Sociology Mind
2011. Vol.1, No.3, 114-120
Copyright © 2011 SciRes. DOI:10.4236/sm.2011.13014
Deindustrialization, Class, and Adolescents: Changing Gender
Attitudes in Middletown*
Lisa Winters1, Wanda Rushing2, Martin Levin2, Troy Blanchard1
1Louisiana State University, Baton Rouge, USA;
2University of Memphis, Memphis, USA.
Received April 28th, 2011; revised May 30th, 2011; accepted April 28th, 2011.
Scholars have become increasingly interested in the role that economic change plays in the processes of self and
collective identification. Previous studies show that the process of deindustrialization in the United States had
specific consequences for individuals with a “working class” labor force identity, particularly in regard to in-
creased financial stability. Using the High School Surveys from Middletown III and IV, collected from local
high school students in Muncie, Indiana in 1977, 1989, and 1999, we examine how local adolescents’ expecta-
tions regarding school and gender were affected by deindustrialization. In this study, we put forth the following
hypotheses: 1) the educational aspirations of adolescents will increase over time, 2) attitudes about gender roles
will become less traditional over time, and 3) students will show a greater recognition of the utility of college
education over time. Finding support for all hypotheses indicates that adolescent attitudes about education and
gender are significantly affected by deindustrialization i n t h e h ypothesized direction.
Keywords: Adolescents, Gender, Education, Deindustrialization
Introduction
Scholars have become increasingly interested in the role of
structural change, specifically economic change, in processes of
self- and collective-identification. One of the most far reaching
economic changes in the U.S during the post-World War II era
was the dramatic deindustrialization that occurred in the early
1970 s and1980 s. The shift of employ ment from higher income
jobs in heavy industrial production to lower-income service oc-
cupations had notable social consequences (Silvestri & Lukasi-
wicz, 1989). This historic transformation created a feminization
of the labor market and challenged preconceived notions about
what it meant to be a member of the industrial labor force. Ac-
cording to Leslie McCall (2001), this type of economic restruc-
turing “lowered the wages and security of most workers even as
it created unprecedented concentrations of wealth” (p. 14).
Although there has been much scholarly inquiry into the con-
sequences of economic change on various communities and
individual industry workers, very few studies have explored
how widespread economic transitions affect adolescent atti-
tudes or high school culture, and even fewer have focused spe-
cifically on the effects of deindustrialization, Why is it impor-
tant to study the identity-formation process of working class
adolescents in the context of deindustrialization? The answer to
this question lies in the unique position of adolescents in rela-
tion to the economy during the 1980s. This generation’s parents
grew up during a time when the United States was experiencing
unprecedented economic growth. Job opportunities soared in
major steel, automotive, and agricultural production industries,
providing working class families with financial security and
prospects of economic mobility. As these students reached high
school, however, and were preparing to enter the work force,
they lived in a time of growing economic instability. The
well-paying factory jobs traditionally found in industry-based
communities-the jobs that many of their fathers had worked in
since they graduated from high school-were quickly disappear-
ing. Young people began to see firsthand the consequences of
deindustrialization on their communities and families, and in
this way, working-class adolescents in the 1980s became wit-
ness to, or “victims” of, one of the most far reaching economic
shifts in the history of the United States (Weis, 1990).
The purpose of this investigation is to assess how adolescents
negotiate their futures in terms of educational goals and gen-
dered roles in a community that was experiencing marked
deindustrialization. In this analysis, we examine data collected
from high school students in the city of Muncie, a small com-
munity in east-central Indiana that experienced marked dein-
dustrialization during the 1970s and 1980s. Muncie is charac-
terized as having a strong manufacturing history, but in recent
decades has altered its economic base in the face of job loss in
the industrial sector. Like so many other manufacturing cities in
the United States, Muncie experienced a plethora of factory
closings during the 1970 s essentially wiping out jobs in the
electronic, steel, glass, and automotive industries that had once
been the backbone of its economy. As the manufacturing in-
dustries were phased out, the economy shifted to being primar-
ily service-oriented, and the community saw an increase in
sales, office work, and other service occupations. Using survey
data collected from Muncie high school students in 1977, 1989,
and 1999, we are able to examine the way local students’ ex-
pectations regarding school, work, and gender roles were al-
tered in response to the community’s ec onomic transformation.
