﬁcally, it could be argued that work conditions do not
lead to psychological distress (through lowered self-efﬁcacy), but rather that PEI levels
of psychological distress lead to a disability in functioning, which in turn, may lead to
difﬁculty in attaining a job affording high levels of self-direction (Link et a1 1993). Thus,
simply observing a relationship between occupational self-direction and self-efﬁcacy
does not necessarily indicate causal direction. Some studies have attempted to test for the
alternative arguments of social causation and social selection with longitudinal data.
Most studies have found support for both social causation and social selection arguments,
such that prior levels of self-efﬁcacy lead to subsequent occupational achievement and
self-direction as well as occupational self-direction leading to self-efﬁcacy (Kohn and

Schooler 1983; Mortimer et a1 1986). Other

53

longitudinal studies, however, have found greater support for the social causation
hypothesis (Link et al 1993; Kohn et al 1990). Future research is needed to elaborate on
the reciprocal nature of the relationship between occupational self-direction and self-

efﬁcacy and distress.

Adolescence and Occupational Self Direction

 

Although there is considerable support for the adult population, the research on
the outcomes of work complexity and autonomy for adolescents is sparse. Yet employed
adolescents are almost universally exposed to conditions of low occupational self-
direction. As discussed previously, most adolescents work in retail and service jobs that
afford few opportunities to exercise authority or attain advancement (Greenberger and
Steinberg 1981) and involve repetitive tasks that require few skills or training
(Greenberger, Steinberg and Ruggiero 1982; Osterman 1989).

As an illustration, fast food restaurants employ primarily adolescents and young
adults (Leidner 1991; Reiter 1996). Research indicates that fast food workers experience
very low occupational self-direction. For instance, Reiter (1996) in her study on “Burger
King”, found that there were very speciﬁc rules for all work tasks (e. g. it should take
exactly 23 seconds to prepare a Whopper), indicating high routinization of work tasks.
Fast food jobs also lacked in substantive complexity as one worker commented “A moron
could learn this job” (p. 150). Furthermore, workers were closely supervised as a
manager was always on the ﬂoor shouting words of so called “encouragement” to make
sure workers were moving quickly. Since many adolescents work at similar service (or

retail) jobs, it is reasonable to assume that the elements of occupational self-direction are

54

often quite low for adolescent workers.

Most studies speciﬁc to adolescents, however, have broken down occupational
self-direction into two categories: autonomy and complexity. Work autonomy
corresponds roughly to Kohn’s notion of “closeness of supervision” (Schwalbe 1985).
This refers to the extent to which workers can independently decide how to perform work
tasks. Work complexity is a combination of Kohn’s concept of “routinization of work
tasks” (can work tasks be performed in a variety of ways?) and “substantive complexity”
(do work tasks require skill, judgment and creativity?) Studies on adolescent workers,
however, have generally examined either work autonomy or work complexity rather than
both in a single study.

The present study asks the question: Is occupational self-direction (speciﬁcally
work autonomy and work complexity) negatively related to adolescents’ psychological
well being, as has been demonstrated for adult populations? Some preliminary evidence
supports this contention. Schulenberg and Bachman (1993) found that adolescents’
psychological well being suffered when they worked at jobs that involved low
complexity over long periods of time. Furthermore, studies demonstrate a positive
impact of work autonomy on self-efﬁcacy. Research indicates that being free from close
supervision and being included in discussions about work tasks increases feelings of
competence and mastery in adolescents (Call 1996; Eccles et al. 1991; Montemayor
1983; Steinberg 1990).

Other research, however, suggests that self-direction at work may actually have a
negative impact on adolescents’ psychological well being. One longitudinal study found

that work autonomy actually increased depressive

55

affect among tenth grade boys (Shanahan et. al. 1991). Shanahan and colleagues
speculated that this ﬁnding was due to the young age of the respondents (10m graders).
Most young adolescents have had little experience in making independent decisions and
being self-directed when performing various tasks. For instance, a young adolescent may
do homework or chores in the home, but his/her teachers or parents are likely to heavily
guide these activities. The young adolescent is likely to confront the expectation for
independent decision making for the ﬁrst time at a part-time job. Such expectations may
be threatening at ﬁrst for those with little experience in taking on such responsibilities.
As one gains experience, however, such autonomy may become less threatening and in
fact begin to have salutary psychological implications, as has been shown for adults. The
question becomes. . .at what point in the life cycle does the experience of work autonomy
cease to be threatening, and begin to have a positive impact on psychological well being?

Currently, the literature on work autonomy and psychological well being contains
a gap between studying early adolescence and adulthood. We know that occupational
self-direction has salutary psychological outcomes for post-college young adults
(Mortimer 1986) and adults (Kohn et al. 1990; Kohn and Schooler 1983). The period of
late adolescence, however, has been neglected in the debate on youth work. To my
knowledge, there are currently no empirical studies on the psychological outcomes of
work autonomy and work complexity for those in late adolescence.4

For scholars interested in adolescent development and the transition to adulthood,
it is important to explore at what point in the life cycle work autonomy and complexity

begin to have a positive impact on psychological well being. The present study will

 

‘ As discussed in a previous section, the present study considers college students to be in the stage of “late
adolescence”

56

attempt to ﬁll this gap by looking at the relationship between work autonomy and
psychological well being during the life cycle period of late adolescence. We would
expect that although work autonomy and complexity during early adolescence is
threatening, that by the time of late adolescence, most individuals will have had sufﬁcient
work experience to begin beneﬁting from more independent work conditions. Thus, I
expect that the relationship between work autonomy/complexity and psychological well

being for older adolescents is similar to that of adults such that:

Hypothesis 3a: Work autonomy and work complexity are both positively related to

psychological well being (see Figure 1)

Hypothesis 3b: The positive relationship between work autonomy and well being is

mediated by self-efﬁcacy (see Figure 2).

Hypothesis 3c: The positive relationship between work complexity and well being is
mediated by self-efﬁcacy (see Figure 2)

Figure 2 Mediating role of self-efficacy

 

  

Occupational Self-
Direction

- Autonomy
0 Complexity

   

4 Self-efﬁcacy - Depression

  
 

 
     
 

57

Demogaphic Diﬂerences in Work Conditions

We have seen that work quality (i.e. work conditions, work/school conﬂict and
work autonomy and complexity) are likely to have important psychological implications
for adolescents. Yet, the psychological impact of work quality may vary by factors such
as age, gender, race, and future occupational goals. Let us explore each in turn.

Age is likely to be an important factor in the psychological impact of youth work.
The relationship between work autonomy and complexity and psychological well being
may vary, based on two factors related to age. First, there may be age differences in the
effects of work within the period of late adolescence. The closer to adulthood that the
adolescent approaches, the more likely that autonomy and/or complexity will become
important to his/her psychological well being (as it is for adults). For instance, a 22 year
old, ready to embark on a professional career may beneﬁt more from work autonomy and
complexity than an 18 year old who is just beginning his/her college career. Thus, I will
test for the interaction effect of age and work autonomy and complexity on psychological
well being.

Second, although we have documented that the majority of adolescents begin
work during early to mid adolescence, there is a signiﬁcant minority of older adolescents
with little or no previous work experience. For them, there are reasons to think that work
autonomy/complexity could have positive or negative effects. As discussed previously,
work autonomy is threatening to early adolescents and has been shown to lead to
psychological distress. One possibility is that older adolescents respond in a similar
fashion since similar to younger workers, they have no previous work experience. The

experience of being thrust into work responsibilities

58

that require high levels of autonomy may be threatening to older adolescents with no
work history.

On the other hand, older adolescents are generally more mature psychologically
and socially than their younger peers, whether or not they have previous work
experience. Research indicates that the ability to understand oneself and our
relationships with others improves throughout the adolescent years (Barenboim 1981;
Selman 1980). Furthermore, by the time of late adolescence, even those without previous
work experience are likely to have participated in other activities involving autonomy and
complexity such as extra-curricular activities, volunteer work, and self directed school
assignments. As a result, those in the stage of late adolescence may beneﬁt from work
autonomy and complexity, despite their lack of work history. Thus, although I do not
hypothesize the direction of effects, the present study will test for differences in the
effects of work autonomy and complexity on psychological well being for those with and
without previous work experience.

Second, some argue that the psychological impact of work autonomy and
complexity for adolescents varies based on the relationship of the part time job to future
occupational goals. Specifically, when the current job is similar to the adolescent’s
image of his/her future worker self, conditions such as low autonomy or low complexity
are less likely to have negative psychological outcomes. For instance, the psychological
impact of waiting on tables is less likely to have a negative impact on a hospitality
business major than an engineering major. Since the hospitality business major sees
him/herself as having a related occupation in the future, the current job (despite its low

autonomy and complexity) may actually contribute

59

positively to the adolescent’s sense of future self. Unfortunately, the present study is
limited in its ability to explore this important issue. An important task for future research
is to explore in detail how respondents view the relationship between their current job
and future occupational aspirations and how this contributes to psychological well being.

Third, race is an important variable to consider when examining the impact of
work conditions on self-efﬁcacy and psychological well being. As discussed previously,
blacks are less likely to attain jobs through contacts. Research has not examined,
however, whether jobs attained through contacts differ in terms of job quality. To my
knowledge, studies have only addressed this issue in terms of monetary differences. For
instance, some studies ﬁnd that found that white youth have contacts to better paying jobs
than black youth (Korenman and Turner 1996; Michigan State Unviversity 1993).
Conspicuously missing from the current research is an examination of racial differences
in work quality and its impact on psychological well being.

The present study will examine these issues related to race. First, we will explore
whether there are differences between jobs held by black and white adolescents in terms
of work conditions, work/school conﬂict, work autonomy, and work complexity.
Second, we shall explore whether the psychological impact of such measures of work
quality differ based on race. Since there is little previous research to guide such a quest,
the present study will not hypothesize direction of effect.

Lastly, gender is likely to be an important variable in the relationship between
work quality and psychological outcomes. Although little attention has been paid to
gender differences in the youth work literature (Yamoor and Mortimer 1990), we will

point out some notable exceptions. First, a handful of

studies indicate that work/school conﬂict and noxious work conditions increase
depression for adolescent boys, but have little impact on adolescent girls (Mortimer,
Shanahan and Call 1996; Shanahan et al. 1991). One explanation is that since there is a
higher expectation that boys will work in adulthood, that difﬁculties at work (i.e.
stressors, conﬂicts) are more likely to threaten boys’ identity and thus have a negative
psychological effect than for adolescent girls (Shanahan, Finch, Mortimer and Ryu 1991).

Others researchers, however, have questioned this explanation. Recent studies
indicate that girls’ occupational aspirations are generally higher than boys (Farmer 1983;
Shapiro and Crowley 1982) and that less than one fourth of girls think they will be full
time homemakers (Ireson and Gill 1988). Furthermore, the majority of college students
(whether male or female) are presumably expecting to work in adulthood since they are
attaining a marketable degree. Thus, the present study will examine whether there
continue to be gender differences in the psychological impact of noxious work conditions
and work/school conﬂict for those in late adolescence.

Work complexity is another measure of work quality that may have different
psychological effects based on gender. Similar to adults, jobs in adolescence are highly
sex-segregated, particularly in early adolescence (Greenberger and Steinberg 1986;
Lewin-Epstein 1981). Speciﬁcally, girls are more likely to perform informal work in
private homes (e.g. babysitting) whereas boys are more likely to work in the formal
sector (e. g. paper carrier, manual labor, restaurant work) (Mortimer et al. 1990).

Some studies indicate that jobs held by adolescent girls tend to be less complex
than jobs held by their male counterparts. For instance, one study that coded the

occupational complexity of youth work based on the

61

Dictionary of Occupational Titles (US. Department of Labor 1986), found that girls’ jobs
were less complex than boys’ jobs in terms of dealing with data and things. With regards
to complexity with people, however, there were no signiﬁcant differences with respect to
gender (Mortimer et al. 1990).

The researchers of the above study were puzzled, however, by the DOT ratings of
boys and girls jobs as equally complex in terms of dealing with people. Speciﬁcally,
Mortimer et al. (1990) argue that the job babysitter (so common for young adolescent
girls) should have a high “complexity with people” rating since it involves intensive
interaction with children as well as full responsibility for childrens’ well being. In
contrast, boys’ jobs (such as manual labor or paper carrier), do not involve interaction
with others as a main job component. It may be that the DOT ratings reﬂects, in part, a
devaluation of “caring work” which has traditionally been women’s work. Since
babysitting primarily involves caring for others, it may be that despite its complexity, it is
not rated as such because of its devalued status.

But whatever the gender differences in level of complexity, another important
issue is whether the psychological impact of complexity varies by gender. Unfortunately,
no studies to date test for such effects. In terms of the other measure of occupational self-
direction (work autonomy), however, there have been a few studies exploring such
Social support refers to actual or perceived resources available from others that increase
the individual’s well being and facilitates the management of stress (McIntosh 1991).
Signiﬁcant others can provide instrumental, informational, and/or emotional assistance to
the individual (House and Kahn 1985). A considerable body of research indicates that

social support has a direct positive impact on

62

psychological well being. Furthermore, research indicates that social support moderates
the relationship between stress and health, such that the deleterious effects of stress on
health outcomes are lessened for those with high levels of social support (see reviews:
Cohen and Wills 1985; House et al. 1988; Kessler and McLeod 1985).

Social support has been demonstrated to buffer the negative relationship between
two types of stressors (life events and chronic strains) and psychological well-being (e. g.
Cohen and Wills 1985). Life events refer to acute changes that require major behavioral
readjustments within a short period of time, whereas chronic strains refer to persistent
demands that require readjustments over prolonged periods of time (see Thoits 1995).

Thus far, the present study has not conceptualized of work quality as “stress.”
However, we ﬁnd this to be a useful conceptualization when considering the impact of
social support on the relationship between work quality and psychological well being.
Aspects of work quality most closely resemble chronic strains since individuals often
have a particular job for many months or years. As a result, workers must deal with
aspects of work including autonomy, complexity, work/school conﬂict, and noxious work
conditions on a consistent (often daily) basis. When these aspects of work are negative,
they can be conceptualized as chronic strains since they require constant readjustment on
the part of the worker. For instance, if Joe experiences time contradictions between work
and school demands, or is exposed to hot and dangerous conditions at work, he must deal
with these aspects of work quality on a consistent basis, which may have psychological
implications. Thus, for the purpose of relating our exploration of social support to the
literature on stress-distress relationships, we will conceptualize of work quality as chronic

strains for the time being.

63

When examining the mitigating role of social support, it is important to recall that
support does not always buffer the relationship between stress and health. What are the
dimensions of social support that have implications for the relationship between stress

and psychological well being?

Source at Support

First, the source of social support has an impact on its’ ability to play a buffering

 

role. Scholars argue that when the source of social support matches the context of the
stressor, that support will have a greater buffering role in the relationship between stress
and health (House 1980). For instance, although social support from one’s family may
mitigate the negative consequences of family stressors, it is likely to have little impact on
the effects of poor work quality. Similarly, receiving support from one’s co-workers is
unlikely to mitigate the negative impact of an individual’s family problems or marital
discord.

Speciﬁc to the work context, scholars have speculated that support from Q

workers (rather than from family or friends) is especially important in lessening the

 

negative impact of poor work quality on psychological well being (Beehr 1985).
Research conﬁrms that support from co-workers has a positive direct effect on
psychological well being (Blau 1981; Fenlason and Beehr 1994; Ganster et al. 1986) as
well as buffers the negative impact of poor work quality on psychological well being
(Fenlason and Beehr 1994; Henderson and Argyle 1985; House 1980; Moore 1985;

Pugliesi 1995).

Although there is considerable evidence for a direct and buffering role of co-
worker social support in the adult work literature, to my knowledge, the role of co-worker
support for adolescent workers has not yet been explored. The following illustrative
example serves to place the concept of contextual support within an adolescent
framework. Suppose 16 year old Brad is a cook at the “Mongolian BBQ.” His job
involves low complexity of work tasks as well as working under extremely hot
temperatures. Although when Brad comes home he receives support from his family, this
cannot directly change the environment in which he works. In contrast, when Brad is at
work, he spends a signiﬁcant amount of time joking and griping about the job with his
co-workers. Although this does not change the actual work that he performs, the
character of the working environment is altered. Although Brad still must perform
routine and pressured tasks, he is among those who directly empathize with his
experience and the camaraderie with his co-workers provides a positive atmosphere that
may buffer the impact of the actual work quality.

A related issue for youth workers is whether one’s co-workers are primarily peers
(adolescents) or non-peers (adult workers). Research indicates that many adolescents
work primarily with those of similar age (Greenberger and Steinberg 1986; Greenberger
1988; Safyer, Leahy and Colan 1995). I would speculate that when an adolescent’s co-
workers are primarily peers, that it is easier to develop an atmosphere of camaraderie as
described above. For instance, if Brad’s co-workers are primarily other students from
college, he is likely to share with them common interests, problems and lifestyles. In
contrast, middle-aged co-workers may be more concerned with their mortgage payments

or their children than with Brad’s concerns of college

65

exams, dorm life, and dating. Thus, a supportive peer atmosphere is more likely to
develop when one’s co-workers are primarily other adolescents. The present study will
examine whether youth workers who work primarily with their peers (other adolescents)
perceive their co-workers to be more supportive than those who work primarily with non-

peer groups (adult workers).

Type at Support

Another dimension of social support that has important implications for
psychological well being is the type of social support. Researchers examining the
psychological implications of work quality generally divide social support into two
global types: emotional and instrumental support (Blau 1981; Ganster et al. 1986;
Kaufmann and Beehr 1986; Thoits 1982). Emotional support refers to caring or listening
empathetically to another person whereas instrumental support refers to performing
concrete tasks for another such as giving advice or physical assistance (Cohen and Wills
1985).

Scholars suggest that emotional and instrumental support have a direct positive
impact on psychological well being, and also replenish the resources depleted by
stressors (Cohen and Wills 1985). Empirical research on work quality (for adults)
conﬁrms that emotional/instrumental support has a direct effect on psychological well
being as well as buffers the negative impact of poor work quality on psychological well
being (Blau 1981; Fenlason and Beehr 1994; Ganster et al. 1986). Since research on
youth work has not explored the role of social support, the present study will follow the

lead of the above studies on adult work quality, and

66

use an emotional measure of social support. An important task for future research is to
compare the impact of various types of social support for adolescent workers’

psychological well being.

Perceived Support

Another important distinction in the social support literature is between perceived
and received support. Received support refers to the actual supportive the individual
receives during times of need. Altemately, perceived support refers to the perception that
one is cared for and that one’s social network will hypothetically provide support if
needed. Research demonstrates that perceived support and received support are m);
highly correlated (Dunkel-Schetter and Bennett 1990).

Evidence suggests that perceived support is important in moderating the
relationship between stress and health. Research demonstrates that the level of perceived
support is a better predictor of adjustment to stressful circumstances than the amount of
support m1 (Henderson, Byme, and Duncan-J ones 1981) or the actual amount of
r_epe_i\_r_e_c_l support (Blazer 1982; Kessler 1992). Thus, it is the perception that others
would provide support during times of need that lessens the negative impact of strains on
psychological well being. Research on work quality has generally used measures of
perceived support, and the present study will follow suite when studying youth workers.
Cost at Social Relationsh§p' s

Thus far, we have spoken of social support as having positive effects (direct or
buffering) on psychological well being. Often overlooked in the literature on social

support, are the costs associated with social

 

67

relationships. In addition to receiving social support, being a member in a social network
often entails support giving as well. While providing support to others in moderation is
ﬁne, excessive support giving may have a negative impact on psychological well being
(Gove, Style and Hughes 1990; Rook 1992; Thoits 1992).

With regards to adolescent workers, those who perceive high levels of co-worker
social support, may also be obligated to provide social support to such co-workers. Such
demands, if excessive, may have a negative impact on adolescents’ psychological well
being. Furthermore, adolescents typically have social relationships outside of work such
as friends and family. Although such relationships provide social support to adolescents
(Barrera 1981; Dombusch 1989; Greenberg et al. 1983), youth employment may interfere
with fulﬁlling demands for support giving to friends and family. Some preliminary
research indicates that adolescents who work have less close relationships with their
peers (Greenberger et al. 1980; Steinberg et al. 1982) and spend less time with family
members (Greenberger et al. 1980).

Future research is needed to clarify whether it is the lack of support giving,
speciﬁcally, that causes this decline in family/friend relationships or whether there are
other factors involved. Furthermore, much more needs to be known regarding how work
interferes with such relationships (e.g. time, energy, etc.). Although this is an important
task, the present study is limited in its ability to test for the costs of social support.

Conclusion

In sum, in order to fully understand how working affects adolescents, we must
explore under what circumstances youth work quality has the greatest impact on

psychological well being. The present study will draw

68

upon previous literatures in exploring the dimensions of social support that have been
shown to buffer the impact of stress on psychological well being. Based on our previous
discussion, we will examine the direct and buffering role of perceived, emotional, co-
worker support on the relationship between work quality and psychological well being
for adolescents. The present study will add to the youth work literature by testing the

following hypotheses (see Figure 1):

Hypothesis 4(a) Perceived Co-worker social support has a positive direct effect on

psychological well being

Hypothesis 4( b ): Perceived Co-worker social support buffers the impact of poor work

quality on psychological well being.

LONG TERM PSYCHOLOGICAL IMPLICATIONS

Thus far we have discussed the relatively short term psychological implications of
work for adolescents. As we have seen, poor work quality tends to have negative
contemporaneous effects on psychological well being. Yet, some scholars argue that
youth work has much longer range implications for psychological well being. There are
two main lines of thought regarding the lagged effect of adolescent work experience on
psychological outcomes: The “Developmental Readiness Hypothesis” and the “Stress

’95

Resistance Hypothesis. Both camps agree with the contention discussed throughout

this essay: that when youth work quality is poor, this leads to negative contemporaneous

 

5 Similar to our discussion of social support, we will again conceptualize of work quality as “chronic
strains” in order to consider the validity of the two competing hypotheses

69

effects on psychological well being. Where they differ is in their predictions of long term
psychological implications from early work experiences.

The Developmental Readiness Hypothesis (Greenberger and Steinberg 1986)
posits that youth work has long term detrimental psychological outcomes. Greenberger
and Steinberg (1986) argue that poor work quality is often experienced before
adolescents have developed adequate coping mechanisms to deal with this situation. Erik
Erikson, a leading ﬁgure in developmental psychology, argued that adolescents require
time for role experimentation, which includes daydreaming and fantasizing about future
possible roles. Erickson contended that adolescents who assume adult roles before they
are psychologically ready to handle them are likely to stunt this “role experimentation”
period and thus limit their identity development (Erikson 1959; 1968).

Proponents of the “Developmental Readiness” hypothesis argue that early work
experience constitutes an activity that is likely to interfere with the adolescents’ time for
role experimentation. Clearly, an adolescent who spends most evenings and weekends
working cannot devote adequate time to complex cognitive processes that require time
for daydreaming and fantasy. Since early work interferes with ﬁtndamental processes of
the adolescents’ successful development into adulthood, Greenberger and colleagues
argue that experiencing work before adolescents are developmentally “ready” is likely to
have negative long term developmental and psychological consequences. Although this
argument is plausible, it largely remains within the theoretical realm, with little empirical
research evidence (Greenberger and Steinberg 1986).

Alternatively, the Stress Resistance hypothesis (Shanahan and Mortimer 1996

posits that while early work experiences are initially

70

detrimental to psychological well being, the long term effect is quite the opposite.
Proponents of this perspective draw upon Hans Seyle’s biological model of the stress
process. Seyle (1976) argues that although biological stressors can in rare cases cause
distress, they more often produce “eustress”, or an increase in resistance to such stressors.
Social scientists applying this model to social phenomena argue that social stressors can
similarly result in an increased stress capacity and a strengthening of coping resources
(see Shanahan and Mortimer 1996).

