2012. Vol.3, No.6, 447-453
Published Online June 2012 in SciRes (
Copyright © 2012 SciRes. 447
Psychological Well-Being, Self-Reported Physical Activity
Levels, and Attitudes to Physical Activity in a Sample of
New Zealand Adolescent Females
Daniel Shepherd*, Chris Krägeloh, Clare Ryan, Grant Schofield
School of Public Health and Psychosocial Studies, Faculty of Health and Environmental Sciences,
Auckland University of Technology, Auckland, New Zealand
Email: {*daniel.shepherd, chris.krageloh, clarya07, grant.schofield}
Received January 26th, 2012; revised February 27th, 2012; accepted April 1st, 2012
Physical activity (PA) is a key component of healthy development, not only physically but also psycho-
logically. The aim of the present study was to measure PA levels and psychological well-being in adoles-
cent females using a cross-sectional design, and to investigate the relationship between the two. Psycho-
logical well-being (self esteem and lack of depression, anxiety, and stress), PA, and established predictors
of PA from the Theory of Reasoned Action (Fishbein & Ajzen, 1975; Madden, Ellen, & Ajzen, 1992;
health consciousness, significant others, priority, perceived barriers, and attitudes) were measured using
148 adolescent females aged 16 to 18 years. Results show a link between depression and level of PA, and
between anxiety and PA. Attitudes towards PA, priority of PA, and perceived barriers to PA were also re-
lated to levels of PA. However, there were no significant associations between psychological well-being
and attitudes towards PA, even though psychological well-being may influence the actual level of activity.
Positive associations between PA and psychological well-being in adolescent females encourage future
studies into causal relationships between the two. The most effective strategies for increasing PA in mid-
dle to late adolescent females may be targeting perceived barriers to PA rather than attitudes.
Keywords: Adolescent Health; Physical Activity; Psychological Well-Being; Attitudes; Barriers; Theory
of Reasoned Action
Existing data indicate good levels of physical activity (PA)
by most New Zealand children, though there is a definite reduc-
tion in participation in adolescence (defined here as 12 to 18
years old), especially for females (Riddoch & Boreham, 1995;
Sallis, Prochaska, & Taylor, 2000; Schofield, Mummery, Scho-
field, & Hopkins, 2007). During mid to late adolescence, fe-
males are at increased risk of inactive lifestyles when compared
with male youth (Hohepa, Schofield, & Kolt, 2004; Schofield
et al., 2007; SPARC, 2003). This is concerning as adolescence
is an important age for girls to stay active, and PA levels often
track into adulthood (Kahn et al., 2008; Robbins, Sikorskii,
Hamel, Wu, & Wilbur, 2009; Schofield et al., 2007). Develop-
mentally, adolescence is characterized by rapid psychological,
psychosocial and physical change. The onset of puberty coin-
cides with unprecedented social demands which often translate
into significant psychological stressors, and females are more
likely to have a more negative body image and lower self-
esteem than males. PA may buffer the negative impacts of pu-
berty, and sustain psychological and physical well-being during
and beyond adolescence (Dugan, 2008).
Increasing PA in adolescent females would lead to better
population health, lower health care costs, and a reduction in
social inequalities in health (Dugan, 2008). To this end sub-
stantial effort has been directed into identifying the barriers that
adolescents face when attempting to undertake PA (Robbins et
al., 2009). Barriers to PA reported by adolescent New Zealand
females included a dislike for performing activities in front of
other class members singularly, perceived incompetence, and
perceptions of peer judgment, all of which inhibit involvement
in physical education classes and attending trials for sport
teams (Hohepa, Schofield, & Kolt, 2006). Additionally, re-
search on social influences and adolescent PA, although some-
what equivocal, suggests that coaches, teachers, siblings, par-
ents, and peers all play a role in shaping PA levels during ado-
lescence. For example, friends may be an important influence
(Seabra, Mendonça, Thomis, Peters, & Maia, 2008), especially
among adolescent girls (Schofield et al., 2007).
The impact of PA on adolescent psychological well-being
has yet to be elucidated. Azar, Ball, Salmon, and Cleland (2008)
reviewed the effects of PA intervention studies on depression in
young women between the ages of 18 and 35, and found that
women who participated in PA were less likely to report de-
pressive symptoms. Additionally, they found that any amount
of PA was associated with fewer depressive symptoms when
compared with no PA (Azar et al., 2008), supporting arguments
against a dose-response relationship between PA and affective
state (Dunn, Madhukak, & O’Neal, 2001).