The Impact of Deindustrialization on Class in
the U.S
*Direct correspondence to Lisa Winters, Louisiana State University, Depart-
ment of Sociology. 126 Stubbs Hall, Baton Rouge, LA 70803-0000

L. WINTERS ET AL. 115
One of the most frequently cited studies concerned with the
impact of deindustrialization in the United States was con-
ducted by Bluestone and Harrison (1982). They define this
process as “widespread, systematic disinvestment in the na-
tion’s basic productive capacity” (p. 6), or as an economic shift
that occurred during the early 1970s, when the manufacturing
industries that dominated the U.S market during the 1950s and
1960s began to be replaced with a new, service-based economy
(fast food, retail). In The Deindustrialization of America (1982),
Bluestone and Harrison detail the consequences of this process
on American families and communities. They argue that when
the booming industrial economy that once promised a higher
standard of living for American workers ceased to grow, it sent
local communities spiraling into economic crises. As capital
shifted from region to region within the United States, or left
the country in search of cheaper means of production, many
factories were forced to close their doors, leaving behind “a
newly emerging group of ghost towns” all across the nation’s
industrial regions (Bluestone & Harrison, 1982: p.6). Subse-
quently, this decline manifested itself in terms of diminished
purchasing power among American families, and a local
economy incapable of “providing people with a simple home
mortgage, a stable job, or a secure pension” (Bluestone & Har-
rison, 1982: p. 4).
The financial impact of deindustrialization on communities
and families was devastating; however, its effects reached far
beyond economic restructuring. According to Lois Weis in
Working Class Without Work (1990), the dramatic shift in the
American economy set in motion a series of social and cultural
changes as well, particularly for the working-class, whose iden-
tity, she argues, wa s historically tied to the type of employment
that was disappearing. In 1985, Weis interviewed forty work-
ing-class, male and female high school students in the commu-
nity of Freeway1—a small manufacturing town experiencing an
economic recession, where the effects of deindustrialization
were causing once predominate factory jobs to decline rapidly.
Attempting to examine the impact of global economic restruc-
turing on the social and educational attitudes of high school
students, Weis probed the students with questions about their
plans post-high school, and their rationale for making those
decisions. She found that many of these working-class students
were struggling to envision their future (Weis, 1990: p. 24).
Not unlike the students Weis interviewed, adolescents in the
United States who came of age in the 1970 s and 1980 s lived
in a world of transition characterized by economic instability.
Adolescents, observing the effects of this instability on their
own families, became increasingly aware of a sluggish domes-
tic economy as they noted job shortages, low wages, and elu-
sive retirement pensions associated with factory closings in
local communities. Weis points out that the high school stu-
dents she interviewed would live out their entire working lives
against the backdrop of transition into a post-industrial society
that would ultimately challenge their working class identities.
She argues that these students held a much different economic
position than their parents and grandparents had at their age,
and that “while they [brought] with them a collective identity of
the ‘old’ working class, they must forge a new identity as the
old industrial order eroded” (Weis, 1990: p. 12).
Identity formation refers to the relational processes of people
as they attempt, either individually or collectively, to position
themselves within a social location by creating and maintaining
spatial distance (Lamont & Molnar: 2002). According to Weis,
working-class identity was constructed around belief in a rigid
sexual division of labor, more specifically, the ability of men to
earn a family wage while women focused on marri age and moth-
erhood. When the U.S industrialized during the 1950s and 60s,
working class men’s identity as breadwinner became increasingly
related to their ability to provide financially for the family and
their willingness to work long hours at the factory (Collins, 1992).
On the other hand, working class women’s identity as home-
maker was tied to service within the domestic sphere, including
cooking and cleaning for the family, and providing physical,
psychological, and emotional care for the husband and children
(Collins, 1992). Belief in a gendered division of labor became so
engrained within this emerging economic system that if the wife
did work outside the household, many husbands felt the need to
justify their wife’s employment, suggesting that it was “some-
thing she does for her own enjoyment or for ‘extras’ they could
otherwise live without” (Epstein, 1981).
The “separate spheres” mentality of the working class would
be challenged, however, when the productive work that had
offered working men a “cloak of respectability” during the
1950s and 60s began to disappear (Weis, 1990: p. 192). Ac-
cording to Weis, “the jobs that served to order the lives, identi-
ties, and political struggles of the working-class no longer ex-
ist[ed] to the extent that they [once had]” (p. 194), and were
systematically phased out and replaced by various occupations
in the service sector. The post-industrial economy challenged
gendered notions of the traditional working class identity by
changing the type and demographic composition of post-in-
dustrial labor. This progression, referred to as a “feminization
of the labor market” by Linda McDowell (1991), ultimately led
to new gendered interpretation of “women” and “men’s” work,
as well as a broader restructuring of the traditional family form
(Adkins, 2001). For example, jobs in the service sector often
required the worker to interact with people and information,
rather than perform “dirty” manual labor traditionally associ-
ated with working-class masculinity. In addition, many service
jobs did not provide an income sufficient to support a family,
shifting the structure of many working class families to
two-income households where women also worked, and forcing
both men and women to re-conceptualize an identity based on a
breadwinner/homemaker mentality.