Thus, in stark contrast to Erikson’s contention that early stressful experiences take
adolescents’ away from important developmental tasks, proponents of the “Stress
Resistance” perspective argue that experiencing early stressful work quality actually
constitutes a valuable developmental experience. Shanahan and Mortimer (1996) argue
that early stressful work experiences cause adolescents to learn effective coping
mechanisms and to begin to deal effectively with poor work quality. Over time, coping
tools developed through early work experiences will buffer the negative impact of poor
work quality on psychological well being in later jobs. Such successful coping, in turn,
leads the adolescent to feel a sense of competency and self-efﬁcacy in the long run.

Although the “Stress Resistance” hypothesis has been demonstrated for biological
stressors, research in the psychosocial realm has been sparse. Social stress researchers
tend to focus on the negative rather than positive impact of stressors. Although
researchers have examined factors that buffer the negative relationship between stress
and health (such as coping and social support resources), they tend to ignore the process

by which stressors actually mobilize such resources (Shanahan and Mortimer 1996).

71

Clearly, the stress resistance hypothesis has yet to be explored in future research.

Thus, the Developmental Readiness and the Stress Resistance hypotheses lead to
very different conclusions regarding the long term psychological implications of early
work experience. As we have seen, research on work quality and psychological well
being in adolescence is still in its infancy stages. The majority of studies only address the
early adolescent period and are cross-sectional in nature. As a result, the lagged effects
of youth work on psychological outcomes have rarely been explored. Yet, this remains a
crucial issue to explore when considering the full implications of youth work. Although
the present study can only address this issue in a preliminary manner, I will test the two

competing hypotheses:

H YPOTHESIS 5(a): Previous poor work quality (i.e. noxious conditions, work/school
conﬂict, low autonomy and low complexity) negatively impacts upon later

psychological well being (while controlling for current work quality and social support)

H YPOTHESIS 5(b) Previous poor work quality (i.e. noxious conditions, work/school
conﬂict, low autonomy and low complexity) positively impacts upon later psychological

well being (while controlling for current work quality and social support)

IMPORTANCE OF THE PRESENT STUDY
Before turning to the methodology employed, 1 would like to say a word about the
signiﬁcance of the present study and what I hope to add to the existing literature on youth

work. First, the present study will contribute to two

72

general literatures. One, this study will add to the general literature on the relationship
between work and psychological well being by clarifying this relationship within a
particular age group (adolescence). Second, this study will contribute to the adolescent
development literature by exploring a largely neglected socialization context (work) on
adolescent development and psychological well being.

More speciﬁcally, the present study has a number of important objectives. First,
this study will attempt to bridge the gap between research on early adolescence and
adulthood by examining the largely neglected period of late adolescence. This will allow
an examination of the continuous effects of work quality on adolescent development as
individuals make the transition into adulthood. Second, the present study will also allow
a preliminary analysis of the long term effects of early work experiences on the
psychological well being of late adolescents. Lastly, we will add to the existing literature
by exploring under what circumstances youth work quality impact upon psychological
well being. Speciﬁcally, we will examine the potentially mitigating factor of co-worker
social support in the relationship between work quality and psychological outcomes.

In addition to making a contribution to academic literatures, the current study has
many important practical implications as well. As parents and policy makers, it is
important to protect the psychological well being of adolescents and to ensure their
healthy development and transition into adulthood. Since having a part-time job during
adolescence has become almost normative, it is crucial to monitor its impacts upon
today’s youth.

Youth workers provide a source of cheap and temporary labor that is necessary in

a service economy. Since employers have a vested

73

interest in attaining the maximum monetary beneﬁt from young workers, we cannot rely
on employers themselves to monitor poor work quality6 (Greenberger and Steinberg
1986). Rather, parents, policy makers, and youth workers themselves need to be aware
of which speciﬁc aspects of job quality are most important for adolescents’ psychological
well being (present and future). This information can be used in a number of ways to
improve the lives of adolescents...

First, parents can monitor their childrens’ jobs to a certain extent. For instance,
parents can ask their child to explain in detail the aspects of poor work quality that s/he

faces. When parents are aware of which speciﬁc job experiences are psychologically

 

harmful, they can warn and encourage their children to take note of poor work quality
and to search for jobs containing relatively few of such conditions. We also need to
explore factors that may buffer the negative impact of poor work quality on
psychological outcomes so that we can provide adolescents with these resources.
Furthermore, an awareness of poor work quality may cause parents to encourage their
children to seek alternative activities to work, such as extra-curricular and volunteer
activities. In contrast to employers of paid employment, the leaders of extra-curricular
activities are not likely to have a vested interest in exploiting young workers, and thus are
more likely to provide positive work experiences.

Information on the psychological impact of work quality can also aid policy
maker’s efforts in improving the lives of young workers. First, child labor laws currently
only restrict hours of work and use of hazardous equipment for young adolescents.

Suppose that the present study ﬁnds that these and other work conditions have a negative

 

6 See earlier discussion on “History of Youth Work”

74

psychological impact for older adolescents. Such information could be used as impetus
for new laws protecting Ldey adolescents and including a wider range of restricted work
conditions. Furthermore, information on the negative psychological impact of work
quality could be used by policy makers and school administrators as impetus for
developing more alternative activities for adolescents that contain positive work
experiences.

Certainly, awareness of poor work quality is only a ﬁrst step in changing the lives
of young workers. Yet, the information gained from the present study (as well as other
studies on youth work) can be used as a tool by parents, policy makers and school
administrators to improve working conditions for adolescents. Given the importance of
the present study for the psychological well being of adolescents, upon learning the
results, we will spend considerable time discussing not only theoretical, but also the

practical implications of such ﬁndings.

75

CHAPTER 2:
METHOD

SAMPLE

As we have seen, the period of late adolescence has been neglected in research on
work and well being. Thus, the present study samples a group of individuals in the
period of late adolescence; undergraduate students. We distributed closed-ended surveys
to 602 undergraduate students in Michigan State University courses during Spring and
Summer semesters 1997. With permission from the instructor, the investigator visited
four Sociology classrooms. After explaining to students the nature of the study and that
their participation was completely voluntary, questionnaires were handed out to all
willing students and returned by the end of the class period.

Although the initial questionnaires included students of all ages, we decided to
include in our analyses only those students between the ages of 18 and 23. It is
reasonable to assume that students in this age group attended college right out of high
school. Since the average time for completion of a Bachelors degree is 4-5 years, we
surmised that students over the age of 23, are likely to be returning or non-traditional
students with previous full-time work experience. The goal of the present study is to
assess the part-time work experiences of an older adolescent sample. Since we deﬁned
“late adolescence” previously as a semi-autonomous stage of life (prolonged
adolescence), young adults with previous full-time work experience would fit into an
adult, rather than an adolescent life cycle category. Furthermore, even if non-traditional
students presently have a part-time job, their impressions of work quality will be biased

by their previous full-time work experiences. For

76

these reasons, we chose to exclude any students over 23 years of age.

All students were given the opportunity to include their name and phone number
for future research. Of the questionnaires that contained contact information, we selected
those that contained complete information regarding work experience. We randomly
telephoned these students, explaining our wish to conduct a follow-up interview based on
their questionnaire. Ten qualitative interviews were conducted in summer 1997,
investigating in more detail the closed ended questions on the survey (see methods
section for more detail).

The present study’s use of a college student sample may remind readers of the
tendency in social science research to use “college sophomore” samples of convenience
and the debate regarding the validity and generalization of such samples. I would like to
defend its use, however, in the present study. First, we are speciﬁcally interested in the
life cycle period of late adolescence. Recall from our previous discussion that we
consider college students to be in this stage of late adolescence since they are semi-
autonomous. In contrast to individuals who become full-time workers right out of high
school, college students are in a prolonged stage of adolescence since they are neither
fully independent or dependent (see p. 11). As a result, our use of a college student
sample is highly desirable since we wish to understand those in this semi-autonomous
stage of life (transition to adulthood), and only wish to generalize to this particular

population,

 

7 We will discuss limitations of our sample in the Discussion Section

77

SURVEY METHODOLOGY

Before discussing the speciﬁc measures used in the present study, let us consider
why survey methodology is appropriate for the questions posed in this study. First, it is
important to consider the scope of our independent variable (work quality) when deciding
upon a suitable methodology. Consider that an individual cannot fully experience work
quality in a day or even in several days, but rather must come to know the work
environment over time. For instance, after my first week as a graduate student at
Michigan State, I was not able to fully evaluate how much work autonomy I would have
nor the noxious conditions I would experience throughout the term. Furthermore, the
experience of work takes time to impact upon psychological well being. For instance, the
time pressures involved in being a graduate student may take a year or more to begin to
affect distress levels. Thus, the context of work and its impact on psychological well
being only becomes knowable over a period of time.

As a result, it is important to measure the variable of work after one has
experienced a job for a reasonable amount of time. Thus, an experiment, for example,
would not be suitable because it could only simulate the conditions of work in a lab
environment perhaps in an hour time period. However close to “real life” these contrived
conditions may be, they cannot simulate the effects of being exposed to such conditions
for many hours per week over a long period of time. A survey, in contrast, can ask
respondents about their previous and current work experiences as well as their
psychological well being. Thus, a survey can capture the long term nature of the job and

its subsequent impact on psychological outcomes.

78

Another important issue to consider is whether a quantitative or a qualitative
study is more suitable for my research questions. First, although the present study is
exploratory in the sense of looking at adolescence, the nature of the relationship between
work quality and psychological well being has been fairly well established in the adult
literatures. Such studies have generally employed survey methodology in exploring these
relationships. Since the present study seeks to compare my results to the general
literature on work and psychological well being, it is important to employ a similar
methodology.

At the same time, however, my study is exploratory in the sense of examining
work and psychological well being for adolescents, a neglected life stage in this
literature. Since the experience of work and psychological well being may be different
for adolescents than for adults, it is also important to provide some context to the tried
and true quantitative measures in the adult literature. As a result, I also employ some
qualitative interviews in order to provide richness and texture to my data as well as to

fully explore any differences in my variables for this unique life stage8

MEASURES
The data presented in Table 1 show the means, standard deviations, ranges and
alphas for the instruments used in analyses. All measures used in this study had

satisfactory reliabilities (Cronbach alpha ranged from .712 to .870).

 

8 Qualitative interviews will be discussed in more detail in the measures section

79

Table 1. Means and Standard Deviations for all Variables in Analysis

 

Scale / Variable Meana S.D. Range Alpha # items
in scale
Autonomy T1 16.077 4.948 5 - 25 .837 5
Autonomy T2 18.704 4.985 5 — 25 .971 5
Challenge T1 10.626 3.905 4 — 20 .755 4
Challenge T2 12.300 4.446 4 - 20 .824 4
Work/School Conﬂict T1 5.045 2.479 3 - 15 .758 3
Work/School Conﬂict T2 6.225 3.218 3 - 15 .826 3
Noxious Conditions T1 9.796 3.777 4 - 20 .768 4
Noxious Conditions T2 9.186 3.589 4 - 20 .712 4
Co-Worker Support T2 7.721 2.790 3 - 12 .870 3
Depression T2 34.040 6.431 18 - 57 .787 14
Race (W=l) .752 .432
Family SES 7.359 2.487
Sex (F=1) .684 .465
Age 20.080 1.965

 

a Proportions are reported for dichotomous items

Recall Accuracy at Work Qualgy'

Respondents are asked to provide information regarding their ﬁrst high school job
held for at least 6 months, as well as current job, if applicable. Although this study is
cross-sectional, respondents are asked retrospectively about their previous work
experience. Before discussing the speciﬁc work measures used, it is important to raise
the issue of whether respondents’ recall of their previous work experience is accurate.

Beginning with Bartlett’s classic 1932 work, researchers have proposed that the
process of remembering is not a passive retrieval of stored information, but rather
involves an active reconstructive process. Research evidence indicates that when people
are asked to recall previous events that they tend to search their memories selectively and
reinterpret past events (Berger 1963; Taylor and Crocker 1981). Furthermore, people

tend to use the present as a benchmark against which

80

to compare past events. Thus, the construction of the past is largely determined by how
different or similar the respondent feels the past is from the present (Ross 1989; Ross and
McFarland 1988). People are not always accurate, however, in their comparisons of past
and present. Research evidence indicates that people tend to exaggerate both the
consistency and/or the difference between past and present events (Bem and McConnell
1970; Conway and Ross 1984).

Despite the difficulties with recalling information, there is evidence that
retrospective reports can, in fact, be adequate indicators of past behaviors. First, recall is
improved when the study asks speciﬁc, rather than more general questions. Maisto et al.
(1982) found that asking alcohol abusers the number of days spent in jail for alcohol
related offenses led to better recall than more general questions on drinking patterns.
Second, asking objective rather than subjective questions is related to improved recall.
Finney (1981) found that objective questions about health status such as number of
hospitalizations produced better recall than asking respondents to recall their previous
health status on a scale from poor to excellent. Thus, these studies demonstrate that
moderate to high recall is possible when retrospective questions are both speciﬁc and
objective.

The present study attempts to employ such techniques for improving recall. First,
we began by asking the respondent a number of speciﬁc and objective questions
regarding their previous work experience. We asked the respondents’ age when they
began their ﬁrst job, the title of the job, the number of hours worked per week, and the
amount of time the respondent stayed at the job. Answers to such questions are likely to

be accurate since they are speciﬁc and leave little room

81

for interpretation. . Thus, through asking speciﬁc and objective questions, we expect a
relatively high recall of respondents’ past work attributes.

Asking these preliminary questions is important in terms of our subsequent
questions on work quality (as we will discuss in a moment). Although these are more
subjective questions, we are conﬁdent that the respondents’ recall will be relatively
accurate for two reasons. One, we ask respondents’ about one speciﬁc job, rather than
about all past job experiences. This allows the respondent to begin thinking concretely
about a particular job experience. For instance, instead of attempting to recall past work
as a lump sum, the respondent will begin to picture a specific place where he/she worked,
the activities performed, and the relationships with particular supervisors and co-workers.
This will aid the respondent greatly in remembering the degree to which he/she
experienced work autonomy or noxious work conditions, etc. Second, we ask about
previous work experience before asking about the respondents’ current job. In this way,
the respondent is less likely to use the current job as a benchmark with which to compare
to the previous job experience. Thus, through the use of preliminary questions on one
speciﬁc job, we expect the respondent to be in an excellent frame of mind to recall this
job accurately. We are conﬁdent that respondents’ recall of past work experiences will

be an adequate measure.

Measures at Work Qualpy’

Let us now turn to the actual work measures used for both past and present work
experiences. See Appendixes A for a complete listing of measures. As discussed

previously, there is a large adult literature on work

82

quality and psychological well being. As a result, there are many well established
measures that have been used successfully in the past and have been shown to have high
reliability and validity. Researchers studying adolescent work quality (most notably
Mortimer and colleagues) have modiﬁed many of these measures slightly for use with an
adolescent population. The present study uses similar measures in order to compare our
results to those of other researchers studying youth work.

Two of our work quality measures are exact replicas of ones used by Mortimer
and colleagues in studies of adolescent workers. In terms of noxious work conditions,
respondents indicate on a 5-point scale from “never” to “almost always” the extent to
which their job includes excessive heat/cold, noise, dirtiness, physical exertion and time
pressure. Examples of questions include “How often was there time pressure on your
job” and “How often were you exposed to excessive heat, cold or noise at work.” Items
were summed, such that higher scores indicate more noxious work conditions (see
Appendix A).

Perceived Work/school conﬂict is measured by asking respondents on a 5-point
scale from “strongly disagree” to “strongly agree” the extent to which they feel there is
contradiction between work and school roles, such as being tired in class and not having
adequate time to complete homework tasks because of work hours. Questions include
“Because of my job, I come to school tired”, “Because of my job I tend to skip class” and
“Because of my job, I come to class unprepared.” Items were summed, such that higher
scores indicate greater perceived work/school conﬂict.

Mortimer and others have demonstrated the above measures to be highly reliable

when used in adolescent samples (e.g. Shanahan et al.

83

1991). The present study uses these measures unaltered in its quest to replicate
Mortimer’s studies with an older adolescent sample. Such instruments, however, are not
without their limitations. First, measures examine perceived rather than actual work
quality. For instance, we cannot be certain that one individual’s perception of “excessive
heat” is the same as anothers’ perception. Still, I contend that such differences are a
matter of degree rather than kind. For instance, a fast food worker often cooks over a
very hot grill. Although one cook may say the heat is excessive “some of the time”,
whereas another says “all of the time”, most cooks would agree the conditions are
uncomfortable to a certain extent. This is because noxious work conditions (particularly
those related to body comfort) are a universally human experience.

The question of perceived vs. actual work quality is particularly relevant for
work/school conﬂict. In our measure, respondents are asked to indicate whether work
responsibilities are to blame for being unprepared or tired in school. Although such
measures are helpful in determining perceived work/school conﬂict, they may be
inadequate in discerning actual work/school conﬂict. For instance, an individual who is
doing poorly in school may use work as a convenient scapegoat, even if there is little real
contradiction between work and school demands. Likewise, an individual experiencing
high levels of conﬂict between work and school may deny that a conﬂict exists if 3/he
must work for his/her livelihood. Thus, measures of perceived work/school conﬂict may
not fully capture actual conﬂicts between work and school demands.

Despite the above limitations, the present study will follow the lead of previous
research in measuring perceived work/school conﬂict and noxious work conditions.

Recall that our aim is to reevaluate the ﬁndings of

84

previous (albeit limited) research on youth work with an older adolescent population.
Since research on the psychological implications of youth work is still in its infancy
stages, it is important that also replicate the measures used in previous studies. In this
way, we will be able to compare our ﬁndings on the psychological implications of work
quality for college students with the results from previous studies on those of high school
age

Having said this, however, we feel that Mortimer and colleagues’ studies are
inadequate in their measurement of occupational self-direction. First, as discussed
previously, studies on adolescent workers (albeit limited) tend to examine only one
component of occupational self-direction (either work autonomy or work complexity).
Mortimer and colleagues chose to examine the work autonomy component. Their
measure of autonomy, however, is limited in that it contains only two general items
asking respondents the extent of control and freedom experienced at work (e. g. Finch et
al. 1991; Shanahan et al. 1991). It is questionable whether young adolescents are able to
fully assess their work experiences in such a global way.

The present study wishes to fully explore both work autonomy and work
complexity. Since studies on adolescent workers appear to be inadequate in this end, and
because our sample contains those making a transition to adulthood, we turn to adult
literatures in attaining complete measures of these constructs. Speciﬁcally, we draw
upon scales of autonomy and complexity that have shown to be highly reliable when used
with adults (e.g. Greenberg and Gunberg 1995). We modify these scales slightly to be
relevant to an older adolescent sample. Although there is no research to date indicating

the reliability of such scales for an older adolescent

85

sample, we use the following measures because of our strong contention that studies on
early adolescent samples are inadequate for our goals.

For work ﬂtonomv, respondents indicate on a 5-point scale from “strongly
disagree” to “strongly agree” the extent to which they feel they can make decisions at
work, control their use of time, and are supervised closely by a boss (five questions total).
Examples of questions include “I had the freedom to decide what to do on my job”, “My
supervisor leaves me alone unless I ask for help” and “I have the freedom to decide what
to do on my job.” Items were summed, with higher scores indicating higher levels of
work autonomy.

For work cmolexitv, respondents indicate on a 5-point scale from “very little” to
“very much” the extent to which they feel there is challenge and variety at their job (two
questions). Respondents also indicate on a 5-point scale from “strongly disagree to
“strongly agree” the extent to which they feel their job requires initial training and/or
prior skills and abilities (two questions). Examples of questions include “My job was so
simple that virtually anybody could handle it with little or no initial training” and “On my
job, I seldom get the chance to use my special skills and abilities”, and “How much
variety is there in your job.” Responses from negatively worded items were inverted, and
then all scores were summed such that higher scores indicate greater levels of work
complexity.

Psycholom' al Measures

As discussed in the literature review, the majority of research on work quality (for

both adolescents and adults) have used two psychological outcome measures: self-

efﬁcacy and psychological distress. In order to

86

compare our study to previous work, we will follow suit in examining both of these
psychological outcomes. We chose to use widely used scales that have been shown to be
highly reliable and valid in past studies on psychological well being. See Appendix B for
a complete listing.

Self-efﬁcacy will be measured with the well known mastery index (Pearlin 1981)
which asks respondents to answer on a 4-point scale the extent to which they feel in
control of their world. Respondents rate how strongly they agree or disagree with
statements such as “There is no way I can solve the problems I have” and “I can do just
about anything I set my mind to (see Appendix B).

Psychological distress is measured with the Center for Epidemiological Studies
Scale (CES-D) which asks respondents on a 5-point scale how they have felt during the
past week. Examples of questions include “I felt that everything I did was an effort”,
“People were unfriendly” and “I felt sad.” This scale has been shown to be highly
reliable and valid in past research (Radloff 1977).

Social Support

As discussed, the present study sought to measure one speciﬁc type of social
support; that received from co-workers. While many social support measures examine
only support from family and friends, the “Provisions of Social Relations Scale” (Turner
1983) divides social support into many areas, including co-worker support. The present
study uses a shortened version of this scale (see Appendix C). Respondents were asked
to indicate on a 4-point scale the amount of support received from co-workers. Questions
include “I have people at work who always take the time to talk over my problems if I

want to”, “I often feel really appreciated by the people

87

I work with” and “I feel close to people at work.” Items were summed such that higher
scores indicate greater perceived co-worker support. Satisfactory reliability and construct
validity for this scale has been established (Turner 1983). Furthermore, we asked
respondents whether the co-workers at their job were primarily other students (peers) or

non-students.

Sociodemographic Variables

The sociodemographic variables in this study are race, gender, family income and
age (see Table 1 for means and standard deviations). Gender and race were dummy
coded such that female=1 and white=1 respectively. In this sample, 75.2% were white
and 68.4% were female. Age was coded in years (mean=20.080). Income in family of
origin was coded on a 10-point scale from “Under $10,000” to “$90,000 and above”
(mean=$60,000-$69,999).

QUALITATIVE INTERVIEWS

The qualitative interviews consist of more in-depth questions regarding all of the
work quality variables (autonomy, complexity, noxious conditions and work/school
conﬂict). A supervised undergraduate student9 conducted the interviews as part of her
requirements for the S.R.O.P. program at Michigan State University"). I had numerous
meetings with this student before she began interviewing in order to describe the research

project’s goals and the purpose of the qualitative interviews. We also discussed her

 

9 Lana Hamilton

1° S.R.O.P. (Summer Research Opportunity Program) allows minority and ﬁrst-generation college students
the opportunity to participate in scholarly research with a faculty member and provides assistance and
guidance in graduate school preparation.

88

methods of interviewing after she had conducted a few of them, particularly with regards
to probing respondents for all of the necessary information.

Respondent names were chosen from the completed questionnaires (described
earlier). At the end of the survey, respondents were offered a space to write their name
and phone number if they wished to participate in future studies. Respondents providing
this information were called and asked to participate in a more in-depth interview.
Respondents were then interviewed at a neutral location (e. g. student union) during
normal business hours. All respondents for this portion of the study read a consent form
which described the questions we would ask, and signed the form granting permission for
us to interview them.

