Journal of Youth and Adolescence, Vol. 35, No. 3, June 2006, pp.

311–319 (

C 2006)

DOI: 10.1007/s10964-006-9036-7

Characteristics of Adolescents Who Report

Very High Life Satisfaction

Rich Gilman1 and E. Scott Huebner2

Revised April 21, 2005; accepted May 3, 2005

Published online: 12 May 2006

This study investigated the characteristics of adolescents who report high levels of life global satis-
faction. A total of 485 adolescents completed the Students’ Life Satisfaction Scale (SLSS) (Huebner,
E. S. (1991). Sch. Psychol. Int. 12: 231–240.) along with self-report measures of intrapersonal, in-
terpersonal, and school-related functioning. Based on their SLSS scores, students were divided into
three groups: “low” (bottom 20% of the distribution), “average” (middle 50%), and “high” (upper
20%). Youth in the high satisfaction group reported significantly higher adaptive functioning on all
dependent variables than youth in the low satisfaction group. Relative to students with average life
satisfaction, students with high life satisfaction reported superior scores on a measure of social stress,
a measure of attitudes toward teachers, and on all measures of intrapersonal functioning. Also, no
adolescents in the high life satisfaction group demonstrated clinical levels of psychological symptoms,
whereas 7% of the average group and 42% of the low satisfaction group reported clinical levels of
symptoms. Taken together, the findings suggested that high life satisfaction is associated with some
mental health benefits that are not found among youth reporting comparatively lower satisfaction
levels.

KEY WORDS: adolescents; well-being; wife satisfaction; psychologist; psychoeducational.

INTRODUCTION wellness, SWB research can thus provide a valuable com-

plement to psychology’s traditional focus on illness and
Subjective well-being research (SWB) focuses on distress (Greenspoon and Saklofske, 2001).
how people evaluate their immediate and ongoing life The components of SWB are debated. Some re-
circumstances (Diener, 2000). Studies investigating SWB searchers conceptualize SWB as comprised of three re-
among youth have increased over the past decade, due to lated, but distinguishable components: positive affect,
calls to expand the definition of positive mental health to negative affect, and global life satisfaction (Lucas and
include more than simply the absence of psychopathol- Diener, 2000). Positive and negative affect refer to the
ogy (Seligman and Csikszentmihalyi, 2000). Given its fo- long-term frequencies of position emotions (e.g., joy,
cus on indicators that facilitate psychological health and pride) and negative emotions (e.g., anger, sadness) ex-
perienced by individuals. Global life satisfaction refers
1 Assistant Professor in the Department of Educational and Counseling to a cognitive appraisal of the overall quality of a per-
Psychology at the University of Kentucky. Research interests include son’s life, based on self-selected standards (Diener et al.,
positive well-being among youth, perfectionism, and peer neglect. He is
1999). Life satisfaction is viewed as a key component
a recent recipient of the APA Division 16 (School Psychology) award
for scholarly contribution. To whom correspondence should be ad- considering that it (a) transcends momentary emotional
dressed at Department of Educational and Counseling Psychology at fluctuations (Diener and Diener, 1996), (b) can influence
the University of Kentucky, Lexington, Kentucky 40506-0017. changes in behaviors (Lewinsohn et al., 1991), and (c) is
2 Professor in the Department of Psychology at the University of South
relatively free of social desirability bias (Diener, 1994).
Carolina. Research interests involve the conceptualization, measure-
Life satisfaction has also been viewed as construct wor-
ment, and implications of positive psychological well-being constructs
among youth. He is currently director of the School Psychology Pro- thy of specific attention because it provides a global in-
gram, and fellow of Division 16 of the APA and the International dex of well-being that is based on criteria determined by
Society for Quality of Life Studies.

