Workaholism, Health, and Self-Acceptance

Christine M. Chamberlin and Naijian Zhang

This study examined the relationships between workaholism, perceived parental workaholism, self-acceptance,
psychological well-being, and physical symptoms among 347 college students. Statistically significant relationships
were found between college students’ perceived parental workaholism and their own workaholism. Also, relationships
between workaholism, self-acceptance, psychological well-being, and physical symptoms were discovered. Students
with higher perceptions of parental workaholism reported lower levels of psychological well-being, lower levels of self-
acceptance, and more physical health complaints. Implications and limitations are discussed.

Workaholism has received considerable attention in the popu- Kitayama (1994) asserted that workaholics focus primar-
lar literature over the past 2 decades, yet there is a dearth of ily on their “career self ” (p. 170) for self-validation, while
empirical investigation on the topic. The lack of empirical ignoring other important aspects of self (i.e., transpersonal
research may be due to the lack of agreement on whether or not self, transcultural–existential self, familial self, and physical
overwork is a problem. It has been suggested by some scholars self). They postulated that this limited focus of self-validation
that workaholism is heavily encouraged by organizations in might result in physical symptoms, alienation from friends
today’s society and is a socially accepted addiction (Fassel, and family, lack of development of the authentic self, and an
1990; Fassel & Schaef, 1991; McMillan & Northern, 1995). existential crisis. Difficulties may arise, particularly when
The term workaholism was first coined in 1971 by Oates, the career self is threatened in some way (e.g., loss of job,
who defined it as an addiction involving an uncontrollable retirement). Fassell (1990) suggested that workaholism may
need or compulsion to work continuously. Since then, worka- contribute to the “progressive numbing of America” (p. 123),
holism has been defined and measured in several ways, with whereby individuals become estranged from their essential
most researchers defining it as an addictive compulsion to self when they are consumed with the addictive busyness
work (Fassel, 1990; Haymon, 1992; Porter, 1996; B. E. Rob- associated with work addiction.
inson & Kelley, 1998). Haymon described workaholics as Overwork and workaholism may cause physical symp-
individuals who are “powerless over their work” (p. 11) and toms (e.g., ulcers, chest pain) and even death (Fassel, 1990;
who suffer from personal losses because of their “uncontrol- Ishiyama & Kitayama, 1994). The term Karoshi, coined in
lable need to work” (p. 11). Workaholics have been compared Tokyo by Tetsunojo Uehata in the 1980s, refers to death or
with alcoholics in that these two groups share some of the permanent disability (e.g., heart disease, deadly asthmatic
same symptoms, such as reality distortion, need to control, attacks, suicide) from overwork (Ishiyama & Kitayama,
denial, anxiety, depression, withdrawal, irritability, and rela- 1994). Research based on two qualitative studies indicated
tionship problems with friends and family members (Pietro- that workaholics may feel too busy to take care of their health
pinto, 1986; B. E. Robinson, 1989; Spruell, 1987). Spence needs (Machlowitz, 1978; Trueman, 1995). Machlowitz’s
and Robbins (1992) developed a battery of questionnaires (1978) results suggested that workaholics did not have ma-
to assess workaholism. They defined workaholism as high jor health problems, whereas Trueman found the opposite.
work involvement and drivenness (due to inner pressures) Spence and Robbins (1992) found a higher number of health
and low enjoyment of work. They also identified a second complaints among workaholics (i.e., individuals with high
type of workaholic (i.e., the enthusiastic workaholic) who, work involvement, with low work enjoyment, and who are
like the regular workaholic, has high work involvement and highly driven from internal pressures) than in some other
drivenness; however, level of work enjoyment is high instead groups of workers (e.g., those with high work enjoyment and
of low. In the current study, workaholism (work addiction) involvement and low internal drivenness). With such limited
was defined based on B. E. Robinson’s (1998a) definition: empirical research on the effects of workaholism on health,
“an obsessive-compulsive disorder that manifests itself the relationship between workaholism and physical health
through self-imposed demands, and inability to regulate remains unclear.
work habits, and an overindulgence in work to the exclusion B. E. Robinson (1998a), who has written extensively on
of most other life activities” (p. 7). the topic of workaholism, claimed that workaholism affects
Researchers and theorists disagree on the potential harmful individuals not only physically but also emotionally, cutting
effects of workaholism and have discussed numerous potential “them off from the rest of the world . . . [causing] them to be
consequences of workaholism in the literature. Ishiyama and in their own cold, dark, lonely world—all alone with room

Christine M. Chamberlin, Department of Counseling Psychology, Ball State University; Naijian Zhang, Department of Counseling
and Educational Psychology, West Chester University of Pennsylvania. Christine M. Chamberlin is now at Keene Middle School,
Keene, New Hampshire. Correspondence concerning this article should be addressed to Christine M. Chamberlin, Keene Middle
School, 17 Washington Street, Keene, NH 03431 (e-mail: [email protected]).

© 2009 by the American Counseling Association. All rights reserved.

Journal of Counseling & Development  ■  Spring 2009  ■  Volume 87 159
 Chamberlin & Zhang