The interviewer asked respondents to elaborate on their experiences at work; how
various indicators of work quality made them feel, and how important it was to them for
such conditions to be altered. For instance, suppose a respondent we will call “Sue”
indicates on the survey that she experiences low levels of control at her job as a fast food
worker. The interviewer would ask her to elaborate on her experiences at work; i.e. what
sgiﬁc experiences made Sue feel that she does not have control at work? Was it the
work itself? Was it her interactions with her supervisors? Sue is also asked whether
having job autonomy is important to her, and how it makes her feel when she is not in
control at work. This question is meant to tap into the impact of work on psychological
well being from the individual’s perspective. Such questions are repeated for each
measure of work quality. The interviewer also inquired about co-worker social support
by asking Sue whether she feels close to people at work and whether that makes a

difference in her work experiences.

89

It is important to keep in mind that the qualitative interviews are not meant as the
bulk of the study, but rather serve as context for the quantitative data. We hope to
elaborate on some of the issues raised in the survey by making an attempt to understand

work conditions in the words and eyes of the respondent.

HUMAN SUBJECTS
This project has been approved by the Human Subjects Review Board (UCRIHS)

at Michigan State University. Subjects were approached in various undergraduate
classrooms and asked for their participation in a project. Before giving students the
surveys, I brieﬂy outlined that the purpose of the project was to learn more about work,
family and well being. I stressed numerous times that participation was completely
voluntary, that respondents could withdraw from the study at any time and without any
speciﬁc reason. Furthermore, I assured potential respondents that their responses would
be kept completely conﬁdential and that results would be presented as grouped data. I
then passed out surveys to those students who indicated a desire to participate. The ﬁrst
page of each survey reiterated my oral comments concerning the voluntary nature of the
project and the conﬁdentiality of responses. For respondents who also participated in the
later interviews, we required that they read and sign a standard consent form outlining
again the above conditions of the study. Thus, to the best of my ability, I feel that
subject’s rights were protected.

As with any project, however, there are potential risks to subjects. In the present
study, respondents were asked personal questions regarding their psychological well

being and work experiences. It is possible that

90

answering such questions could have brought up painful memories of psychological
issues or difﬁculties at work. On the other hand, for some respondents, discussing
negative job conditions or psychological issues may have actually been a positive
experience. Adolescents are rarely asked about job conditions or how work experiences
makes them feel. Thus, for some respondents, answering such questions may have
provided an important psychological outlet.

Furthermore, the present study has important benefits for respondents and for
adolescents as a whole. As discussed previously, attaining a fuller understanding of the
implications of youth work is an important ﬁrst step in changing poor work quality. We
must understand the mechanisms by which work affects well being and explore factors
that may buffer such impacts. Since this study’s ﬁndings can be used to improve the
situation of youth workers (including respondents in the study), I would argue that the

beneﬁts of participation far outweigh the costs.

91

CHAPTER 3:

RESULTS

REGRESSION ASSUMPTIONS

Our analysis strategy involves additive and interactive regression models to test
our hypotheses. Before embarking upon multiple regression, however, we must ﬁrst
determine whether the assumptions of regression are valid in our data. We ﬁrst tested for
normality and homoscedasticity of residuals for each dependent variable (depression and
self-efﬁcacy). For each dependent variable, we developed a full model including all
independent variables that would be used in any/all of the following analyses as
predictors.

For both depression and self-efﬁcacy, results indicated that the residuals were
normal and homoscedastic. In the case of depression, only one case was more than 3
standard deviations from the mean (case=3.5) and the maximum Cook’s D was <1 (max
Cook’s D = .067). Furthermore, the skew of the residual and the original variable of
depression were identical (.34), which was well within the limits of acceptability. For the
self-efﬁcacy scale, there was only one outlier (case: -3.38), the maximum Cook’s D was
<1 (max Cook’s D: .074), and the skew of the residual (skew: -.41) and the original
variable ( -.51) were similar and well within the acceptable range for normality.

Next, we tested for homoscedasticity of residuals. We performed an ANOVA
comparing the standardized residual (for depression or self-efﬁcacy) by three equal
groups of the standardized predictors (of the full model). The Levene’s test revealed

non-signiﬁcance for both depression and self-efﬁcacy,

92

indicating that the variances were equal in the population. Thus, we have conﬁrmed that
the regression assumptions of homoscedasticity and normality of residuals are not
violated in our data.

Our next step was to test for the regression assumption of linearity. We ﬁrst
divided each independent variable into three equal groups. Next, we performed a series
of ANOVAs, examining depression or self-efﬁcacy by levels of each independent
variable. We examined the sum of squares for linearity and deviation from linearity and
their signiﬁcance levels. Results indicated that none of the independent variables had any
signiﬁcant deviation from linearity. This indicates that any results that we ﬁnd in our
subsequent models will be of a linear nature. Thus, all of the regression assumptions
have been met in our data.

Lastly, for all models used subsequently, we tested for multi-collinearity of our
independent variables. We examined the Variance Inﬂation Factor (VIP) for all
predictors in the model. The general rule of thumb was that any VIFs >3 were cause for
concern. In these cases, we mean corrected the independent variable by subtracting the
mean from all values of the predictor. When an interaction term had an unacceptably
high VIF level, we mean corrected for each variable that made up the interaction term,
and then re-created the interaction. In all cases, mean correcting brought the VIFs to an
acceptable level (usually <2).

EMPLOYMENT STATUS AND WORK HOURS

The primary objective of the present study was to explore the psychological

implications of work conditions for youth workers. Before exploring this central

question, however, we wanted to take a snapshot of the

93

characteristics of our sample in terms of employment rates and average number of hours
worked per week. Table 2 displays job characteristics for respondents’ current job as
well as job held for 6 months or more during high school (if applicable). Similar to
previous estimates, the majority of our respondents had a job during high school (70.9%)
as well as currently (55.8%). For those who are/were employed, the average number of
hours worked per week was 15.6 hours/week during high school and 16.1 hours/week at
current job (college).

It is interesting to note some signiﬁcant demographic differences in employment
rates and work hours (see Table 2). We performed the appropriate statistical signiﬁcance
test (t-test or ANOVA) on each sociodemographic comparison (race, gender, income and
grade) for high school and college employment and work hours.

Consistent with previous research (Fordham 1996; Giordano 1993), our study
finds that during high school, minorities are less likely to work than whites (58.4% vs.
75.1%) (t: -3.93; p<.001), whereas in college, there is no signiﬁcant difference in rates of
employment. Furthermore, we ﬁnd that adolescents from wealthy families, males and
younger students are signiﬁcantly leg likely to be employed in college than their
low/middle income (F: 14.984; p<.001), female (t: -2.65; p<.01), and older student (t:-
2.59; p<.01) counterparts. In terms of work hours, the only signiﬁcant difference is that
on average, juniors/seniors in college work more hours (mean=19.l hours/week) than
their Freshman/Sophomore counterparts (mean=13.7 hours/week) (t: -5.21; p<.001).

We also performed regression analyses to conﬁrm the above ﬁndings. First, we
examined two regression equations predicting employment status (for ﬁrst job and then

for current job) by all demographic variables. Results

94

are generally consistent with our previous ﬁndings. Speciﬁcally, during high school,
minorities are less likely to work than whites (B=.167; p<.001) and during college, males
(B=.108; p<.05) higher SES students (B=-.046; p<.001), and younger students (B=.018;
p<.10) are less likely to work than their respective counterparts.

Next, we regressed work hours (for ﬁrst job and then for current job) on all
demographic variables. We were concerned that perhaps the type of job might inﬂuence
the number of hours worked. Thus, we controlled for all work quality variables
(autonomy, complexity, work/school conﬂict, and noxious conditions) in each regression
equation. Results are consistent with our previous ﬁndings (from t-tests and ANOVAs)
indicating that only one demographic variable (age for college students) signiﬁcantly
predicted work hours. Speciﬁcally, older college students tend to work more hours

per/week than their younger counterparts (B=1.420; p<.001).

95

Table 2. Job Characteristics of Sample"

Race
Sex

Family
Income

Grade

 

 

 

 

 

 

 

 

 

 

 

Employed Hours Worked
High Current High Current
School School

Mean St. dev. Mean St. dev.
White 75.1 59.5 15.386 6.776 16.218 10.028
Nomwhite 584*“ 61.6 16.396 8.232 15.680 8.237
Male 69.9 51.5 15.600 7.208 16.852 9.031
Female 72.4 63.4" 15.684 7.104 15.811 9.738
Under $10-39,999 68.4 75.7 15.746 7.228 16.625 6.601
$40-79,999 74.7 65.1 16.397 6.975 16.956 8.898
580+ 69.1 488*" 14.594 7.210 14.698 9.555
Before High School 7.5 13.791 11.999
Fresh / Sophomore 35.0 55.6 15.903 6.530 13.714 7.679
Junior / Senior 28.4 66.5 ** 15.695 6.105 l9.163*** 11.077
Total 70.9 55.8 15.612 7.130 16.095 9.637
N 602 560 426 327

 

* p < .05 ** p<.01 *** p<.001
at-test or ANOVA preformed for each comparison

96

 

PSYCHOLOGICAL CONSEQUENCES OF WORK QUALITY

Given our hypotheses, we now turn our attention to those in our sample who are
currently working at part-time jobs. We wish to explore the impact of work quality on
adolescents’ psychological well-being. We hypothesized that measures of work quality
such as noxious work conditions, low autonomy and complexity and work/school conﬂict
would decrease self-efficacy and increase psychological distress.

The ﬁrst step in testing the impact of work quality on subsequent psychological
well-being was to regress depression and self-efﬁcacy on various work conditions. Table
3 presents the results of two additive models predicting psychological distress by
measures of work quality, controlling for sociodemographic variables. As shown in the
ﬁrst model for psychological distress, three out of the four work conditions signiﬁcantly
predict depression. Speciﬁcally, complexity (B: -. 199; p<.05) decreases psychological
distress while work/school conﬂict (B: .294; p<.05) and noxious work conditions (B:
.492; p<.001) increase psychological distress.

Thus, adolescents who feel they perform simple, repetitive tasks at work or who
are exposed to extremely hot, cold or dangerous conditions at work are more likely to
suffer from psychological distress and reduced self-efﬁcacy. Furthermore, adolescents
who perceive high levels of contradiction between their school and work roles are more
likely to suffer adverse psychological consequences. Interestingly, contrary to our
hypothesis, work autonomy has no discemable effect on psychological well-bein g for the
adolescents in our sample. Thus, the degree to which adolescents have decision making

control over how they perform their job appears to

97

have no ill psychological effect. Still, there is considerable overall support for our
hypotheses that as work quality becomes increasingly noxious, there are negative

psychological consequences for adolescents.

Table 3. Reggssion of Work Quality Predicting Psychological Distressa

 

Work Quality 1 2
Complexity -. 199 (-. 139)* -.030 (-.021)
Autonomy .122 ( .096) .046 (.036)
Work / School Conﬂict .294 ( .151)* .253 (.130)*
Noxious Conditions .429 ( .275)*** .201 (.112)
Race (W=1) .320 ( .021) .268 (.017)
Sex (F=1) 1.389 ( .098) 1.476 (.105)*
Family SES -.143 (-.058) -.048 (-.019)
Age .036 ( .036) .103 (.033)
Self-Efﬁcacy -1.090 (-.492)***
Constant 26.317 48.350

R2 .142 .348

N 269 267

 

aStandardized follow Unstandardized coefﬁcients in parentheses
*p <.05 ** p < .01 *** p < .001

In Hypotheses 3b and 3c, we speculated that the positive relationship between
work autonomy and work complexity and psychological well-being would be mediated
by self-efﬁcacy. The ﬁrst step in testing these hypotheses was to examine the impact of
complexity and autonomy on self-efﬁcacy. Table 4 displays the results of an additive
model regressing self-efﬁcacy on work quality.

As shown, complexity (B: .149; p<.001) has a signiﬁcant positive impact on self-
efﬁcacy. Thus, as the complexity of work tasks increase, the adolescent experiences
higher levels of self-efﬁcacy. Furthermore, although not hypothesized, note that work
stressors have a highly signiﬁcant negative impact on self-efﬁcacy (B: -.267; p<.001).

Thus, as adolescents are exposed to excessive levels of

98

heat, cold or noise, their sense of control decreases signiﬁcantly. In contrast, work
autonomy does n_ot significantly predict self-efﬁcacy as hypothesized. Mirroring the
results for psychological well-being, work autonomy appears to be unimportant in terms
of psychological consequences.

The next step in testing the mediating effect of self-efﬁcacy was to control for
efﬁcacy in the equations predicting psychological distress. If we turn back to Table 3,
equation two, we note that self-efﬁcacy has a signiﬁcant negative impact on depression
(B: -1.090; p<.001). As we would expect, as self-efﬁcacy increases, psychological
distress decreases. Furthermore, when self-efﬁcacy is in the equation, the relationship
between complexity and psychological distress, which was signiﬁcant in equation one
(B: -.199; p<.05) goes to non-signiﬁcance in equation two. This provides evidence that
self-efﬁcacy does, in fact, mediate the relationship between complexity and
psychological distress. In other words, as the complexity of adolescent’s work tasks
increases self-efficacy, which in turn has a positive impact on psychological well-being.

Furthermore, although not hypothesized, self-efﬁcacy appears to mediate the
relationship between noxious work conditions and psychological distress. As shown in
equation two, noxious work conditions have a signiﬁcant negative relationship with self-
efﬁcacy (B: -.267; p<.001). Furthermore, the relationship between noxious conditions
and distress which was highly signiﬁcant in equation one (B: .492; p<.001) goes to non-
signiﬁcance when controlling for self-efﬁcacy in equation two. Thus, noxious conditions
have an indirectly negative impact on psychological well-being through self-efficacy.

Thus, there is considerable support for Hypotheses 1-3 that poor work quality

measures such as low complexity, high work/school

99

conﬂict, and noxious work conditions have negative consequences for adolescents’
psychological well-being. Furthermore, the work quality measures of low complexity
and noxious work conditions exert their inﬂuence indirectly through decreasing the
adolescents’ sense of competency (self efﬁcacy) which then leads to psychological

distress.

ﬂble 4. Regssion of Work Quality Predicting
Self-Efﬁcacya

 

 

Work Conditions

Complexity .149 (_231)***
Autonomy -.060 (-.104)
Work/School Conﬂict -.052 (-.059)
Noxious Conditions -.267 (-.328)***
Race (W=1) .113 (.016)
Family SES .110 (.110)
Sex (F=l) .007 (.007)
Age .063 (.063)
Constant 19903

R‘ .165

N 272

 

z“Standardized follow Unstandardized coefﬁcients in parentheses
*p <.05 ** p < .01 *** p < .001

100

C O- WORKER SOCIAL SUPPORT

Our next question involved the role of co-worker social support in the relationship
between work quality and psychological well-being. But ﬁrst, who are the adolescents’
co-workers? The present study found that the majority (67.1%) of employed adolescents
work primarily with other college students, whereas only 32.9% work with primarily
non—students. We performed a regression analysis to determine whether adolescents felt
more social support from peer co-workers than non-peer co-workers. We regressed
perceived social support on co-worker type, while controlling for all work quality
measures (autonomy, complexity, work/school conﬂict, and noxious conditions) and
demographic variables. Contrary to our predictions, respondents who worked primarily
with peers were no more likely than those working with non-peers to perceive social
support (B=.007; non-significant).

Our next step in the analysis was to test for main and interactive effects of co-
worker social support on psychological distress. We hypothesized that co-worker social
support would have a direct positive impact on psychological well-being (Hyp. 4a) as
well as buffer the negative impact of work quality on psychological well-being (Hyp. 4b).
Table 5 presents the results of an additive and interactive model predicting psychological
distress. In equation one, we see that co-worker social support decreases psychological
distress (B: -.248; p<.10)11 Thus, adolescents who perceive their co-workers to be
supportive are less likely to suffer from psychological distress. This provides support for

Hyp. 4a.

 

“ albeit this effect only approaches acceptable signiﬁcance at the .10 level

101

Next, we tested the hypothesis (4b) that co-worker support mitigates the negative
impact of poor work quality on psychological well-being. We created four interaction
terms by multiplying co-worker social support by each measure of work quality (noxious
conditions, autonomy, complexity and work/school conﬂict). Only one interaction term
signiﬁcantly predicted psychological distress as indicated in equation two (Table 5). As
shown, the interaction of co-worker social support and work complexity predicted
psychological distress (B: -.O70; p<.05). This means that having a job that includes
complex work tasks decreases psychological distress only for those with high levels of
co-worker social support. Thus, co-worker social support only has a positive impact on
psychological well-being for adolescents who have highly complex jobs. We will discuss

the implications of this ﬁnding in the discussion section.

Table 5. Regr_ession of Co-Worker Supp_ort Predicting Psychological Distressa

 

1 2
Complexity -.164 (-.114)+ -. 164 (-.l 14)+
Autonomy .155 (.122)+ .153 (.120)+
Work I School Conﬂict .292 (.150)* .299 (.154)*
Noxious Conditions .499 (.278)*** .486 (.271)***
Race (W=l) .659 (.042) .622 (.040)
Family SES -.141 (-.057) -.105 (-.042)
Sex (F=l) 1.495 (.106)+ 1.501 (.106)+
Age .057 (.018) .042 (.013)
Co-Worker Support -.248 (-.109)+ -.254 (-.11 l)+
Support * Complexity -.070 (-.133)*
Constant 24.868 21.417
R2 .156 .173
N 266 265

 

aStandardized follow Unstandardized coefﬁcients in parentheses
+p < .10 *p <.OS ** p < .01 *** p < .001

DEMQQRAPHIC DIFFERENCES

102

Up to this point, we have controlled for sociodemographics while examining the
psychological implications of work quality. Next, we will explore demographic
differences in work quality and their impact on psychological well-being.

First, we explored whether there were any signiﬁcant differences based on race,
gender, age, SES or prior work experience in the content of work quality. We regressed
each measure of work quality (autonomy, complexity, work/school conﬂict, and noxious
conditions) on each demographic variable of interest. There were only two signiﬁcant
demographic differences in work quality. Both socio-economic background (B=.224;
p<.05) and age of respondent (B=.295; p<.05) were positively related to work
complexity. Thus, college students from wealthier backgrounds were more likely to be in
a job affording challenge and variety. Furthermore, as students move through the college
experience, they are more and more likely to experience complexity at work. Overall,
however, results indicate that work quality does mp vary based on sociodemographic
factors. Being black or white, female or male, younger or older, from a rich or poor
family background, or one’s prior work experience has little discemable impact on the
conditions experienced at work.

Although there are few sociodemographic differences in work quality, it may be
that the psychological ianagt of such quality varies on the basis of demographic
variables. We performed a series of additive and interactive regression models to test for
such differences. We created an interaction term for each measure of work quality
multiplied by each demographic variable (race, age, gender, SES, prior work experience).
We then regressed psychological distress on each model including one interaction term.

In order to rule out the possibility that the type of job

103

has an impact, we controlled for all aspects of work quality (autonomy, complexity,
work/school conﬂict, and noxious conditions) in each model. Only two interactions
signiﬁcantly predicted adolescents’ psychological well-being. As shown in Table 6, the
interaction terms of noxious conditions*race and noxious conditions*age signiﬁcantly
predicted psychological distress.

This means that noxious work conditions have a greater negative impact on
psychological well-being for particular levels of race and age. Subgroup regression
analysis reveals that blacks and younger students are more prone to psychological distress
when experiencing noxious work conditions, even when controlling for other aspects of
work quality. Thus, even when experiencing comparable levels of uncomfortable work
conditions (e.g. excessive heat, cold or noise), blacks are more likely than whites to be
psychologically distressed. Similarly, younger adolescents are more susceptible than
older adolescents to be psychologically distressed when experiencing noxious work

conditions.

104

Table 6. Demoggaphic Differences in the Irﬂraa of Workﬂuality on

Psychological Distress

 

1 2
Complexity -.203 (-.142)* -. 185 (-.130)*
Autonomy .137 (.108) .128 (.101)
Work/School Conﬂict .261 (.134)* .313 (.161)**
Noxious Conditions .513 (.287)*** .518 (.290)***
Race (W=1) .272 (.018) .438 (.028)
Family SES -.l49 (-.060) -.l49 (-.061)
Sex (F=1) 1.340 (.095) 1.308 (.093)
Age .053 (.017) -.l71 (-.054)
Race * Conditions -.489 (-.117)*
Age * Conditions -.117 (-.128)*
Constant 29.464 29.93
R2 .157 .155
N 268 268

 

“Standardized follow Unstandardized coefﬁcients in parentheses

*p <.05 ** p < .01 *** p < .001

LONG TERM IMPLICATIONS OF WORK QUALITY

A ﬁnal goal of the present study was to make a preliminary assessment of the

long term psychological implications of work quality. As you may recall, we measured
work quality at respondents’ current job as well as their ﬁrst job for 6 mo. or more (if

applicable). As a result, we can perform some limited analysis on this retrospective data.

First, we assessed whether work quality changed from time one to time two. We

105

higher levels of work autonomy (t= -5.65; p<.001),

compared the means for each measure of work quality by when the job was held (high
school or currently). We performed a t-test to assess the signiﬁcance of differences.
Results indicate that overall, work conditions signiﬁcantly improve from time one to time

two. Speciﬁcally, as compared to the ﬁrst job, adolescents’ current job tends to contain

complexity (t= -4.90; p<.001) and lower noxious conditions (t= 2.45; p<.01). On the
negative side, however, work/school conﬂict tends to be higher in the second job (-6.45;
p<.001).

Thus, work quality tends to be more negative in the ﬁrst job. As a result, we
wished to explore the psychological impact of such conditions on current psychological
well-being. Our intention was a preliminary assessment of the long term psychological
implications of youth work during early adolescence.

Table 7 presents the results of a regression of T1 work quality (high school job)
predicting T2 psychological distress (current). Although the present study was unable to
control for T1 psychological distress, we did control for T2 work quality. Only one
measure of work quality, noxious work conditions, has a signiﬁcant long term impact.
As shown, noxious work conditions at T1 increase psychological distress at T2 (B=.303;
p<.05). Thus adolescents who were exposed to excessively hot cold or noisy conditions
at a high school job are more likely to suffer from subsequent psychological distress
(during college).

Throughout the present study, noxious work conditions have been a recurring
theme in having very important psychological implications. These results indicate that
noxious work conditions are harmful for psychological health not only in the short run,
but also may have important long range implications. Certainly, this result must be taken
with caution as a result of the retrospective nature of the data. Still, the possibility that
noxious work conditions have long term psychological implications is an important

preliminary ﬁnding.

106

Table 7. Regression of Psychologicﬁal Distress on T1 Work Quality“

 

 

 

T2 Psych Distress
Complexity -.016 (-.010)
. . Autonom .093 (.072)
T1 WO‘k condlt‘ons Work / Scyhool Conﬂict .358 (.139)
Noxious Conditions .303 (.181)*
Complexity —.201 (-. l40)*
. . Autonom .071 (.056)
T2 work C°“d’”°“s Work / Scyhool Conﬂict .117 (.060)
Noxious Conditions .406 (.225)**
Race (W=1) -.437 (-.027)
Family SES -.066 (-.027)
Sex (F=1) 1.517 (.105)
Age -.192 (-.039)
Constant 27.774
R“ .188
N 232

 

“Standardized follow Unstandardized coefﬁcients in parentheses
*p <.05 ** p <.01*** p < .001

107

CHAPTER 4:

DISCUSSION

The present study’s objective was to examine the impact of work quality on
psychological well being for those in late adolescence. In general, ﬁndings indicate that
similar to Mortimer’s groundbreaking research on young adolescents (as well as studies
on adult populations), that poor work quality has a negative psychological impact on
older adolescent workers. Our results indicate that three out of four measures of work
quality have a signiﬁcant negative impact on psychological well-being.