311
0047-2891/06/0600-0311/0

C 2006 Springer Science+Business Media, Inc.
312 Gilman and Huebner

individuals rather than researchers (Diener, 2000). Frisch life satisfaction judgments may not only be important in
(1999) argues that life satisfaction measures should be and of themselves; they may also reflect or determine
integrated with symptom-oriented measures on a routine important adaptive outcomes.
basis in clinical and health psychology in order to provide However, very little research has specifically inves-
more comprehensive, universal outcome evaluations. tigated adolescents who report high levels of global life
Research has found that positive global life satisfac- satisfaction. Relative to the study of antecedents and/or
tion is normative among adults (Diener and Diener, 1996) determinants of individual differences in adolescent life
and adolescents (Huebner, 2004). Incorporating Lazarus’ satisfaction, there is a paucity of research focusing on
(1991) theory of coping as a conceptual framework, main- variables presumed to be correlates or consequences of
taining a positive outlook on life (i.e., a positive “set life satisfaction, including high levels of life satisfaction.
point”) may serve as a signal against the backdrop of Such studies are necessary to determine how youth re-
ongoing cognitive appraisals, affective states, and envi- porting a high global satisfaction set point fundamentally
ronmental circumstances. This set point may function as differ from individuals reporting more normative levels
a comparison standard against which potential negative of global satisfaction and the implications of such differ-
emotional and behavioral reactions are made (Headey and ences. Information obtained from such studies may help
Wearing, 1992). illuminate factors that should be tested in longitudinal
Although studies have shown that departures from studies as correlates or benefits associated with individ-
positive SWB levels correlate with a number of maladap- ual differences in adolescents’ life satisfaction. Further,
tive outcomes (e.g., Dear et al., 2002; Furr and Funder, although it is clear that positive life satisfaction judg-
1998), much less is known of potential psychological, ments confer some benefits for adults in general (Diener
psychoeducational, and psychosocial benefits related to and Seligman, 2002), most individuals report moderate
maintaining significantly high levels of well-being. In or “average” levels of life satisfaction (Cummins, 1998).
one of the few studies to directly investigate variables re- In this manner, moderate levels of life satisfaction may
lated to differing SWB levels, Diener and Seligman (2002) represent the optimal level necessary for adaptation. If
found that “very happy” adults (i.e., highest 10% of sev- such is the case, youth who report high life satisfaction
eral SWB measures) reported significantly less psycho- would garner no greater psychological, psychosocial, and
logical distress, more positive social relations, and were psychoeducational benefits than youth who report average
more extraverted than adults whose SWB scores were in satisfaction levels. If this possibility was empirically sup-
the extreme low end of the continuum. Similar findings ported, the findings would challenge calls to encourage
have also been reported in Friedman et al. (2002). the promotion of global life satisfaction to optimal levels
Child and youth life satisfaction research has focused for all individuals (Diener, 2000).
on the study of life satisfaction as an outcome variable, that The purpose of this study was thus to examine the
is, most studies have investigated assumed determinants characteristics of adolescents reporting high levels of
of individual differences in life satisfaction. Such stud- global life satisfaction. The characteristics investigated
ies have revealed a wide ranging network of associated included a wide range of school-related (e.g., perceived
variables (see Gilman and Huebner, 2003 for a review). school experiences, grade-point average, structured ex-
These variables include family, peer, neighborhood, and tracurricular activity participation), interpersonal (e.g., re-
self-related (e.g., personality, cognitive attributions), and lations with parents, social stress), and intrapersonal (e.g.,
activity (e.g., participation in structure extra curricular hope, depression, and anxiety) variables, all of which have
activities) variables. Demographic variables (e.g., gender, been related to life satisfaction in previous studies (see
socioecomic status) appear to play a very modest role Gilman and Huebner, 2003).
at best in child or youth global life satisfaction reports
(Huebner et al., 2000). Adolescent life satisfaction re-
ports have been correlated with a variety of risk behaviors
(e.g., risky sex behavior, alcohol and drug use) as well, Study Hypotheses
although the directionality of the relationships has not
been determined. Importantly, preliminary studies have 1. Similar to Diener and Seligman’s (2002) study
suggested that youth life satisfaction reports mediate the with adults, it was expected that high life satis-
relationship between stressful life events and internalizing faction would be beneficial for adolescents. Youth
behaviors (McKnight et al., 2002) and moderate the re- reporting such satisfaction levels would report
lationship between stressful life events and externalizing significantly higher levels of positive functioning
behaviors (Suldo and Huebner, 2004). Thus, adolescent on the academic, interpersonal, and intrapersonal
Life Satisfaction Levels and Adolescence 313

measures compared to youth reporting low levels peers, and the instruments were administered in counter-
of global life satisfaction. balanced order. Further, at least one teacher/administrator
2. Adolescents reporting high levels of life satisfac- and research assistant was assigned to a specific testing
tion would also report more positive academic, location in order to monitor the students’ behavior and to
interpersonal, and intrapersonal functioning than help answer questions. These strategies were designed to
youth reporting average levels of life satisfaction. reduce potential order and/or social desirability effects.
Finally, the first author was available to answer any stu-
dent questions.
METHOD