only for other tasks to be completed” (p. 6). Ishiyama and significant correlation between workaholism and burnout in
Kitayama (1994) suggested that work addiction keeps people their study of 240 teachers. Workaholism was related to the
from facing the existential task of defining who they are. depersonalization and emotional exhaustion aspects of burn-
Workaholism has been correlated with depression out but not to the personal accomplishment component.
(Haymon, 1992), anxiety (Haymon, 1992), perfectionism Although researchers have examined many possible emo-
(Machlowitz, 1978; Spence & Robbins, 1992), and anger tional consequences of workaholism, few have investigated
(Haymon, 1992), as well as stress (Jackson, 1992; Spence & overall psychological well-being. Two studies examined the
Robbins, 1992) and burnout (Nagy & Davis, 1985). well-being of workaholics, both with very narrow operational
Spence and Robbins (1992) measured job stress among a definitions of well-being. Jackson (1992) defined emotional
random sample of 291 social workers with academic positions well-being as being present-oriented in approach to life, just
and found that workaholics (i.e., those high in work involvement one component of self-actualization as measured by the Per-
and drivenness from inner pressures and with low enjoyment sonal Orientation Inventory (POI; Shostrom, 1974, as cited in
of work) scored much higher on job stress, perfectionism, and Jackson, 1992). She found that workaholics had lower levels
nondelegation of responsibility than work enthusiasts (i.e., of well-being (i.e., present orientation) than nonworkaholics.
those with low drivenness and high enjoyment of work and It is questionable whether the measure Jackson chose to assess
work involvement). Cluster analyses revealed the emergence well-being, the POI, actually measures well-being because this
of several other types of worker profiles. Disenchanted work- measure is narrowly focused on time orientation. Emotional
ers (above average on drivenness and below average on work well-being has also been operationally defined as feeling ful-
involvement and enjoyment), enthusiastic workaholics (above fillment as opposed to frustration (Machlowitz, 1978). This
average on work involvement, enjoyment, and drivenness), and seems to be a broader definition of emotional well-being, yet
male workaholics displayed considerably more job stress than it is still limited. Machlowitz interviewed 34 workaholics
the male relaxed workers (above average on enjoyment and and found the majority of them (27) to be fulfilled. Although
below average on work involvement and drivenness), unengaged a sense of fulfillment may be one aspect of well-being, it is
workers (below average on work involvement, enjoyment, and doubtful that it encompasses it fully. Research is needed to
drivenness), and work enthusiasts. examine the relationship between workaholism and a more
The findings relating workaholism and burnout have been comprehensive measure of psychological well-being.
mixed. Burke and Matthiesen (2004) studied workaholism Workaholism not only affects the individual work addict
among 211 Norwegian journalists. The authors considered but also penetrates the lives of the workaholic’s coworkers,
three workaholism types (work enthusiasts, work addicts, friends, and family members. Workaholics can negatively
enthusiastic addicts) and also included numerous personal affect all family members, who may develop a set of mental
demographic and work situation characteristics, measures of health problems of their own (B. E. Robinson & Chase, 2000).
positive and negative affect, burnout components, and absen- Interpersonal relationships may be peripheral to work in the
teeism. The three burnout components (exhaustion, cynicism, lives of workaholics (Machlowitz, 1978; Trueman, 1995),
and professional efficacy) were measured by the Maslach which can lead to marital conflict, divorce, and psychologi-
Burnout Inventory–General Survey (MBI-GS) developed by cal consequences for children (Carroll & Robinson, 2000;
Schaufeli, Leiter, Maslach, and Jackson (1996). The results of Machlowitz, 1978; B. E. Robinson, Carroll, & Flowers, 2001;
the study indicated that the three workaholism types showed B. E. Robinson & Kelley, 1998). Dysfunctional patterns
similarities on personal and work situation characteristics, similar to such patterns in alcoholic families (e.g., enabling,
but work enthusiasts expressed substantially more positive denial, high expectations of perfection for children) appear
affect and considerably less negative affect, exhaustion, and in workaholic families (B. E. Robinson, 1998b).
cynicism than one or both of the other workaholism types. B. E. Robinson and Post (1995) found that workaholism
However, there was no difference among the workaholism interfered with intimate relationships and was positively cor-
types in terms of professional efficacy or absenteeism. Cox related with family dysfunction. In a random sample of 326
(1982) also studied burnout in relation to workaholism, using female spouses, spouses of workaholics had a higher level of
an earlier version of the MBI (Maslach & Jackson, 1981). marital estrangement, less positive affect toward husbands, and
Workaholism was found to be only minimally related to higher external locus of control than spouses of nonworkaholics
burnout (i.e., feeling of emotional exhaustion, frequent dep- (B. E. Robinson et al., 2001). In their study, Bonebright, Clay,
ersonalization, and a low sense of personal accomplishment) and Ankenmann (2000) discovered that workaholics had more
in the 40 psychology graduate students who were surveyed. work–life conflict than nonworkaholics, and their results suggest
More hours worked per week were found to be predictive of that workaholism is associated with higher conflict with the work-
perceived emotional exhaustion, and female students tended aholic’s home life. Furthermore, in two studies comparing adult
to demonstrate more workaholic patterns than male students. children of workaholics with adult children of nonworkaholics
Results of this study must be interpreted cautiously, given the and adult children of alcoholics, the adult children of workaholics
small sample size. Nagy and Davis (1985) found a statistically had a higher incidence of depression and parentification (i.e., a

160 Journal of Counseling & Development  ■  Spring 2009  ■  Volume 87

Workaholism, Health, and Self-Acceptance

role reversal between parent and child) than the adult children strengths, they had more difficulty accepting their shortcom-
of nonworkaholics (Carroll & Robinson, 2000). Moreover, adult ings than did the nonworkaholics. Another unresolved issue
children of workaholic fathers have been found to have more therefore, concerns workaholics’ level of self-esteem. Given
anxiety, depression, and external locus of control (B. E. Robinson the great deal of theoretical discussion suggesting that worka-
& Kelley, 1998) than the adult children of nonworkaholics. B. E. holics have low self-worth and self-esteem, it is surprising that
Robinson and Kelley expected that adult children of workaholics little research exists on this topic. Because self-acceptance
would also have a lower level of self-concept than the adult chil- was the main aspect of self-esteem correlated to workaholism
dren of nonworkaholics; however, they did not find this in their in Jackson’s study, further investigation of self-acceptance is
study. This lack of correlation may be due in part to the type of needed to corroborate this finding.
measure used to assess self-concept. For instance, the scale used
may not have tapped the construct of self-acceptance, which may Purpose of This Study
be more highly correlated to workaholism than other measures Although many studies have found relationships between
of self-esteem (Jackson, 1992). Further investigation is needed workaholism and psychological difficulties, the question
to better understand the consequences of parental workaholism of whether or not workaholism is harmful still remains.
on children, including whether or not children of workaholics Empirical data are scattered because there are only a few
become workaholics themselves. studies examining each construct (e.g., anxiety, depression,
Theories and the existing empirical literature explaining the stress), with very few researchers having investigated overall
function and development of workaholism can be categorized psychological well-being (Jackson, 1992; Machlowitz, 1978)
into two areas, with the development of workaholism being and self-acceptance (Jackson, 1992) in workaholics. Dupuy
influenced by internal factors (e.g., feelings of low self-worth, (1978) offered a more comprehensive definition and measure
insecurities, and avoidance of painful feelings, fears, and of psychological well-being (i.e., the General Well-Being
intimacy) and/or external factors (e.g., family, social, and [GWB] scale) than those offered by Machlowitz (1978) and
organizational messages). Regarding external factors, research Jackson. His measure, the GWB assesses overall positive feel-
has demonstrated that workaholics believed that their parents ings about one’s inner personal state and includes the scales
had high expectations of them (Dougherty, 1989) and that their Anxiety, Depression, General Health, Positive Well-Being,
parents’ love was dependent on their (i.e., the children’s) suc- Self-Control, and Vitality.
cess (Machlowitz, 1980). Furthermore, institutions, such as Clinical observations and theoretical speculation suggest
the educational system, the political system, and the church, that workaholism may lead to physical health problems and
as well as the competitive nature of industrialized cultures may consequences to the workaholic’s family members, yet re-
contribute to the development of workaholism (Fassel, 1990). search is limited in these areas, especially regarding physical
A study of seven U.S. organizations found the encouragement health consequences. Physical symptomatology in children
of codependent and workaholic behavior and the “work ‘til of workaholics has been particularly overlooked, which
you drop” message very much alive in corporate America could suggest either a lack of relationship or an unfortunate
(McMillan & Northern, 1995). oversight. Finally, the likelihood of workaholism developing
Internal factors may also contribute to the development in children of workaholics is unclear, with few studies having
of workaholism. For instance, work may serve the function explored this issue. Machlowitz (1980) concluded from her
of increasing the workaholic’s self-esteem (Porter, 1996) and research study that some workaholics may have followed in
may be the workaholic’s primary source of self-validation the footsteps of their workaholic parents. Fassel (1990) as-
(Ishiyama & Kitayama, 1994) and personal identity (Trueman, serted that the children of workaholics may be conditioned to
1995). Porter suggested that workaholics “receive ego- become workaholics, or they may go to the other extreme and
enhancement from work involvement to the extent that external “become paralyzed and incapable of any action” (p. 78).
rewards can temporarily bolster feelings about self ” (p. 74). To address the aforementioned unresolved theoretical is-
Therefore, workaholics seek accomplishments so that others sues, the current study examined the relationships between
and, ultimately, they will have evidence that they are valuable. workaholism, psychological well-being, physical complaints,
Empirical investigations in this area have been sparse. Trueman and self-acceptance. This study simultaneously examined the
interviewed 8 workaholics and found that their personal identity well-being of students who were workaholic and the wellness
was formed through their work; in other words, the workaholics of students who perceived their parents as being workaholic.
in that study relied on their work to define them. In her study We hypothesized the following:
on self-acceptance and self-regard among workaholics, Jackson
(1992) found the workaholics in her sample of 112 individuals 1. There would be no relationship between perceived
to have lower levels of self-acceptance than the nonworkahol- (by adult children) parental workaholism and adult
ics, although there were no differences in levels of self-regard children’s workaholism. (Note. The adult children in
(i.e., a second component of self-esteem). Results indicated the current study were the students mentioned in these
that although workaholics were able to like themselves for their hypotheses.)