Exploring these issues of class and gender in a post-industrial
context, Weis navigated the collective consciousness of the
“sons and daughters of the white industrial proletariat”, noting
that these students were, in fact, very aware of the process of
deindustrialization, how it affected their families, and how it
could possibly affect their own futures. She argues that students
at “Freeway High” were seeing the utility of a high school di-
ploma and the value of a college degree more than ever before,
expressing a desire to continue their education rather than be-
come employed in the manual-labor sector where many of their
fathers had worked since high school. In addition, Weis ex-
plores the identity formation of males and females in relation to
this movement and examines how their views of “appropriate”
gender roles shifted with the changing economic structure. She
discovered that, although many male students still envisioned
1To preserve anonymity and protect the identities of the individuals she
interviewed , Weis cr eated th e hypoth etical na me “Freeway ” for th e commu-
nity wher e her invest igation took place; ho wever, sh e notes that i t is located
in the Northeast United States (Weis, 1990).

L. WINTERS ET AL.
116
themselves living in patriarchical households, female students
were challenging this traditional arrangement by envisioning
themselves as financially independent before getting married or
rearing children. Noting that ideas about work, school, and
family are all tied to the construction of gender, Weis concludes
that deindustrialization shaped and molded the gendered identi-
ties of these working class adolescents.
In addition to Weis’ work, the Middletown Studies, arguably
one of the best known community studies, and a classic within
sociology, examined the effects of economic change for indi-
vidual and community well-being. Beginning in 1924, the ini-
tial investigators, Robert Staughton Lynd, his wife Helen
Merrell Lynd, and a team of researchers, conducted an ethno-
graphic study of an entire town to understand the American
urban experience of a community “experiencing rapid social
change” (Caplow et al., 2004). The Lynds’ success with their
first major publication concerning this community, Middletown:
A Study in Modern American Culture (1929), prompted a fol-
low up study, with the Lynds returning to Muncie in 1935 and
subsequently publishing a second work entitled Middletown in
Transition: A Study in Cultural Conflicts (1937). These two
studies document behavioral and attitudinal changes that occur
within communities specifically within the context of industrial
development and modernization. Middletown, the comprehen-
sive, longitudinal field study, still fascinates scholars today,
more than eighty years later.
Fifty years after the Lynds’ initial research, Theodore Cap-
low, Howard Bahr, and Bruce Chadwick conducted the first
major replication of the Lynds’ Middletown research. Funded
by the National Science Foundation, the Middletown III studies
began in 1975. Caplow and his team, which included 20-plus
doctoral and post-doctoral students, housewives, and various
interviewers (Hoover, 1990), used statistical analysis to take the
pulse of the community with a total of nine surveys: the Family
Roles Survey (1977), the High School Survey (1977, 1989), the
Community Survey (1978), the Government Services Survey
(1978), the Kinship Survey (1978), the Neighborhood Survey
(1978), the Religion Survey (1978), the Women’s Occupational
Survey (1978), and the Recreation Survey (1982). Twenty years
later, Middletown IV replicated the two most important surveys
conducted by the Caplow team, the Community Survey and the
High School Survey.
Whereas the Lynds’ earlier studies of Muncie allowed them
to take a “snapshot” of the community as it experienced indus-
trial growth, Caplow et al. (2004) documents the changes that
occurred with Muncie’s industrial decline. This comprehensive
documentation of the cultural practices and social attitudes of
the citizens of Muncie during these economic transitions pro-
vides a rich data source for examining the impact of these proc-
esses on individuals. Moreover, the High School Surveys tap
into the social attitudes of students growing up in the midst of
this economic transition, allowing for an investigation of how
deindustrialization specifically affects adolescents.
Caplow and his team have reported changes in Muncie’s ado-
lescent attitudes over time on a range of issues; however, these
reports have not exhausted the survey data. For example, Cap-
low’s most comprehensive work using the Middletown III and IV
High School Survey’s—“The Middletown Replications: 75
Years of Change in Adolescent Attitudes, 1924-1999”—dis-
cusses trends in high school student views on issues of religion,
patriotism, social justice, civil rights, information source credibil-
ity, and international relations. However, his analysis does not
include how students’ educational goals, occupational aspiratio ns,
or other aspects of high school culture may have evolved over
time. Given the changing economi c context of Mu ncie during t he
time frame that the Middletown III and IV data were collected
(1977-1999), and the way processes of deindustrialization have
been shown to affect adolescent’s attitudes about school, work,
and identity (Weis, 1990),we believe an analysis that takes ad-
vantage of survey data on these topics is long overdue.
Documenting Deindustrialization in Muncie:
Muncie is an ideal site for investigation because of its well
documented history as a manufacturing community, and its
subsequent shift into a post-industrial society where the econ-
omy is based on service, rather than production. A review of
the city’s history reveals that, after the discovery of natural gas
in 1886 just 12 miles north of Muncie, the population of this
small community grew exponentially as it transitioned from an
almost exclusively agricultural area into an aggressive indus-
trial city. Lured by the promise of an endless gas supply, free
land, railroad facilities, and cheap wage labor from surrounding
farms, factories from further east, such as Ball Brothers Glass
Manufacturing Company, relocated to Muncie and employed
thousands of workers (Lynd, 1929). Its population ballooned
from 11, 345 in 1890 to 73,320 in 1988, an increase of 546.3%,
making Muncie the second fastest growing city in the state next
to Indianapolis (Morton, 1990).