The formation of positive attitudes towards PA during ado-
lescence may be an important step towards their actual partici-
pation in activity. Smoll and Schutz (1980) reported that atti-
tudes towards PA explained 20% of variance of PA in a sample
of 16- to 18-year old youth. Godin and Shephard (1984) found
that the attitudes of 13- to 15-year-olds towards PA, their pre-
vious experience with PA as such, and their current habits ex-
*Corresponding autho
plained almost 50% of variance of participation in PA. How-
ever, others (e.g., Thompson & Humbert, 2003) appeal for
more research into understanding the attitudes and behaviors of
adolescents regarding PA. A useful framework here is the The-
ory of Reasoned Action, a theory of attitude-behavior relation-
ships that links attitudes, subjective norms (akin to perceived
social pressure), and behavioral intentions to volitional beha-
vior (Fishbein & Ajzen, 1975; Madden, Ellen, & Ajzen, 1992),
which has also been successfully applied to PA (Biddle &
Mutrie, 2008). Volitional behavior is posited to result from
behavioral intentions, which in turn are said to arise out of a
combination of a person’s attitude toward performing the be-
havior in question and the perceived social pressure (subjective
norm) put upon them to perform that behavior (Sparks, Guthrie,
& Shepherd, 1997).
There have been few studies investigating the association
between PA and psychological well-being in adolescent fe-
males (Azar et al., 2008). The present exploratory study exam-
ined whether levels of PA in a sample of New Zealand adoles-
cent girls are associated with indicators of psychological well-
being (i.e., depression, anxiety, stress, and self-esteem). Addi-
tionally, known predictors of PA in adults, including attitudes
to PA, the influence of others, other preferred activities, per-
ceived barriers, and health consciousness, were also examined
within the framework of the theory of reasoned action (Fishbein
& Ajzen, 1975; Madden, Ellen, & Ajzen, 1992).
Participants were 148 females, aged between 16 to 18 years
old, attending a New Zealand all-girls secondary school in
Auckland, New Zealand. The school principal was contacted by
the researchers and gave permission to conduct this study at the
school. Participants were not offered any incentive to partici-
pate, and assent was obtained from each prior to the research
commencing. Questionnaires were distributed during class time,
and participants were asked to read a participant information
sheet prior to completing the questionnaires, which included
definitions of PA. To minimize effects of peer pressure, indi-
viduals were asked to spread out as much as possible across the
classroom and were asked to complete the questionnaires
without consultation with their peers. The protocol of the study
was approved by the Auckland University of Technology Ethics
Self-Report Physical Activity. The New Zealand Physical
Activity Questionnaire-Short Form (NZPAQ-SF; Mackay,
Schofield, & Schluter, 2007; SPARC, 2004) is a paper-and-pen
questionnaire that consists of eight items designed to assess le-
vels of PA in the New Zealand population. The NZPAQ-SF re-
turns a final PA score and categorizes individuals according to
one of three PA groups: relatively i nactiv e (<2.5 hours of activ-
ity per week: PAQ1); relatively active (2.5 - 4.9 hours of activ-
ity per week: PAQ2); and highly active (5+ hours of activity per
week: PAQ3).
Psychological Well-being. The Depression, Anxiety, Stress
Scale-42 (DASS-42; Lovibond & Lovibond, 1995) consists of
42 items, divided into three subscales: depression, anxiety, or
stress. The depression subscale corresponds closely to the ac-
cepted criteria for mood disorders, while the anxiety subscale
corresponds most closely to the symptom criteria for the vari-
ous anxiety disorders, with the exception of generalized anxiety
disorder (Lovibond & Lovibond, 1995). The DASS-42 consists
of statements to which participants respond using a four-point
Likert scale, ranging from 0 (Did not apply to me at all) to 3
(Applied to me very much, or most of the time). Additionally,
Rosenberg’s Self-Esteem Scale (RSES; Rosenberg, 1965), a
brief, ten-item, unidimensional measure of global self-esteem,
was administered. Responses to the statements contained in the
RSES are made on a four-point scale (0 - 3), ranging from
strongly agree to strongly disagree. Total scores computed
from the RSES range from 0 to 30, with higher scores indicat-
ing higher self-esteem.