Journal of Counseling & Development  ■  Spring 2009  ■  Volume 87 161

 Chamberlin & Zhang

2. There would be a negative relationship between Procedure

students’ workaholism and students’ psychological
well­-being. Students were recruited from undergraduate and graduate
3. There would be a positive relationship between classes to participate in the current study. One method of
students’ workaholism and students’ physical health recruitment consisted of posting a sign-up sheet for the study
complaints. titled “Work Patterns and Well-Being” on the experimental
4. There would be a negative relationship between stu- sign-up board for the Department of Counseling Psychol-
dents’ workaholism and students’ self-acceptance. ogy. This sign-up sheet instructed the students to meet in a
5. Student workaholics (those with medium to high risk specified classroom where the study was fully explained and
for workaholism) would report less psychological they were given an informed consent form. Students’ ques-
well-being, more physical health complaints, and less tions about the study were answered before they signed the
self­-acceptance than nonworkaholic students (students consent form. They were notified that their identity would
with low risk for workaholism). be confidential and that they were free to withdraw from
6. Students who perceived their parents to be workaholic the study at any time. The principal researcher (first author)
would report less psychological well-being and less distributed a packet of questionnaires to each student to com-
self-acceptance than students with perceived non- plete, instructing them to follow the directions in the packet
workaholic parents. and to ask the researcher questions at any time. During the
7. Students who perceived their parents as workaholic session, the researcher remained seated quietly in the front
would not report any more physical health com- of the classroom, available for questions. After completion
plaints than students with perceived nonworkaholic of the questionnaires, students were given a debriefing form
parents. regarding the nature of the study, questions were answered,
and students were given a research credit slip.
Given the high social acceptance and social encourage- Another method of recruitment involved going to graduate-
ment of work addiction, the denial of its potential hazards to level classes to request volunteers. These students completed
the self and the family, and the limited amount of empirical a survey and a consent form outside of class time and returned
research on the topic, it seems crucial to further investigate them in separate envelopes to the principal investigator’s
the correlates of workaholism. The more information that is mailbox at the university.
compiled regarding the relationship between workaholism and
Instruments
well-being, the more likely that individuals and organizations
may make subsequent healthful changes. The Work Addiction Risk Test (WART; B. E. Robinson, 1989)
was used to measure student workaholism. The WART consists
Method of 25 work habit descriptors, which are rated on a 4-point
scale ranging from always to never true. One total score is
Participants generated, with higher scores indicating a higher degree of
We collected 347 surveys from volunteer undergraduate and workaholism (B. E. Robinson, 1996a). Studies have reported
graduate students at a midwestern university. Four surveys that the WART has test–retest reliability over a 2-week time
were eliminated from analysis because they were either in- frame (r = .83; B. E. Robinson, Post, & Khakee, 1992), split-
valid or duplicates. Sixty-four participants indicated on the half reliability (r = .85; B. E. Robinson & Post, 1995), internal
demographic page that they had an illness and therefore were consistency (r = .88; B. E. Robinson, 1999), face validity (B.
eliminated from analysis because of possible confounding of E. Robinson & Post, 1994), and content (B. E. Robinson &
their answers on the health symptom measure. Furthermore, Phillips, 1995) and concurrent (B. E. Robinson, 1996b, 1999)
scales with less than 90% item completion were eliminated validity. For the sample in the current study, a Cronbach’s alpha
from analysis. of .89 was found for the WART. “It is important to remember
Participants thus were 279 students, 249 of which were that a test is not reliable or unreliable . . . Thus, authors should
undergraduate students and 29 were graduate students (1 provide reliability coefficients of the scores for the data be-
student did not specify status). Ages of participants ranged ing analyzed even when the focus of their research is not
from 18 to 50 years (M = 22.14, SD = 4.81), with the major- psychometric” (Wilkinson & APA Task Force on Statistical
ity of participants under age 24 (88%), female (70%), and Inference, 1999, p. 596).
Caucasian American (85%). Participants who identified The Children of Workaholic Parents Screening Test
themselves as minorities were African American (11%), (CWST; B. E. Robinson & Carroll, 1999) was used to mea-
Hispanic/Latino American (2%), Asian American (1%), and sure perceived parental workaholism. The CWST consists of
other (1%). Furthermore, a large proportion of the sample 30 questions (answered by checking yes or no) that measure
studied psychology-related social sciences (43%). Students respondents’ perceptions, experiences, attitudes, and feelings
participated strictly on a volunteer basis. concerning their parents’ work habits. The yes ratings were