Sample Measures

Four hundred ninety students in two schools from Global Life Satisfaction
two separate districts in two Southeastern states partic-
ipated in the study. Both schools contained grades 6 The Students’ Life Satisfaction Scale (SSLS:
through 12. There were 83 sixth-graders, 72 seventh- Huebner, 1991) is a 7-item self report measure designed to
graders, 68 eighth-graders, 75 ninth-graders, 68 tenth- assess global life satisfaction for youth in grades 3 through
graders, 66 eleventh-graders, and 57 twelfth-graders. 12. Students rated their agreement to the items on a 6-point
Fifty-four percent of the sample was male, and the Likert scale (1: strongly disagree, 2: disagree, 3: slightly
sample was comprised of 87% Caucasian (versus 2% disagree, 4: slightly agree, 5: agree, 6: strongly agree).
Asian-American, 2% African-American, 3% Hispanic- Scoring of the SLSS involves reverse-keying negatively
American, and 6% “Other”) students. The mean age of worded items, adding all item responses, and dividing by
the sample was 14.45 (SD = 2.12). Approximately 4% of the total number of items. Higher scores denote higher
the sample was considered to represent the lower socio- global satisfaction. Previous studies using the SLSS have
economic status, as measured by free/reduced lunch eligi- demonstrated adequate psychometric properties, includ-
bility. Because this research was aimed at individual dif- ing coefficient alphas in the 0.80 range, evidence of
ferences in satisfaction among a general (i.e., non-clinical) convergent and discriminant validity, and support of the
sample of youth, students known to be diagnosed with underlying factor structure via exploratory and confirma-
learning and/or emotional disorders were excluded from tory analysis (see Gilman and Huebner, 2000). In this
the study. study, the internal consistency of the SLSS was 0.88. The
mean score was 4.78 (SD = 1.05), consistent with what
has been reported in previous research (McKnight et al.,
Procedure 2002).
The Behavior Assessment System for Children
Approval to collect data was secured from the uni- (BASC: Reynolds and Kamphaus, 1992) was used to as-
versity institutional review board, both school districts sess a wide range of academic, interpersonal, and intrap-
and from each school’s principal. Only those students ersonal dimensions. The BASC is a 186-item self-report
who returned a completed parent consent form and pro- instrument that contains four adaptive subscales (which
vided written assent were eligible to participate in the measure positive adjustment) and ten clinical subscales
study. Criterion for admission to one public school was (which measure maladjustment). Items comprising each
participation in university research projects, and thus all subscale are dichotomously score (yes/no) and are then
students participated in the study (N = 341). The other summed to yield a standardized T-score. All BASC pro-
public school (N = 320) did not have this admission cri- tocols were computer scored in this study, which yields
terion. One hundred and forty-nine students from this standardized T-scores for each subscale but does not com-
school returned both permission forms, yielding a 47% pute reliability estimates. Nevertheless, the mean scores of
consent rate. Preliminary analyses found no statistically all BASC subscales were similar to what is reported in the
significant mean score differences on any of the variables BASC manual. In addition, the BASC also contains three
across school setting. Thus, the data were combined as one validity indices to detect careless and specious respond-
group. ing. While the vast majority (99%) reported T-scores that
The eligible students from each school gathered in were within the acceptable range, protocols for five youth
their cafeteria and completed the instruments. All students had one or more of the validity indices in the Extreme
were instructed to sit at least two seats apart from their Caution.
314 Gilman and Huebner

In addition to select BASC variables (described be- Interpersonal Mental Health Indicators
low), other variables in the study were as follows:
The Parent Relations and the Interpersonal Relations
subscales of the BASC (12 and 18 items, respectively)
Academic Variables were used to assess the perceived quality of interactions
with respect to parents, peers, and adults, while the So-
Grade Point Average (GPA) cial Stress subscale (18 items) was used to assess the
degree of chronic stress that the youth experiences when
Cumulative GPA was obtained by asking students interacting with others. Both the Interpersonal Relations
to report their GPA since their entry into either middle and Parent Relations are adaptive subscales, with higher
school or high school. Mean GPA across all grades was T-scores denoting greater satisfaction in their relation-
computed because it was believed to be more reliable ships with others. The Social Stress subscale is a clinical
than GPAs obtained from one or two selective courses. indicator, with higher T-scores indicating greater levels
Self-reported GPAs ranged from 1.20 to 4.0 (M = 3.36, of tension in relating to others. Coefficient alphas are
SD = 0.58). Analyses of the distribution of scores revealed adequate, with internal consistencies ranging from 0.77
a slight negative skew, although the skew and kurtosis of (Parent Relations) to 0.88 (Social Stress) for adolescents
the distribution did not significantly depart from normal. (Reynolds and Kamphaus, 1992). The mean score for the
Although district policy precluded gathering actual GPA, Parent Relations subscale was 49.37 (SD = 10.36) and
self-reported GPA was almost identical to each school’s 51.45 (SD = 9.06) for the Interpersonal Relations sub-
grade-point distribution. scale. The mean score for the Social Stress subscale was
48.58 (SD = 9.14).

Structured Extracurricular Activities (SEAs)

