Content uploaded by Piotr K Oles
Author content
All content in this area was uploaded by Piotr K Oles on May 04, 2017
Content may be subject to copyright.
Does running a first marathon influence general
self-efficacy and positive orientation?
ANNA GORCZYCA*, TOMASZ JANKOWSKI**, PIOTR OLES***
(*) State Development Bank of Poland
(**) Institute of Psychology, John Paul II Catholic University of Lublin
(***) Faculty of Psychology, SWPS University of Social Sciences and Humanities Poland
The aim of this study was to measure the potential increase in general self-effi-
cacy (GSE) and positive orientation (PO) in the context of a first marathon run.
Self-efficacy and positive orientation, defined as a general tendency to evaluate self,
life and future in a positive way were measured at three time points: at the begin-
ning of preparation, just after completing the marathon, and two months after com-
pleting the marathon. Eighty adults (23 women) completed the General Self-Effi-
cacy Scale (GSES; Schwarzer & Jerusalem, 1995) and Positivity Scale (PS; Caprara
et al., 2012). A latent trajectory model was used to test the hypothesis that self-effi-
cacy and PO would be higher immediately after the marathon and two months later
than at the start of preparation. The marathon had a significant effect only in
women. Slopes indicating rates of change in self-efficacy and PO were also corre-
lated. The second autoregressive model suggested that PO was a causal factor in
changes in GSE over time. Results are discussed in the light of relevant contempo-
rary literature.
KEY WORDS: Marathon; Positive orientation; self-efficacy.
Introduction
In most people physical activity promotes physical and mental health; it
increases their potential, improves wellbeing and establishes a solid base for
positive self-beliefs (Bezjak & Langga-Sharifi, 1991; Warner et al., 2014).
Physical activity and sports training can have a significant influence on self-
efficacy (Martin, 1995), defined as “people’s beliefs about their capabilities
to exercise control over events that affect their lives” (Bandura, 1989, p.
1175). One of most effective things that can increase the level of self-efficacy
Preparation of this article was supported by Grant NN106 428240 from the National Cen-
ter of Science, Poland
Correspondence to: Piotr Oles; oles@kul.pl)
Int. J. Sport Psychol., 2016; 47: 1-00
doi: 10.7352/IJSP 2016.47.000
for people is training in preparation for a marathon. Self-efficacy is an impor-
tant psychological variable, which can influence goal orientation related to
the race (performance vs outcome; Martin, 1995) as well as personal out-
comes of the race (finishing time; Gayton, Matthews, & Burchstead, 1986).
Preparation for distance running requires considerable time, and also
effort of increasing intensity over longer distances and for a longer duration.
During such training, the individual may notice changes in their endurance
and resistance to exhaustion or pain (Johnson, Stewart, Humphries, &
Chamove, 2012). Thus, training for a marathon can prove that it is possible
to train successfully in physical and mental skills. There are a few results sup-
porting this statement. For example, self-efficacy measured in participants
before and after a 15-week marathon training programme increased signifi-
cantly while the level of positive emotions slightly decreased (Samson, Sol-
mon, & Stewart, 2013). GSE beliefs may change through an interaction
between the changes in physical capabilities, which are the result of intensive
training, and beliefs about personal skills. Preliminary support for this claim
was provided by Samson (2014), who showed that participants mentioned
physiological/emotional states as a main source of information contributing
to the evolution of self-efficacy.
Prima facie it seems reasonable to suggest that proving that one has the
ability to run 42 kilometres constitutes an important life event unless one has
already notched up many other sporting achievements. A marathon takes
some hours and great physical effort that needs sufficient training and
demands mental skills, such as endurance and resistance to pain or exhaus-
tion. Moreover, the financial costs of participation in and travelling to
marathon events is not trivial: respondents in Wicker and Hallmann’s study
(2013) declared a willingness to spend up to 1500 €to travel to an overseas
marathon. All these facts suggest that participating in marathons has strong
motivations. People have a variety of expectations and goals related to
marathons, both primary (e.g., finishing time) and secondary (health orienta-
tion, psychological coping, life meaning, etc.; Masters, Ogles, & Jolton,
1993). One of the most frequent goals for running a long race, particularly
for the first time, is to enhance self-esteem (Masters & Ogles, 1995). As peo-
ple engage in marathons to improve how they feel about themselves, we can
reasonably expect that finishing a race changes self-evaluation attitudes.