162 Journal of Counseling & Development  ■  Spring 2009  ■  Volume 87

Workaholism, Health, and Self-Acceptance

summed to obtain a total score, with higher scores being more student health center because of physical health problems.
indicative of perception of parental workaholism. B. E. Robin- In addition, students rated their physical health on a scale
son and Carroll reported an internal reliability for the CWST from 1 to 7 (1 = poor health, 7 = excellent health). They also
of r = .92 at pretest and .93 at a 2-week posttest, and test–retest rated the following statements on a 5-point Likert-type scale
reliability (t = .68, p < .01) in a sample of 207 undergradu- ranging from 1 (strongly disagree) to 6 (strongly agree): “I do
ate students. Furthermore, concurrent validity with the Beck not have time to get sick,” “I usually get sick (e.g., a cold or
Depression Inventory (r = .46, p < .01) and the Parentification the flu) during school vacations,” and “I do not have time to
Questionnaire (r = .60, p < .01) has been demonstrated (B. E. take care of my health needs.” A Cronbach’s alpha of .92 was
Robinson & Carroll, 1999). A Cronbach’s alpha of .90 was calculated for the CHIPS with the current sample.
calculated for the CWST with the current sample. Self-acceptance was measured by the Self-Acceptance
The GWB developed by Dupuy (1978) was used to measure Scale (SAS; Phillips, 1951). This scale consists of 25 items
psychological well-being. This 18-item scale measures six (e.g., “I feel inferior as a person to some of my friends”),
dimensions with the following subscales: Anxiety, Depres- which are rated on a 5-point Likert-type scale ranging from
sion, General Health, Positive Well-Being, Self-Control, and 1 (not at all true) to 5 (completely true). One total score
Vitality, which are intended to measure the degree to which is obtained, with higher scores indicating lower levels of
an individual has positive feelings about his or her “inner self-acceptance. In the current study, the items were reverse
personal state.” For the first 14 items (e.g., “Have you been scored, so that higher scores indicated higher levels of self-ac-
anxious, worried, or upset?”), respondents use a scale of 1 to 6 ceptance. Test–retest reliability (r = .84; J. P. Robinson, 1973)
(anchor descriptors vary for each question) to rate the intensity and convergent validity (rs = .55 to .73) with other measures
or frequency of symptoms. The last 4 items (e.g., “How much of self-acceptance (Omwake, 1954) have been demonstrated.
energy, pep, vitality have you felt?”) use a 0 to 10 rating scale In addition, predictive validity has been established between
with different adjectives (e.g., no energy at all, listless to very the SAS and scales measuring acceptance toward others (rs
energetic, dynamic) on each end of the scale. For each item, = .41 to .74; Omwake, 1954; Phillips, 1951).
respondents are instructed to rate their answer based on how The five measures used in the survey were randomly as-
they felt during the past month. One total score is obtained, sorted and attached to a demographic information sheet, which
with higher scores indicating higher levels of psychological elicited the following information from students: age, sex, eth-
well-being. The GWB has demonstrated test–retest reliability nicity, grade point average, year in school, relationship status,
(rs = .68 to .85) over 3 months (Fazio, 1977; Monk, 1981), and number of credit hours currently scheduled. An estimate
internal consistency (rs = .88 to .95; Fazio, 1977; Himmelfarb of the total number of hours spent per week in the following
& Murrell, 1983), and concurrent validity (i.e., depression, areas was also obtained: homework, class time, other school-
r = .69: anxiety, r = .64; Fazio, 1977). For the sample in the related tasks, paid employment, volunteer work, household
current study, Cronbach’s alphas were calculated at .90 for tasks, childcare, exercise, and leisure activity. Students were
the GWB. also asked to rate the following statements on a 5-point Likert-
The Cohen-Hoberman Inventory of Physical Symptoms type scale ranging from 1 (strongly disagree) to 6 (strongly
(CHIPS) developed by Cohen and Hoberman (1983), a 33- agree): “I enjoy school” and “I enjoy my job(s)”. To assess
item list of common physical symptoms (e.g., colds, nausea) socioeconomic status (SES), questions were asked about
was used to measure physical health complaints. Students rated students’ parents concerning the following: educational level,
each item on a 5-point Likert-type scale (ranging from 0 = not present income, relationship status, and current occupation
bothered to 4 = extreme bother) according to how much the were also asked. Age, sex, ethnicity, SES, and year in school
symptom had bothered or distressed them during the preceding were analyzed in the preliminary analyses for correlations
2 weeks. Internal reliability for the CHIPS is .88, and this mea- with the dependent variables, while the other variables were
sure significantly correlated with the Center for Epidemiologic included in the survey for possible future analysis.
Studies Depression Scale (r = .44, p < .001) and the use of
student health facilities in two college student samples during Design and Data Analyses
the 5 weeks following scale completion (.22 and .29; Cohen Workaholism of the student and the perceived level of pa-
& Hoberman, 1983). A total score is obtained by summing the rental workaholism were the independent variables, while
ratings of each item, with higher total scores indicating higher psychological well-being, physical health complaints, and
levels of physical health complaints. self-acceptance were the dependent variables. Each inde-
Students were also asked to list other physical complaints pendent variable consists of two levels, nonworkaholism
they had experienced that were not on the scale and to rate and workaholism, creating the following four categories: (a)
them in the same fashion on a 5-point scale. Furthermore, low risk for perceived parental workaholism, (b) medium to
they indicated the following for the 4 months preceding the high risk for perceived parental workaholism, (c) low risk
study: (a) the number of days they had been sick and (b) the for student workaholism, and (d) medium to high risk for
number of times they had seen a physician or been to the student workaholism.