In this ﬁnal section, we will explore the implications of our speciﬁc quantitative
ﬁndings. We will discuss the impact of each measure of work quality on psychological
well-being, as well as explain the role of self-efﬁcacy, co-worker support and
demographic variables on this relationship. Throughout the discussion, we will include
insights from our 10 qualitative interviews. As discussed previously, these interviews are
not meant to form the bulk of this study, but rather serve as useful background to our
quantitative ﬁndings. As such, we will include any useful quotes from such interviews

under the relevant discussion section.

NOXIOUS WORK CONDITIONS

We ﬁnd noxious work conditions to be of extreme importance for adolescents’
psychological well being. First, previous research ﬁnds that most adolescents work in
sales/retail jobs that include noxious conditions (Aronson et al. 1996). Not surprisingly,
the majority of adolescents in our sample reported noxious work conditions at their job.

For instance, 58.5% felt time pressure at work and

108

50.7% reported that they were exposed (sometimes or always) to excessive heat, cold or
noise. For instance, one respondent explained “In the winter, the store was so cold that I
got headaches from cold air and fans being right above my head.” Another reported “it
was extremely hot; blazing hot sun.” and another said “I had to stand the whole time.”
Thus, noxious work conditions are an uncomfortable reality for many older adolescent
workers.

The present study finds that noxious work conditions have a negative impact on
psychological well being, providing strong support for our first hypothesis. The more
adolescents feel they are exposed to excessive heat, cold, noise or time pressures, the
more likely they are to be psychologically distressed. This conﬁrms the results of
Shanahan et al. (1991), who explored work conditions in a younger adolescent
population. Since we used the same measures of noxious work conditions, we can be
conﬁdent that noxious work conditions are distressing to older (as well as younger)
adolescents. Furthermore, such results are hardly surprising given the long tradition of
research indicating the distressing effect of noxious work conditions on adult workers
(e.g. House et al. 1986; Kohn and Schooler 1983).

Thus, noxious work conditions have important psychological implications for all
age groups, including those in late adolescence. Why is the ﬁnding of noxious work
conditions as distressing so consistent across age groups? One possibility is that having
comfortable working conditions is a basic human requirement for well-being. Maslow’s
theory of hierarchical need satisfaction (1954) posits that human needs are arranged
hierarchically such that lower-order needs such as food, shelter and safety must be

satisﬁed before higher-order needs such as friendship,

109

esteem and self-actualization can be attained. In Maslow’s theory, the most alienating
jobs are those that do not provide for minimum levels of physiological and safety needs.
To be sure, many of the noxious job conditions explored in the present study are part of
such basic human needs as constant body temperature, shelter from noise pollution and a
reasonable pace of effort expenditure. These conditions are, by their very nature, noxious
to humans, no matter their age.

Thus, contrary to conventional wisdom”, the fact that adolescent jobs are
temporary does n_qt protect them from the negative psychological impact of noxious work
conditions. Even if the adolescent is aware that exposure to excessive heat/cold, noise or
time pressures will likely end once s/he attains a professional job, it is still
psychologically distressing to work under these noxious conditions at present. Thus,
noxious work conditions are psychologically distressing to adolescents as they are for
adults. It is a profoundly “human” distress.

Our conceptualization of noxious conditions as a human distress brings to mind a
previous discussion on the arguments of Lazarus and colleagues (see p. 18). Recall that
Lazarus contends that events are not inherently stressful, but rather it is the individual
who subjectively gives events their meaning. Lazarus (1991) argues that speciﬁc to the
work context, it is not useful to identify stressors that have a negative impact on most
workers because stress is ultimately an individual phenomena. Whether or not a work
event will be experienced as psychological distress depends upon the individual
subjective assessment, which varies based on individual differences in disposition and

life experiences (Lazarus 1991).

 

“2 see previous discussion on “Salience of Youth Work”

110

The present study, in contrast, sees noxious work conditions as an objective
reality that lead to psychological distress in the majority of workers (regardless of
personal disposition or subjective interpretation). As discussed, noxious conditions fall at
the bottom of Maslow’s hierarchy of needs and thus form a profoundly human distress.
As a result, we have seen that the psychological impact of noxious conditions are not
likely to vary based on age or the temporary nature of one’s job. Likewise, the impact of
noxious conditions are unlikely to vary based upon personal disposition or life
experiences. It is difﬁcult to imagine, for instance, two workers exposed to an extremely
hot furnace, with one deﬁning such conditions as noxious, while the other does not.
Instead, being exposed to uncomfortable conditions is a physical reality and profoundly
distressing from a human standpoint. Thus, the present study maintains its position that
noxious conditions are objective stressors that lead to psychological distress.

Furthemore, our strongest ﬁnding in the present study is that noxious conditions
are associated with psychological distress. Other measures of work quality, such as
work/school conﬂict and work complexity were also signiﬁcantly associated with
psychological distress, but such results were not quite as statistically strong (see Results
section). The present study made every effort to measure all aspects of work quality
objectively (see p. 79). Still, noxious conditions are a physical reality, whereas
work/school conﬂict and work complexity are social realities, and thus somewhat more at
risk for subjective interpretation.

Lazarus (1991) contends that psychological distress is nothing more than the
subjective interpretation of work events. If this were true, then we would expect the less

“objective” stressors like work/school conﬂict and work complexity to have a greater

 

111

impact on psychological distress (since according to Lazarus, these measures would, in
fact, be confounded with psychological distress). But instead, we ﬁnd that noxious
conditions (the more objective stressor) has an even stronger statistical relationship with
psychological distress than work/school conﬂict or work complexity (measures which are
more at risk for confounding with psychological distress). This comparison gives us
conﬁdence that the present study has not simply measured individual variation in the
interpretations of events. Rather, we argue that all our measures of work quality, to the
best of our ability, are objectively measured and are separate from psychological distress.

The next question of interest in the present study is. . .even if noxious work
conditions are related to psychological distress, what are the effects of such conditions
over the long run? In other words, even if exposure to noxious conditions is
contemporaneously harmful, is the adolescent’s psychological well-being affected over
the long haul? Our study ﬁnds that noxious work conditions do, in fact, have long term
psychological implications. Speciﬁcally, we ﬁnd that noxious work conditions
experienced at a high school job have a negative impact on current psychological well-
being (even when controlling for current noxious work conditions). Thus, individuals
who experience excessive heat/cold, noise or time pressures at work during early
adolescence are more at risk for psychological distress in late adolescence.

Certainly, this result must be taken with caution as a result of retrospective data.
Still, our results provide some support for the “Developmental Readiness Hypothesis.”l3

Recall that Greenberger and Steinberg (1986) argue that noxious work conditions

experienced before adolescents have developed adequate coping mechanisms, are likely

 

'3 See previous discussion on “Long Term Psychological Implications”

112

to result in detrimental long term psychological outcomes. Since we measured noxious
work conditions experienced during a high school job, it is plausible that respondents
were not able, at that time of early adolescence, to cope effectively with such noxious
conditions (that subsequently increased psychological distress during late adolescence).

In contrast, there is little support for the alternative “Stress Resistance
Hypothesis” advocated by Shanahan and Mortimer (1996). Noxious work conditions do
not have positive long term psychological implications. Contrary to Mortimer and
colleagues predictions, noxious work conditions do not constitute a valuable
developmental experience, nor do they appear to mobilize coping resources. Instead, our
study indicates that, similar to adults, noxious work conditions have negative
contemporaneous as well as long term psychological implications for adolescents.

An important question for future research is why noxious work conditions during
early adolescence have long term psychological implications. There are a number of
possibilities. First, noxious work conditions may have a direct negative impact on later
psychological well-being. This seems unlikely, though, since it is hard to imagine that
exposure to excess heat at age ﬁfteen, for instance, would directly cause psychological
distress at age twenty.

A more likely possibility is that noxious work conditions cause contemporaneous
psychological distress in early adolescence (e. g. Shanahan et a1. 1991) which then
continues into late adolescence. Adolescence is a crucial time period for the formation of
personality (Erickson 1959), identity, and psychological well-being (Dombusch 1989).
Research indicates that depression in adolescence constitutes as important factor adult

psychological distress (Fleming and Offord 1990;

113

Peterson et al. 1993). Since mental health in adolescence has such long term
consequences, young adolescents damaged psychologically by noxious work conditions,
may be unable to recover psychologically regardless of later positive work experiences.
It is crucial to monitor all experiences (including noxious work conditions) that cause
psychological distress in early adolescence, since interference with the development of
psychological well-being may cause depression in later adolescence and adulthood.
Unfortunately, the present study does not include the variable of depression
during early adolescence. As a result, we are unable to test the mediating role of T1
depression in the relationship between T1 noxious work conditions and T2 depression (as
described above). Certainly, this is an important task for future longitudinal studies that
are able to control for depression during early adolescence. In the meantime, our study
provides preliminary support that noxious work conditions are important not only for
contemporaneous psychological well-being, but also have long term implications for
psychological distress. Noxious work conditions are indeed important for mental health

throughout the adolescent life cycle stage.

WORK / SCHOOL CONFLICT

Another aspect of work quality with important psychological implications is the
contradiction between work and school roles. We ﬁnd that the perception of
contradictory demands between work and school has a negative impact on psychological
well-being. This replicates the results of other studies (albeit of early adolescence) that
high levels of perceived work/school conﬂict leads to psychological distress (Finch et al.

1991; Shanahan et al. 1991). Thus, the present study

114

provides evidence that perceived work/school conﬂict has important psychological
implications not only for young adolescents, but for older adolescents as well.

Interestingly, our qualitative interviews tell a somewhat different story. Those
students who perceived contradictions between work and school roles generally did not
view this situation as psychologically distressing. Respondents who reported
work/school conﬂict were asked “how does it make you feel when your job interferes
with school?” Many respondents seemed to imply that the question itself wasn’t relevant.
For instance, one respondent replied “I can’t do anything about it; I have to work” while
another said “It’s just the way it is.” Other respondents reported that work/school
conﬂict had no psychological impact; as one respondent commented “it doesn’t bother
me.”

One wonders whether work/school conﬂict truly had no psychological impact on
such respondents, or whether they felt the need to justify their work schedules. Since
many college students contribute to their tuition or living expenses (Greenberger 1988),
these respondents may have been working out of economic necessity. As such, it would
be difﬁcult to concede that work/school conﬂict had any negative psychological
ramiﬁcations since the only alternative to such conﬂict was not attending college (which
would have even more negative implications). In other words, adolescents may not want
to consider potentially damaging psychological implications of work/school conﬂict
when they see little opportunity for such conﬂict to change.

Keep in mind that the qualitative interviews measured respondents’ perceptions of

the impact of work/school conﬂict on psychological well-being. In contrast, the

quantitative surveys asked questions pertaining to

115

work/school conﬂict and psychological distress separately and as such, are probably more
accurate in assessing the actual impact of work/school conﬂict (which leads to
psychological distress). Still, it is interesting to consider that college students may have a
tendency to deny any negative psychological impact of work/school conﬂict. Such
tendencies make it all the more important that we uncover the actual psychological
implications of contradictory demands between work and school roles.

The present study has done the important work of replicating the results of
previous studies with an older adolescent sample. Although we have conﬁrmed that
work/school conﬂict is related to psychological distress for older adolescents, our
measure of work/school conﬂict (taken from previous studies) is not ideal. Recall from
our previous discussion that researchers have generally used perceived rather than actual
work/school conﬂict. As a result, we cannot be certain how or even if work and school
roles actually contradict for adolescents. We do not yet know what sorts of
contradictions between work and school demands are most harmful for psychological
well-being. We will introduce a number of worthy pursuits in this regard for future
research.

First, work and school roles may contradict in terms of at time dimension. Since
daily time is ﬁnite, hours spent engaging in work tasks diminish time potentially
available for schoolwork. Although studies find little association between work hours
and school achievement (Mortimer and Shanahan 1991, 1994), some scholars suggest
that adolescents compensate for time lost to work by taking less demanding coursework
or decreasing time spent in other pursuits such as extra-curricular activities, leisure, social

events, and relaxation (Carr 1996; Greenberger 1988;

116

Steinberg and Dombusch 1991). Despite such speculations, no study to date has
performed a comprehensive breakdown of adolescents’ time use (Mortimer et al. 1996).
Yet in order to fully understand work/school conﬂict in terms of a time dimension, we
must explore the exact nature of adolescents’ time use in terms of work, school and other
activities. This is an important task for future studies.

A second factor that may cause work/school conﬂict (and its impact on
psychological distress) is the tinting of work shifts in relation to school activities. The
postindustrial economy has seen a substantial increase in service jobs characterized by
irregular shifts and temporary labor; work that is particularly well suited to adolescent
workers”. For instance, since the mid 19503, the number of part-time employees has
increased at an average annual rate of four percent and the number of temporary
employees has increased from 20,000 to 629,000 (see Negrey 1990). Contingent
employees (including adolescents) are attractive to employers who wish to balance their
demand for work and supply of workers as efﬁciently as possible. Part-time or
temporary employees are often required to have a “ﬂexible” schedule, based on the
availability of work, rather than on employee availability. This means that employee
schedules change from week to week based on employer need (Negrey 1990).

For instance, Ester Reiter performed a fascinating qualitative study on work
experiences at one of the largest service industries today: fast food. She worked at a
“Burger King” restaurant for 10 mo. in order to explore employee experiences in a
generally part-time, contingent environment. First, she discovered that the training

manual for new managers at “Burger King” restaurants instructed “Do not schedule to

 

" See previous discussion on “A Brief History of Youth Work"

117

accommodate employees—schedule to the needs of your restaurant. Schedule on a daily
basis using at least hourly increments” (Reiter 1996). These words were put into practice
in a number of ways. First, work schedules changed on a weekly basis and were posted
only one day prior to the subsequent week to be worked. Second, managers often
expected employees to stay later than their scheduled shift. One woman explained
“When I start to leave at the scheduled time, they tell me not to because it’s too busy.”
Third, managers often put heavy pressure on employees to work unscheduled shifts when
the restaurant is understaffed. For instance, one worker said he was afraid to answer the
phone because it was very difﬁcult to refuse the heavy pressure when managers requested
he work extra hours. Clearly, “ﬂexibility” was squarely in the hands of the employers.
The timing of work shifts as described above is all too common for adolescent
part-time workers. What is the impact of such “ﬂexible” (i.e. unpredictable) schedules on
adolescents? One main consequence is difﬁculty in planning any non-work activities,
including school work. Although college students only spend 12-15 hours/week in class,
many additional hours must be spent reading, writing papers, and studying for exams.
Suppose Mike, a college student, waits on tables at a local restaurant. When he began his
job, he wrote down his course schedule, indicating the shifts he was free to work.
Although his work schedule changes every week, he is never required to work during his
class time. But suppose Mike has an exam coming up two weeks from today. If he only
receives his work schedule one week in advance, he cannot count on being able to study
the night before the exam. Furthermore, even if he is not scheduled to work that day, he

may still be called in last minute to work an unscheduled shift.

118

Thus, the unpredictability of many part-time work schedules allows no leeway for
contingencies in the school schedule. In other words, it is not so much that hours at work
detract from hours spent studying, but it is the timing of work hours that is crucial in
contradicting with school work. If Mike worked the same shifts every week, he would be
able to plan his studying around work. But ﬁnding out his work schedule at a moment’s
notice is likely to interfere with the work he must do outside of the classroom. Thus, an
important task for future research exploring work/school conﬂict, is to consider the
timing of work hours in relation to school work, and how such contradictory demands
impact upon psychological well-being.

A third factor to consider is whether work and school are contradictory in terms of
energy levels. Work/school conﬂict is likely to be of particular concern for those in late
adolescence (especially college students) since many are faced with heavy school
demands. College courses often demand considerable energy and effort outside of the
classroom in order to succeed academically. Thus, when college students work a shift at
a physically demanding job, do they have energy left over to do their homework
assignments? Recall from our previous discussion15 that school requirements involve
more class time for high school students and more homework time for college students.
Since homework generally demands higher energy levels, we suggested that work/school
conﬂict would increase from high school to college. And respondents in our study did, in
fact, report signiﬁcantly higher perceived work/school conﬂict in college than in high

school. Of course, we cannot be certain that energy level is the root cause of this increase

 

‘5 See previous discussion on “Role Conﬂict”

119

in work/school conﬂict. Future research is needed that speciﬁcally asks respondents if
and when they feel drained from working and/or studying and how these interact.

In sum, the present study adds to the literature in its ﬁnding that perceived
work/school conﬂict negatively impacts psychological well-being for older adolescents.
The next step is to explore more speciﬁcally how work and school demands actually
contradict. We have suggested three avenues for future research including a
comprehensive breakdown of time use, the timing of work shifts, and energy levels16
The ﬁrst two avenues are more objective in the sense that respondents merely need to
answer how they spend their time or how their work scheduling operates, and the
researcher can compare the psychological implications of various breakdowns of time use
and/or work schedules. The avenue of energy levels is more subjective, since one must
ask respondents whether they are tired from work or school activities. Still, the nature of
the questions is much more speciﬁc than simply asking whether respondents perceive
work/school conﬂict. Given the speciﬁcity of these three avenues, they are likely to
reveal actual, rather than perceived work/school conﬂict. More importantly, exploring
the speciﬁc ways in which work and school roles contradict will allow a fuller

understanding of why work/school conﬂict causes psychological distress in adolescents.

OCCUPATIONAL SELF DIRECTION
Another aspect of work quality is occupational self-direction (or alienated labor)
which refers to the extent to which workers have control over their labor and utilize their

skills and capacities on the job. As discussed previously, research on adolescent part-

 

” Certainly, this is not an exhaustive list. but rather serves as a beginning impetus for future research.

120

time workers (including the present study) examine two aspects of occupational self-
direction: work complexity and work autonomy. Since most adolescents work in retail
and service jobs that afford few opportunities for work autonomy and/or complexity (e. g.
Greenberger and Steinberg 1981; Greenberger et al. 1982), do these conditions have a

negative impact on psychological well-being?

Work Complexgy'
In terms of work complexity, we find support for Hypothesis 3a. Speciﬁcally,

low work complexity is negatively related to adolescents’ psychological well-being.
Speciﬁcally, when adolescents feel work tasks are repetitive and do not allow the use of
skills and abilities, they are at risk for experiencing psychological distress. These results
replicate the ﬁndings from previous studies on young adolescents (e. g. Schulenberg and
Bachman) and adults (e.g. Kohn and Schooler 1983). Thus, work tasks that are varied
and require skill and creativity have positive psychological implications throughout the
adolescent (as well as adult) life cycle stages.

Next, the present study took a step beyond previous research on adolescent part-
time workers by exploring the role of self-efﬁcacy in the relationship between work
complexity and psychological distress. Based on previous research on adults (Kohn and
Schooler 1983; Kohn et al. 1990), we hypothesized that work complexity is indirectly
related to psychological well-being through self-efﬁcacy. Indeed, we ﬁnd support for
Hypothesis 3c, such that that self—efﬁcacy mediates the negative relationship between low
work complexity and psychological distress. Speciﬁcally, an adolescent whose work

tasks are repetitive and do not require particular

121

skills/abilities tends to experience a decrease in his/her assessment of personal
competency or effectiveness. This lowering of self-efﬁcacy, in turn, leads to
psychological distress.

It is interesting that repetitive work tasks that do not allow adolescents to use their
skills and abilities lead to feelings of low competency or self-efﬁcacy in their everyday
life. This provides support for Kohn’s generalization theory which states that individuals
learn or generalize the “lessons of the job to outside-the-job realities (Kohn 1981: p. 290).
In other words, similar to adults, when adolescents are not treated as if they were
competent at work, they begin to feel less competent in general. This ﬁnding ﬂies in the
face of the conventional wisdom that since youth work is temporary, adolescents are able
to distance themselves from the worker role. As a result, (30 the logic goes), work
complexity should have no impact on global feelings of efﬁcacy or psychological well-
being17 Yet our study provides strong evidence that the worker role is, in fact, salient to
older adolescents, since work complexity has negative psychological implications.

One reason why work complexity is salient for older adolescent workers
(especially college students) is that they are only a few years away from attaining a
professional job. As a result, many older adolescents spend a great deal of time
imagining their future “possible self” in terms of the adult worker role (Mortimer and
Johnson 1997). Even though an adolescent’s present job is quite different from the job
likely attained after graduation, the very act of working is still likely to be salient in terms
of its perceived connection to later adult work experiences (Clausen 1993; Mortimer and

Johnson 1997).

 

'7 See previous discussion on “Salience of Youth Work”

122

As an illustration, suppose Jack is a social science major who works part-time for
a market research company doing telephone interviewing. Jack’s work tasks are low in
complexity since he is required to use the same survey (and thus the same format) in all
of his telephone calls. Furthermore, although Jack has taken coursework in statistics and
methodology, his job does not afford any opportunity to use his skills in terms of
questionnaire design or data analysis. When Jack began his job, he rationalized that in a
few years, he would be the person performing the more complex tasks, perhaps even at a
similar company. But as time went on, and as Jack was continually denied the
opportunity to participate in the design or analysis of surveys, he began to question his
competency in these complex tasks. Although he may still believe he will attain a more
complex job in adulthood, Jack’s present job certainly does not inspire his conﬁdence in
his work skills and abilities (and thus causes lowered self-efﬁcacy). This example
illustrates the saliency of work complexity for older adolescents. Low work complexity
(even at a temporary job) is likely to lower the adolescent’s self-efﬁcacy as s/he imagines
his/her future self in an adult job.

Second, low work complexity during late adolescence may have detrimental long
term psychological implications. Scholars contend that the development of self-efﬁcacy
during adolescence is central in making a successful transition to adulthood (Dombusch
1989; Mortimer et al. 1996). Since college students are in the transitory phase between
adolescence and adulthood, experiences such as low work complexity are likely to
sacriﬁce this developmental period. Thus, instead of increasing one’s sense of
competency during late adolescence, those who spend many hours performing repetitive,

low-skilled tasks actually decrease (or at least fail to

123

develop) the sense of efﬁcacy necessary for adulthood roles. This lowered sense of
efﬁcacy, in turn, is related to psychological distress in older adolescents (as shown in the
present study).

Previous research demonstrates that psychological distress during adolescence is
an important risk factor for adulthood depression (e. g. Fleming and Offord 1990). Since
low work complexity is indirectly related to psychological distress during adolescence,
this negative psychological state may continue into adulthood. Although work
complexity during adolescence may have no direct effect on adulthood depression, the
psychological distress generated from low work complexity may have signiﬁcant
implications if it occurs during the delicate developmental period of late adolescence. In
other words, it may be difﬁcult to recover (or gain in the ﬁrst place) self-efﬁcacy or
psychological well-being during adulthood that is sacriﬁced during adolescence (due to
low work complexity). Thus, low work complexity not only has contemporaneous
effects on psychological distress (as demonstrated in the present study), but may also
have detrimental psychological implications into adulthood.

The present study serves as great impetus for future longitudinal research on work
complexity and psychological distress in adolescent and adult populations. Contrary to
conventional wisdom, we demonstrate that low work complexity has a negative impact
on self-efﬁcacy and psychological well-being during late adolescence (similar to adult
populations). Given the critical nature of the “transition to adulthood” period in terms of
developing self-efﬁcacy and psychological well-being, it is vital that future research
monitor the long term impact of low work complexity during adolescence on adult

psychological functioning. Work complexity is,

124

indeed, a crucial aspect of work quality for adolescent (and possibly adult) psychological

well-being.