Running marathons can also change indicators of hedonic well-being. The
results of Wilson, Morley and Bird’s study (1980) suggest that runners and
joggers have fewer depression symptoms, less anger, less confusion, and more
vigour than people who do not exercise at all. Based on these results, we
expect that running marathons can influence another indicator of hedonic
2A. Gorczyca, T. Jankowski, P. Oles
well-being, namely satisfaction with life, a general assessment of life quality
with regard to criteria chosen by oneself (Diener, 1984). Thus, we argue that
running a first marathon as an important life event can change one’s beliefs
about one’s potential, one’s life in general (success and satisfaction with life)
and one’s future achievements (optimism, self-efficacy).
Recent studies reviewed by Caprara (2009) suggest that three important
variables investigated in the context of well-being, namely self-esteem, opti-
mism and satisfaction with life, can be explained by a latent dimension which
accounts for the most of their common variance. Caprara named this factor
PO and defined it as a basic predisposition to adaptive beliefs about one’s
self, life, and future (Caprara, Steca, Alessandri, Abela, & McWhinnie,
2010). In this context, the aim of this study was to investigate changes in GSE
and PO related to running a first marathon. We asked the question if an indi-
vidual’s first successful marathon run, after a period of intensive training,
increased GSE or PO.
We therefore hypothesised that self-efficacy and PO (as an evaluation of
participating in a demanding race) will be higher after a first marathon run
than at the start of preparation for that marathon run and that increases in
self-efficacy and PO will last for at least two months. We chose two months
as a period of time sufficient to reduce the physiological and transient psy-
chological effects of participating in this extremely demanding run (see, for
example, the results of Nicolas, Banizette and Millet’s study from 2011,
which suggest that some effects of an ultramarathon, such as stress and
recovery, can be observed up to several weeks after the run). In this way, we
tended to check if the postulated increase of PO and/or self-efficacy over-
comes a psychological state and have more to do with a relatively stable mod-
ification of important personal beliefs. We were also interested in whether
the putative increase in self-efficacy would influence PO. Such a postulate is
supported both theoretically and empirically. We claim that an increase in
self-efficacy is an effect of a marathon training programme while the self-
esteem enhancement can be the result of fulfilling positive expectations and
beliefs about own abilities. Thus, a change in self-esteem is conditioned by a
previous change in self-efficacy. Some studies have also shown that increas-
ing self-efficacy with respect to emotional control and interpersonal relation-
ships improved PO in adolescents (Caprara, Alessandri, & Barbaranelli,
2010). Caprara (2010) argued that increases in self-efficacy as a result of
learning new skills and gaining new and important competencies could mod-
ify PO.
As mentioned earlier, several studies have reported increases in self-effi-
cacy beliefs during marathon training (e.g., Samson, Solmon, & Stewart,
Marathon, Self-efficacy and Positive Orientation 3
2013; Samson, 2014). Other studies investigated the relationship between
marathon performance or endurance training and self-efficacy beliefs (Gay-
ton, Matthews, & Burchstead, 1986; Martin, 1995; Okwumabua, 1985;
Scholz, Nagy, Schüz, & Ziegelmann, 2008). Some studies showed that grit,
pain resistance and self-efficacy are crucial to successful marathon running
(Freund et al., 2013; Johnson, Stewart, Humphries, & Chamove, 2012). We
also found one report of an increase in self-efficacy following long distance
running (Boudreau, 2010). Thus, we extended this research by investigating
whether successful first-time participation in a marathon would change PO
as well as GSE.
Several studies suggest that women can be motivated to run a marathon
in a different way to men. For example, Vertinsky (2000) suggested that,
because of cultural stereotypes, women who decide to run a marathon are
more determined to finish it than men are, a claim that is supported by ear-
lier results of Masters and Lambert’s study (1989). Because of motivational
differences between men and women, it is plausible to expect gender differ-
ences in the results of our study. Therefore, we decided to include gender
into the model as an important independent variable.
Method
PARTICIPANTS
One hundred and thirty-one participants (34 women) enrolled in the study; 80 (23
women) had completed their first marathon and answered the questionnaires three times.
Note that completing a first marathon was an inclusion criterion for being in the study. The
participants were adults aged between 20 and 54 years old, M age = 35.10 years, SD = 7.75.
They were recruited over the internet from various parts of the country, mainly central Poland.
Most participants were graduates (74%), married (55%), Roman Catholic (75%), and lived in
towns with a population of over 20,000 (80%).
MEASURES
General self-efficacy scale. This is a short questionnaire intended to measure GSE,
defined as a general belief or confidence in one’s ability to act effectively in difficult or chal-
lenging situations (Schwarzer & Jerusalem, 1995). The GSES consists of 10 items and
responses are given on a four-point Likert scale ranging from 1 – not at all true to 4 – com-
pletely true. The GSES has high internal consistency, Cronbach’s
α
= 0.82.