Journal of Counseling & Development  ■  Spring 2009  ■  Volume 87 163

 Chamberlin & Zhang

Preliminary analyses involved an analysis of the means and did women and reported fewer physical health complaints (t =
standard deviations of the independent and dependent vari- –2.06, df = 277, p < .05). No statistically significant differences
ables; Pearson’s r correlations between the dependent variables between groups were found for SES.
and a reliability analysis of the scales were used to measure
the dependent variables. Furthermore, Pearson’s r correlations Primary Analyses
and independent samples t tests were conducted to explore Four Pearson product–moment correlations were performed
group differences on the independent and dependent variables to test Hypotheses 1 through 4. Hypothesis 1 stated that there
in terms of age, school status, gender, ethnicity, and SES. would be no relationship between students’ perceived parental
Initial analyses consisted of four Pearson’s r correlations. workaholism and students’ workaholism. This hypothesis was
The first examined Hypothesis 1, which explored the rela- not supported. In fact, a statistically significant correlation was
tionship between the two independent variables (i.e., student found (r = .20, p < .001), with the higher the student perception
workaholism and perceived parental workaholism). The of parental workaholism, the higher the level of the student’s
second, third, and fourth correlations examined Hypotheses workaholism. Hypotheses 2, 3, and 4 were supported. That is,
2, 3, and 4, which considered the relationships between stu- higher student workaholism scores (i.e., higher scores on the
dent workaholism, psychological well-being, physical health WART) were related to lower student psychological well-being
complaints, and self-acceptance. (r = –.36, p < .000), students with higher workaholism scores
Further analyses included two, 2 by 3 between-subjects reported considerably more physical health complaints (r =
multivariate analyses of variance (MANOVAs) to exam- .30, p < .000), and students with higher workaholism scores
ine Hypotheses 5, 6, and 7. The first MANOVA examined reported less self-acceptance (r = –.38, p < .000).
Hypothesis 5, which was concerned with the differences
between workaholic and nonworkaholic students in levels Secondary Analyses
of the dependent variables (i.e., psychological well-being, Table 2 shows the means and standard deviations for the
physical health complaints, and self-acceptance). The second two student workaholic groups on the measures of the de-
MANOVA was used to analyze Hypotheses 6 and 7, involving pendent variables. To analyze Hypotheses 5 through 7, two
how level of perceived parental workaholism was related to 2 by 3 between-subjects MANOVAs were used. The purpose
the three dependent variables. of the first MANOVA was to examine the interaction effects
between the two levels of student workaholism (Group 1 =
Results low risk for workaholism, Group 2 = medium to high risk
for workaholism) and the three dependent variables (i.e.,
Preliminary Analyses
psychological well-being, physical health complaints, and
Table 1 shows the means and standard deviations for independent self-acceptance). The two groups for the independent vari-
and dependent variables. Results of the Pearson product–moment able (i.e., student workaholism) were formed on the basis of
correlations indicated that age of student was significantly cor- past research indicating that individuals with scores of 56 or
related in a negative direction with physical health complaints below on the WART are not workaholic, while those with
(r = –.16, p < .006) and positively correlated with psychological scores between 57 and 66 are at medium risk for workahol-
well-being (r = .12, p < .045). Undergraduate students reported ism, and individuals scoring above 66 are at high risk (B. E.
notably more physical health complaints (t = 2.49, df = 276, p Robinson, 1998b, 1999). The mean score on the WART for
< .02) and lower self-acceptance (t = –2.65, df = 275, p < .01) the group of students in the current study (M = 53.20, SD =
than did graduate students. Considerably more graduate students 10.60) was similar to past findings for participant scores by
scored within the medium to high workaholic group than did the B. E. Robinson (e.g., in his 1999 study, the mean WART score
undergraduate students (χ2 = 4.54, df = 1, p < .03). Men reported was 55.5). Therefore, B. E. Robinson’s criteria were used to
much higher levels of psychological well-being (t = 2.47, df =
272, p < .02) and overall health (t = 3.32, df = 277, p < .001) than Table 2
Means and Standard Deviations for the Two
Table 1 Student Workaholic Groups on the Dependent
Measures
Means and Standard Deviations for Independent
and Dependent Variables (N = 279) Student Workaholism Groups
M SD
Variable M SD
Variable 1 2 1 2
Student workaholism 53.20 10.60
Perceptions of parental workaholism 5.45 5.40 Self-acceptance 93.24 83.50 16.11 16.65
Self-acceptance 89.76 16.86 Psychological well-being 72.93 62.01 14.20 14.90
Psychological well-being 68.87 15.34 Physical health complaints 19.64 30.07 16.72 20.93
Physical health complaints 23.48 19.06
Note. 1 = low risk for workaholism (n = 168); 2 = medium to high
Overall health self-rating 5.17 1.11
risk for workaholism (n = 102).