Indeed, the Industrial Revolution had descended on this
small community in East Central Indiana, and for the first half
of the 20th century, Muncie was the center for glass and steel
manufacturing in the state of Indiana. Despite its early success
as an industrial city, however, by the latter part of the 20th cen-
tury, its manufacturing base had begun to deteriorate, and by
the late 1960s, the community’s economy began to undergo a
major transition. Lavish practices with the city’s seemingly
“endless” supply of its most valuable resource-natural gas-led
to its disappearance, forcing many factories to shut down, re-
duce their workforce, or relocate (Caccamo, 1992). Heavy in-
dustry began to be phased out, starting in 1962, when the Ball
Glass plant closed its doors. Other major factories followed suit
over the next two decades. Most of Muncie’s largest employers,
including Indiana Steel and Wire, General Motors, and Delco-
Remy, disappeared, signaling that the community’s best days of
industrialization were over (Caccamo, 1992).
By 1999, the city of Muncie was no longer a manufacturing
city. Its population had declined to less than 68,000 and it be-
came one of many rust-belt cities struggling to reinvent itself
economically in the face of deindustrialization. With the growth
of white-collar and service sector jobs education and healthcare
have become the bedrock of the Muncie economy (Caccamo,
1992). Ball State University and Ball Memorial Hospital/ Car-
dinal Health Systems are the two leading employers in the
county, and no manufacturing operation cracks the top five in
terms of the number of workers employed. The leading manu-
facturing company, Borg Warner Automotive, once Muncie’s
largest employer, closed in April 2009 (Muncie-Delaware
County Economic Development Alliance, 1997; http://www.
bsudailynews.com/features/dn-decade-borgwarner-s-closing-
representative-of-changing-economy-1.2125015).
Examination of census data2 collected on Delaware County
2County Business Patterns [CDROM]. Washington, D.C. U.S Dept o
f
Commerce, Bureau of the Census, Data User Services (1986-2001).

L. WINTERS ET AL. 117
(of which Muncie is the county seat) confirms a number of
trends reflective of Bluestone and Harrison’s (1982) argument
discussed earlier, and demonstrates the pattern of deindustriali-
zation, as it began in the 1960s, accelerated throughout the
1970s, and reached devastating levels by the 1980s. For example,
occupational data gathered for Delaware County reveals a strik-
ing decrease in the percentage of the workforce population em-
ployed in non-service industries (Agriculture, Construction,
Manufacturing) from 46% in 1977, to 32% in 1996. Data also
suggest that, as non-service occupations traditionally coded as
“blue-collar work” declined throughout this period, Delaware
county experienced a simultaneous increase in the percentage of
the workforce employed in service industries (Transportation,
Communications, and Utilities, Wholesale and Retail Trade,
Finance, Insurance, and Real Estate, and Services), rising from
54% in 1977, to 69% in 1997. Increased employment in service
industries and the relative decline in employment in non-service
industries demonstrate a shift toward a post-industrial society, an
era that is “characterized by struggle over the symbolic realm of
information and the production of culture” (Weis, 2004: p. 9).
The changing occupational structure of Muncie in terms of
gender also must be considered, as an increase in female em-
ployment is an expected trend in communities undergoing rapid
deindustrialization. Census data3 document this trend in the
Muncie Standard Metropolitan Statistical Area (SMSA), later
referred to in the census as the Metropolitan Statistical Area
(MSA). The percent of females in the workforce increased from
34% to 60% from 1960 to 2000, suggesting that the idea of
women working outside of the domestic sphere not only was
becoming more socially acceptable, but also was becoming an
economic necessity under the community’s restructured econ-
omy. As the employment base of the working class eroded
throughout this period, male and female workers, either entering
the workforce for the first time or displaced from closing facto-
ries, were encouraged to look toward the service sector for em-
ployment (McKibben, 1987). These jobs, despite growing at a
rapid rate, did not pay as well as the old manufacturing jobs and
had a substantially lower potential for salary increase (McKib-
ben, 1987). McKibben (1987) argues that the “downward mobil-
ity of a significant sector of workers cause[ed] a polarization of
the work force” (p. 6), making it almost impossible for working
class families to maintain their standard of living with only one
wage earner. Consequentially, it became necess ary for wom en to
enter the workforce and earn a supplementary income.
Based on this previous research that examines trends in so-
cial attitudes of adolescents in a changing economic context
(Bettis, 1996; Davis & Pearce, 2007; Weis, 1990), we test the
following hypotheses:
1) The educational aspirations of adolescents in Muncie will
increase over time, an effect which will be more pronounced
for females.
2) Student attitudes regarding appropriate gender roles for
men and women will become less traditional over time, an ef-
fect which will be more pronounced for females.
3) Students will show a greater recognition of the importance
of having a college education over time, an effect which will be
more pronounced for females.