Work Autonomy

In contrast to our ﬁndings on work complexity, the present study ﬁnds no
signiﬁcant impact of work autonomy on psychological well-being. We were surprised by
this, given our hypothesis that work autonomy would be positively related to
psychological well-being for older adolescents (see Hypothesis 3a). Recall that the
present study ﬁlls a gap between studying the psychological impact of work autonomy on
young adolescents and adults. As discussed previously, Shanahan and colleagues (1991)
ﬁnd that high work autonomy actually decreases psychological well-being among young
adolescents (10th graders). The authors speculate that since young adolescents have
limited experience with decision making, they feel threatened by high work autonomy,
and thus experience psychological distress. Studies on later life cycle stages, however,
ﬁnd that work autonomy is no longer threatening and in fact has a positive impact on
psychological well-being for post-college young adults (Mortimer 1986) and adults
(Kohn et a1. 1990; Kohn and Schooler 1983).

The present study speculated that during the life cycle stage of late adolescence,
individuals would have sufﬁcient experience with decision making activities to no longer
be threatened by work autonomy. As a result, we hypothesized that older adolescents
would beneﬁt psychologically from work autonomy, similar to adults. But instead, we

ﬁnd no impact of work autonomy on psychological

125

well-being for older adolescents. Certainly, this indicates that by the time of late
adolescence, work autonomy is not perceived as a threat, as it is for younger adolescents.
Yet, individuals do not appear beneﬁt psychologically from work autonomy until the
post-college years. This is puzzling since older adolescents do beneﬁt psychologically
from our other measure of occupational self-direction (work complexity). Why is work
complexity more important than work autonomy in terms of psychological well-being for
older adolescents?

One possibility is that older adolescents’ expectations for occupational self-
direction are similar to their current school experiences. At a large university like
Michigan State, (with class sizes upwards of 200 students), the majority of school
assignments involve reading, taking multiple choice tests, and the occasional paper. The
material covered in such assignments is typically quite complex. Since new material is
consistently presented, assignments are hardly routinized. Furthermore, students must
have a high degree of skill and/or ability to grasp challenging concepts. Thus, college
students are quite accustomed to school tasks involving high complexity.

In contrast, college students typically have little autonomy or decision making
power in completing their school tasks. Professors decide upon students’ reading
assignments and often outline explicitly what students will be tested upon or the topic of
a writing assignment. Furthermore, many college students are not autonomous in other
aspects of college living. For instance, many students live in dormitories where they are
partially supervised, and thus cannot make independent decisions about the “house
rules.” Furthermore, estimates indicate that about half of college students are ﬁnancially

dependent on their parents (National Center for

126

Education Statistics 1993), and as such, are unlikely to have full decision making control
over ﬁnancial matters.

Thus students at large universities are accustomed to situations involving high
complexity but low autonomy. When such adolescents work at part-time jobs, however,
they often experience low autonomy and low complexity. Since college students are
accustomed to complex school work, making the switch to routinized work tasks that do
not require any special ability is particularly difﬁcult, especially when the adolescent
feels s/he has gained skills in school that could be applied to more complex work tasks.
Given this discrepancy between school and work tasks, it is not surprising that the present
study found a relationship between low work complexity and psychological distress.

Autonomy, on the other hand, is not something college students regularly
encounter in school or living situations (as discussed above). Often it is not until young
adulthood that individuals make fully independent decisions such as buying a car or
house or deciding upon a general lifestyle. As a result, older adolescents may not expect
to be able to make independent decisions in a job situation. Although high work
autonomy is not threatening (as it is for younger adolescents), it is also not an expected
feature of work. Thus, the denial of autonomy at work would be unlikely to cause
psychological distress. Thus, older adolescents’ experience with low autonomy in other
aspects of life may explain the present study’s ﬁnding that low work autonomy has no
impact on psychological well-being.

In addition to exploring the impact of work autonomy on psychological well-
being, we also examined its effect on self-efﬁcacy. In contrast to Hypothesis 3b, the

present study found no signiﬁcant impact of work

127

autonomy on self-efﬁcacy for older adolescents. The explanation for this ﬁnding could
be similar to the above argument, that since older adolescents are unaccustomed to high
work autonomy, that it has no impact on self~efﬁcacy (similar to its null effect on
psychological well-being). Yet, we are still puzzled by this ﬁnding because it stands in
stark contrast to previous research as well as our own study’s qualitative ﬁndings. Let us
discuss each in turn.

First, a number of previous studies on young adolescents indicate work autonomy
increases self-efﬁcacy (Call 1996; Eccles et al. 1991; Steinberg 1990). Thus, in contrast
to the negative impact of work autonomy on psychological well-being, making
independent decisions at work has a positive impact on self-efﬁcacy for young
adolescents. Even though adolescents may be threatened by work autonomy (which
causes psychological distress), it still appears to increase their sense of competency or
efﬁcacy. For instance, suppose that Sue works at a retail clothing store. She is given
ample leeway in deciding how to display clothing on the racks and in dealing with
customer complaints. Although Sue may ﬁnd this freedom threatening and thus is often
psychologically distressed, she observes herself making these independent decisions.
Even if work autonomy is psychologically difﬁcult for Sue, she begins to believe that she
is capable of and competent in such work tasks, thereby increasing her sense of efﬁcacy.

In addition to research on young adolescents, research indicates that work
autonomy increases self—efﬁcacy for adults (e. g. Kohn and Schooler 1983). When we
discussed the impact of work autonomy on psychological well-being, it was reasonable to
suppose that the effect of work autonomy goes from negative (young adolescence) to

neutral (late adolescence) to positive (adulthood). In

128

the case of self-efficacy, however, we have no reason to suspect that the positive impact
of work autonomy would disappear between the period of young adolescence and
adulthood. If work autonomy causes a sense of competency in young adolescence, we
would assume that this would continue throughout the adolescent life cycle period into
aduhhood.

Another reason we are suspicious of the present study’s null ﬁnding regarding the
impact of work autonomy on self-efﬁcacy is because our qualitative interviews tell a very
different story. We asked respondents how it made them feel when they were unable to
make decisions as to how to carry out work tasks. Overwhelmingly, respondents’
descriptions seemed to indicate feelings of low self-efﬁcacy. Many used words to
describe their feelings such as “helpless”, “a loss of control”, “inferior”, and “incapable”
One respondent went on to explain “it makes me feel like they don’t think I know what
I’m doing.” Another said “They tried to break your spirit. I felt trapped and incapable of
doing the wor ” Given the multitude of similar responses, it seems clear that low work
autonomy does, in fact, have a negative impact on older adolescents’ sense of efﬁcacy.

Certainly, we cannot take too much stock in our qualitative results, given they are
based only on 10 respondents. Still, why are our ﬁndings so discrepant? One
consideration is that our survey questions measure sense of efﬁcacy in general, whereas
our qualitative interviews ask about respondents psychological reaction to the speciﬁc
condition of low work autonomy. In contrast to Kohn’s generalization theory, perhaps
older adolescents do not generalize their experiences of low work autonomy to other

aspects of life. Instead, perhaps they only suffer lowered efﬁcacy with regards to work

tasks and experience no decline in general self-

129

efﬁcacy. Unfortunately, the present study cannot explore this issue fully. Thus, an
important task for future research is to compare and contrast the impact of work

autonomy on speciﬁc and general self-efﬁcacy in older adolescents.

EMOTIONAL LABOR

In addition to work complexity and autonomy, we discovered another aspect of
alienated labor that may be important for older adolescents’ psychological well-bein g:
“Emotional Labor.” As discussed previously‘“, the rise in the service economy since
WWII has brought with it an increase in what Arlie Hochschild (1983) refers to as
“Emotional Labor” This refers to work tasks that require the worker to “induce or
suppress feeling in order to sustain the outward countenance that produces the proper
state of mind in others.” (p. 7). Estimates indicate that between one-half and one third of
US. workers have jobs that require signiﬁcant levels of emotional labor. This is
particularly signiﬁcant for adolescent part-time workers, since the majority work in retail
or service sectors (Aronson et al. 1996) where emotion work is most required
(Hochschild 1983).

Although we did not ask directly about emotion work, during the course of the
qualitative interviews, many respondents described work situations that seemed to
suggest that emotional labor was a central factor in their work tasks. For instance, when
one respondent was asked how low work complexity made her feel, she answered instead
by saying “What was really the most challenging was dealing with supervisors and rude

customers. Just trying to stick it out and stay there was the most challenging part.” For

 

‘8 See previous discussion on “Alienated Labor and Occupational Self-Direction”

130

this respondent, containing her natural (negative) emotional reactions when others were
rude was her work. This closely echoes Hochschild’s contention that in postindustrial
econorrries, the capacity for emotional labor (rather than working with “things) has
become the central task of today’s service jobs.

Another respondent commented “I refuse to lie to people to make a sale. You
can’t turn yourself on and off. If I’m not a dishonest person, how can I lie to make a
sale? How can I be dishonest half the day and honest the other?” Such comments
closely parallel Hochschild’s study (1983) on Delta airlines, where she found that ﬂight
attendants were required to smile at moment’s notice, no matter what their internal
emotional state. Clearly, the respondent in the present study felt a great deal of cognitive
dissonance because of the inconsistency in her behavior at work and her private feelings.
Hochschild argues that when workers must constantly manage their emotions, that they
eventually lose their very capacity to feel and become alienated from the services they
provide. It appears that this unfortunate process has already begun for the above
mentioned respondent.

How many older adolescents have similar experiences at work? Probably the
majority since so many adolescents work in sectors (retail and service) that require
emotional labor. What is the impact of such emotional labor on psychological well-
being? To be sure, the respondents in the present study appeared distressed in their
descriptions about emotional labor, but the present study has no way of testing for this
empirically. It is unfortunate that the present study did not have the foresight to ask
questions regarding emotional labor and its psychological implications. However, our

discovery proves to be extremely promising ground for

131

future research. To my knowledge, no studies to date have examined the impact of
emotion work on adolescent part-time workers. This is an extremely exciting and

important line of research for the future.

CO- WORKER SOCIAL SUPPORT

The present study makes an important contribution by exploring the impact of co-
worker social support on older adolescent workers. Although Mortimer and colleagues
have examined the negative impact of work quality on adolescent psychological well-
being, they have yet to explore the role of co—worker social support in this relationship.
No study to date has explored the potentially positive impact of co-worker social support
on adolescent workers’ psychological well-being as well as the possible mitigating role
co-worker support plays in the relationship between low work quality and psychological
distress. The present study draws upon adult literatures in this quest.

First, our results indicate that perceived emotional support from one’s co-workers
has a positive impact on older adolescents’ psychological well-being. This ﬁnding
provides strong support for Hypothesis 4(a). Furthermore, we have conﬁrmed the results
of previous studies (albeit on adults) indicating that co-worker social support has a direct
positive impact on psychological well-being (Blau 1981; Fenlason and Beehr 1994).
Thus, when older adolescents feel they can talk over their problems and feel close with
people at work, they are more likely to experience psychological well-being.

We had expected the most supportive co-workers to be members of one’s peer
group. We reasoned that college students would share more in common in terms of

interests, problems and lifestyles with other students

132

rather than with older workers. As a result, we expected older adolescent co-workers to
f orrn a camaraderie that would be protective of psychological well-being19 We failed to
find, however, any difference in perceived social support based on whether respondents
worked primarily with peers or non-peers. Thus, co-worker support is important for
older adolescents’ psychological well-being regardless of the source of support.

Should we conclude that peer co-workers are no more supportive than older adult
workers for adolescents? I would caution against this hasty conclusion. First, the present
Study is a pioneer in examining the psychological implications of co-worker support, and
thus demands replication. Second, our measure of co-worker social support may not
adequately capture the type of support provided by peers. We followed the lead of

previous studies on adults in our measure of co-worker social support. And indeed, we
ﬁnd that older adolescents view adult workers as providing the same level of this type of
support as peers. It is possible, however, that peer co-workers develop a sense of
camaraderie that is supportive but does not always involve things such as talking over
one’s problems. For instance, older adolescent co-workers may “joke around” or discuss
college courses or mutual acquaintances. Although adolescents may not always feel
emotionally “close” to their co-workers, this sort of camaraderie among peer workers
may still be perceived as supportive and thus have positive psychological implications. It
is important that future research follow up on the present exploratory study by examining
more speciﬁcally how older adolescents interact with both peers and non-peers at work.
In addition to examining direct effects, we also explored the potential moderating

role of co-worker support in the relationship between work quality and psychological

 

'9 See previous discussion on “Source of Support"

133

distress (see Hypothesis 4b). We ﬁnd that co-worker social support moderates the
relationship between work complexity and psychological well-being. This means that
high work complexity only has positive psychological implications for adolescents with
high levels of co-worker social support. Recall in a previous discussion, we speculated
that older adolescents beneﬁt psychologically from high work complexity given that they
are accustomed to complexity in their school work. Yet skills attained through complex
school work may not be directly transferable to work. For instance, even if an adolescent
is quick to grasp difﬁcult concepts in biology, this does not mean s/he is automatically
able to participate in a meeting on how to best market a company’s product.

Older adolescents may need support from more experienced co-workers in order
to fully transfer their skills and abilities from a school to a work context. For instance,
supportive co-workers may listen to concerns, help the adolescent “problem-solve”, make
introductions to other staff members, and encourage the adolescent to take on even more
complex tasks. Thus, a more experienced co-worker may be particularly helpful in
mentoring the adolescent into the world of work. This support, in turn, is likely to
increase psychological well-being. Those without such support, in contrast, may be less
likely to transfer their skills to a work context, and thus may not attain any psychological
beneﬁt even when performing complex work tasks.

It is equally important that we did not ﬁnd any other signiﬁcant interactions of co-
worker support and aspects of work quality. Regardless of co-worker social support,
noxious work conditions and work/school conﬂict had negative relationships with
psychological well-being. Talking over concerns with co-workers did not help to

alleviate the psychological consequences of

134

experiencing uncomfortable/dangerous work conditions or contradictory demands
between work and school.

This is not too surprising since discussing such concerns is unlikely to result in
any real change in terms of these aspects of work quality. For instance, exposure to
excessive heat is a profoundly human distress, and adolescents are unlikely to feel better
through talking it over with others. Furthermore, work and school demands will continue
to contradict (even when discussing the situation with others) and thus will continue to
cause psychological distress. Thus, although receiving co-worker social support may be
temporarily cathartic, it does not buffer the negative relationship between noxious

conditions and work/school conﬂict and psychological well-being.

DEMOGRAPHIC DIFFERENCES

The present study examines demographic differences in aspects of adolescent
part-time work and their psychological implications. Since this is not the main focus of
our study, we will speculate on our ﬁndings, but will not provide exhaustive
explanations.

First, we looked at differences in employment status and work hours. Consistent
with previous research (Fordham 1996; Giordano 1993), we ﬁnd that during high school,
minorities are less likely to work than whites. This ﬁnding is consistent with the “spatial
mismatch hypothesis”; the idea that since the majority of low-skilled job growth has been
in the suburbs, minorities (who are more likely to live in inner city), are less able to
obtain employment (e. g. Fordham 1996). Furthermore, previous research indicates that

minorities are at a disadvantage in attaining

135

employment because of the lack of personal contacts and job referrals (e. g. Korenman
and Turner 1996).20 Certainly, we cannot deﬁnitively conclude the existence of “spatial
mismatch” since we did not collect data on rates of seeking employment by black and
white adolescents. Thus, an important task for future research is to fully examine the
underlying causes of racial differences in adolescent employment rates.

Second, we speculated that racial differences in employment status would be
lessened or eliminated for older adolescents attending college. Since the majority of
college students (both minorities and whites) live on or near a college campus (National
Center for Education Statistics 1993), this would presumably eliminate the effects of
spatial mismatch. Consistent with our expectations, there were no signiﬁcant differences
in employment status by race for college students. Furthermore, previous research
indicates that minority students are actually more likely to work than white students
during college (Michigan State University, 1993). Thus, it may be that once minorities
live in the same area as whites (i.e. during college), blacks no longer have signiﬁcant
difﬁculty in ﬁnding and attaining employment. Again, we must qualify this conclusion
since we do not have data indicating rates of seeking employment during college. A
complete examination of the underlying reasons for racial differences in employment
rates is an important topic for future inquiry.

Furthermore, the present study ﬁnds that male college students are signiﬁcantly
less likely to be employed than female college students. This ﬁnding is in contrast to
previous research indicating. no employment status differences by gender (Mortimer et al.

1990; US. Department of Labor 1987). Recall that the majority of adolescent jobs are

 

2° See previous discussion on “Demographic Differences in Work Status.”

136

contained in the retail and service sectors. Research indicates that women are more likely
to be employed in such sectors than men (US Census Bureau 1992). Since most
adolescent jobs are in such sectors, college students may mimic their adult counterparts in
the sense that female college students are more likely to take such jobs than male college
students.

Another ﬁnding of interest is the signiﬁcant difference in employment status by
age. Speciﬁcally, younger college students are less likely to be employed than older
students. Furthermore, older college students work more hours, on average, at a part-
time job than younger college students. As college savings accounts run dry over time,
and since college tuition costs increase when one becomes an upperclassman, college
students may have to increasingly work to make ends meet. Furthermore, as college
students get closer to graduation, they are more likely to envision themselves as a
“worker” rather than a “student.” As a result, older adolescents may be more likely to
attain a part-time employment in preparation for their future adult work role.

The high cost of college tuition may also explain the ﬁnding that during college,
students from higher socio-economic backgrounds are less likely to work than their low
SES counterparts. Recall that previous research indicates the majority of young
adolescents spend their earnings on material goods rather than towards long term goals
(including saving for college expenses). Given the high cost of tuition and housing, it is
reasonable to assume that college students are likely to spend a good portion of their
income on these expenses (Greenberger 1988) rather than solely on consumer goods (like
high school students). As a result, it is likely that only those from wealthy families are

able to afford not to work (assuming they receive

137

ﬁnancial assistance from their parents). Whereas college students from high SES
backgrounds can earn sufﬁcient spending money during the summer months to last
throughout the school year, lower SES students must work during the school year simply
to make ends meet

Although the above explanation is plausible, this brings up the larger issue of our
measurement of socio-economic status for college students. There are a number of
limitations in the present study’s measurement of SES that should be addressed in future
research. First, although I speculate above that college students spend much of their
money earned from part-time jobs on college expenses, the present study did not
explicitly test for this. Regardless of family of origin SES, we need to deﬁnitively
determine why college students are working (i.e. what do they plan to spend their money
on?). Furthermore, we need to test for the congruence in adolescents’ stated reasons for
working and the ways in which money is actually spent. Thus, an important task for
future research is to construct a comprehensive breakdown of adolescents’ earned and
spent income.

Furthermore, as discussed previously, college students are in a semi-autonomous
stage of life. Younger adolescents in high school live with their families and as such are
ﬁnancially dependent upon them for the necessities of life (food, shelter, etc.). In
contrast, those in college often live away from home, and thus may be fully, partially, or
not at all reliant on their family of origin for ﬁnancial support. Instead, college students’
ﬁnancial status may include savings from their own work, ﬁnancial aid, or gifts from

family/friends. The present study’s sole reliance on family SES makes a number of

138

assumptions that do not take the adolescents’ personal income/social class into account.

The ﬁrst assumption is that the socio-economic status of family of origin is
directly transferable to resources devoted to the college aged child. Our measure of
family SES implicitly assumes that higher SES families will devote more ﬁnancial
resources to their children than lower SES families. This is not necessarily the case. For
instance, even though a higher SES family may be able to afford ﬁnancially to devote
considerable resources to their college aged child, they may choose, on principle, that
their child should be ﬁnancially independent at this stage of life. Or, a lower SES family
may make great ﬁnancial sacriﬁces in other areas of life to be able to devote considerable
ﬁnancial resources to their college-aged child. Thus, simply measuring family of origin
income does not adequately capture the extent to which adolescents receive ﬁnancial
resources.

Another limitation of our family SES measure is its failure to capture the variety
in family conﬁgurations. Recall that we asked respondents to indicate the income level
of their family of origin. Yet this measure assumes the adolescent comes from a nuclear
family, since its asks about only one family of origin. This measure of SES may be
acceptable for high school students, since many live with only one family (and thus the
basic necessities are generally covered by that one family). But for college students who
are semi-autonomous and generally live away from home, it is possible to receive
ﬁnancial resources from multiple families and stepfamilies. For instance, if John’s
parents are divorced, he may receive resources from both his mom (who is single) and

from his Dad and stepmom. Thus, our measure of family SES is limited in that it forces

139

John to “choose” upon which family to base his socio-economic status background.

Third, our measure of family SES assumes that older adolescents’ social class is
deﬁned solely by their family. Yet older adolescents are often earning signiﬁcant
ﬁnancial resources themselves, and many are paying for college and living expenses.
Furthermore, many receive ﬁnancial aid (loans and scholarships) from government,
business, and universities rather than from their families. Since older adolescents are in a
semi-autonomous stage of life, we must measure social class in ways that reﬂect this.
We must begin to see older adolescents as having their own personal income status,
rather than their SES being determined solely by family resources.

An important task for future research is to construct a socio-economic status
measure that reﬂects older adolescents’ personal income. This measure would involve
ﬁnancial resources from a number of different sources. These would include the
adolescents’ own work earnings, resources from family (or multiple families), resources
from friends, ﬁnancial aid (loans/scholarships) from government, business or universities,
and previous savings. This measure would also ask adolescents the extent to which they
use their ﬁnancial resources for living and/or college expenses vs. luxury items. Such a
measure of socio-economic status would be rich in detail and thus extremely useful when
exploring differences in youth work based on older adolescents’ independent social class
status.

Next, the present study tested for demographic differences in work quality. We
ﬁnd few signiﬁcant differences in terms of work quality (noxious conditions,
work/school conﬂict, work complexity, or work autonomy). The only signiﬁcant

demographic differences occur for work complexity.

140

Speciﬁcally, older adolescents and those from wealthier backgrounds are more likely to
experience work complexity. Thus, employers may view older and/or adolescents from
rich backgrounds as being more capable of performing complex work tasks.
Alternatively, older and/or wealthier adolescents may seek jobs containing more complex
work tasks. For instance, an adolescent from a wealthier background may have parents
who encourage him/her to seek employment containing higher complexity. The main
point, however, is that overall, whether the adolescent is from a rich or poor background,
black or white, male or female, younger or older, or has prior work experience appears to
have little impact on most aspects of work quality. It seems that the playing ﬁeld is
relatively level in terms of work quality for college students.

The question of interest in the present study, however, is whether work quality
has an impact on psychological well-being. Thus, we tested for demographic differences
in the psychological implications of work quality. And in this, we ﬁnd two interesting
demographic differences. Given the importance of noxious work conditions for
psychological well-being in our previous results, it is not surprising that it is here that we
ﬁnd signiﬁcant demographic differences as well.

First, we ﬁnd a signiﬁcant interaction of noxious work conditions and age on
psychological well-being. This means that younger college students are more likely than
older college students to experience psychological distress when exposed to noxious
work conditions. This ﬁnding harkens back to the “Developmental Readiness
Hypothesis” developed by Greenberger and Steinberg 1986. As discussed previously,

this hypothesis says that low work quality experienced before the adolescent has had a

141

chance to develop adequate coping skills is likely to have long term detrimental
psychological implications.

Thus, younger college students may be less prepared to cope with uncomfortable
work conditions such as excessive heat/cold, noise and dirtiness. Although older college
students may ﬁnd such noxious conditions just as uncomfortable, over time, they may
learn ways to cope such that their psychological well—being is not sacriﬁced. For
instance, when faced with extreme heat at work, older adolescents may be more prepared
than their younger counterparts by bringing extra drinking water or cooler clothing.
Furthermore, they may cope by internally visualizing a cooler environment. Since it
takes time to develop such coping techniques, younger adolescents are more at risk for
psychological distress when experiencing noxious conditions.