164 Journal of Counseling & Development  ■  Spring 2009  ■  Volume 87

Workaholism, Health, and Self-Acceptance

form the two groups in this study. Students with scores of 56 Table 3
or below on the WART were categorized as nonworkaholic
Means and Standard Deviations for the Two
(low risk for workaholism group; n = 168), while students
Perceived Parental Workaholic Groups on the
scoring above 56 were categorized in the workaholic group
Dependent Measures
(medium to high risk for workaholism; n = 102).
Hypothesis 5 stated that workaholic students would report Perceived Parental Workaholism
Groups
less psychological well-being, more physical health com-
plaints, and less self-acceptance than nonworkaholic students. M SD
The students in the medium to high risk for workaholism group Variable 1 2 1 2
scored an average of 10 points below students in the low risk Self-acceptance 92.74 85.39 17.48 15.30
Psychological well-being 71.45 66.05 15.73 14.62
for workaholism group on self-acceptance, approximately Physical health complaints 22.19 28.66 19.51 20.60
11 points lower on psychological well-being, and approxi-
Note. 1 = low risk for parental workaholism (n = 94); 2 = medium to
mately 10 points higher on physical health complaints. The
high risk for parental workaholism (n = 94).
interaction between students with a high workaholic risk and
those with a low workaholic risk was statistically significant dependent measures. A MANOVA was used to examine the
(Pillai’s trace, F = 14.51, p < .0001, R² = .14), indicating that interaction effects between the two levels of perceived parental
statistically significant differences existed between students workaholism (Group 1 = low student perceptions of parental
with low workaholism scores and those with medium to high workaholism, Group 2 = medium to high student perceptions
workaholism scores in terms of the dependent variables. of parental workaholism) and the three dependent variables
Statistically significant main effects for self-acceptance, (i.e., self-acceptance, psychological well-being, and physical
F(1, 268) = 22.64, p < .0001, R² = .14, and psychological health complaints). The two groups for the independent vari-
well-being, F(1, 268) = 36.17, p < .0001, R² = .21, were found. able perceived parental workaholism were formed based on B.
This indicates that students in the workaholic group scored E. Robinson’s (1999) method of forming groups for student
much lower on measures of self-acceptance and psychological workaholism. That is, scores below the mean (M = 5.45, SD
well-being than students in the nonworkaholic group. Main = 5.40) were categorized as students’ perceived low risk for
effects for physical health complaints were also statistically parental workaholism, whereas scores between the mean and 1
significant, F(1, 268) = 20.36, p < .0001, R² = .11, but they standard deviation above the mean were categorized as students’
were not interpretable without further analysis because of a perceived medium risk for parental workaholism, and scores
violation of equality of error variance, F(1, 268) = 10.24, p < above 1 standard deviation of the mean were labeled students’
.002. This variance suggests that the group with the smaller perceived high risk for parental workaholism. As with the first
number of students (i.e., the workaholic group) had the largest MANOVA, the medium and high risk groups were combined
amount of variance in scores on the physical health complaints to form one group. Thus, the two levels examined in the cur-
measure, whereas the group with the larger number of students rent MANOVA were students’ perceived low risk for parental
(i.e., the nonworkaholic group) had the least amount of variance workaholism (Group 1) and students’ perceived medium to high
in physical health complaints scores. Therefore, the chance for risk for parental workaholism (Group 2).
Type I error was increased, making further analysis necessary Box’s test of equality of covariance matrices indicated a vio-
before interpretations could be made. lation of homogeneity of variance (F = 2.43, p < .02). Therefore,
Having equal numbers of students for the two groups formed the random sampling down procedure previously described
to assess the effect for the independent variable student worka- in the first MANOVA was used. The group of students who
holism has been recommended as a way to make homogeneity perceived their parents as low risk for parental workaholism
of variance more conditionally robust, thus decreasing the was sampled down from n = 179 to n = 94. The students who
potential for Type I error (Stevens, 1996). Therefore to control perceived their parents as having medium to high levels of
for error variance, random sampling down of the nonworkaholic risk for workaholism, scored an average of 7 points lower on
group from n = 168 to n = 102 was conducted. After sampling self-acceptance, approximately 5 points lower on psychologi-
down, a violation in error variance was still noted, F(1, 202) = cal well-being, and approximately 6 points higher on physical
14.15, p < .0001. However, the test of significance was condi- health complaints than did the students who perceived their par-
tionally robust given that equal ns existed in both groups, with ents in the low workaholic risk group. The interaction between
the Type I error potential likely to be less than .05. A statisti- students who perceived a low risk for parental workaholism
cally significant main effect for physical health complaints was and students with perceptions of medium to high parental risk
found, F(1, 202) = 19.33, p < .0001, R² = .09, with students in for workaholism was statistically significant (Pillai’s trace, F =
the workaholic group scoring considerably higher on physical 3.77, p < .01; h² = .06), indicating that statistically significant
health complaints than students in the nonworkaholic group. differences existed in terms of the dependent variables between
Table 3 shows the means and standard deviations for stu- students with low scores and medium to high scores of percep-
dent perceptions of parental workaholism (two groups) on the tions of parental workaholism.

Journal of Counseling & Development  ■  Spring 2009  ■  Volume 87 165

 Chamberlin & Zhang

Statistically significant main effects were found for self- emotional symptoms such as depression, anxiety, or low self-
acceptance, F(1, 186) = 9.42, p < .002, R² = .05; psychological esteem (B. E. Robinson & Kelley, 1998)—symptoms similar
well-being, F(1, 186) = 5.95, p < .016, R² = .03; and physical to those that might be found in children of alcoholics. Research
health complaints, F(1, 186) = 4.89, p < .02, R² = .03. Results had not suggested that children of workaholics would also
indicated that students who perceived their parents’ level of exhibit more physical health complaints. This finding, how-
workaholism to be in the medium to high risk category had ever, makes sense if one considers the idea of wholism within
notably lower self-acceptance scores, lower psychological health psychology, in which the mind is thought to influence
well-being scores, and a higher number of physical health the body. Thus, it may be that individuals’ psychological
complaints than did students who perceived their parents symptoms, such as lower levels of psychological well-being
to be in the lower risk for workaholism category. Thus, Hy- and lower levels of self-acceptance, are influencing their
pothesis 6 (i.e., students who perceived their parents to be health, causing them to actually have or merely perceive more
workaholic would report less psychological well-being and physical health problems. There was also a positive correla-
less self-acceptance than students who perceived their parents tion between student workaholism and perceptions of parental
as nonworkaholic) was supported by these results. However, workaholism, which might indicate that the students with
Hypothesis 7 (i.e., students who perceived their parents to perceived workaholic parents may experience more physical
be workaholics would not report any more physical health symptoms than the other group merely because of their own
complaints than did students who perceived their parents to workaholic tendencies.
be nonworkaholic) was not confirmed.
Limitations

Discussion Limitations of the current study involve the nature of the

sample and the methods of measurement. Generalizability
Results from the current study indicate that (a) workaholism of the results could be questionable because the majority of
may be associated with physical health problems and lower students were Caucasian undergraduates from counseling
levels of both psychological well-being and self-acceptance psychology courses. It is possible that the students’ level of
and (b) the psychological well-being of adult children of workaholism as well as the perception of their parents work
workaholic parents can be affected by the parents’ workaholic patterns could change as the students age and have more ex-
behaviors. Whether or not workaholism leads to these negative perience with the workforce. Some students may not yet have
consequences cannot be determined because directionality and individuated from their family and thus may be more likely to
causality cannot be concluded from the current study. repeat learned workaholic behaviors. Ideally, future research
Two hypotheses were not supported. Contrary to expecta- would examine individuals who are employed full-time, per-
tion, students with higher workaholism scores on the WART haps measuring levels of workaholism at different stages in
were found to perceive their parents as more workaholic. The life and career. It could be argued, however, that workaholic
idea that these variables would not be related was based on tendencies may already be formed at college age, particularly
Fassel’s (1990) assertion that children of workaholics may ei- for older students, because many students work in addition
ther (a) become workaholics themselves through conditioning to taking classes. There is also the possibility of the effect of
or (b) go to the other extreme of inaction and paralysis in their a demand characteristic for students to be highly productive
work. Based on this assertion, we thought that no correlation during their training years that needs to be considered. This
would exist because students with both types of reactions may have contributed to the higher incidence of workaholism
would be represented in the current sample. However, students in the graduate students than in the undergraduate students.
who perceived their parents as more workaholic actually did The demand characteristic may also be a factor in individuals
score higher on their own level of workaholism. This supports who are out of school and in the initial stage of their career,
Machlowitz’s (1980) finding that some of the workaholics in because they may experience high pressures to prove them-
her study followed in the footsteps of their workaholic parents selves through their performance.
as well as B. E. Robinson’s (1998b) assertion that children The limited systematic validation of the scales used to measure
of workaholics tend to become high achievers to gain their self-acceptance and physical health complaints is another area
parents’ approval. The results could also mean that students for caution for this study. This limited validation leads to a lack
who are more workaholic themselves merely tend to perceive of norms with which to compare the current populations’ scores
their parents as being more workaholic—the parents may not in order to test for practical and clinical significance. In addition,
actually be workaholic. the use of self-report measures introduces bias. Furthermore, the
The second unexpected finding was that adult children of lack of the agreement on a unified definition of workaholism may
perceived workaholics had more physical health complaints influence the generalizability of the results. Moreover, dichoto-
than the students in the group who perceived their parents as mous groups with interval scores on the WART were formed on
nonworkaholic. Much of the theoretical literature concerning the basis of past research. Such formation of groups may discard
children of workaholics focuses on the children developing variance and lower the reliability of the results.