Data and Methods
This research is a secondary data analysis of the Middletown
III and IV High School Surveys. The data were made available
through the Inter-University Consortium for Political and So-
cial Research (ICPSR) at the University of Michigan
(http://www.ICPSR.umich.edu)4. Because it may be possible to
identify respondents in such a detailed community-level study,
access to this particular data set is restricted by ICPSR and
permission to use it was granted only after documenting local
Institutional Review Board (IRB) approval of the proposed
project and successfully completing an Application for Re-
stricted ICPSR Data Files and an acceptable Data Protection
Plan (see Appendix). These outline the research project and
specifically describe the computing environment where the data
will be analyzed, and how the data will be stored, encrypted,
transmitted, and eventually destroyed after the completion of
the research.
The original questionnaires, each consisting of approxi-
mately 60 items on school life, occupational aspirations, family
background, and student attitudes on various personal, political,
and social issues, were administered to cross-sections of the
student population of Muncie’s public high schools during
1977, 1987, 1999. The surveys were approximately eight pages
in length and consisted of questions either framed on a 4 point
Likert scale ranging from “strongly agree” to “strongly dis-
agree”, or with pre-determined response categories, with no
formal “uncertain” option was offered as an alternative. For the
1977 and 1987 studies, school officials and parents were pas-
sively notified by mail about what the study entailed. By 1999,
however, regulations about parental consent had become more
stringent and the surveys could not be administered until writ-
ten permission for each participant was granted and on file at
the school. These new regulations resulted in a decrease in the
response rate for the 1999 survey year. All surveys were ad-
ministered by the students’ homeroom teachers and were de-
signed to be completed within a 50 minute time period. The
response rates for the 1977, 1987, and 1999 questionnaires
were 81%, 82%, and 71%, respectively (Caplow et al., 2004).
The potential study sample is made up from the entire stu-
dent population of Muncie’s public high schools in during 1977,
1989, and 1999 (N = 7443). In keeping with Weis’ decision to
interview only high school juniors because they represent the
class of students most likely to be concerned with the decision
of whether or not to attend college (Weis, 1990), the data ana-
lyzed in this study include only the responses of upperclass-
men.5 Based on respondent’s self-reported class, only juniors
4Caplow, Bahr, Chadwick, Call, & Hicks (2007). COMPILATION OF
MIDDLETOWN III AND MIDDLETOWN IV DATA, 1977-1999 [MUN-
CIE, INDIANA] [Computer File]. ICPSR04604-v2. Provo, Utah: Brigham
Young University [producer]. Ann Arbor, MI: Inter-university Consortium
for Political and Social Researc h [d ist ributor], 2007-10-16 .
5The decision to exclude high school freshmen and sophomores from the
analysis is further sup
p
orted by an analysis of the data by class in school,
which revealed that there was a significant difference in the responses o
f
j
uniors and seniors from other students in the sample. This finding is in
line with Weis’ (1990) theory that high school juniors and seniors are at a
key point of decision about whether to continue their education after high
school. Weis argues that, because college entrance exams (such as the
PSATs and SATs) are administered during junior year, these students would
have already begun to formulate their plans after high school, while fresh-
men and sophomore students may not yet feel pressed to make decisions
about attendin g college.
3Census of Population: Social and Economic Characteristics, Indiana. Wash-
ington, D.C., U.S Dept of Commerce, Economics and Statistics Administra-
tion. Bureau of the Census.

L. WINTERS ET AL.
118
and seniors within each decade are included in the analysis.
This results in the elimination of 4,098 lowerclassmen, bringing
the total number of eligible respondents to 3,345.
Dependent Variables. There are six dependent variables in
this analysis; two measuring the student’s educational aspira-
tions, two measuring adolescent attitudes about appropriate
gender roles for men and women, and two that we interpret
both as a measure of the student’s attitudes about the impor-
tance of a college education. We measure the student’s Educa-
tional Aspirations with two items. The first item is a response
to the question “How far do you want to go in school?” and was
coded into two categories: “High school or less” and “some
college or college.” The second item is a response to the ques-
tion “What do you plan to do after high school?” and responses
were coded into two categories: “Education beyond high
school” and “No education beyond high school.”
We also measure student attitudes about gender wit h two items.
The first item is a response to the question “What do you feel is
the ideal role for a woman today-wife, mother, career woman, or
some combination of these?” Responses were coded into two
categories: “Includes Career” and “Does not include career”. The
second item is a response to the question “Agree or Disagree: A
woman should give up her job or profession after she marries”
and was coded into two categori es “Strong Agree or Agree” and
“Strongly Disagree and Disagree”. We measure student attitudes
about the importance of a college education with two items hav-
ing identical response categories: “How satisfactory do you be-
lieve your father’s education is?” and “How satisfactory do you
believe your mother’s education is?” Responses to these items
were coded as “Satisfied” or “Not Satisfied”.
Independent and Control Variables. The key independent
variables in this analysis are gender and year in which the sur-
vey was taken. Gender was recoded into a dummy variable for
females (coded 1), with males (coded 0) as a reference category.