Attitudes. The Attitudes Towards Physical Activity Scale
(ATPAS; SPARC, 2003) contains 46 statements scored using a
five-point Likert scale, and measures influences on PA. The
ATPAS probes factors that have been identified to influence
PA behavior, and includes subscales representing health con-
sciousness, attitudes towards PA, significant others, priority of,
and barriers to PA. The health consciousness variable investi-
gates factors regarding exercising and health, including living a
healthy lifestyle, making changes in daily routines in order to
prevent health problems, and following recommended activity
guidelines. The variable attitudes towards PA examines par-
ticipant views of PA its advantages. Significant others explores
the influences of peers, family, teachers and important others
on their intention to engage in PA, while priority of PA meas-
ured the priority of engaging in PA by assessing other possible
activities participants may prefer to engage in, such as internet
use, video games, and watching the television. The sub-scale
barriers to PA include lack of energy, costs associated with PA
and the availability of facilities, feelings of safety and the ef-
forts associated with being physically active.
Data Analysis
Data screening and analyses were undertaken using the Sta-
tistical Package for the Social Sciences (SPSS, v.17). Analysis
commenced with an evaluation of each scale’s psychometric
properties, including tests for floor and ceiling effects (item
means and standard deviations), internal consistency (Cron-
bach’s alpha), and to validate dimensionality (corrected item-
total correlations and principal components analysis). Inferen-
tial tests included Multivariate Analysis of Variance (MANO-
VA), one-way Analysis of Variance (ANOVA) and Multiple
Linear Regression (MLR) analyses, all undertaken in accor-
dance with Tabachnick and Fidell’s (2007) guidelines. Five
cases were excluded from the analysis as they constituted mul-
tivariate outliers as defined by extreme Mahalanobis distances.
The Theory of Reasoned Action asserts that a person’s be-
havioral intention depends both on the person’s attitude about
the behavior and upon subjective norms, and expresses an equa-
lity between behavioral intention and the weighted sum of atti-
tude and subjective norm, that is:
behavioral intention
attitudesubjective normWW
W1 and W2 are empirically derived weights, and behavioral
intention is the intention to be physically active, and is a com-
posite variable calculated from four items asking questions such
as “I intend to participate in activity as much as I can each
Copyright © 2012 SciRes.
Copyright © 2012 SciRes. 449
The overall integrity of the scales was satisfactory, and a
missing data analysis revealed that less than half a percent of
data was missing. The average reported duration of PA was 9.2
hours per week (see Table 1(a)). Means (M), standard devia-
tions (SD), and Cronbach’s alphas (αc) for the five ATPAS
subscales are presented in Table 1(b). Mean scores on each of
the DASS-42 subscales (see Table 1(c)) are comparable to
those obtained from normal adult populations and lower than
reported clinical norms (Lovibond & Lovibond, 1995). Princi-
pal components analysis (PCA) undertaken on each of the three
DASS subscales all returned single-factor solutions. The inter-
nal consistency of the DASS-42 has been reported elsewhere as
αc = .71 for depression, αc = .79 for anxiety and αc = .81 for
stress (Brown, Chorpita, Korotitsch, & Barlow, 1997), and the
estimates reported in Table 1c compare favorably with these. A
PCA performed on the RSES confirmed its unidimensionality,
and its internal consistency was sufficiently high (c = .89).
The sample mean was 20.63 (Table 1(d)) and thus above the
cut-off score of 15, below which self-esteem is considered to be
low (Rosenberg, 1965).
Zero-order correlation statistics (Pearson’s r) were calculated
to assess the relationship between psychological well-being
(self-esteem, lack of depression, anxiety, and stress) and influ-
ences to PA as measured using the ATPAS (Table 2). There is
no statistical evidence indicating a relationship between atti-
tudes towards PA and psychological well-being (p > .05). Sta-
tistically significant moderate (0.3 - 0.6) positive correlations
were noted between the four psychological well-being mea-
sures and barriers to PA, along with a number of smaller (<.3)
To reduce the experiment-wide error rate, a simple MA-
NOVA was conducted with PAQ group as the between-groups
factor and the four latent psychological well-being variables
constituting the dependent variables (DVs). Small to medium
correlations existed between the four DVs at each of three PAQ
levels, and a Bartlett’s test of sphericity (2(9) = 703.251, p
< .001) and a Box’s M test of equality of covariance matrix (F
= .85, p = .653) further confirmed the viability of a MANOVA.