166 Journal of Counseling & Development  ■  Spring 2009  ■  Volume 87

Workaholism, Health, and Self-Acceptance

Particular caution needs to be taken when applying these may need to be confronted, and new positive cues and prompts
results cross-culturally because people from different cultures for the nonworkaholic behavior will need to be developed.
may view workaholism and its effects differently. The definition The findings of this study indicate that parents’ workaholic
chosen for this study is from B. E. Robinson’s (1989) work, behavior does have an impact on their children. Therapists
which describes workaholism as a type of obsessive-compulsive should be aware that children of workaholics may feel driven
disorder (OCD). This raises some diagnostic questions. For toward perfectionism, self-reliance, approval seeking, and
instance, if workaholism is actually a type of OCD, does this overresponsibility, and they should be screened for anxiety and
mean that the person must meet the Diagnostic and Statistical depression (B. E. Robinson & Kelley, 1998). B. E. Robinson
Manual of Mental Disorders (4th ed.; American Psychiatric and Kelley asserted that family-of-origin work is essential and
Association, 1994) criteria for OCD, or are there separate can include genograms and inner-child work. B. E. Robinson
diagnostic criteria that should be used? Can individuals who and Kelley have instructed parents to provide unconditional
do not meet official OCD criteria still be workaholic? Finally, regard for their children as opposed to measuring the child’s
the current research is correlational in nature, as is the majority worth by what the child achieves. In addition, counselors can
of research examining workaholism. Future research needs to teach the child or adult child of the workaholic how to relax,
be conducted that is predictive in nature so that causality and have fun, be more flexible and spontaneous, live more in the
directionality of results can be more accurately inferred. moment, be less self-critical, and develop his or her inner,
nurturing voice.
Implications Future research is needed to examine the stages that a
There are implications of the current study for families, indi- workaholic may go through in his or her addiction process.
viduals, spouses, and coworkers. If workaholics tend to report Furthermore, identifying how work addiction affects people at
more physical health complaints, lower psychological well-being, varying developmental points in their life (e.g., at retirement
and lower self-acceptance than nonworkaholics, their own level versus as a new parent) would be of interest.
of happiness is in jeopardy, as well as the happiness of people Results from this study in no way suggest that hard work or
around them. The excessive demands workaholics place on ambition in itself is detrimental. Work can bring a great deal of
themselves may lead to more health complaints and inhibited joy, fulfillment, and sense of purpose to people’s lives; it is neces-
functioning at work, school, or home. This nonaccepting atti- sary for the survival of humankind. It is when individuals become
tude toward oneself could transfer to being overly critical and dependent on work in a way that has negatively affected their life
demanding of others. The implications of these findings are (e.g., higher incidences of physical complaints and psychiatric
important for therapists and organizational psychologists to be symptoms, low level of self-acceptance) or the lives of those
aware of in their practice. Interventions may need to be made in around them that psychological intervention may be needed.
the work environment or in the family, and individuals may pres- More social attention to this issue is also crucial if the
ent to treatment at varying stages of this addiction. Treatment necessary types of organizational changes are to be made.
may be indicated when workaholics are experiencing unpleasant Employers need to be made aware of the potential downside
consequences from their work patterns or if significant others to encouraging workaholic behaviors and to take measures
or employers are negatively affected. to make their organizations healthier environments in which
When working with clients on issues related to the workaholic to work. As with other addictions, it is easy to blame the
behavior, counselors may help clients explore consequences and individual for his or her behavior, while larger economic
rewards of the workaholic behavior and educate them regarding the and political influences are ignored. The driving economic
potential effects of workaholic behavior (e.g., health consequences) and political forces behind societal patterns contributing to
and the positive effects of leisurely activities (e.g., taking time to workaholic behavior must be identified and addressed because
relax and rejuvenate will help with later productivity). Because the health of workers and families depends on it.
the workaholic behavior also negatively affects the quality of life
of others, particularly individuals close to the workaholic, coun- References
selors may include interventions by family members, coworkers,
or friends that challenge the workaholic with the specific ways in American Psychiatric Association. (1994). Diagnostic and statistical
which his or her behavior is affecting others. manual of mental disorders (4th ed.). Washington, DC: Author.
Counselors may help the workaholic to identify his or Bonebright, C. A., Clay, D. L., & Ankenmann, R. D. (2000). The
her individual reinforcement history. For example, specific relationship of workaholism with work-life conflict, life satis-
messages received in the workaholic’s family of origin need faction, and purpose in life. Journal of Counseling Psychology,
to be explored, as well as current family messages. Family 47, 469–476.
therapy could be very useful during this stage, so that family Burke, R. J., & Matthiesen, S. (2004). Workaholism among Nor-
members can become aware of how they may be reinforcing wegian journalists: Antecedents and consequences. Journal
the workaholic behavior. Clients in the workaholic’s home and of the International Society for the Investigation of Stress, 20,
work environment who are reinforcing workaholic behavior 301–308.