Intrapersonal Mental Health Indicators
Students were asked to list the specific extracurric-
ular activities that they have participated in since their Hope
enrollment in either middle school or high school. Based
upon previous SEA definitions (Mahoney, 2000), only The Children’s Hope Scale (Snyder et al., 1997) was
those activities that contained some semblance of struc- used to assess youths’ capacity to (a) develop specific
ture and active participation were included in the anal- strategies to reach personal goals (pathways) and (b) ini-
yses. Passive activities such as television viewing were tiate and sustain the motivation for using those strate-
excluded. All extracurricular activities were coded as sim- gies (agency). Although the majority of hope research
ple frequency counts and analyzed independently. The has utilized adult samples, recent findings have demon-
total number (678), and the median number of SEAs strated the unique contribution of hope to positive mental
(3.00, SD = 2.33) was consistent with previous research health among youth, including its theoretical relationship
(Gilman, 2001). with life satisfaction (Snyder et al., 2003). The scale con-
sists of six items based on a 7-point Likert scale, with
three items each comprising the Pathways and Agency
Negative Perceived School Experiences subscales. Scores range from 6 to 42 for each item. All
items are summed to yield a mean Total score. Because
The BASC Attitudes to School subscale (12 items) the pathways and agency components are necessary to
and the Attitudes to Teachers subscale (13 items) were sustain successful goal pursuit (Irving et al., 1998), only
administered to assess youths’ general opinion of their the Total hope score was analyzed in this study. Previous
school experiences and quality of teachers and/or school studies with the Children’s Hope Scale revealed adequate
administrators. Both subscales measure school maladjust- psychometric properties, including good internal consis-
ment, and thus higher T-scores denote greater negative tencies ranging between 0.72 and 0.86 for the subscale
experiences with the schooling process and/or school per- and Total scale, and evidence of convergent and discrimi-
sonnel. The BASC manual reports internal consistency nant validity (see Snyder et al., 2003 for a review). In this
estimates ranging from 0.79 to 0.82 for adolescents. The study, the coefficient alpha for the Total hope score was
mean score for the Attitudes to School subscale in this 0.88, and the mean score was 26.29 (SD = 5.59).
study was 51.72 (SD = 9.88) and 49.87 (SD = 9.68) for Self-Esteem, Locus of Control, Depression and Anx-
the Attitudes to Teachers subscale. iety were all measured via select BASC subscales. The
Life Satisfaction Levels and Adolescence 315

Table I. Intercorrelations of all Variables in the Study

1 2 3 4 5 6 7 8 9 10 11 12 13
1. Global —
satisfaction
2. Attitude to −0.41∗∗ —
school
3. Attitude to −0.33∗∗ 0.58∗∗ —
teachers
4. GPA 0.32∗∗ −0.32∗∗ −0.32∗∗ —
5. SEAs 0.17∗∗ −0.27∗∗ −0.27∗∗ 0.41∗∗ —
6. Social stress −0.58∗∗ 0.37∗∗ 0.25∗∗ −0.11∗ 0.01 —
7. Interpersonal 0.37∗∗ −0.37∗∗ −0.12∗ 0.05 0.15∗ −0.52∗∗ —
relations
8. Parent 0.52∗∗ −0.43∗∗ −0.44∗∗ 0.29∗∗ −0.13∗ −0.49∗∗ 0.27∗∗ —
relations
9. Self-esteem 0.49∗∗ −0.29∗∗ −0.18∗∗ 0.15∗ 0.00 −0.59∗∗ 0.39∗∗ 0.39∗∗ —
10. Hope 0.67∗∗ −0.43∗∗ −0.26∗∗ 0.38∗∗ 0.22∗∗ −0.43∗∗ 0.40∗∗ 0.43∗∗ 0.42∗∗ —
11. Anxiety −0.34∗∗ 0.19∗∗ 0.20∗∗ −0.01 0.08 0.57∗∗ −0.13∗ −0.24∗∗ −0.43∗∗ −0.26∗∗ —
12. Depression −0.62∗∗ 0.47∗∗ 0.31∗∗ −0.24∗∗ −0.12∗ 0.67∗∗ −0.42∗∗ −0.55∗∗ −0.58∗∗ −0.52∗∗ 0.40∗∗ —
13. Locus of −0.51∗∗ 0.43∗∗ 0.46∗∗ −0.16∗ −0.14∗ 0.62∗∗ −0.23∗∗ −0.55∗∗ −0.38∗∗ −0.43∗∗ 0.49∗ 0.60∗∗ —
control
Note. SEAs = structured extracurricular activities, GPA = self-reported grade point average.
∗ p < 0.05; ∗∗ p < 0.01.

Self-Esteem subscale (8 items) assesses both physical and study, with the skewness and kurtosis all within accept-
global characteristics of self-esteem. The Locus of Con- able limits (i.e., less than 2.0; Lomax, 2001). Pearson
trol subscale (16 items) assesses the degree to which youth bivariate correlations and descriptive statistics were then
believe that events are controlled by external events or computed for all measures. As shown in Table I, the in-
others. Both the Depression (19 items) and Anxiety tercorrelations were significant but also moderate, with
(20 items) subscales are designed to measure chronic global satisfaction significantly and positively related to
levels of each construct. The Self-Esteem subscale is an all BASC adaptive subscales (i.e., Interpersonal Relations,
indicator of positive adjustment, while the other three Parent Relations, Self-Esteem) as well as hope, SEA par-
subscales are clinical subscales. The BASC manual re- ticipation and GPA. Global satisfaction was also a signif-
ports that the alpha coefficients range from 0.81 (Locus icant negative but moderate correlate with various forms
of Control) to 0.89 (Depression) among adolescent youth of psychological and interpersonal distress. Such findings
(Reynolds and Kamphaus, 1992). The mean scores indicate that the variables in the study, although sharing
for the four subscales were: Self-Esteem (M = 49.19, some degree of variance, nevertheless assess separable
SD = 8.91), Locus of Control (M = 50.13, SD = 9.88), constructs (see Stevens, 1996).
Depression (M = 49.24, SD = 7.77), and Anxiety Similar to previous studies that have coded unidi-
(M = 48.31, SD = 9.74). mensional SWB ratings into discrete categories (Irving
et al., 1998), the highest and lowest 20% of the SLSS
global mean score distribution was selected to reflect the
Designation of Life Satisfaction Groups “high” and “low” end of the satisfaction continuum. In
comparison with these groups, youth from the middle
Examination of the scoring distribution of all mea-
50th percentile of the satisfaction distribution were se-
sures was conducted prior to group placement to assess for
lected as reporting “average” global satisfaction. Eighty-
outliers and to test for multivariate normality. All scores
eight youth (46 males, 42 females) were placed in the
were first transformed to standardized z-scores. Based on
low satisfaction group, 252 youth (140 males, 112 fe-
recommendations by Stevens (1996), z-scores greater than
males) were placed in the average satisfaction group, and
± 4.00 were excluded from the data. Five youth had scores
98 youth (52 males, 46 females) were placed in the high
that exceeded this criterion. These youth were the same
satisfaction group. The mean (and standard deviation) of
youth who had one or more BASC validity indices in the
the global satisfaction score for each category was: Low
Extreme Caution range and thus these youth were ex-
(M = 2.85, SD = 0.69), Average (M = 4.61, SD = 0.40),
cluded from further analyses. There was no significant
and High (M = 5.72, SD = 0.21). No significant gender
departure from normality on any of the variables in the
316 Gilman and Huebner