There was a small but significant multivariate effect of the
grouped DVs in relation to PAQ grouping (Wilks Lambda
= .867, F(8,266) = 2.464, p = .014), indicating that psychologi-
cal well-being is related to group membership. Levene’s tests of
equality of variances were then performed prior to conducting
univariate F tests. For each of the four DVs the null hypothesis
that the within-groups variability is equitable across the three
PAQ groups was supported (p > .100). The univariate F tests’s
revealed significant differences across the three PAQ groups for
depression (F(2,136) = 4.625, p = .011) and anxiety (F(2,136) =
4.625, p < .001), but not for stress (F(2,136) = 2.609, p = .077)
or self-esteem (F(2,136) = .561, p = .572). Subsequent post-
hoc tests employing Bonferroni inequalities showed that those
in the relatively inactive group (i.e., PAQ1) reported signifi-
cantly higher levels of depression than those in the highly ac-
tive group (p = .014) but not those in the relatively active group
(p = .071). Additionally, those in the relatively inactive group
had a significantly higher self-reported anxiety mean than both
the relatively active (p = .003) and highly active (p = .026)
PAQ groups.
Five influences on PA, namely attitude, health consciousness,
significant others, priority, and barriers, were examined for
associations with PAQ category. A 3 (PAQ group) × 5 (influ-
ences) MANOVA was undertaken to examine the association
between the grouped DVs (i.e., influences) and PAQ group.
Small but significant correlations existed between the five DVs
at each of three PAQ levels. Two indices, the Bartlett’s test of
sphericity (2(14) = 158.367, p < .001) and Box’s M (F = 19.81,
p = .949), confirmed that the data satisfied the homogeneity of
covariance assumption. The MANOVA revealed a significant
multivariate effect (Wilks Lambda = .846, F(10,264) = 2.303, p
= .013). Following non-significant Levene’s tests a battery of
Table 1.
Summary statistics for the a) NZPAQ-SF; b) ATPAS; c) DASS-42; and d) RSES. Participants are divided into three activity groups: relatively inac-
tive (<2.5 hours per week; PAQ1), relatively active (2.5 - 4.9 hours per week; PAQ2), and highly active (>5 hours per week; PAQ3).
PAQ 1 (n = 40) PAQ 2 (n = 27) PAQ 3 (n = 72)
Scale # of items Overall c M SD
Scale cM SD
Scale c M SD
Scale c
Physical activity 9 - 1.118 .819 - 3.629 .614 - 15.305 13.962 -
Health consciousness 11 .718 46.878 4.905 .765 49.557 4.388 .642 49.384 4.704 .724
Attitudes 12 .724 46.878 4.924 .706 44.740 4.034 .667 44.945 4.539 .735
Significant others 7 .816 21.512 4.739 .691 21.667 4.574 .799 21.534 5.009 .808
Priority 5 .721 14.049 3.420 .623 12.815 3.235 .647 49.384 3.479 .768
Barriers 8 .681 23.400 4.499 .574 20.296 5.179 .695 19.425 4.551 .636
(c) DASS-42
Depression 14 .901 2.372 1.461 .786 1.619 1.131 .903 1.647 1.359 .727
Anxiety 14 .754 2.415 1.171 .802 1.574 1.002 .803 1.647 .959 .657
Stress 14 .869 9.195 7.527 .880 6.037 5.488 .841 6.959 5.927 .861
(d) RSES
Self esteem 10 .870 20.875 5.312 .843 19.593 5.645 .878 19.972 5.192 .884
Table 2.
Zero-order correlations for indices of psychological well-being (column variables) and influences of PA (row variables).
Depression Anxiety Stress Self-esteem
Attitudes –.107 .05 .013 –.099
Health consciousness –.196* .035 –.133 –.124
Significant others .175* .132 .227** .264**
Priority .226** .179* .135 .178*
Barriers .392** .248** .333** .339**
*p < .05 (2-tailed); **p < .001 (2-tailed).
five univariate F tests were undertaken with appropriate alpha
adjustments. For three of the five DVs, attitude towards PA
(F(2,136) = 4.043, p = .02), priority (F(2,136) = 4.111, p
= .018), and barriers to PA (F(2,136) = 9.299, p < .001), there
were significant differences across PAQ groups. Subsequent
univariate pairwise comparisons applying Bonferroni proce-
dures revealed that the highly active PAQ group had a more
positive attitude towards PA than the relatively inactive group
(p = .029), and also placed a higher priority on PA (p = .014).
In terms of perceived barriers the relatively inactive group re-
ported more barriers to activity than either of the relatively
active (p = .026) or highly active (p < .001) PAQ groups, but no
significant differences were noted between the relatively active
and highly active groups.