Second, we ﬁnd a signiﬁcant interaction of race and noxious conditions on
psychological well-being. This means that black college students are more likely than
white college students to suffer psychological distress when experiencing noxious work
conditions. We saw previously that there are no racial differences in levels of noxious
conditions. Thus, this ﬁnding indicates that black college students are more
psychologically reactive than whites when experiencing uncomfortable and/or dangerous
conditions such as excessive temperatures, noise, dirtiness and time pressures. How can
we explain this ﬁnding?

One possibility is that blacks are more sensitive to noxious conditions based on
previous experiences. African Americans are disproportionately represented among
lower SES populations. Those who live in poor areas are more likely to experience

negative environmental conditions such as housing

142

with insufﬁcient heating/cooling, noise and light pollution, and poor toxic substance
management. For instance, toxic waste dumps are more likely to be located in poor,
inner city areas. As an illustration, the recent demolition of the Hudsons building in
downtown Detroit left behind toxic substances estimated to be at dangerous levels.
Those who lived near the demolition site (mostly poor blacks) were simply instructed to
stay indoors, while it took many weeks for the mess to be cleaned up sufﬁciently. As a
result of situations such as these, blacks (particularly those from urban backgrounds) are
more likely than whites to be keenly aware of the harmful effects of toxic environments.

When black college students (from poor, urban backgrounds) work at part-time
jobs, the experience of noxious conditions is likely to be far more salient than to a white
college student from the suburbs, for instance. Having seen ﬁrst-hand the effects of toxic
environments, many black college students view noxious work conditions as a very real
danger or threat. In contrast, white college students, who are more likely to have
suburban backgrounds, may view such conditions as simply temporarily uncomfortable
rather than posing any real threat. It is not surprising, then, that black college students
are more likely to suffer from psychological distress when experiencing noxious work
conditions.

The present study ﬁnds no signiﬁcant gender differences in either level of work
quality or in the psychological implications of work quality. Speciﬁcally, males and
females held jobs with equal levels of work autonomy, work complexity, noxious
conditions and work/school conﬂict. Furthermore, the males and females in our sample
had similar psychological reactions when experiencing low work quality. We were

somewhat surprised by these ﬁndings since many

143

previous studies on youth work have uncovered signiﬁcant gender differences (e. g.
Mortimer, Shanahan and Call 1996; Shanahan et al. 1991). How can we explain this
discrepancy?

First, it is important to note that previous studies overwhelmingly examine those
in the stage of early adolescence. For instance, the ﬁnding that jobs held by adolescent
females are generally less complex than jobs held by adolescent males (U .S. Department
of Labor, 1986) refers to adolescents of high school age. Research indicates that during
the stage of early adolescence, jobs are likely to be sex-segregated (Greenberger and
Steinberg 1986), with girls more likely to perform informal work in private homes (e. g.
housework, babysitting) whereas boys are more likely to work in the formal sector (e. g.
paper carrier, restaurant work, etc.) (Mortimer et a1. 1990). These differences in the
sector of the economy in which boys and girls work, are likely to result in gender
differences in work complexity and other aspects of work quality during early
adolescence.

For college students, in contrast, the playing ﬁeld in more level in terms of the
types of jobs adolescents hold. First, as adolescent girls move into the period of late
adolescence, they are more likely to hold jobs in the formal sector (Mortimer et a1. 1990).
Furthermore, since the students in our sample attend college away from home, and
private homes are a good distance from the college campus, it is unlikely that girls would
have the opportunity to work in the private sector. Instead, both males and females are
likely to work in similar retail and/or food/drinking establishments on or near the college

campus. Since there is little differentiation in jobs (both males and females work in the

144

formal sector), it is not surprising that levels of work quality would be similar, regardless
of gender.

Second, in contrast to previous research, the present study ﬁnds no gender
differences in the impact of work quality on psychological well-being. Previous research
on high schoolers demonstrates that noxious conditions and work/school conﬂict increase
depression for boys, but not girls (Mortimer, Shanahan and Call 1996; Shanahan et al.
1991). The authors speculated that since there is a higher expectation that boys will work
in future, difﬁculties associated with work were more likely to threaten boys’ identity and
psychological well-being.

Although this explanation may be valid for a general high school sample, the
importance of work is likely to be independent of gender for a college sample. Females
with high career aspirations are likely to select themselves into college. Thus, the
majority of college students (whether male or female) are presumably expecting to work
in adulthood since they are attaining a marketable degree. In fact, even in the general
adolescent population, research indicates that less than one fourth of girls think they will
be full-time homemakers (Ireson and Gill 1988). As a result, the entry into the worker
role (during adolescence) is likely to be equally salient for all college students, and thus
difﬁculties at work will lead to psychological distress equally for males and females.
When viewed in this light, our null ﬁndings on gender make sense since the extent and

meaning of the worker role converges during late adolescence.

145

PRACTICAL IMPLICA TIONS

Now that we have reviewed the present study’s important ﬁndings, let us say a
word about the practical implications of our results. It is vital that we use our ﬁndings
not only to advance the literature on youth work, but also to improve the lives of older
adolescent workers. As parents and policy makers, we must protect adolescents’
psychological well-being and ensure their healthy development and transition into
adulthood. As a caveat, it is unlikely that we can signiﬁcantly alter the extent to which
adolescents work. As discussed, complex market factors have caused a vast increase in
youth work since WWII. Thus, for the foreseeable future, youth work is here to stay.
We can, however, monitor the aspects of work quality that the present study ﬁnds to be
most harmful for older adolescents’ psychological well-being.

Most efforts to date on improving the experience of part-time work for
adolescents have focused almost exclusively on limiting the number of hours worked.
For instance, in the United States, 14 and 15 year olds may only work 3 hours/day after
school, for a weekly maximum of 18 hours/day (Beyer 1994). Yet, recent research
indicates that it may not be work hours per se, but rather the quality of work that has a
greater impact on psychological well-being. At present, there is a signiﬁcant time lag in
research on work quality and policies reﬂecting this focus.

Furthermore, most efforts to date have focused almost exclusively on young
adolescents (high school students). For instance, the Fair Labor Standards Act (FLSA),
passed in 1938, restricts the hours and settings in which children under 18 years of age
can be employed (Beyer 1994). Yet many of these regulations were enacted during a

time period when the majority of 16 and 17 year olds

146

were not attending school. Such regulations have often not been updated since that time
period to reﬂect the current concerns of older adolescents still attending school
(Steinberg: as interviewed by NPR, 1999). Speciﬁc to the present study, older
adolescents who attend college continue to combine work and school roles, and are
subjected to many of the same aspects of low work quality as younger adolescents. We
cannot continue to ignore the phenomena of older adolescent part-time work in our
debates on policy. The present study will suggest areas that are particularly important
(based on our empirical ﬁndings) when enacting policy on older adolescent workers.

The main ﬁnding of the present study is that noxious work conditions have a
strong impact on older adolescents’ psychological distress. Similar to research ﬁndings
on young adolescents and adults, we ﬁnd that older adolescents who are exposed to
excessive heat, cold or noise are at greater risk for psychological distress. One method of
reducing adolescents’ exposure to noxious conditions (and thus improve psychological
well being) is to construct new and enforce existing health and safety regulations.

Generally, the health and safety of young workers has been addressed in the
public health and medical ﬁelds (e.g. American Academy of Pediatrics, Committee on
Environmental Health 1995; NIOSH 1996) but has been largely ignored in the
psychosocial ﬁelds (Castillo 1999). Numerous studies and reports indicate the negative
physical health implications for adolescents working under noxious or dangerous work
conditions. For instance, estimates indicate that adolescents sustained 21,620 work-
related injuries and illnesses serious enough to require at least one day away from work
in the US. in 1993 (CDC 1996). Common injuries include lacerations, sprains,

contusions and burns. Furthermore, each year in the

147

United States, nearly 70 people less than 18 years old die from injuries at work (Derstine
1997).

The present study is important in that it illustrates the need for psychological and
physical health policy makers to join together in the ﬁght against noxious work
conditions. Clearly, health and safety issues have negative implications for both mental
and physical health. One way to improve work conditions for adolescents (and indeed,
for all workers) is to improve compliance with existing health and safety regulations. For
instance, research indicates that 38-86% of adolescent work-related injuries are
associated with prohibited activities (Castillo et al. 1994). Activists have suggested
methods to improve compliance with health and safety laws such as increasing monetary
penalties for violations, increasing personnel for enforcement of laws (Child Labor
Coalition 1993), and the periodic review of regulations to ensure that they reﬂect current
health and safety standards (American Public Health Association 1995).

Furthermore, we need to create new regulations pertaining to noxious conditions
for older adolescent workers. At present, older adolescents are not protected by child
labor laws (since these pertain to workers under 18 years of age). For instance, child
labor laws restrict young adolescents from working in particularly dangerous
occupations. Since adolescents are still maturing physically and psychologically certain
occupational hazards are considered more dangerous for adolescents than for more
experienced adult workers (National Research Council 1993).

Yet, as we discussed previously, older adolescents who attend college are in a
semi-autonomous stage of life. Although their physical development may be near

complete, certainly their psychological development is

148

not. For instance, similar to younger adolescents, college students may not have the
experience or psychological maturity to fully recognize the dangers inherent in noxious
work conditions. Most of the child labor laws were during a time period (19403) when
far fewer adolescents attended college than do today. It is vital that we recognize the
changing demographic fabric of our society, and expand child labor laws to include older
adolescents as well. The present study indicates that older adolescents’ psychological
well-being is at risk when experiencing noxious work conditions, and we must draft new
health and safety laws to reﬂect this important ﬁnding.

Second, the present study ﬁnds that low work complexity causes psychological
distress in older adolescents. This is a somewhat more challenging issue to attack. The
public is generally in support of reducing noxious work conditions, since it easy to see
their harmful effects. Work complexity, on the other hand, is not a condition that all
persons value highly. Particularly in the case of adolescent work, the conventional
wisdom is that repetitive tasks that require little skill or ability are harmless for
adolescents since they will eventually attain a more professional job. Some go as far to
suggest that starting at the bottom and performing such repetitive tasks constitute a
beneﬁcial developmental experience.

Yet, the present study ﬁnds that low complexity is, in fact, harmful for older
adolescents’ psychological well—being. We need to begin small by asking employers to
include a few more complex tasks in the work day for older adolescents. For instance, a
telephone interviewer could spend one hour per day doing preliminary data analysis.

Although the majority of the adolescent’s workday would still include low complexity,

149

the opportunity for complex work on a limited basis may still have signiﬁcant
psychological beneﬁts.

How can we exert pressure on employers to enact such policies? One way is
through school placement ofﬁces that match up students with local employers. Such
offices can exchange their free advertisement of job openings to students with a promise
by the employer to provide a base level of work complexity. Such school ofﬁces can
point out the beneﬁts of such policies to employers. If we return to our market research
example, the company could be convinced that training the older adolescent in data
analysis would beneﬁt them over the long term. In this way, older adolescent workers
could perform preliminary data analysis and free the higher paid data analysts to perform
more sophisticated analysis activities. Thus, ﬁnding ways to convince employers that
providing some level of complexity for older adolescent workers is really in their best
interests is a vital step in ﬁghting for improved work quality for such workers.

Third, we find that work/school conﬂict has a signiﬁcant negative impact on older
adolescents’ psychological well-being. Fighting for change in this regard seems to be the
most challenging task of those mentioned thus far. As discussed previously, the rise in
the service economy has increased the degree of temporary labor and “ﬂexible” work
schedules. As we have seen, this “ﬂexibility” is firmly in the hands of employers, and as
such, work often contradicts with school demands for older adolescents. This is a
difﬁcult issue to tackle because employers have a vested interest in enacting such
practices since ﬂexible schedules allow them to increase their efﬁciency signiﬁcantly.
Again, we must rely on school placement ofﬁces to exert pressure on employers to

increase their sensitivity to school schedules. This is

150

hardly a complete solution, however, since workers (regardless of age) are confronting
contradictions between work and other role demands as a result of work “ﬂexibility.”
This is an issue that is tied to larger structural forces that will be quite resistant to change.

Fourth, we ﬁnd that co-worker social support is important in increasing older
adolescents’ psychological well-being. Furthermore, co-worker social support buffers
the negative relationship between low work complexity and psychological well-being.
We must exert pressure on employers to enact mentoring programs in which adolescent
workers are matched with more experienced workers. The mentor will be required (as
part of their job), to guide the adolescent in performing more complex work tasks,
making introductions to other staff members, and providing council and advice. This
would ultimately beneﬁt employers since adolescent workers would be better trained and
more able to perform complex work tasks. Further research is needed that does a
cost/beneﬁt analysis of implementing such programs. Such studies are vitally needed in
order to tackle the difﬁcult task of convincing employers that such mentoring program
will increase efﬁciency over the long term.

Thus far we have discussed solutions based upon employers improving work
quality in their businesses (with pressure from school placement ofﬁces). Another
method of increasing psychological well-being among older adolescent workers,
however, is the adolescent him/herself being extremely selective when choosing a part-
time job. It is vitally important that we widely disseminate the results of the present
study (and similar others) that indicate the harmful psychological implications of poor
work quality. Increasing awareness among adolescent workers is a vital ﬁrst step in

changing poor work quality.

151

Speciﬁcally, older adolescents who are aware of the negative psychological
implications of noxious conditions, work/school conﬂict, and low work complexity will
be more likely to seek out part-time jobs that contain better work quality. Furthermore,
when encountering poor work quality, adolescents who are armed with knowledge and
awareness will be more likely to confront employers and ﬁght for better working
conditions. Employers often get away with providing poor work quality because
adolescents ﬁgure that this is just “the way it is.” To be sure, this method is limited given
the multitude of adolescent jobs that contain poor work quality and the resistance of
employers to changing such conditions. Still, an awareness will increase older
adolescents’ resistance to “putting up” with poor work quality, and given enough time,
will likely lead to some base level of response by employers in terms of improving work
quality.

Such individual awareness may even lead to collective efforts to improve work
quality for older adolescent workers. Presently, young workers are the most signiﬁcantly
underrepresented groups by unions in most developed countries (Gallagher 1999). The
change from a manufacturing to a service economy has brought with it (in the United
States) a decrease in union representation. Until unions can devise new strategies for
organizing service and retail workplaces, young people are likely to be in non-unionized
environments, since they tend to work in these industries. Furthermore, young people
tend to “job hop”, making organizing efforts even more difﬁcult for adolescent workers.

Despite the obstacles, there have been a number of union efforts to improve
working conditions for adolescent workers. Many of these efforts have been in Canada,

where overall union membership for adult workers has

152

remained steady (in contrast to the United States’ declining rates). For instance, in the
19703, union organizing efforts in McDonald’s, Ponderosa, and Winco Steak —n- Burger
outlets in London, Ontario were successful in signing up the majority of workers in the
outlets they targetted. Unfortunately, not all the outlets were organized, and thus the
labor relations board decided that the union did not have a majority. Still, these efforts
gave hope that young workers could, in fact, be organized (Reiter 1996).

An impressive bargaining effort for young workers occurred in the United States
during the early 19803. The Association of Community Organizations for Reform Now
(ACORN) launched an independent union, the United Labor Union (ULU) in four major
cities. Their efforts speciﬁcally targeted young, exploited, low-wage workers. In
successful drives to sign up young people, union organizers explained to adolescent
workers that they deserved “dignity” and “respect” from their employers. A3 a result of
their efforts, many fast food restaurants (especially in Detroit where ACORN was
located) raised wages, created minimum shifts, and gave paid holidays (Reiter 1996).

Although this union was eventually decertiﬁed in the late 19803, such efforts left
a positive legacy for the ability to organize young people into unions. First, it became
clear that young people were, in fact, interested in union membership if they felt it served
their interests. As one Detroit organizer put it “A lot of these fast food operators think
they can treat their teenage employees like garbage just because they’re young and
inexperienced. The kids are fed up with this kind of treatment and that’s why they’re
joining our union.” (Reiter 1996).

Second, although ACORN focused primarily on increasing wages for young

workers, there is reason to suspect that other aspects of

153

work quality should be targeted in the future. Certainly, wages are a critical issue for
adults who are supporting themselves and possibly dependents as well. However, as we
have seen, most adolescents who work part-time come from middle-class backgrounds,
and generally do not use their money for long term savings. As a result, it is reasonable
to suspect that increases in wages are not likely to have a signiﬁcant impact on
adolescents; psychological well-being. The present study found a number of aspects of
work quality that are, in fact, psychologically harmful for adolescent workers such as
work/school conﬂict, low work complexity, and noxious conditions. Since ACORN
successfully appealed to young people’s sense of wanting “respect” from employers, this
could easily be translated into an appeal for more complex work tasks for instance. It is
important that future union efforts work focus on the ﬁndings from studies such as the
present one in making decisions about which battles to ﬁght.

Lastly, parents can play an important role in limiting the negative psychological
implications of poor work quality for older adolescents. First, we need to disseminate our
results not only to adolescents, but also to their parents. The conventional wisdom
among most parents is that youth work is inherently beneﬁcial. We must increase
awareness among parents that while youth work certainly has some beneﬁts, many of its
aspects have negative psychological implications. Parents must be made aware of which
speciﬁc aspects of work quality are most detrimental to their children’s’ psychological
well-being. Although parents cannot completely control their children’s’ working
activities, they can inquire about the work quality at their son or daughter’s workplace.

This increased awareness among parents may transfer to their children in that adolescents

154

will be more likely to consider various aspects of work quality when choosing a part-time
job.

Furthermore, parents who are aware of the negative psychological implications of
poor work quality are less likely to encourage their children to work. Since much of the
conventional wisdom indicates that work is inherently good for adolescent development,
many parents have actually pushed their children to work. Armed with greater
awareness, however, parents may decide to increase their ﬁnancial assistance to their
son/daughter and suggest alternative activities such as volunteer work or extra-curricular
activities. Certainly, the high cost of college tuition may preclude this possibility for
lower or even middle income families. Still, parental awareness of the negative
psychological implications of youth work may change the extent to which adolescents are

pressured to have a part—time job.

LIMITATIONS AND DIRECTIONS FOR FUTURE RESEARCH

As with all research, the present study is not without its limitations. Throughout
the discussion section, we have noted speciﬁc limitations and questions for future
research within each section on work quality. At this point, I would like to discuss two
general limitations that are pervasive throughout the entire study. Given the exploratory
nature of the present study, these limitations can serve as an important impetus for future
research.

First, the present study is limited in its use of cross-sectional data. Throughout
our discussion, we have conceptualized of work quality as causing psychological well-

being. Although this is theoretically grounded, we

155

cannot deﬁnitively establish the causal direction of this relationship in an empirical sense.
Our main barrier in testing for causality is that we have no measure of prior
psychological well-being. Although we ask respondents about their previous work
experiences retrospectively, questions regarding prior levels of psychological distress are
likely to be heavily confounded with current levels of distress. Yet, we must have this
information in order to fully test for the alternative argument that psychological well-
being causes aspects of work quality. There are two ways in which this argument can be
theoretically conceptualized. . .

First, adolescents who are psychologically distressed may seek out jobs with
lower work quality. Or put another way, adolescents who are psychologically distressed
may not expend the extra effort to attain a job with high work quality. For instance,
suppose Sue is suffering from psychological distress. She sees a “help wanted” sign at a
restaurant and applies immediately because she doubts she can ﬁnd much better (as a
result of her depression). Sue’s job has extremely noxious work conditions as well as
low complexity and autonomy. Furthermore, her job hours contradict with studying for
college exams. As a result, Sue suffers from further bouts of depression, which ironically
enough, actually cause her to stay at her current job even longer.

Another possibility is that psychological distress leads to the perception that one
is experiencing lower work quality. In other words, whether or not work quality is poor,
adolescents who are psychologically distressed will perceive it to be 30. Let us return to
our example of Sue who is psychologically distressed. One of Sue’s co-workers at the
restaurant is a fellow college student, Tim (who has high psychological well-being).

Although Sue and Tim have similar work experiences,

156

Sue allows her depression to cloud her thoughts in the sense that she perceives poor work
quality whereas Tim does not. These perceptions of poor work quality, in turn, cause Sue
to become increasingly psychologically distressed.

Note that in both of the above examples, psychological well-being and work
quality (or perceived work quality) work in reciprocal fashion. In other words,
psychological distress and poor work quality form a downward spiral where each
increases the other over time. Our present study only takes a single snapshot of this
relationship, and thus cannot assess reciprocal effects. We do not know, for instance,
whether prior psychological well-being affected current work quality. Furthermore, we
do not know how current work quality will impact upon psychological distress in
adulthood. Thus, an important and exciting task for future research is to assess such
relationships using data that tracks individuals from early adolescence through adulthood.
The present study provides a fertile starting ground for such longitudinal studies covering
the late adolescent period.

A second limitation in the present study is its use of a college student sample.
Recall that we wish to examine those in a semi-autonomous life cycle phase. As
discussed previously, college students ﬁt this bill quite well since they are neither fully
independent nor dependent and continue to combine work and school roles. Thus it is not
the use of a college student sample per se that we feel is the present study’s limitation.

Rather, the limitation of our sample lies in the lack of m of college students.
Our data come from a single university whose students’ experiences are not necessarily
representative of all colleges and universities. First, Michigan State University is a very

large university with over 40,000 students, the majority

157

of whom live on or near campus. The massive size of this university has created a small
“town” of sorts (East Lansing) that caters primarily to students. As a result, there are a
plethora of low level service jobs both on and near campus such as fast food restaurants,
cafeterias, and student oriented retail stores. Furthermore, employers have a large sample
of cheap student labor from which to ﬁll these positions.

In contrast, a smaller college or a commuter campus would not have as great a
need for student oriented services on or near campus. As a result, students at such
colleges would be forced to seek employment in the surrounding community. In the case
of Michigan State University, the Lansing area has many highly skilled state government
workers, and has little need for college student workers. At a smaller college, however,
students may be more skilled than the residents in the surrounding community, and thus
may be able to land a job of relatively high quality. Furthermore, employers cannot
afford to be too choosy since they only have a small sample of students to draw upon to
ﬁll positions. As a result, students at smaller colleges may be more likely to attain a job
with higher work quality than a student at a large student-oriented university.

Second, Michigan State University draws its student population from relatively
wealthy suburban areas in Michigan. Although many students pay for a portion of their
education, their wealthy backgrounds may impact upon motives for working. Similar to
ﬁndings for high school students (e. g. Bachman 1983), college students may work more
to buy material goods rather than economic necessity. Furthermore, although Michigan
State is ofﬁcially a land grant university, it draws only a fraction of students from rural

(or urban) communities. As a result, students from suburbia may have very different

158

orientations towards work than those from rural or urban communities.

To be sure, there are many other differences based on type of university and
student background than I’ve covered here. My intent is not to provide an exhaustive
review of such differences, but rather to clarify the extent to which our sample is
generalizable. The results of the present study can only be reasonably generalized to
relatively large, self contained universities in the United States. Having said this, our
results are not without merit when speculating on a variety of college student
experiences. Rather, our study examines general processes of the impact of work quality
on psychological well-being for those in late adolescence. As such, the present study is
an important starting point for research on the work experiences for all those making the

transition to adulthood.

CONCLUSION

The present study has made many important contributions. We have done the
important work of replicating Mortimer and colleagues’ groundbreaking research on
youth work using an older adolescent sample. It is vital that we examine the impact of
work quality across the life span, including the “transition to adulthood” period. Similar
to Mortimer’s research, we ﬁnd that poor work quality such as noxious conditions,
work/school conﬂict, and low work complexity have negative effects on psychological
well-being for older adolescents. We have also moved beyond Mortimer’s work on early
adolescence in ﬁnding that co-worker social support is important for psychological well-

being in older adolescents. Furthermore, we discovered that “emotional labor” is a

159

signiﬁcant aspect of youth work and an exciting area for future research on younger and
older adolescents.