Journal of Counseling & Development  ■  Spring 2009  ■  Volume 87 167

 Chamberlin & Zhang

Carroll, J. J., & Robinson, B. E. (2000). Depression and parenti- Nagy, S., & Davis, L. G. (1985). Burnout: A comparative analysis of
fication among adults as related to parental workaholism and personality and environmental variables. Psychological Reports,
alcoholism. The Family Journal, 8, 360–367. 57, 1319–1326.
Cohen, S., & Hoberman, H. M. (1983). Positive events and social Oates, W. (1971). Confessions of a workaholic: The facts about work
supports as buffers of life change stress. Journal of Applied Social addiction. New York: World.
Psychology, 13, 99–125. Omwake, K. T. (1954). The relation between acceptance of self and
Cox, D. R. (1982). The workaholic behavior pattern and the experi- acceptance of others shown by three personality inventories.
ence of burnout: A correlative study (Doctoral dissertation, United Journal of Consulting Psychology, 18, 443–446.
States International University, 1982). Dissertation Abstracts Philips, E. L. (1951). Attitudes towards self and others: A brief ques-
International, 43 (02), 519. tionnaire report. Journal of Consulting Psychology, 15, 79–81.
Dougherty, S. E. (1989). Analysis of a selected number of female work- Pietropinto, A. (1986). The workaholic spouse. Medical Aspects of
aholics (Doctoral dissertation, The George Washington University, Human Sexuality, 20, 89–96.
1989). Dissertation Abstracts International, 50 (06), 1642. Porter, G. (1996). Organizational impact of workaholism: Sugges-
Dupuy, H. J. (1978, October). Self-representations of general psycho- tions for researching the negative outcomes of excessive work.
logical well-being of American adults. Paper presented at the annual Journal of Occupational Health Psychology, 1, 70–84.
meeting of the American Public Health Association, Los Angeles. Robinson, B. E. (1989). Work addiction. Deerfield Beach, FL: Health
Fassel, D. (1990). Working ourselves to death. New York: HarperCollins. Communications.
Fassel, D., & Schaef, A. W. (1991). A feminist perspective on work Robinson, B. E. (1996a). Concurrent validity of the Work Addiction
addiction. In N. Van Den Bergh (Ed.), Feminist perspectives on Risk Test as a measure of workaholism. Psychological Reports,
addictions (pp. 199–211). New York: Springer. 79, 1313–1314.
Fazio, A. F. (1977). A concurrent validational study of the NCHS Robinson, B. E. (1996b). The psychosocial and familial dimensions
General Well-Being Schedule. (Vital & Health Statistics, Series of work addiction: Preliminary perspectives and hypotheses.
2, No. 73; DHEW Publication No. HRA 78–1347). U.S. Depart- Journal of Counseling & Development, 74, 447–452.
ment of Health, Education, and Welfare, National Center for Robinson, B. E. (1998a). Chained to the desk. New York: New York
Health Statistics. University Press.
Haymon, S. W. (1992). The relationship of work addiction and Robinson, B. E. (1998b). The workaholic family: A clinical perspec-
depression, anxiety, and anger in college males (Doctoral dis- tive. The American Journal of Family Therapy, 26, 65–75.
sertation, Florida State University, 1992). Dissertation Abstracts Robinson, B. E. (1999). The Work Addiction Risk Test: Development
International, 53, 5401B. of a tentative measure of workaholism. Perceptual and Motor
Himmelfarb, S., & Murrell, S. A. (1983). Reliability and validity of Skills, 88, 199–210.
five mental health scales in older persons. Journal of Gerontol- Robinson, B. E., & Carroll, J. J. (1999). Assessing the offspring of
ogy, 38, 333–339. workaholic parents: The Children of Workaholics Screening Test.
Ishiyama, F. I., & Kitayama, A. (1994). Overwork and career-centered Perceptual and Motor Skills, 88, 1127–1134.
self-validation among the Japanese: Psychosocial issues and Robinson, B. E., Carroll, J. J., & Flowers, C. (2001). Marital es-
counseling implications. International Journal for the Advance- trangement, positive affect, and locus of control among spouses
ment of Counseling, 17, 167–182. of workaholics and spouses of nonworkaholics: A national study.
Jackson, D. L. (1992). Correlates of physical and emotional health The American Journal of Family Therapy, 29, 5, 397–410.
among male and female workaholics (Doctoral dissertation, Uni- Robinson, B. E., & Chase, N. (2000). High-performing families:
versity of Oregon, 1993). Dissertation Abstracts International, Cause, consequences, and clinical solution. Alexandria, VA:
53 (10), 5446. American Counseling Association.
Machlowitz, M. M. (1978). Determining the effects of workahol- Robinson, B. E., & Kelley, L. (1998). Adult children of workahol-
ism (Doctoral dissertation, Yale University, 1979). Dissertation ics: Self-concept, anxiety, depression, and locus of control. The
Abstracts International, 40 (01), 480. American Journal of Family Therapy, 26, 223–238.
Machlowitz, M. (1980). Workaholics: Living with them, working Robinson, B. E., & Phillips, B. (1995). Measuring workaholism:
with them. Reading, MA: Addison-Wesley. Content validity of the Work Addiction Risk Test. Psychological
Maslach, C., & Jackson, S. E. (1981). The measure of experienced Reports, 77, 657–658.
burnout. Journal of Occupational Behavior, 3, 99–113. Robinson, B. E., & Post, P. (1994). Validity of the Work Addiction
McMillan, J. J., & Northern, N. A. (1995). Organizational code- Risk Test. Perceptual and Motor Skills, 78, 337–338.
pendency: The creation and maintenance of closed systems. Robinson, B. E., & Post, P. (1995). Split-half reliability of the Work
Management Communication Quarterly, 9, 6–45. Addiction Risk Test: Development of a measure of workaholism.
Monk, M. (1981). Blood pressure awareness and psychological well- Psychological Reports, 76, 1226.
being in the Health and Nutrition Examination Survey. Clinical Robinson, B. E., Post, P., & Khakee, J. F. (1992). Test-retest reliability of
Investigative Medicine, 4, 183–189. the Work Addiction Risk Test. Perceptual and Motor Skills, 74, 926.

168 Journal of Counseling & Development  ■  Spring 2009  ■  Volume 87

Workaholism, Health, and Self-Acceptance

Robinson, J. P. (1973). Measures of social psychological attitudes. Spruell, G. (1987). Work fever. Training and Development Journal, 2, 47.
Ann Arbor: University of Michigan, Institute for Social Stevens, J. (1996). Applied multivariate statistics for the social sci-
Research. ences (3rd ed.). Mahwah, NJ: Erlbaum.
Schaufeli, W. B., Leiter, M. P., Maslach, C., & Jackson, S. E. (1996). Trueman, A. L. (1995). Socialization and workaholism: A lifestyle
The Maslach Burnout Inventory-General Survey. Palo Alto, CA: perspective (Master’s thesis, Dalhousie University, 1996). Dis-
Consulting Psychologists Press. sertation Abstracts International, 34 (04), 1700.
Spence, J. T., & Robbins, A. S. (1992). Workaholism: Definition, Wilkinson, L., & APA Task Force on Statistical Inference. (1999).
measurement, and preliminary results. Journal of Personality Statistical methods in psychology journals: Guidelines and ex-
Assessment, 58, 160–178. planations. American Psychologist, 54, 594–604.

Journal of Counseling & Development  ■  Spring 2009  ■  Volume 87 169