Table 3 displays the results of four simultaneous multiple
linear regression analyses undertaken after screening for nor-
mality, linearity, homoscedasticity, and independence of re-
siduals. The criterion variable was behavioral intention, and
four Theory of Reasoned Action models were tested: the three
PAQ groups and also pooled data. All four models returned
acceptable overall fits: relatively inactive (F(2,38) = 29.67, p
< .001), relatively active (F(2,24) = 5.332, p < .012), highly
active (F(2,70) = 18.586, p < .001), and pooled data (F(2,138)
= 56.54, p < .001). Table 3 reports both unstandardized (B) and
standardized coefficients () along with a variance accounted
for statistic (R2). The sub-scales, attitudes towards PA and sig-
nificant others were the independent variables, and in a linear
combination were used to predict behavioral intention (see
Equation (1)).
For the relatively inactive group, the two independent vari-
ables effectively account for the variance in behavioral inten-
tion. Note that for the standardized beta coefficients attitudes
towards PA is significantly different from zero across the four
regression analyses, whereas the significant others variable is
not significant. This implies that attitudes towards PA can be
used to predict behavioral intention, but significant others does
not possess such utility. For the pooled data, the zero-order
correlation between behavioral intention and actual PA was
small but also significant (r = .206, p = .014).
Approximately 29% of participants surveyed engaged in less
than 2.5 hours of at least moderately intense PA per week.
However, 19.5% of those surveyed were relatively active, par-
ticipating in 2.5 to 4.9 hours of activity per week, and 52%
were highly active, with five or more hours of activity per week.
These results echo previous research undertaken in New Zea-
land, which showed that the proportion of participants who
were physically active were 70%, while 30% of adults were
insufficiently active (SPARC, 2003). Thus, PA in adolescent
New Zealand females may possibly be indistinguishable from
adult levels.
The small but significant association between PA and two
dimensions of psychological well-being (i.e., depression and
anxiety) mirrors an emerging body of evidence indicating that,
for both adolescents and adults, increased PA is positively as-
sociated with better mental health (Sallis et al., 2000; Tey-
chenne, Ball, & Salmon, 2008). Sallis et al. (2000) report an
inverse relationship between depression and adolescents’ levels
of PA. Reviews of the literature found, for adults, strong evi-
dence for a positive relationship between levels of occupational
and leisure time PA and reduced symptoms of depression
(Dunn et al., 2001; Fox, 1999). Additionally, our results sup-
port previous findings that PA is associated with a reduction in
anxiety (Fox, 1999). Social cognitive and self-efficacy theory
(Bandura, 1986, 1997) suggest that the relationship between PA
and mental health is mediated by feelings of self-efficacy and
self-concept. Annesi (2005) states that, when a person senses
accomplishment in a physical task, self-concept improves and
better coping with psychological stressors ensues. This, in turn,
would improve mood and feelings of overall well-being. How-
ever, our finding of no association between PAQ grouping and
stress suggests that PA may not function as a mechanism for
stress relief in adolescent females.
In relation to influencess of PA, there was an association
between activity prioritization and level of PA, with those in
the highly active PAQ group reporting higher mean priority
scores than those in the relatively inactive and relatively active
PAQ groups. This finding cannot be explained by the higher
PAQ group being more health conscious, as this was statisti-
cally equivalent across the three PAQ groups. In a qualitative
study of adolescent girls and barriers to PA (Kientzler, 1999), it
was noted that the most frequently reported barrier to PA was
conflict with other activities, and in tandem with others (Koe-
zuka, Koo, Allison, Adlaf, Dwyer, Faulkner, & Goodman,
2006), our data provide quantitative support for this assertion.
The lack of association between PAQ group and the influ-
ence of others was unexpected, and reports of a strong influ-
ence of others on an individual’s exercising behavior (Schofield
et al., 2007; Seabra et al., 2008) has not been replicated here.
However, a review by Sallis et al. (2000) identified that females
were not only less likely to be physically active when compared
to males, but were also less influenced by significant others and
role models who were endorsing PA. After puberty, girls have
greater fat mass and less muscle mass than boys, and while
these biological differences may, in part, account for some of
the gender and developmental variation in PA, social and envi-
ronmental influences may be equally or more important (Garcia,
Broda, Frenn, Coviak, Pender, & Ronis, 1995).
Copyright © 2012 SciRes.
Table 3.
Unstandardized and standardized coefficients for the two sub-scales attitudes to PA and significant others that make up the Theory of Reasoned Ac-
tion for the three PAQ groups (1 = relatively inactive, 2 = relatively active, 3 = highly active) and for the entire sample. Model fit statistics (R and R2)
are included.