placement difference was found across the satisfaction were in the expected direction and were moderate at
groups, Kruskal–Wallis = 0.36, p > 0.05. It should be best. The highest degree of shared variance was found
noted that not every participant fully completed the BASC, on the Total hope score for all three satisfaction groups,
which resulted in a listwise deletion of some items and ren- with these correlations significant (at the 0.05 level) and
dering an incomplete score on the corresponding subtests. ranging from 0.35 (average satisfaction group) to 0.47
The BASC data were analyzed independently to account (for both the low and high satisfaction groups). In ad-
for the lower sample size, as the number of students in the dition to the hope construct, correlations among the low
low satisfaction group was reduced to 72 students in the satisfaction group ranged from 0.13 (Interpersonal Rela-
low satisfaction group (37 males, 35 females), 225 youth tions) to − 0.45 (Depression). Among the average satis-
in the average satisfaction group (129 males, 96 females), faction group, correlations ranged from − 0.02 (Anxiety)
and 87 youth in the high satisfaction group (44 males, to − 0.25 (Social Stress), and the correlations for the high
43 females). satisfaction group ranged from .01 (GPA) to − 0.23 (At-
A 2 (gender) × 3 (satisfaction groups) multivariate titude to School). These findings suggest that although
analyses of variance (MANOVA) was conducted to deter- there was some degree of overlap, the associations were
mine response differences across all dependent variables not of a magnitude to warrant concern that the dependent
with respect to gender. There was no significant gender variables were assessing the same construct. Further, the
effect, Pillai’s Trace = 0.08, F(12, 346) = 0.28, p > finding that the correlations were in the expected direc-
0.05, nor was there a significant gender-by-satisfaction tions across all three satisfaction groups provides support
group interaction, Pillai’s Trace = 0.02, F(12, 346) = for the sensitivity of global satisfaction to assess various
0.30, p > 0.05, indicating that males and females re- dimensions at different levels.
ported equivalent levels of each dependent variable within Two separate MANOVAs were conducted to account
each satisfaction group. A separate 8 (grade) × 3 (satisfac- for the different numbers of youth who completed the
tion groups) MANOVA found a significant grade effect, instruments. In the first analysis, the three life satisfac-
Pillai’s Trace = 0.55, F(12, 333) = 2.41, p < 0.05, tion groups served as the independent variables and the
although the multivariate effect size (ή2 ) was consid- BASC subscales served as the dependent variables. A
ered small (0.08). With the significance level Bonferroni- separate MANOVA was conducted, with the global sat-
corrected to .004 (0.05/12) to control for Type I errors, isfaction groups again serving as the independent vari-
follow-up univariate tests revealed that SEA participation able and hope, GPA and SEAs serving as the dependent
was the only significant finding, F(7, 361) = 3.16, p < variables. For the BASC variables, a significant main ef-
0.004, with youth in the highest grades (i.e., grades 11 fect was found, Pillai’s Trace = 0.48, F(9, 374) = 13.43,
and 12) reporting that they participated in more SEAs p < 0.001, η2 = 0.24. A significant main effect was also
than the youth in the earliest grades represented in the found for the remaining variables in the study, Pillai’s
study (i.e., grades 6 and 7) (M = 4.63, SD = 0.41, and Trace = 0.37, F(3, 407) = 30.87, p < 0.001, η2 = 0.19.
M = 2.51, SD = 0.38, respectively). Although there was All variables were then categorized and labeled as
a statistically significant grade-by-satisfaction group in- school, interpersonal, or intrapersonal variables to facili-
teraction, Pillai’s Trace = 0.88, F(14, 338) = 1.91, p < tate ease of interpretation, and analysis of each univariate
0.05, ή2 = 0.07, no significant univariate F-test was found F-test was Bonferroni corrected to 0.003 to correct for
after again correcting the alpha to .004. In general, con- Type I errors. Table II reports the means and standard
sidering that (a) there were no gender differences, (b) deviations. On every measure in the study, youth report-
the multivariate effect sizes for the significant grade and ing high global satisfaction reported significantly higher
grade-by-satisfaction group results were quite small, and scores than youth reporting low global satisfaction. The
(c) there were no grade differences across the vast major- strongest magnitude of difference was found on depres-
ity of variables, the data were analyzed collectively. sion and hope. In addition, relative to youth reporting
average global life satisfaction, youth in the high satisfac-
tion group reported significantly higher scores on mea-
RESULTS sures of hope, self-esteem, and (internal) locus of control
as well as lower scores on measures of social stress, anx-
Intercorrelations calculated separately for each sat- iety, depression, and (negative) attitudes toward teachers.
isfaction group found that with the exception of two non- Students reporting average global satisfaction were simi-
significant correlations specific to the high satisfaction lar to youth in the high satisfaction group on measures of
group (Parent Relations, r = − 0.11 and Self-Esteem, general attitudes towards school, GPA, SEA involvement,
r = − 0.09), all correlations with global satisfaction parent relations, and interpersonal relations. Finally, the
Life Satisfaction Levels and Adolescence 317