In sum, the present study has made a signiﬁcant contribution to two general
literatures. First, we have added to the general literature on the relationship between
work and psychological well-being by clarifying this relationship within a particular age
group (late adolescence). Second, this study has contributed to the adolescent
development literature by exploring a largely neglected socialization context (work) on
adolescent development and psychological well-bein g. Lastly, the present study’s results
have many important practical implications for improving the lives of older adolescent

workers.

160

APPENDICES

161

APPENDIX A: WORK UALITY MEASURES

Noxious Work Conditions

1. How often is there time pressure on your job?
1 2 3 4

Never Sometimes

2..How often are you exposed to excessive heat, cold or noise at work?
1 2 3 4

Never Sometimes

3. I have too much work to do everything well
1 2 3 4

Not at all true Somewhat true

4. I feel drained of energy when I get off work
1 2 3 4

Not at all true Somewhat true

Perceived Work/School Conﬂict

1. Because of my job, I come to school tired
I 2 3 4

Strongly disagree Somewhat disagree Neither agree Somewhat agree
Nor disagree

2. Because of my job, I tend to skip classes

1 2 3 4

Strongly disagree Somewhat disagree Neither agree Somewhat agree
Nor disagree

3.Because of my job, I come to class unprepared

1 2 3 4

Strongly disagree Somewhat disagree Neither agree Somewhat agree
Nor disagree

162

5

Almost Always
5

Almost Always

Very true

Very true

5

Strongly agree

5
Strongly agree

5

Strongly agree

Work Autonomy

1.How much freedom do you have on your current job? That is, how much do you
decide on your own what you do on the job?

1 2 3 4 5
Very little. there are Some freedom to Very much. 1 make
few decisions 1 can make Make decisions many decisions

2.My supervisor leaves me alone unless I ask for help

1 , 2 3 4 5
Strongly disagree Somewhat disagree Neither agree Somewhat agree Strongly agree
Nor disagree

3. My job allows me to control my own work pace

1 2 3 4 5
Strongly disagree Somewhat disagree Neither agree Somewhat agree Strongly agree
Nor disagree

4.It is basically my responsibility to decide how my job gets done

1 2 3 4 5
Strongly disagree Somewhat disagree Neither agree Somewhat agree Strongly agree
Nor disagree

5.1 have the freedom to decide what to do on my job

1 2 3 4 5
Strongly disagree Somewhat disagree Neither agree Somewhat agree Strongly agree
Nor disagree

Work Complexgy'

1. How much challenge is there on your job?
1 2 3 4 5

Very little Some challenge A great deal of challenge
Challenge

2. How much variety is there in your job?

1 2 3 4 5
Very little, 1 do pretty Some variety Very much. 1 do many
much the same thing over different things
and over

3. My job is so simple that virtually anybody could handle it with little or no initial

training
1 2 3 4 5

Strongly disagree Somewhat disagree Neither agree Somewhat agree Strongly agree
Nor disagree

4.0n my job, I seldom get the chance to use my special skills and abilities

I 2 3 4 5
Strongly disagree Somewhat disagree Neither agree Somewhat agree Strongly agree
Nor disagree

163

APPENDIX B: PSYCHOLOGICAL MEASURES

Selt-Eﬂacy Scale
1=Strongly Disagree 2=Somewhat Disagree 3=Somewhat Agree 4=Strongly Agree '

OWNP‘MPP’N!‘

There is no way I can solve the problems I have
Sometimes I feel I’m being pushed around in life

I have little control over the things that happen to me

I can do just about anything I set my mind to

What happens to me in the future mostly depends on me
I often feel helpless in dealing with the problems of life
I am able to do things as well as most other people

I feel that I do not have much to be proud of

I am inclined to think I am a failure

10. I am satisﬁed with myself
11. I wish I could have more respect for myself

Depression (CES-D scale)

1=Never 2=Hardly Ever 3=Some of the Time 4=Most of the Time 5=Always

OWNE‘MPP’NT‘

I felt depressed

I felt that anything I did was an effort

My sleep was restless

I was happy

I felt lonely

People were unfriendly

I enjoyed life

I didn’t feel like eating. My appetite was poor
I felt sad

10. I felt that people disliked me
11. I could not get “going”

12. I felt that I just had to get drunk
13. I felt like school was going well
14. I felt happy to be in school

164

APPENDIX C: SOCIAL SUPPORT SCALE

=Very Much Like My Experience 2=Much Like My Experience
3=Somewhat Like My Experience 4=Not at all Like My Experience

1. I feel close to the people at work
2. I have people at work who would always take the time to talk over my problems

should I want to
3. I often feel really appreciated by the people I work with

**Are your co-workers primarily other students (peers) or non-students?

a) primarily other students
b) primarily non-students

165

REFERENCES

165

REFERENCES

American Academy of Pediatrics, Committee on Environmental Health. 1995. “The
Hazards of Child Labor.” Pediatrics 95:311-313.

American Psychiatric Association, Committee on Nomenclature and Statistics. 1980.
Diagnostic and Statistical Manual of Mental Disorders (3rd ed.). Washington, DC:
American Psychiatric Association.

American Psychiatric Association. 1987. Diagnostic and Statistical Manual of Mental
Disorders, Revised Third Edition. Washington DC: American Psychiatric Association.

American Public Health Association. 1995. “Protection of Child and Adolescent
Workers—Policy Statement Adopted by the Governing Council of the American Public
Health Association, November 2, 1994.” American Journal of Public Health. 85:440-
442.

Anderson, J .C., S. Williams, R. McGee, P.A. Silva. 1987. “DSM III Disorders in
Preadolescent Children: Prevalence in a Large Sample From the General Population.”
Archives of General Psychiatry 44:69-76.

Aneshensel, Carol 8., RR. Frerichs and V.A. Clark. 1981. “Family Roles and Sex
Differences in Depression.” Journal of Health and Social Behavior, 222379-93.

Aronson, Pamela J ., J .T. Mortimer, C. Zierman, and M. Hacker. 1996. “Generational
Differences in Early Work Experiences and Evaluations.” In J.T. Mortimer and MD.
Finch (eds. ), Adolescents, Work, and Family: An Intergenerational Developmental
Analysis. London: Sage Publications.

Bachman, Jerald G. 1987. “Adolescence: An Eye on the Future.” Psychology Today,
6(8).

Bachman, Jerald G. 1983. “Premature Afﬂuence: Do High School Students Earn Too
Much?” Economic Outlook USA; Survey Research Center, Institute for Social Research.
10(3) Summer: 64-71.

Bachman, Jerald 0., DE. Bare, and EL Frankie. 1986. “Correlates of Employment
Among High School Seniors” (Monitoring the Future Occasional Paper No. 20). Insitute
for Social Research, Ann Arbor, Michigan.

Bachman, Jerald G., L.D. Johnston, and P. O’Malley. In Press. “Recent Findings from
Monitoring the Future: A Continuing Study of the Lifestyles and Values of Youth.” In F.
Andrews (ed.), Research on the Quality of Life. Ann Arbor, MI: Institute for Social
Research.

167

Bandura, Albert. 1982. “The Self and Mechanisms of Agency.” Pp. 3-39 in J. Suls (ed)
Psychological Perspectives on the Self, vol 1. NJ: Erlbaum.

Barenboim, Carl. 1981. “The Development of Person Perception in Childhood and
Adolescence: From Behavioral Comparisons to Psychological Constructs to
Psychological Comparisons.” Child Development. 52:129-44.

Bartlett, EC. 1932. Remembering: A Study in Experimental and Social Psychology.
London: Cambridge University Press.

Barton, P. 1989. Earning and Learning: The Academic Achievement of High School
Juniors with Jobs. Princeton, NJ: Educational Testing Service.

Beehr, TA. 1985. “The Role of Social Support in Coping with Organizational Stress” Pp.
375-398 in T.A. Beehr and RS. Bhagat (Eds) Human Stress and Cognition in
Organizations: An Integrated Perspective. NY: Wiley.

Bell, Daniel. 1973. The Coming of Post-Industrial Society. NY: Basic Books.

Bem, DJ. and H.K. McConnell. 1970. “Testing the Self-Perception Explanation of
Dissonance Phenomena: On the Salience of Premanipulation Attitudes.” Journal of
Personality and Social Psychology 14:23-31.

Berger, PL. 1963. Invitation to Sociology: A Humanistic Perspective. Garden City, NY:
Doubleday.

Beyer, D. 1994. “Child Labor in the 90’s: Trouble Spots and Timely Solutions. In _,
National Child Labor Committee (Ed ) Essay in Child Labor in the 903: How Far Have
We Come? (pp. 26-30). NY: National Child Labor Committee.

Billings, AG. and RH. Moos. 1982. “Work Stress and the Stress-Buffering Roles of
Work and Family Resources.” Journal of Occupational Behavior, 3:215-32.

Blau, G. 1981. “An Empirical Investigation of Job Stress, Service Length and Job Strain”
Organizational Behavior and Human Performance 27:279-302.

Blauner, Robert. 1964. Alienation and Freedom. Chicago: University of Chicago.

Blazer, D. 1982. “Social Support and Mortality in an Elderly Community Population.”
American Journal of Epidemiology. 115:684-94.

Brief, Arthur P. and J .M. George. 1991. “Psychological Stress and the Workplace: A

Brief Comment on Lazarus’ Outlook” in FL. Perrewe (Ed.) Handbook on Job Stress
(Special Issue). Journal of Social Behavior and Personality, 6(7): 15-20.

168

Broadbent, DE. and D. Gath. 1981. “Symptom Levels in Assembly Line Workers.” Pp.
243-252 in G. Salvendy and M.J. Smith (Eds) Machine Pacing and Occupational Stress.
London: Taylor and Francis.

Broman, Clifford L., V.L. Hamilton, W.S. Hoffman, R. Mavaddat. 1995. “Race, Gender,
and the Response to Stress: Autoworkers’ Vulnerability to Long-Term Unemployment.”
American Journal of Community Psychology. 23(6):813-842.

Bromet, Evelyn J ., M.A. Dew, D.K. Parkinson, and H.C. Shulberg. 1988. "Predicting
Effects of Occupational and Marital Stress on the Mental Health of a Male Workforce."
Joumal of Organizational Behavior 9: 1-13.

Brown, George W., A. Bifulco and TO. Harris. 1987. “Life Events, Vulnerability, and
Onset of Depression: Some Reﬁnements.” British Journal of Psychiatry 150:30-42.

Brown, George W., L. Lemyre and A. Bifulco. 1992. “Social Factors and Recovery from
Anxiety and Depressive Disorders: A Test of Speciﬁcity.” British Journal of Psychiatry
161:44-54.

Call, Kathleen T. 1996. “The Implications of Helpfulness for Possible Selves.” In J .T.
Mortimer and MD. Finch (eds), Adolescents, Work, and Family: An Intergenerational
Developmental Analysis. London: Sage Publications.

Carnegie Council on Policy Studies in Higher Education. 1980. Giving Youth a Better
Chance. San Francisco: Jossey-Bass.

Carr, Rhoda Vl, J .D. Wright, and C]. Brody. 1996. “Effects of High School Work
Experience a Decade Later: Evidence from the National Longitudinal Study.” Sociology
of Education 69:66-81.

Castillo, Dawn. 1999. “Occupational Safety and Health in Young People.” Pp. 159-200
in J. Barling and E.K. Kelloway (Eds) Young Workers: Varieties of Experiences.
Washington DC: American Psychological Association.

Castillo, Dawn, D. Landen and LA. Layne. 1994. “Occupational Injury and Deaths of 16
and 17 year olds in the United States.” American Journal of Public Health, 84: 646-649.

Centers for Disease Control and Prevention (CDC). 1996. “Work Injuries and Illnesses
Associated with Child Labor—United States, 1993.” Morbidity and Mortality Weekly
Report. 45:464-468.

Child Labor Coalition. 1993. Child Labor Update and Recommendations for Action.
Wasington DC: National Consumers’ League.

Clausen, John A. 1993. American Lives: Looking Back at the Children of the Great
Depression. NY: Free Press.

169

Cohen, Sheldon and TA. Wills. 1985. “Stress, Social Support, and the Buffering
Hypothesis.” Psychological Bulletin 98:310-57.

Conway, M. and M. Ross. 1984. “Getting What you Want By Revising What You Had.”
Journal of Personality and Social Psychology 47:738-748.

Cooley, C. 1902. Human Nature and the Social Order. NY: Scribners.

D'Amico, Ronald. 1984. "Does Working in High School Impair Academic Progress?"
Sociology of Education. 57:157-164.

Derstine, B. 1997. Job-Related Fatalities Involving Youths, 1992-1995. Fatal Workplace
Injuries in 1995: A Collection of Data and Analysis (US. DOL Rep. NO. 913).
Washington DC: US. Department of Labor.

Dohrenwend, Bruce P. and BS. Dohrenwend. 1969. Social Status and Psychological
Disorder: A Causal Inquiry. NY: Wiley-Interscience.

Dohrenwend, Bruce R, B.G. Link, R. Kern, P.E. Shrout and J. Markowitz. 1990.
“Measuring Life Events: The Problem of Variability Within Event Categories.” Stress
Medicine 6: 179-187.

Dombusch, Sanford M. “The Sociology of Adolescence” Annual Review of Sociology
15:233-59.

Dunkel-Schetter, C. and T. Bennett. 1990. “Differentiating the Cognitive and Behavioral
Aspects of Social Support.” Pp. 267-96 in ER. Sarason, I.G. Sarason and GR. Pierce
(Eds) Social Support: An Interactional View. NY: Wiley.

Dupre, John and R. Gagnier. 1996. “A Brief History of Work” Journal of Economic
Issues 30(2):553-559.

Eccles, J .S., C.M. Buchanan, C. Flanagan, A. Fuligni, C. Midgley, and D. Yee. 1991.
“Control versus Autonomy During Early Adolescence” Journal of Social Issues, 47:53-
68.

Edwards, Richard. 1979. Contested Terrain: The Transformation of the Workplace in the
Twentieth Century. NY: Basic Books.

Elder, Glen H., Jr. 1974. Children of the Great Depression. Chicago: University of
Chicago Press.

Elder, Glen H. and A.M. O’Rand. 1995. “Adult Lives in a Changing Society.” Pp. 452-
475 in KS. Cook, G.A. Fine, and J .8. House (eds) Sociological Perspectives on Social
Psychology. Boston: Allyn and Bacon.

170

Erikson, Erik H. 1959. “The Problem of Ego Identity.” Psychological Issues 1:101-164.
_. 1963. Childhood and Society. NY: Norton.

__.1968. Identity: Youth and Crisis. NY: Norton.

Erikson, Kai. 1987. “On Work and Alienation.” American Sociological Review, 51:1-8.

and SP. Vallas, eds. 1990. The Nature of Work: Sociological Perspectives.
New Haven, CT: Yale University Press.

Farmer, HS. 1983. “Career and Homemaking Plans for High School Youth” Journal of
Counseling Psychology 30(1): 40-45.

Farmer, Mellissa M. and KP. Ferraro. 1997. “Distress and Perceived Health:
Mechanisms of Health Decline” Journal of Health and Social Behavior, 39(Sept.):298-
31 1.

Fenlason, Kristofer and T.A. Beehr. 1994. “Social Support and Occupational Stress:
Effects of Talking to Others” Journal of Organizational Behavior 15: 157-175.

Finch, Michael D., J .T. Mortimer, M.J. Shanahan, and S. Ryu. 1991. “Work Experience
and Control Orientation in Adolescence.” American Sociological Review. 56:597-611.

Finch, Michael D. and J .T. Mortimer. 1996. “Future Directions for Research on
Adolescents, Work, and Family.” Pp. 221-236 in J .T. Mortimer and MD. Finch (eds)
Adolescents, Work, and Family: An Intergenerational Developmental Analysis. London:
Sage Publications.

Finney, HQ 1981. “Improving Reliability of Retrospective Survey Measures: Results of
a Longitudinal Field Survey.” Evaluation Review 5:207-229.

Fordham, Si gnithia. 1996. Blocked Out: Dilemmas of Race, Identity, and Success at
Capital High. University of Chicago Press: Chicago and London.

French, John R.P., R.D. Caplan, and R. Van Harrison. 1982. The Mechanisms of Job
Stress and Strain. NY: Wiley.

Fromm, Erich. 1968. The Revolution of Hope. NY: Harper and Row.
Gallagher, Daniel. 1999. “Youth and Labor Representation.” Pp. 235-258 in J. Barling

and E.K. Kelloway (Eds) Young Workers: Varieties of Experience. Washington DC:
American Psychological Association.

171

Ganster, D.C., M.P. Fusilier and B.T. Mayes. 1986. “Role of Social Support in the
Experience of Stress at Work.” Journal of Applied Psychology 71:102-110.

Gecas, Viktor. 1982. “The Self Concept.” Annual Review of Sociology 8: 1-33.

.1989. “The Social Psychology of Self-Efficacy” Annual Review of Sociology
15:291-316.

Gecas, Viktor and MA. Seff. 1989. “Social Class, Occupational Conditions, and Self-
Esteem.” Sociological Perspectives. 32(3):353-364.

Ginzberg, Eli. 1977. The Job Problem. Scientiﬁc American 237:43-51.

Giordano, Peggy C., S.A. Cemkovich and A. DeMaris. 1993. “The Family and Peer
Relations of Black Adolescents.” Journal of Marriage and the Family. 55:261-276.

Gove, Walter R. 1972. “The Relationship Between Sex Roles, Mental Illness, and Marital
Status.” Social Forces, 51:34-44.

Gove, Walter R. 1978. “Sex Differences in Mental Illness Among Adult Men and
Women: An Evaluation of Four Questions Raised Regarding the Evidence on the Higher
Rates of Women.” Social Science and Medicine. 12B: 187-98.

Gove, Walter R., C.B. Style, and M. Hughes. 1990. “The Effect of Marriage on the Well-
Being of Adults.” Journal of Family Issues, 1124-35.

Greenberg, Edward S. 1986. Workplace Democracy: The Political Eﬂects of
Participation. Ithaca, NY: Cornell University Press.

Greenberg, Edward S. and L. Grunberg. 1995. “Work Alienation and Problem Alcohol
Behavior.” Journal of Health and Social Behavior. 36(March):83-102.

Greenberger, Ellen. 1988. “Working in Teenage America.” Pp. 21-50 in J. Mortimer and
KM. Borman (Eds) Work Experience and Psychological Development Through the Life
Span. Boulder, Colorado: Westview Press, Inc.

. 1984. “Children, Families and Work” Pp. 103-122 in ND Reppucci, L.A.
Weithom, E.P. Mulvey and J. Monahan (Eds) Children, Mental Health and the Law.
Beverly Hills, CA: Sage.

 

Greenberger, Ellen and L.D. Steinberg. 1981. “The Workplace as a Context of
Socialization for Youth.” Journal of Youth and Adolescence, 10:185-210.

and L.D. Steinberg. 1986. When Teenagers Work: The Psychological and Social
Costs of Adolescent Employment. NY: Basic Books.

172

Hauser, ST. and HA. Levine. 1994. “Relatedness and Autonomy in Adolescence: Links
with Ego Development and Family Interactions.” The American Society for Adolescent
Psychiatry: The Relational Context of the Ego: University of Chicago Press.

Henderson, M. and M. Argyle. 1985. “Social Support by Four Categories of Work
Colleagues: Relationships between Activities, Stress and Satisfaction” Journal of
Occupational Behavior 62229-239.

Henderson, 8., D. Byme and P. Duncan-J ones. 1981. Neurosis and the Social
Environment. NY: Academic Press.

Hochschild, Arlie R. 1983. The Managed Heart: Commercialization of Human Feeling.
Berkeley: University of California Press.

Houben, G.J. 1991. “Production Control and Chronic Stress in Work Organizations.” Int.
Journal of Health Services 21(2):309-327.

House, James S., V. Strecher, H.L. Metzner and CA. Robbins. 1986. "Occupational
Stress and Health among Men and Women in the Tecumseh Community Health Study."
Journal of Health and Social Behavior, 27:62-77.

, L.D. Steinberg, and M. Ruggiero. 1982. “A Job is a Job...Or is it?” Work and
Occupations, 9:79-96.

, L.D. Steinberg, and A. Vaux. 1981. “Adolescents Who Work: Health and
Behavioral Consequences.” Developmental Psychology, 17:691-703.

 

House, James S. 1980. Occupational Stress and the Mental and Physical Health of
Factory Workers. Ann Arbor, MI: Institute for Social Research Report Series.

Ireson, C. and S. Gill. 1988. “Girls’ Socialization for Work.” Pp. 132-148 in A.H.
Stromberg and S. Harkess (Eds). Women Working, 2nd Edition. Mountain View, CA:
Mayﬁeld.

Johansson, G. “Psychoneuroendocrine Correlates of Unpaced and Paced Performance.”
Pp. 277-286 G. Salvendy and M.J. Smith (Eds) Machine Pacing and Occupational
Stress. London: Taylor and Francis.

Josselson, Ruthellen.. 1980. “Ego Development in Adolescence” Pp. 188-210 in J.
Adelson (Ed.) Handbook of Adolescent Psychology. NY: Wiley-Interscience Publication.

Kaplan, Stuart L., G.K. Hong and C. Winhold. 1984. “Epidemiology of Depressive

Symptomatology in Adolescents” Joumal of the American Academy of Psychiatry
23(1):91-98.

173

Karasek, Robert A. And T. Theorell. 1990. Healthy Work: Stress, Productivity, and the
Reconstruction of Working Life. NY: Basic Books.

Kashani, J.H., G.A. Carlson, N .C. Beck, E.W. Hooper, C.M. Corcaran, J .A. McAllister,
C. Fallati, T.K. Rosenberg and J .C. Reid. 1987. “Depression, Depressive Symptoms and
Depressed Mood Among a Community Sample of Adolescents.” American Journal of
Psychiatry 144:932-4.

Kaufmann, GM. and T.A. Beehr. 1986. “Interactions Between Job Stressors and Social
Support: Some Counterintuitive Results.” Journal of Applied Psychology 71:522-526.

Keithly, Diane C. and RA. Deseran. “Households, Local Labor Markets, and Youth
Labor Force Participation.” Youth and Society 26(4):463-492.

Kessler, Ronald, R.H. Price and CB. Wortman. 1985. “Social Factors in
Psychopathology.” Annual Review of Psychology, 36:531-72.

Kessler, Ronald C., K.A. McGonagle, S. Zhao, C.B. Nelson, M. Hughes, S. Eshleman, H.
Wittchen, K.S. Kendler. 1994. “Lifetime and 12-Month Prevalence of DSM-III-R
Psychiatric Disorders in the United States: Results From the National Comorbidity
Study.” Archives of General Psychiatry, 5128-19.

Kessler, Ronald C. and J. McRae, Jr. 1981. “Trends in the Relationship Between Sex and
Psychological Distress: 1957-1976.” American Sociological Review, 46:443-52.

Kessler, Ronald C., J .B. Turner and J .8. House. 1988. “Effects of Unemployment on
Health in a Community Survey: Main, Modifying and Mediating Effects” Journal of
Social Issues 44(4):69-85.

Kohn, Melvin L. 1981. “Personality, Occupation and Social Stratiﬁcation: A Frame of
Reference” Pp. 267-297 in D]. Treiman and RV. Robinson (eds) Research in Social
Stratiﬁcation and Mobility, Vol. 1 Greenwich, CT: J AI Press.