PAQ1 (R = .781, R2 = .609)
Std. Error Intercept B t p
Attitudes .25 1.952 0.78 7.67 >.001
Significant others .15
–.031 –.02 –.20 .841
PAQ2 (R = .539, R2 = .290)
Std. Error Intercept B t p
Attitudes .38 1.249 0.55 3.32 .003
Significant others .20
–.156 –.13 –.77 .449
PAQ3 (R = .652, R2 = .425)
Std. Error Intercept B t p
Attitudes .19 1.329 .66 6.83 >.001
Significant others .12
–.021 –.02 –.18 .858
Pooled data (R = .671, R2 = .442)
Std. Error Intercept B t p
Attitudes .012 .130 .688 10.484 >.001
Significant others .014
–.014 –.075 –1.142 .256
Barriers to PA were associated with PAQ group, with those
in the relatively inactive PAQ group reporting higher average
scores on questions probing barriers to PA than either of the
relatively active and highly active PAQ groups, echoing inter-
national studies (Robbins et al., 2009). One quantitative study
that explored barriers to PA among adolescent girls (Robbins,
Pender, & Kazanis, 2003), reported that self-consciousness and
concerns about appearance during PA was a major barrier,
while lack of interest and motivation was also identified as a
substantial barrier. This later finding is interesting as impaired
motivation is a common symptom of depression, which varied
across the three activity groups in our data.
Our results indicate that participants with more positive atti-
tudes towards PA also engage in higher levels of actual activity.
Previous studies have found that the formation of a positive
attitude towards PA is an important step towards actual partici-
pation in activity (Annesi, 2005; Smoll & Schutz, 1980), and
our data appear consistent with these findings. However, an
interesting finding to emerge from this study is that while there
is a relationship between perceived barriers to PA and psycho-
logical well-being, no such relationship exists between attitudes
and psychological well-being. The implication here is that, for
female adolescents experiencing impaired psychological well-
being, the perceived value of PA may not be in doubt, but the
perceived barriers to PA may be seen as more insurmountable
as the severity of the psychological impairment increases. Thus,
rather than focusing interventions on attitude change, a better
strategy might be to reduce actual and/or perceived barriers
identified by those experiencing psychological difficulties.
The Theory of Reasoned Action (TRA), equating a person’s
intention to engage in PA with a linear combination of attitude
and subjective norm, was successfully applied to the data. Our
results clearly show that of the two TRA components, attitude
has an influence on behavioral intention whilst the subjective
norm component does not. The statistically significant rela-
tionship we found between intention and attitude has prece-
dence in the literature (Chatzisarantis & Hagger, 2005; Cour-
neya & Friedenreich, 1999; Courneya, Vallance, Jones, & Re-
iman, 2005; Shen, McCaughtry, & Martin, 2008) and supports
the notion that intentions to engage in PA can be influenced by
targeting attitudes alone (Wood, 2008). The ability of the TRA
to predict behavioral intention was satisfactory, with r2 values
ranging from .3 to .6. However, behavioral intention, as pre-
dicted by the model, was itself a weak predictor of actual PA.
This disparity between predicted behavioral intention scores
and actual PA indicates that while attitudes play a significant
role in determining intentions to engage in PA, there are other
factors that must be accounted for. Our data suggests a negative
relationship between barriers and level of PA, and therefore
measures of behavioral intention incorporating measures of
perceived barriers would enhance predictive power. Though not
applied in our study, the Theory of Planned Behavior (Ajzen,
1985; Madden, Ellen, & Ajzen, 1992) is an extension of the
TRA that, in addition to the attitudes and subjective norms
terms that make up the TRA, contains the additional concept of
perceived behavioral control, or control beliefs.
Primary limitations of this study were that the sample size
was small, inhibiting analytical options, and that a self-report
inventory was employed. Self-report inventories, although they
are often the only practical way to measure PA levels in repre-
sentative population samples, have a known tendency to over-
report activity and under-report sedentary behaviors—a ten-
dency that may be increasing because of increasing social de-
sirability bias (SPARC, 2004). The sample limits the generali-
zation of results, and further data need to be collected from
other scholastic contexts in order to determine if the findings
we report are typical. Additionally, the present study was ob-
servational, which therefore precludes inferences about cause
and effect relationships, but the results are consistent with the
findings from other studies indicating that PA can positively
Copyright © 2012 SciRes. 451
affect psychological well-being (Penedo & Dahn, 2005). Fur-
thermore, we measured behavioral intention and actual behav-
ior on the same occasion, thus assuming that the intention to
engage in PA had not changed in the recent past.