Table II. Means (and Standard Deviations) of the Three Global Satisfaction Groups on School, Interpersonal, and Intrapersonal Variables
Low Average High F Sig. ES
School variables
Attitude to school 58.58 (9.54)a 50.67 (9.17)b 47.66 (8.52)b 30.42 0.001 0.14
Attitude to teachers 55.90 (8.64)a 49.43 (9.76)b 46.14 (8.05)c 22.90 0.001 0.11
GPA 3.01 (0.65)a 3.42 (0.54)b 3.49 (0.49)b 19.44 0.001 0.08
SEAs 2.37 (1.76)a 3.48 (2.34)b 3.31 (2.09)b 7.53 0.001 0.04
Interpersonal variables
Social stress 56.56 (9.80)a 47.55 (7.92)b 42.72 (4.58)c 64.83 0.001 0.25
Parent relations 38.50 (11.76)a 51.30 (8.68)b 54.05 (5.44)b 73.25 0.001 0.28
Interpersonal relations 47.15 (11.13)a 52.40 (7.27)b 54.36 (4.05)b 19.08 0.001 0.09
Intrapersonal variables
Self-esteem 42.11 (9.00)a 50.97 (8.14)b 54.09 (5.90)c 49.60 0.001 0.21
Hope 20.46 (4.84)a 27.02 (4.49)b 30.25 (3.51)c 111.61 0.001 0.35
Anxiety 52.43 (9.65)a 48.31 (9.34)b 42.95 (7.75)c 22.13 0.001 0.10
Depression 59.55 (10.63)a 47.81 (5.65)b 45.13 (3.06)c 113.97 0.001 0.37
Locus of control 58.30 (10.24)a 49.18 (9.04)b 44.41 (5.79)c 51.98 0.001 0.21
Note: df for each univariate F-test is 2, 383 for the BASC scales and 2, 410 for grade-point average (GPA), structured extracurricular activities (SEAs),
and hope. Mean scores not sharing the same subscript are significant at the 0.01 level.

scores of youth with average global life satisfaction were demic, interpersonal, and intrapersonal functioning than
significantly different from those for the youth in the low youth reporting low life satisfaction. Youth reporting high
satisfaction group on all measures, with scores reported global satisfaction reported more positive relationships
by the low satisfaction group reflecting more maladaptive with others (including peers and parents), less intraper-
functioning. sonal distress (such as anxiety and depression), higher
Although mean scores reported in the above anal- levels of hope, and a greater sense of personal control
yses revealed significant differences across global satis- than youth reporting low global satisfaction. Such findings
faction groups, they may conceal the number of youth in suggest that high levels of life satisfaction and various in-
the high or average satisfaction groups who may have re- dices of positive behavioral and psychological adjustment
ported elevated T-scores that would be considered “clin- are interrelated. The present findings also parallel previ-
ically” significant. Analysis of the number of youth re- ous studies investigating global satisfaction levels among
porting clinically significant T-scores may provide an- adults (Dear et al., 2002; Diener and Seligman, 2002),
other perspective with which to investigate the psycho- suggesting that this interrelationship begins as early as
logical and behavioral ramifications of having high (ver- adolescence. Further, and with respect to school variables,
sus low and average) levels of global satisfaction. For students with high life satisfaction reported more positive
this analysis, the BASC Social Stress, Depression and school experiences, a greater frequency of SEA participa-
Anxiety subscales (referred to as the SAD triad) were tion, and higher GPAs than students with low satisfaction.
used considering that elevated T-scores (65 or higher) These findings reveal that high life satisfaction is associ-
on all three subscales indicate “severe emotional distur- ated with positive academic experiences and also suggest
bance characterized by depression with great tension and important conceptual connections between life satisfac-
severe, acute distress” (Reynolds and Kamphaus, 1992, tion and school context factors, not suggested by other
p. 64). Sixteen youth (7%) in the average satisfaction studies (e.g., Huebner, 1991). Thus, future researchers
group and 30 youth (42%) in the low satisfaction group may wish to expand the range of school-related variables
scored in the clinically significant range on the SAD triad. that are investigated in studies of life satisfaction in youth
In comparison, no youth from the high satisfaction group (e.g., teacher attitudes, children’s school identification,
reported clinically significant T-scores on the triad. academic engagement) (see Baker et al., 2003).
Results of this study also support the importance of
assessing the full range of life satisfaction judgments, in-
DISCUSSION cluding distinctions between average and higher levels
of life satisfaction. As hypothesized, youth who reported
As expected, the results of this study demonstrated high satisfaction also reported significantly less social
that youth reporting high global life satisfaction also re- stress and intrapersonal distress, and significantly more
ported significantly higher scores on all measures of aca- positive perceptions of teachers than youth who reported
318 Gilman and Huebner