. 1990. “Unsolved Issues in the Relationship Between Work and Personality.” Pp.
36-68 in The Nature of Work, edited by K. Erikson and SP. Vallas, New Haven, CT:
Yale University Press.

, A. Naoi, C. Schoenbach, C. Schooler, and KM. Slomczynski. 1990. “Position in
the Class Structure and Psychological Functioning in the United States, Japan, and
Poland.” American Journal of Sociology. 4:24-52.

and C. Schooler, in collaboration with J. Miller, K.A. Miller, C. Schoenbach, and

R. Schoenberg. 1983. Work and Personality: An Inquiry into the Impact of Stratiﬁcation.
Norwood, NJ: Ablex.

174

Kutcher, SP. and P. Marton. 1989. “Parameters of Adolescent Depression” Psychiatric
Clinics of North America 12(4):895-918.

Landsbergis, Paul A. 1988. "Occupational Stress Among Health Care Workers: A Test
of the Job Demands Control Model." Journal of Organizational Behavior, 9:217-39

Lazarus, Richard S. 1991. “Psychological Stress in the Workplace” in P.L. Perrewe (Ed.)
Handbook on Job Stress (Special Issue). Journal of Social Behavior and Personality,
6(7):1-13.

Lazarus, Richard S. and S. Folkman. 1984. Stress, Appraisal and Coping. NY: Springer.

Leidner, Robin. 1991. “Serving Hamburgers and Selling Insurance: Gender, Work, and
Identity in Interactive Service Jobs” Gender and Society, 5(2): 154-177.

. 1987. “Transactional Theory and Research on Emotions and Coping.” In L. Laux
and G. Vossel (Eds) Personality in Biographical Stress and Coping Research. European
Journal of Personality, 1:141-169.

Lewin-Epstein, Noah. 1981. Youth Employment During High School. Washington:
National Center for Education Statistics.

Link, Bruce 6., MC. Lennon, and B.P. Dohrenwend. 1993. “Socioeconomic Status and
Depression: The Role of Occupations Involving Direction, Control, and Planning.”
American Journal of Sociology. 98(6): 1351-87.

Mainquist, Sheri and D. Eichom. 1989. “Competence in Work Settings.” Pp. 327-67 in
Adolescence and Work: Inﬂuences of Social Structure, Labor Markets, and Culture,
edited by D.Stem and D. Eichom. Hillsdale, NJ: Lawrence Erlbaum.

Maistro, S.A., L.C. Sobell, A.M. Cooper, M.B. Sobell. 1982. “Comparison of Two
Techniques to Obtain Retrospective Reports of Drinking Behavior from Alcohol
Abusers.” Addictive Behaviors 7:33-38.

Manning, Wendy D. 1990. “Parenting Employed Teenagers.” Youth and Society. 22: 184-
200.

Margolis, Lewis H. And D.C. Farran. 1984. "Unemployment and Children."
International Journal of Mental Health. 132107-24.

Marsh, H.W. 1991. “Employment During High School: Character Building or a
Subversion of Academic Goals?” Sociology of Education, 64: 172-189.

Marx, Karl. [1844] 1964. Selected Writings in Sociology and Social Philosophy.
Translated by T.B. Bottomore. NY: McGraw-Hill.

175

Maslow, Abraham H. 1954. Motivation and Personality. NY: Harpers and Brothers.

McIntosh, Nancy J. 1991. “Identiﬁcation and Investigation of Properties of Social
Support” Journal of Organizational Behavior, 12:201-17.

MacLeod, Jay. 1995. Ain’t No Makin ’ It: Aspirations and Attainment in a Low-Income
Neighborhood. Boulder: Westview Press

McFarlane, M.D., A. Bellissimo and GR. Norman. 1995. “The Role of Family and Peers
in Social Self-Efficacy: Links to Depression in Adolescence” American Journal of
Orthopsychiatry 65(3):402-410.

McGee, R., M. Feehan, S. Williams, F. Partridge, P.A. Silva and J. Kelly. 1990. “DSM
III Disorders in a Large Sample of Adolescents.” Journal of the American Academy of
Child Psychiatry 29:611-19.

Mead, George H. 1934. Mind, Self, and Society. Chicago: University of Chicago Press.

Menaghan, Elizabeth G. and ES. Merves. 1984. “Coping with Occupational Problems:
The Limits of Individual Efforts.” Journal of Health and Social Behavior 25:406-23.

Michigan State University. 1993. Student Employment Ofﬁce of Career Development
and Placement Services. Study on Student Employment.

Miller, Karen A., M.L. Kohn, and C. Schooler. 1986. "Educational Self-Direction and
Personality." American Sociological Review 512372-90.

Mirowsky, John. 1994. “The Advantages of Indexes over Diagnoses in Scientiﬁc
Assessment.” Pp. 261-290 in W.R. Avison and I.H. Gotlib (Eds) Stress and Mental
Health: Contemporary Issues and Prospects for the Future. NY: Plenum Press.

Mirowsky, John and CE. Ross. 1986. “Social Patterns of Distress. " Annual Review of
Sociology. 12:23-45.

. 1989. Social Causes of Psychological Distress. NY: Aldine de Gruyter.

. 1989. “Psychiatric Diagnosis as Reiﬁed Measurement.” Journal of Health and
Social Behavior, 30(March):11-25.

Montemayor, R. 1983. “Parents and Adolescents in Conﬂict: All Families Some of the
Time and Some Families All of the Time.” Journal of Early Adolescence, 3:83-103.

Moore, Helen. 1985. “Job Satisfaction and Women’s Spheres of Work.” Sex Roles.
13:663-78.

176

Mortimer, Jeylan T. 1988. Introduction, Pp. 1-15 in J .T. Mortimer and KM. Borman
(Eds) Work Experience and Psychological Development Through the Life Span. Boulder,
Colorado: Westview Press, Inc.

Mortimer, Jeylan T., and MD. Finch. 1986. “The Effects of Part-Time Work on
Adolescents’ Self-Concept and Achievement.” In K. Borman and J. Reisman (Eds)
Becoming a Worker. Norwood, NJ: Ablex.

Mortimer, Jeylan T. and MD. Finch. 1996. “Work, Family, and Adolescent
Development.” In J.T. Mortimer and MD. Finch (eds), Adolescents, Work, and Family:
An Intergenerational Developmental Analysis. London: Sage Publications.

Mortimer, Jeylan T., M.D. Finch and G. Maruyama. 1988. “Work Experience and Job
Satisfaction: Variation by Age and Gender” Pp. 109-55 in J .T. Mortimer and K. Borman
(Eds) Work Experience and Psychological Development Through the Life Span. Boulder:
Westview Press.

Mortimer, Jeylan T., M.D. Finch, M. Shanahan and S. Ryu. 1992. “Work Experience,
Mental Health and Behavioral Adjustment in Adolescence.” Journal of Research on
Adolescence 2(1):25-57.

Mortimer, Jeylan T., M.D. Finch, S. Ryu, M.J. Shanahan and KT. Call. 1996. “The
Effects of Work Intensity on Adolescent Mental Health, Achievement, and Behavioral
Adjustment: New Evidence from a Prospective Study” Child Development 67: 1243-
1261.

Mortimer, Jeylan T., M.D. Finch, T.J. Owens and M. Shanahan. 1990. “Gender and Work .
in Adolescence” Youth and Society 22(2):201-224.

Mortimer, Jeylan T. and MK. Johnson. 1997. “Adolescent Work and the Transition to
Adulthood” Paper presented at the session ‘Student Employment and Subsequent Life
Outcomes’ at the 1997 Annual Meeting of the Amercian Sociological Association,
Toronto.

Mortimer, Jeylan and J. Lorence. 1979. “Occupational Experience and the Self-Concept:
A Longitudinal Study.” Social Psychology Quarterly 422307-23.

Mortimer, Jeylan T., J. Lorence, and BS. Kumka. 1986. Work, Family, and Personality.
Norwood, NJ: Ablex Publishing Corporation.

Mortimer, Jeylan T. and M.J. Shanahan. 1991. “Adolescent Work Experience and
Relations with Peers.” Paper presented at the annual meeting of the American
Sociological Association, Cincinnati.

Mortimer, Jeylan T. and M.J. Shanahan. 1994. “Adolescent Work Experience and Family
Relationships.” Work and Occupations, 212 369-384.

177

National Center for Education Statistics 1993. “Proﬁle of Undergraduates in US.
Postsecondary Education Institutions: 1989-90.” US. Department of Education, Oﬁ‘ice of

Educational Research and Improvement.

National Commission on Youth. 1980. The Transition of Youth to Adulthood: A Bridge
too Long. Boulder: CO: Westview.

National Institute for Occupational Safety and Health. 1996. National Occupational
Research Agenda (DHHS NIOSH Publication No. 96-115). Cincinnati, OH: US.
Department of Health and Human Services, Public Health Service, Centers for Disease
Control and Prevention., NIOSH.

National Research Council. 1993. Pesticides in the Diets of Infants and Children.
Washington DC: National Academy Press.

Naoi, Michiko and C. Schooler. 1990. “Psychological Consequences of Occupational
Conditions among Japanese Wives.” Social Psychology Quarterly. 53(2): 100-1 16.

Neff, James A. 1984. “Race Differences in Psychological Distress: The Effects of SES,
Urbanicity, and Measurement Strategy.” American Journal of Community Psychology,
122337-51.

Negrey, Cynthia. 1990. “Contingent Work and the Rhetoric of Autonomy.” Humanity
and Society, 14(1): 16-33.

Noble, David. 1984. Forces of Production: A Social History of Industrial Automation.
NY: Knopf.

Ortman, Patricia E. 1988. “Adolescents’ Perceptions of and Feelings About Control and
Responsibility in Their Lives.” Adolescence. 232913-24.

Osterman, Paul. 1989. Getting Started: The Youth Labor Market. Cambridge: MIT Press.

Pearlin, Leonard I. 1983. "Role Strains and Personal Stress." Pp. 1-32 in Psychosocial
Stress: Trends in Theory and Research, edited by HE. Kaplan. NY: Academic Press.

. 1991. “The Study of COping: An Overview of Problems and Directions” Pp. 261-
76 in J. Eckenrode (Ed) The Social Context of Coping NY: Plenum.

Pearlin, Leonard I., M.A. Lieberman, E.G. Menaghan and J .T. Mullan. 1981. “The Stress
Process.” Journal of Health and Social Behavior, 222337-56.

Phillips, Sarah and K. Sandstrom. 1990. “Parental Attitudes Toward Youth Work.” Youth
and Society, 22(2): 160-183.

178

President’s Science Advisory Committee, Panel on Youth. 1972. Youth: Transition to
Adulthood. Chicago: University of Chicago Press.

Pugliesi, Karen. 1988. “Employment Characteristics, Social Support, and the Well-Being
of Women.” Women and Health, 14235-58.

. 1995. “Work and Well-Being: Gender Differences in the Psychological
Consequences of Employment.” Journal of Health and Social Behavior. 36257-71.

Reiter, Ester. 1996. Making Fast Food. Montreal & Kingston: McGill-Queen’s
University Press.

Regier, Darrel A, J .E. Helzer, M.M.Weissman, H. Orvaschel, E. Gruenberg, J .D. Burke,
and DA. Regier. 1984. “Lifetime Prevalence of Speciﬁc Psychiatric Disorders in Three
Sites.” Archives of General Psychiatry, 41:949-958.

Reiter, Ester. 1996. Making Fast Food. Montreal and Kingston: McGill-Queen’s
University Press.

Riessman, Catherine K. 1990. Divorce Talk: Women and Men Make Sense of Personal
Relationships. New Brunswick, NJ 2 Rutgers University. Press.

Ritter, Christian. 1988. “Social Supports, Social Networks, and Health Behaviors.” Pp.
149-61 in D.S. Gochman (Ed.) Handbook of Health Behavior Research. NY: Plenum

Publishing Corporation.

Roberts, R.E., J .A. Andrews, P.M. Lewinsohn and H. Hops. 1990. “Assessment of
Depression in Adolescents Using the Center for Epidemiologic Studies Depression
Scale.” Journal of Consult. Clin. Psychology 22122-128.

Roberts, R.E., P.M Lewinsohn and J .R. Seeley. 1991. “Screening for Adolescent
Depression: A Comparison of the CES-D and the BDI” Am. Acad. Child Adolesc.
Psychiat. 30:58-66.

Robins, L.N., J .E. Helzer, J. Croughan, and K. Ratcliff. 1981. “National Institute of
Mental Health Diagnostic Interview Schedule: Its History, Characteristics, and Validity.
Archives of General Psychiatry, 382381-89.

9’

Robins, L.N. and DA. Regier (eds) 1991. Psychiatric Disorders in America: The
Epidemiologic Catchment Area Study. New York, NY: Free Press.

Rook, Karen. 1992. “Detrimental Aspects of Social Relationships: Taking Stock of an

Emerging Literature.” Pp. 157-69 in H.O.F. Veiel and U. Baumann (Eds) The Meaning
and Measurement of Social Support. NY: Henri sphere.

179

Rosenberg, Morris. 1979. Conceiving the Self. NY: Basic Books.

Rosenberg, Morris. 1981. “The Self Concept: Social Product and Social Force.” In M.
Rosenberg and R. Turner (Eds) Social Psychology: Sociological Perspectives, pp. 593-
624. NY: Basic Books.

Rosenberg, Morris, C. Schoenbach, C. Schooler and F. Rosenberg. 1995. “Global Self-
Esteem and Speciﬁc Self-Esttem: Different Concepts, Different Outcomes” American
Sociological Review, 60:141-156.

Rosenﬁeld, Sarah. 1989. “The Effects of Women’s Employment: Personal Control and
Sex Differences in Mental Health. ” Journal of Health and Social Behavior. 30:77-91.

Ross, M. 1989. “The Relation of Implicit Theories to the Construction of Personal
Histories.” Psychological Review 96:341-357.

Ross, Catherine E. and J. Mirowsky. 1989. “Explaining the Social Patterns of
Depression: Control and Problem Solving---or Support and Talking.” Journal of Health
and Social Behavior. 302206-19.

Ross, Catherine E., J. Mirowsky, and J. Huber. 1983. “Dividing Work, Sharing Work,
and In-Between: Marriage Patterns and Depression.” American Sociological Review,
48:809-823.

Ross, M. and C. McFarland. 1988. “Constructing the Past: Biases in Personal
Memories.” Pp. 299-314 in D. Bar-Tal and A. Kruglanski (Eds) Social Psychology of
Knowledge. Cambridge, England: Cambridge University Press.

Rotter, J. 1966. "Generalized Expectancies for Internal Versus External Control of

Reinforcement." Psychological Monographs. 80: 1-28.
Rutter, Michael, J. Tizard and K. Whitmore. 1981. Education, Health and Behavior.
Huntington, NY: Krieger.

Sarason, Irwin 6., BR. Sarason (Eds) 1985. Social Support: Theory, Research, and
Applications. Boston: Martinus Nijhoff.

Schill, W.J., R. McCartin and K. Meyer. 1985. “Youth Employment: Its Relationship to
Academic and Family Variables.” Journal of Vocational Behavior 26: 155-163.

Schoenbach, V.J., B.H. Kaplan, R.C. Grimson and B.H. Wagner. 1982. “Use of a
Symptom Scale to Study the Prevalence of a Depressive Symptom in Young
Adolescents” American Journal of Epidemiology 116:791-800.

Schooler, C. and A. Naoi. 1988. “The Psychological Effects of Traditional and of
Economically Peripheral Job Settings in Japan.” American Journal of Sociology. 942335-
55.

180

Schulenberg, J. and J .G. Bachman. 1993. Long Hours on the Job? Not so Bad for Some
Adolescents in Some Types of Jobs: The Quality of Work and Substance Use, Affect and
Stress. Paper presented at the annual meeting of the Society for Research on Child
Development, New Orleans.

Schwalbe, Michael L. 1985. “Autonomy in Work and Self-Esteem.” The Sociological
Quarterly 262519-35.

Seeman, Melvin. 1967. "On the Personal Consequences of Alienation in Work. "
American Sociological Review. 322273-85.

Seligman, Martin. ER 1975. Helplessness: On Depression, Development, and Death.
San Francisco: Freeman

Selman, Robert. 1980. The Growth of Interpersonal Understanding: Developmental and
Clinical Analyses. NY: Academic Press.

Selye, Hans. 1976. The Stress of Life. NY: McGraw Hill.

Shanahan, Michael and J .T. Mortimer. 1996. “Understanding the Positive Consequences
of Psychosocial Stressors.” Pp. 189-209 in B. Markovsky, M. Lovaglia and R. Simon
(Eds) Advances in Group Processes (Vol. 13). Greenwich, CT 2 J AI Press Inc.

Shapiro, D. and J .E. Crowley. 1982. “Aspirations and Expectations of Youth in the
United States: Part 2. Employment Activity” Youth and Society 14:33-58.

Silver, Roxane L., C. Boon and M.H. Stone. 1983. “Searching for Meaning in
Misfortunate: Making Sense of Incest.” Journal of Social Issues 39(2):81-102.

Simon, Robin W. 1995. “Gender, Multiple Roles, Role Meaning, and Mental Health”
Journal of Health and Social Behavior 362182-94.

Smith, Adam. 1937. An Inquiry into the Nature and Causes of Wealth of Nations. 1776.
Reprint. NY: Modern Library.

Spenner, Kenneth I. and LB. Otto. 1985. “Work and Self-Concept: Selection and
Socialization in the Early Career.” Pp. 197-235 in Research in Sociology of Education
and Socialization, vol. 5, edited by AC. Kerckhoff. Greenwich, CT 2 J AI Press.

Steel, Lauri. 1991. “Early Work Experience Among White and Non-White Youth. ”
Youth and Society 22:419-447.

181

Steinberg, L. 1990. “Autonomy, Conﬂict and Harmony in the Family Relationship. Pp.
255-276 in SS. Feldman and GR. Elliott (Eds), At the Threshold: The Developing
Adolescent. Cambridge, MA: Harvard University Press.

Steinberg, L. 1999. Interviewed on “Talk of the Nation” program on National Public
Radio. January 4, 1999.

Steinberg, Laurence and SM. Dombusch. 1991. “Negative Correlates of Part-Time
Employment During Adolescence: Replication and Elaboration.” Developmental
Psychology. 27(2)2304-313.

Steinberg, Laurence, S. Fegley and S. Dombusch. 1993. “Negative Impact of Part-Time
Work on Adolescent Adjustment: Evidence from a Longitudinal Study” Developmental
Psychology 292171-180.

Stets, Jan E. 1992. “Interactive Processes in Dating Aggression: A National Study.”
Journal of Marriage and the Family 542165-77.

Stevens, William M. 1979. Our Children Should be Working. Springﬁeld, IL: Charles C
Thomas.

Super, DE. 1990. “A Life-span, Life Space Approach to Career Development.” Pp. 197-
261 in D. Brown, L. Brooks and Associates (Eds) Career Choice and Development. San
Francisco, CA: Jossey-Bass.

Sutherland, V. and M.J. Davidson. 1993. “Using a Stress Audit: The Construction Site
Manager’s Experience in the UK” Work Stress 7(3)2273-286.

Taylor, SE. and J. Crocker. 1981. “Schematic Bases of Information Processing.” In B.T.
Higgins, C.P. Herman and MP. Zanna (Eds) Social Cognition: The Ontario Symposium
(Vol. 1, pp. 89-134). Hillsdale, NJ 2 Erlbaum.

Thoits, Peggy A. 1982. “Conceptual, Methodological and Theoretical Problems in
Studying Social Support as a Buffer Against Life Stress” Joumal of Health and Social
Behavior 23: 145-149.

Thoits, Peggy A. 1992. “Identity Structures and Psychological Well Being: Gender and
Marital Status Comparisons.” Social Psychology Quarterly 55(3)2236-56.

. 1994a. “Identity-relevant Events and Psychological Symptoms: A Cautionary
Tale.” Journal of Health and Social Behavior 36:72-82.

.1994b. “Stressors and Problem-Solving: The Individual as Psychological
Activist.” Journal of Health and Social Behavior 35: 143-59.

 

182

. 1995. “Stress, Coping, and Social Support Processes: Where Are We? What
Next?” Journal of Health and Social Behavior (Extra Issue):53-79.

Turner, R. Jay. 1983. “Direct, Indirect and Moderating Effects of Social Support Upon
Psychological Distress and Associated Conditions.” Pp. 105-55 in H. Kaplan (ed.)
Psychosocial Stress: Trends in Theory and Research, NY: Academic Press.

Turner, R. Jay and W.R. Avison. 1992. “Innovations in the Measurement of Life Stress:
Crisis Theory and the Signiﬁcance of Event Resolution.” Journal of Health and Social
Behavior. 33236-50.

US. Department of Labor. 1987. Employment and Eaming 34(10). Washington: US.
Government Printing Ofﬁce.

US. Department of Labor, Employment and Training Administration. 1986. Dictionary
of Occupational Titles (4‘h ed.). Washington DC: US. Government Printing Ofﬁce.

Van Maanen, John and E. Schein. 1979. “Toward a Theory of Organizational
Socialization.” Research in Organizational Behavior 12209-64.

Vaux, Alan. 1988. Social Support: Theory, Research and Intervention. NY: Praeger.

Veiel, Hans CF. and U. Baumann. 1992. “The Many Meanings of Social Support.” Pp.
1-7 in H.O.F. Veiel and U. Bauman (eds) The Meaning and Measurement of Social
Support. NY: Hemisphere Publishing.

Vulcan, Beatrice. 1968. “American Social Policy Toward Youth and Youth
Unemployment.” In M. Herman, S. Sadofsky, and B. Rosenberg (Eds), Work, Youth, and
Unemployment. NY: Thomas Y. Crowell.

Walker, R. 1990. “Three-Day ‘Child Watch’ Sweep Finds 7,000 Minors Allegedly
Working Illegally.” Education Week, p.5.

Warr, Peter. 1987. Work, Unemployment, and Mental Health. Oxford: Clarendon Press.

Wells, V.E., G.L. Klerman and E.Y. Deykin. 1987. “The Prevalence of Depressive
Symptoms in College Students.” Social Psychiatry 22:20-28.

Wheaton, Blair. 1980. “The Sociogenesis of Psychological Disorder: An Attributional
Theory.” Journal of Health and Social Behavior 21:100-124.

. 1994. “The Domains and Boundaries of Stress Concepts” Paper presented at the
1994 Meetings of the American Sociological Association, Los Angeles, CA.

Whitaker, A., J. Johnson, D. Shaffer, J .L. Rapaport, K. Kalikow, B.T. Walsh, M. Davies,
S. Braiman and A. Dolinsky. 1990. “Uncommon

183

Troubles in Young People” Archives of General Psychiatry 472487-96.

Wilson, William J. 1997. When Work Disappears: The World of the New Urban Poor.
NY: Alfred A. Knopf, Inc.

Wrigley, J. 1986. “Compulsory School Laws: A Dilemma with a History. In J. Simon
and D. Stipek (Eds), Reconsidering Compulsory Schooling for Adolescents: Studies in
Social Science, Education, and Law. NY: Academic Press.

World Health Organization. 1991. “Mental Health and Behavioral Disorders (including
Disorders of Psychological Development). In: International Classiﬁcation of Diseases—
10’" Edition. Geneva, Switzerland: World Health Organization; Chap. 5. Diagnostic
Criteria for Research, Draft for Field Trials.

Yamoor, Catherine and J.T. Mortimer. 1990. “Age and Gender Differences in the Effects

of Employment on Adolescent Achievement and Well-Being” Youth and Society 22 (2)
Dec.2 225-240.

184

"Illllllllllllllllllllli