Survey year was defined by creating two dummy variables. The
first, labeled 1989 was coded 1 for the 1989 survey responses
and 0 for all other survey years. The second, labeled 1999 was
coded 1 for the 1999 survey responses and 0 for all other survey
years. This coding scheme, of course, made 1977 the reference
category. Where appropriate, we include control variables for
both mother and father’s education, measured by the following
items: “How far did your father go in school?” and “How far
did your mother go in school?” To prepare the data for analysis
with binary logistic regression, responses were coded into two
categories: “high school or less” and “some college or college”.
Findings
To test hypotheses concerning educational aspirations, atti-
tudes about gender roles, and recognition of the value of higher
education, we employ binary logistic regression to determine if
there is a significant difference in the probability that students
would respond in this manner within each survey year, control-
ling for gender and parent’s education where appropriate. De-
scriptive statistics for each of the variables in the model appear
in Table 1.
Results from the logistic regressions are reported in Table 2
and 3. The first hypothesis, which predicts that the educational
aspirations of adolescents in Muncie will increase over ti me from
1977 to 1999, is strongly supported (see Table 2A). Controlling
Table 1.
Descriptive statistics for high school juniors and seniors in Muncie,
Indiana by survey year (percent).
Survey Year 1977 1989 1999
Gender
Males 49 53 51
Females 51 47 49
Mother Went to College 31 39 55
Father Went to College 36 39 51
Aspiring to Atte nd College 73 89 94
Planning to Pursue E ducation Beyond HS 57 68 71
Ideal Role for Women Includes Career 70 82 83
Women Should Give Up Jo b After Marriage 10 8 10
Father’s Education Satisfactory 89 82 81
Mother’s Education Satisfactory 91 88 86
Table 2.
Binary logistic regression results for educational aspirations and gender
role attitudes of high school juniors and seniors in Muncie, Indiana.
A1 Aspired
Education
B1
Plans after HS C2 Ideal Role D2 Give up Job
Coeffi-
cient Odds Ratio Coeffi- cient Odds Ratio Coeffi- cient Odds Ratio Coeffi- cient Odds Ratio
19893 1.047***2.8500.412***1.510 .729*** 2.072 .366*1.442
19993 1.429***4.1740.421***1.523 .671*** 1.956 .057 1.059
Gender4 0.265*1. 3030.559***1.748 1.124* ** 3.076 1.497***4.470
Father’s
Educa-
tion1 1.102***3.0110.891***2.437 .243* 1.275 .492**1.636
Mother’s
Edcation11.165***3.2060.765***2.150 .397*** 1.488 .214 1.239
*p < .05. **p < .01. ***p < .001. Note: 1reference category: high school or less;
2reference category: traditional gender attitude; 3reference category: 1977; 4reference
category: males.
Table 3.
Binary logistic regression for satisfaction with parents’ education level
for high school juniors and seniors in Muncie, Indiana whose parents
have a high school education or less.
Father’s Education
Satisfactory1 Mot h er’s Education Satisfactory1
CoefficientOdds
Ratio Coefficient Odds Ratio
19892 –.536*** .585 –.159** .853
19992 –.456*** .634 –.176*** .839
Gender3 –.012 .988 –.141 .868
*p < .05. **p < .01. ***p < .001. Note: 1reference category: not satisfied;
2reference category: 1977; 3reference categor y: males
for gender and parent’s education, the odds that students in
Muncie aspired to go to college increased by 185% in 1989
over what they were in 1977 (p < .001), and increase by over
317% in 1999 compared to 1977 (p < .001). Likewise, the odds
that students planned to attend college increased 51% in 1989
over what they were in 1977 (p < .001), and 52% in 1999 com-
pared to 1977 (p < .001) (see Table 2B). As predicted, these
results were significantly different by gender, with females
being 30% more likely than males to aspire to go to college (p
< .05) and 74% more likely to plan to attend college (p < .001)
within the 22 year period from 1977 to 1999. These findings are
in the expected direction and suggest that there is a significant
difference in the educational aspirations of adolescents over
time, and that this effect is stronger for females.
The second hypothesis, which predicts that attitudes about
gender roles will become less traditional over time from 1977

L. WINTERS ET AL. 119
to 1999, is partially supported. As reported in Table 2C, the
odds that adolescents in Muncie believed that a woman’s ideal
role included career increased by 107% in 1989 over what they
were in 1977 (p < .001) and the odds that they disagreed that a
woman should give up her job or profession after getting mar-
ried increased by 44% in 1989 from what they were in 1977 (p
< .05) (Table 2D). These results are in the expected direction
and suggest that there is a significant difference in attitudes
about gender over time, and that this effect is stronger for fe-
males, with the odds of female students being almost 208%
higher than their male counterparts to agree that the ideal role
of woman includes career, and 347% higher than males to dis-
agree that woman should give up her job within the 22 year
period from 1977 and 1999 (p < .001).