To summarize, the present study found links between levels
of PA and depression and anxiety in adolescent females. Atti-
tudes to PA was a strong predictor of an individual’s intention
to engage in PA, but not the actual levels of PA itself. While
levels of health consciousness and attitudes to PA did not differ
between the inactive and active participants, inactive partici-
pants reported higher perceived barriers. These data indicate
that the most effective manner for school-based interventions
aiming to increase PA may therefore be to directly target barri-
ers rather than addressing attitudes. To this end, school districts
should consider the physical education environment alongside
minimal activity requirements, so that the quality of the activi-
ties serves to reduce barriers, rather than participation.
Ajzen, I. (1985). From intentions to actions: A theory of planned be-
haviour. In J. Kuhl, & J. Beckmann (Eds.), Springer series in social
psychology (pp. 11-39). Berlin: Springer Verlag.
Annesi, J. J. (2005). Improvements in self-concept associated with reduc-
tions in negative mood in preadolescents enrolled in an after-school
physical activity program. Psychological Re p ort s, 97, 400-404.
Azar, D., Ball, K., Salmon, J., & Cleland, V. (2008). The association
between physical activity and depressive symptoms in young women:
A review. Mental Health and Physical Activity, 1, 82-88.
Bandura, A. (1986). Social foundations of thought and action: A social
cognitive theory. New Jersey: Prentice-Hall.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York:
Biddle, S., & Mutrie, N. (2008). Psychology of physical activity: De-
terminants, well-being, and interventions. London; New York: Routl-
Brown, T. A., Chorpita, B. F., Korotitsch, W., & Barlow, D. H. (1997).
Psychometric properties of the Depression Anxiety Stress Scales
(DASS) in clinical samples. Behaviour Research and Therapy, 35,
79-89. doi:10.1016/S0005-7967(96)00068-X
Chatzisarantis, N. L. D., & Hagger, MS. (2005). Effects of a brief in-
tervention based on the theory of planned behavior on leisure-time
physical activity participation. Journal of Sport Exercise Psychology,
27, 470-487.
Courneya, K. S., & Friedenreich, C. M. (1999). Utility of the Theory of
Planned Behavior for understanding exercise during breast cancer
treatment. Psycho-Oncology, 8, 112-122.
Courneya, K. S., Vallance, J. K. H., Jones, L. W., & Reiman, T. (2005).
Correlates of exercise intentions in non-Hodgkin’s lymphoma survi-
vors: An application of the theory of planned behavior. Journal of
Sport and Exercise Psychology, 27, 335-349.
Dugan, S.A. (2008). Exercise for preventing childhood obesity. Physi-
cal Medicine and Rehabilitation Clinics of North America, 19, 205-
216. doi:10.1016/j.pmr.2007.11.001
Dunn, A. L., Madhukak, H. T., & O’Neal, H. A. (2001). Physical activ-
ity dose-response effects on outcomes of depression and anxiety.
Medicine & Science in Sports & Exercise, 33, S587-S597.
Fishbein, M, &. Ajzen I. (1975). Belief, attitude, intention, and behav-
iour: An introduction to theory and research. Reading, MA: Addi-
Fox, K. R. (1999). The influence of physical activity on mental well-
being. Public Health Nutrition, 2, 411-418.
Garcia, A. W., Broda, M. A. N., Frenn, M., Coviak, C., Pender, N. J., &
Ronis, D. L. (1995). Gender and developmental differences in exer-
cise beliefs among youth and prediction of their exercise behavior.
Journal of School Health, 65, 213-219.
Godin, G., & Shephard, R. J. (1984). Normative beliefs of school chil-
dren concerning regular exercise. Journal of School Health, 54,
443-445. doi:10.1111/j.1746-1561.1984.tb08909.x
Hohepa, M., Schofield, G., & Kolt, G. (2004). Adolescent obesity and
physical inactivity. New Zealand Medical Journal, 117, U1210.
Hohepa, M., Schofield, G., & Kolt, G. (2006). Physical activity: What
do high school students think? Journal of Adolescent Health, 39,
328-336. doi:10.1016/j.jadohealth.2005.12.024
Kahn, J. A., Huang, B., Gillman, M. W., Field, A. E., Austin, S. B.,
Colditz, G. A., & Frazier, A. L. (2008). Patterns and determinants of
physical activity in U.S. adolescents. Journal of Adolescent Health,
42, 369-377. doi:10.1016/j.jadohealth.2007.11.143
Kientzler, A. L. (1999). Fifth- and seventh-grade girls’ decisions about
participation in physical activity. The E lementary School Journal, 99,
391-414. doi:10.1086/461932
Koezuka, N., Koo, M., Allison, K. R., Adlaf, E. M., Dwyer, J. J. M.,
Faulkner, G., & Goodman, J. (2006). The relationship between sed-
entary activities and physical inactivity among adolescents: Results
from the Canadian Community Health Survey. Journal of Adolescent
Health, 39, 515-522. doi:10.1016/j.jadohealth.2006.02.005
Lovibond, S. H., & Lovibond, P. F. (1995). Manual for the depression
anxiety stress scales (2nd ed.). Sydney: Psychology Foundation.