average satisfaction levels. These findings suggest that reporting higher levels of life satisfaction. In addition,
there may be “critical cutoffs” related to global life sat- a unique finding of this study was that high global life
isfaction that are associated with particular psychological satisfaction was associated with greater adaptation across
and behavioral benefits that are not found among youth re- various facets of academic, interpersonal, and intraper-
porting average or normative levels of global satisfaction. sonal functioning than comparatively lower satisfaction
Further support was provided by the differential frequen- levels, including even average satisfaction. This finding
cies of students who self-reported “clinically significant” is consistent with studies that have investigated various
levels of behavior problems. levels of other SWB constructs, including hope (Snyder,
Although the study yielded a number of important re- 2002), positive affect (Friedman et al., 2002), and hap-
sults, limitations are to be noted. First, the study was cross- piness (Diener and Seligman, 2002) and their relation-
sectional, which precluded determining the direction of ship with mental health indicators. Taken together, these
the relationship between life satisfaction and the depen- findings suggest further support for the notion that rel-
dent measures in this study. It may be that life satisfaction atively high levels of global life satisfaction may serve
influences interpersonal, intrapersonal, and/or academic as a key psychological strength towards optimal human
behaviors. Nevertheless, it may also be the case that inter- functioning.
personal, intrapersonal, and/or academic behaviors influ-
ence the levels of life satisfaction of these youth. Studies
of an experimental and/or longitudinal nature are neces-
REFERENCES
sary to determine the direction of causality. Second, the
data were based on self-reports, which may yield method
Anderman, E. M., and Johnston, J. (1998). Television news in the class-
variance concerns. Third, some youth may have been less room: What are adolescents learning? J. Adolesc. Res. 13: 73–100.
than honest in their responses. Although the study at- Baker, J. A., Dilly, L. J., Aupperlee, J. L., and Patil, S. A. (2003).
tempted to control for the latter concern by including only The development of school satisfaction: Schools as psychologically
healthy environments. Sch. Psychol. Quart. 18: 206–221.
BASC protocols with acceptable validity indices, it could Cummins, R. A. (1998). The second approximation to an international
not account for some youth who may have maintained standard for life satisfaction. Soc. Indic. Res. 43: 307–334.
a “defensive denial of distress” (Friedman et al., 2002, Dear, K., Henderson, S., and Korten, A. (2002). Well-being in Australia:
Findings from the National Survey of Mental Health and Well-Being.
p. 357). Using multiple raters (e.g., parents, peers) and Soc. Psychiatr. Psychiatr. Epidemiol. 37: 503–509.
methods (e.g., physiological data) is the necessary next Diener, E. (1994). Assessing subjective well-being: Progress and oppor-
step in this line of research. Finally, self-reported GPA tunities. Soc. Indic. Res. 31: 103–157.
Diener, E. (2000). Subjective well-being: The science of happiness and
was collected, which again may introduce the possibil- a proposal for a national index. Am. Psychol. 55: 34–43.
ity of response bias. Although (a) the procedure used to Diener, E., and Diener, M. (1996). Most people are happy. Psychol. Sci.
obtain the data was quite controlled, (b) the skew and kur- 7: 181–185.
Diener, E., and Seligman, M. E. P. (2002). Very happy people. Psychol.
tosis of self-reported GPA were within the normal range Sci. 13: 81–84.
of score distributions, (c) self-reported GPA has been Diener, E., Suh, E., Oishi, S., Lucas, R. E., and Smith, H. L. (1999).
shown to be a reliable and valid measure in other studies Subjective well-being: Three decades of progress. Psychol. Bull.,
125: 276–302.
(Anderman and Johnston, 1998), and (d) the data obtained Friedman, E. T., Schwartz, R. M., and Haaga, D. A. F. (2002). Are the
were identical to the actual GPA distribution from each very happy too happy? J. Happiness Stud. 3: 355–372.
school, obtaining permanent products (e.g., actual GPA) Frisch, M. B. (1999). Quality of life assessment/intervention and the
Quality of Life Inventory (QOLI). In Maruish, M. R. (ed.), The Use of
is necessary to corroborate these findings. Finally, the Psychological Testing for Treatment Planning and Outcome Assess-
results of this study were based on a sample that was pri- ment (2nd edn.). Lawrence Erlbaum, Hillsdale, NJ, pp. 1227–1331.
marily composed of Caucasian students from the USA. Furr, R. M., and Funder, D. (1998). A multimodal analysis of personal
negativity. J. Pers. Soc. Psychol. 74: 1580–1591.
Additional studies with students from different racial and Gilman, R. (2001). Relationship between life satisfaction, social inter-
cultural backgrounds are necessary to determine the gen- est, and participation in extracurricular activities among adolescent
eralizability of the findings. students. J. Adolesc. 30: 749–767.
Gilman, R., and Huebner, E. S. (2000). Review of life satisfaction mea-
This study reflected key strengths as well. Most sures for adolescents. Behav. Change 17: 178–195.
notable was the significant response differences across Gilman, R., and Huebner, E. S. (2003). A review of life satisfaction
various school, interpersonal and intrapersonal variables research with children and adolescents. Sch. Psychol. Rev. 18: 192–
205.
by students reporting different levels of global satis- Greenspoon, P. J., and Saklofske, D. H. (2001). Toward an integration
faction. Consistent with previous findings (Gilman and of subjective well-being and psychopathology. Soc. Indicators Res.
Huebner, 2003), youth reporting low life satisfaction re- 54: 81–108.
Headey, B., and Wearing, A. (1992). Understanding Happiness: A
ported higher levels of interpersonal and intrapersonal dis- Theory of Subjective Well-Being. Longman Cheshire, Melbourne,
tress, and less positive academic experiences than youth Victoria, Australia.
Life Satisfaction Levels and Adolescence 319