These findings are also consistent with trends found by Weis
(1990) and Davis and Pearce (1992); however, it is important to
note a slight decline from 1989 to 1999 (as compared to 1977)
in the probability that Muncie adolescents agreed that the ideal
role of woman includes career. Although a contingency table
analysis revealed a significant increase between the percentages
of students agreeing with this ideal across each survey year, the
regressions suggest that, from 1989 to 1999, there is a 12%
decrease in the odds that students supported women having a
career from each decade compared to 1977. Along the same
lines, there is also a decrease in the probability that students
believe a woman should give up her job after marriage from
1989 to 1999, dropping from 31% to 6%—a result that is not
significantly different from 1977.
The third hypothesis, which predicts that students will show
greater recognition of the importance of a college education over
time, is also supported. As shown in Table 3, which includes
only responses from students whose mother and father’s educa-
tion was high school or less, the odds of agreeing that their fa-
ther’s education was satisfactory at high school or less were
42% lower in 1989 than in 1977, and 37% lower in 1999 than in
1977. Likewise, the odds of students agreeing that their mother’s
education is satisfactory at high school or less were 15% lower
in 1989 as compared to 1977, and 16% lower in 1999 than in
1977. The findings suggest that, over time, students are not only
aspiring or planning to attend college, but are viewing the pros-
pect of obtaining an education beyond high school as a necessity.
These findings do not indicate that gender has an effect on this
variable, but show that by 1999, having an education of high
school or less is unsatisfactory for men and women alike. Of
course, because this analysis was restricted to adolescents whose
parents had high school educations of less, parental education
was not included as control variables in the equations.
There is, of course, concern that the results reported above
could possibly be biased toward students who lived in intact
households, as all respondents who did not report both the
mother’s education and father’s education were considered
“missing” and were not included. Given the greater likelihood
that adolescents who come from homes where one parent is not
present (as a result of divorce, death, etc.) would not know both
parents education level, an additional analysis of the missing and
included cases was necessary to assess possible bias. A break-
down of the included cases for mother and father’s education by
household type reveals that approximately 61% of the cases
included in the sample come from intact households, while 39%
come from single parent (18%), step-parent (15%) or some other
type (5.6%) of household. Given these results, we do not expect
any great bias in regard to household type in the results.
Discussion and Conclusion
Weis found that many adolescents in a de-industrialized
community (Freeway) understood the utility of college training
and aspired to continue their education after high school gradua-
tion. Findings from the present study indicate that students in
Muncie, a community apparently very similar to “Freeway” in
terms of its economic evolution, also became increasingly aware
of the value of an education, and both aspired and planned to
attend college in increasing proportions. In addition, students in
Muncie (particularly females) showed similar trends in regard to
changing attitudes about traditional gender roles and can argua-
bly be said to have experienced a similar “critical moment of
critique” against male dominance and patriarchy associated with
white, working class families (Weis, 1990: p. 54). The findings
presented in this paper clearly support all three hypotheses,
demonstrating that from 1977 to 1999, adolescents in Muncie
had increased educational aspirations, showed greater recogni-
tion of the importance of obtaining a college education, and
became less-traditional in terms of their attitudes about appro-
priate gender roles for men and women-effects which were more
pronounced for female students.
The implications of the findings regarding deindustrialization
and adolescent attitudes about education and gender are particu-
larly important for understanding the identity construction proc-
esses of working class adolescents. Many cultural sociologists,
including Lamont (2000) and Somers (1994), argue that sym-
bolic boundaries, such as the gendered notion that “men’s work”
is in the public sphere while “women’s work” is in the domestic
realm, are weighted by the environment in which they are em-
bedded. They insist that groups mobilize to define who they are
in relation to broad structural changes that challenge idealized
conceptions of individual and collective identity. The findings in
this analysis support Weis’ conclusions that the process of
deindustrialization essentially challenged working class values
by wiping out jobs traditionally associated with working class
identity. The low-wage service occupations offered in the
post-industrial “Freeway” forced many male students to con-
sider continuing their education after high school and to
re-define their ideas about the value of women earning a wage in
the public work sector. When Weis returned to Freeway 15
years later to re-interview the same students as young adults, she
found it was the males who had decided to “cash in” their “he-
gemonically based masculine persona” who fared better and
lived financially stable lives (Weis, 2004: p. 90). She argues this
occurred because “settled” jobs offering the most stability
tended to be those associated with the most schooling, tradition-
ally thought of as feminine, and also those which are most likely
to demand “a partner who earns roughly comparable money”
(Weis, 2004: p. 90). This is an important finding, particularly
because it demonstrates how symbolic boundaries around class
and gender are directly related to working class mobility.
Historically, working class identity also has been related to
the construction of racial identity, e.g., “whiteness” (Roediger,
1991). Understandings about race, like those about gender and
class are affected by dramatic economic and political shifts.