Mackay, L. M., Schofield, G. M., & Schluter, P. J. (2007). Validation
of self-report measures of physical activity: A case study using the
New Zealand physical activity questionnaire. Research Quarterly for
Exercise and Sport, 78, 189-196.
Madden, T. J., Ellen, P. S., & Ajzen, I. (1992). A comparison of the
theory of planned behavior and the theory of reasoned action. Per-
sonality and Social Psychology Bulletin, 18, 3-9.
Penedo, F. J., & Dahn, J. R. (2005). Exercise and well-being: A review
of mental and physical health benefits associated with physical activ-
ity. Current Opinion in Psychiatry, 18, 189-193.
Riddoch, C. J., & Boreham, C. A. G. (1995). The health-related physi-
cal activity of children. Sports Medicine, 19, 86-102.
Robbins, L. B., Pender, N. J., & Kazanis, A. S. (2003). Barriers to
physical activity perceived by adolescent girls. Journal of Midwifery
and Women’s Health, 48, 206-212.
Robbins, L. B., Sikorskii, A., Hamel, L. M., Wu, T.-Y., & Wilbur, J.
(2009). Gender comparisons of perceived benefits of and barriers to
physical activity in middle school youth. Research in Nursing &
Health, 32, 163-176. doi:10.1002/nur.20311
Rosenberg, M. (1965). Society and the adolescent self-image. New
Jersey: Princeton University Press.
Sallis, J. F., Prochaska, J. J., & Taylor, W. C. (2000). A review of cor-
relates of physical activity of children and adolescents. Medicine &
Science in Sports & Exercise, 32, 963-975.
Sasse, S. K., Greenwood, B. N., Masini, C. V., Nyhuis, T. J., Fleshner,
M., Day, H. E. W., & Campeau, S. (2008). Chronic voluntary wheel
running facilitates corticosterone response habituation to repeated
audiogenic stress exposure in male rats. Stress, 11, 425-437.
Schofield, L., Mummery, K.W., Schofield, G., & Hopkins, W. (2007).
The association of objectively determined physical activity behavior
among adolescent female friends. Research Quarterly for Exercise
and Sport, 78, 9-15. doi:10.5641/193250307X13082490460175
Seabra, A. F., Mendonça, D. M., Thomis, M. A., Peters, T. J., & Maia,
J. A. (2008). Associations between sport participation, demographic
and socio-cultural factors in Portuguese children and adolescents.
European Journal of P u bli c H ea lth , 18, 25-30.
Copyright © 2012 SciRes.
Copyright © 2012 SciRes. 453
Shen, B., McCaughtry, N., & Martin, J. (2008). Urban adolescents’
exercise intentions and behaviors: An exploratory study of a trans-
contextual model. Contemporary Educational Psychology, 33, 841-
858. doi:10.1016/j.cedpsych.2007.09.002
Smoll, F. L., & Schutz, R. W. (1980). Children’s attitudes toward phy-
sical activity: A longitudinal analysis. Journal of Sport Psychology, 2,
SPARC (2003). SPARC facts: Results of the New Zealand sport and
physical activity surveys (199 7 - 2 0 0 1 ). Wellington: SPARC.
SPARC (2004). The New Zealand physical activity questionnaires:
report on the validation and use of the NZPAQ-LF and NZPAQ-SF
self-report physical activity survey instruments. SPARC: Wellington.
Sparks, P., Guthrie, A. A., & Shepherd, R. (1997). The dimensional
structure of the perceived behavioral control construct. Journal of
Applied Social Psychology, 27, 418-438.
Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics
(5th ed.). Boston, MA: Allyn and Bacon.
Teychenne, M., Ball, K., & Salmon, J. (2008). Physical activity and
likelihood of depression in adults: A review. Preventive Medicine, 46,
397-411. doi:10.1016/j.ypmed.2008.01.009
Thompson, A. M., & Humbert, M. L. (2003). Factors influencing
physical activity and sedentary behaviors in children and youth.
Medicine & Science in Sports & Exercise, 35, S161.
Wood, M. E. (2008). Theoretical framework to study exercise motiva-
tion for breast cancer risk reduction. Oncology Nursing Forum, 35,
89-95. doi:10.1188/08.ONF.89-95