Huebner, E. S. (1991). Initial development of the Students’ Life Satis- McKnight, C., Huebner, E. S., and Suldo, S. M. (2002). Relation-
faction Scale. Sch. Psychol. Int. 12: 231–240. ships among stressful life events, temperament, problem behavior,
Huebner, E. S. (2004). Research on assessment of life satisfaction of and global life satisfaction in adolescents. Psychol. Sch. 39: 677–
children and adolescents. Soc. Indicators Res. 66: 3–33. 687.
Huebner, E. S., Drane, W., and Valois, R. F. (2000). Levels and demo- Reynolds, C. R., and Kamphaus, R. W. (1992). Manual for the Behav-
graphic correlates of adolescent life satisfaction reports. Sch. Psychol. ioral Assessment System for Children. American Guidance Service,
Int. 21: 281–292. Circle Pines, MN.
Irving, L. M., Snyder, C. R., and Crowson, J. J. (1998). Hope and coping Seligman, M. E. P., and Csikszentmihalyi, M. (2000). Positive psychol-
with cancer by college women. J. Pers. 66: 195–214. ogy: An introduction. Am. Psychol. 55: 5–14.
Lazarus, R. S. (1991). Emotion and Adaptation. Oxford University Press, Snyder, C. R. (2002). Hope theory: Rainbows in the mind. Psychol.
New York. Inquiry 13: 249–275.
Lewinsohn, P., Redner, J., and Seeley, J. (1991). The relationship be- Snyder, C. R., Lopez, S. J., Shorey, H. S., Rand, K. L., and Feldman,
tween life satisfaction and psychosocial variables: New perspectives. D. B. (2003) Hope theory, measurements, and applications to school
In Strack, F., Argyle, M., and Schwartz, N. (eds.), Subjective Well- psychology. Sch. Psychol. Quart. 18: 122–139.
Being. Plenum Press, New York, pp. 193–212. Snyder, C. R., Hoza, B., Pelham, W. E., Rapoff, M., Ware, L., Danovsky,
Lomax, R. G. (2001). An Introduction to Statistical Concepts for M., Highberger, L., Rubinstein, H., and Stahl, K. J. (1997). The
Education and Behavioral Sciences. Lawrence Erlbaum, Mahwah, development and validation of the Children’s Hope Scale. J. Pediatr.
NJ. Psychol. 22: 399–421.
Lucas, R. E., and Diener, E. (2000). Subjective emotional well-being. In Stevens, J. (1996). Applied Multivariate Statistics for the Social Sciences
Lewis, M., and Haviland, J. M. (eds.), Handbook of Emotions (2nd (3rd edn.). Lawrence Erlbaum, Mahwah, NJ.
edn.). Guilford Publications, New York, pp. 325–337. Suldo, S. M., and Huebner, E. S. (2004). Does life satisfaction
Mahoney, J. L. (2000). School extracurricular activity participation as a moderate the effects of stressful life events on psychopatholog-
moderator in the development of antisocial patterns. Child Dev. 71: ical behaviors during adolescence? Sch. Psychol. Quart. 19: 93–
502–516. 105.