One of the major limitations of this study is not being able to
test Weis’ conclusions in regard to race and the way working

L. WINTERS ET AL.
120
class identity, particularly for adolescent males, is constructed
around a racialized “other”. Weis reported that the identity of
male students at “Freeway” High School was entangled with
“strident racism”, specifically against the small but growing
African American population in that community. Given the
similar demographics of Muncie, it would be interesting see if
the same racial tensions would have been expressed. Unfortu-
nately, when Middletown III and IV were conducted, questions
that address attitudes about racism and white privilege were
deemed insensitive and removed from the High School Surveys;
however, if one could identify a data source that would allow
for such an analysis, determining the effect of deindustrializa-
tion on adolescent attitudes about race would be an interesting
new direction for this research.
Another shortcoming of the current study is its inability to
conclude with a high degree of certainty that it was actually the
process of deindustrialization that produced the aforementioned
trends and changes in high school culture, not some other broad
structural shift or social movement that determined these out-
comes. The nature of these data and the method employed sim-
ply does not permit determination of the exact reasoning behind
students increased educational aspirations or transformed atti-
tudes about gender to the degree that Weis’ qualitative tech-
nique does. However, we can conclude that in the community
of Muncie, Indiana, which draws striking similarity to “Free-
way” in terms of its manufacturing history and notable dein-
dustrialization, trends in high school student attitudes about
education and gender seemingly parallel to what Weis wit-
nessed during the same time frame.
References
Adkins, L. (2001). Cultural feminization: “Money, sex and power” for
women. Signs, 26, 669-695. doi:10.1086/495625
Bettis, P. (1996). Urban students, liminality, and the postindustrial
context. Sociology of Education, 69, 105-125.
doi:10.2307/2112801
Bluestone, B., & Harrison, B. (1982). The de-industrialization of
America. New York: Basic Books.
Caccamo, R. (2000). Back to middletown: Three generations of socio-
logical reflections. Pal o A lto , California: Standford University Press.
Collins, P. H. (2000). Black feminist thought: Knowledge, conscious-
ness, and the politics of empowerment. New York: Routledge.
Caplow, C., Bahr, H. M., & Call, V. R. A. (2004). The middletown
replications: 75 years of change in adolescent attitudes, 1924-1999.
Public Opinion Quarterly, 68, 287-313. doi:10.1093/poq/nfh016
Davis, S., & Pearce, L. (2007). Adolescent work-family gender ideolo-
gies and educational expectations. Sociological Perspectives, 50,
249-271. doi:10.1525/sop.2007.50.2.249
Epstein, C. F. (1992). Tinkerbells and pinups: The construction and
reconstruction of gender boundaries at work. In M. Lamont & M.
Fournier (Eds.), Cultivating differences: Symbolic boundaries and
the making of inequality (pp. 232-256). Chicago, Illinois: University
of Chicago Press.
Hoover, D. W. (1990). Middletown revisited. Ball State monograph
number thirty-four. Muncie, Indiana: Ball State University.
McKibben, J. N. (1987). Job gains versus wage gains—Indiana’s dein-
dustrialization continues. Indiana Business Review, 62, 5-7.
Morton, M. J. (1990). Indiana a century ago: Comparing 1890 to 1988.
Indiana Business Rev i ew , 65, 1-6.
Lamont, M. & Fournier, M. (1992). Cultivating differences: Symbolic
boundaries and the making of inequality. Chicago, Illinois: Univer-
sity of Chicago Press.
Lamont, M. (2000). The dignity of working men: Morality and the
boundaries of race, class, and immigration. Camridge, MA: Harvard
University Press, New Yor k: Ru ss el l S ag e F ou nd a ti on .
Lamont, M., & Molnar, V. (2002). The study of boundaries in the so-
cial sciences. An nua l Review of Sociology, 28, 167-95.
doi:10.1146/annurev.soc.28.110601.141107
Lynd, R. S., & Lynd, H. M. (1937). Middletown in transition. New
York: Harcourt, Brace & World, Inc.
Lynd, Robert S. & Lynd, Helen M. (1929). Middletown: A study in
American culture. New York: Harcourt, Brace & Company.
McDowell, L. (1991). Life without father and Ford: The New gender
order of Post-Fordism. Transactions of the Institute of British Geog-
raphers, New Series, 16, 400-419.
McCall, L. (1988). Complex inequality: Gender, class, and race in the
new economy. New York: Routledge.
Roediger, D. R. et al. (1991). The wages of whiteness: Race and the
making of the American working class. Memphis, TN: Verso.
Silvestri, G., & Lukasiewicz, J. (1989). Projections of Occupational
Employment, 1988-2000. Monthly Labor Review, 112, 42-65.
Somers, M. (1994). Reclaiming the epistemological “Other”: Narrative
and the social constitution of identity. In C. Calhoun (Ed.), Social
theory and the politics of identity (pp. 37-99). Cambridge, MA:
Blackwell.
Weis, L. (1990). Working class without work: High school students in a
de-industrializing economy . New York: Routledge.
Weis, L. (2004). Class reunion: The remaking of the American white
working class. New York: Routledge.