Positivity scale. This is a brief, eight-item questionnaire with responses given on a five-
point Likert scale ranging from 1 – strongly disagree to 5 – strongly agree. The items measure
self-esteem, satisfaction with life and optimism. The scale has a one-factor structure and this
4A. Gorczyca, T. Jankowski, P. Oles
Marathon, Self-efficacy and Positive Orientation 5
common factor explains over 50% of variance in scores. The internal consistency of the PS is
adequate, Cronbach’s
α
= 0.75-0.78 (Caprara et al., 2012).
PROCEDURE
All participants were asked to answer two questionnaires three times. The first assess-
ment took place at the start of their preparations for their first marathon (March 2013). The
second assessment took place a day or two after they had run their first marathon (the first
week in May 2013) and the third in July 2013, two months after they had completed their first
marathon. The assessments were thus two months apart.
Results
DATA ANALYSIS
We hypothesised that there would be two effects. The first was that, rel-
ative to the start of training (baseline: t0), there would an increase in the run-
ners’ PO and GSE, immediately after finishing the race (post-marathon: t1)
and two months later (follow-up assessment: t2). The second was that
changes in GSE would have a longitudinal effect on PO. We were also inter-
ested in whether gender modified the rate of change in PO and GSE.
There are two main approaches to the analysis of longitudinal data:
latent trajectory modelling (LTM) and autoregressive modelling (AM) and
these are used to address different theoretical problems. LTM is used to test
the hypothesis that separate temporal trajectories lead to significant changes
in participant scores (Preacher, Wichman, MacCallum, & Briggs, 2008). In
our study LTM was used to test a model hypothesising changes in PO and
GSE over time; it also enabled us to correlate PO and GSE intercepts and
rates of change, and to predict latent means and slopes from gender. AM, on
the other hand, can be used to estimate the time-specific autoregressive and
cross-lagged effects of one variable on another (Selig & Little, 2012); we used
AM to check whether baseline GSE influenced PO immediately after finish-
ing the marathon and whether GSE measured immediately after the race
influenced PO several weeks later. Although Bollen and Curran’s (2004)
autoregressive latent trajectory model provides a way of combining LTM and
AM into one model, we decided to apply LTM and AM separately. There
were two reasons for this: first, we wanted to evaluate several different effects
and using two separate models makes results easier to interpret and second,
we had a relatively small sample which would have made it more difficult to
specify more complex models.
One hundred and thirty-one participants completed the PO and GSE
measures at baseline, 105 at t1and 80 at t2, giving a total of 51 dropouts. We
checked whether the dropouts differed from other participants at the base-
line using Student’s t-test. This indicated that the dropouts had similar GSE
to those who completed all the assessments (t(129) = 1.5, p= ns) but had
higher PO at baseline (t(129) = 2.53, p< .05). Only participants who com-
pleted all three assessments were included in subsequent analyses. We con-
sider the differences between dropouts and included participants in the Dis-
cussion section.
Model fit was evaluated using relative chi-square ratio (c2/df; ratios of
less than 3 were taken as an indication of acceptable fit), the comparative fit
index (CFI; Bentler, 1990; values greater than 0.90 indicated acceptable fit),
incremental fit index (IFI; Bolen, 1990; values greater than 0.90 indicated
acceptable fit), and the root mean square error of approximation (RMSEA;
Browne & Cudeck, 1993; values less than 0.08 indicated acceptable fit). All
analyses were carried out using IBM SPSS 22 and AMOS.22 software.
Descriptive Statistics and Correlations
Table 1 displays means, standard deviations, skewness, kurtosis and the
correlation matrix for PO and GSE at different time points.
Mean PO and GSE were moderately high and, importantly, skewness
and kurtosis were fairly low (< 1 in all but one case), so use of normal theory
maximum likelihood estimation was permissible.
6A. Gorczyca, T. Jankowski, P. Oles
TABLE I
Descriptive Statistics And Correlation Matrix For Positive Orientation And Self-Efficacy In Three Time
Points
Self-efficacy
Time1 Time2 Time3 Time1 Time2 Time3
mean 30.38 32.04 31.80 31.65 32.16 31.83
sd 4.40 3.10 3.21 3.94 2.61 3.13
skewness -.63 -.25 -.60 -.36 -.10 -.27
kurtosis .51 -.09 1.55 .52 -.41 .94
PO1
PO2 ,568**
PO3 ,643** ,543**
GSE1 ,496** ,186 ,258*
GSE2 ,502** ,467** ,444** ,584**
GSE3 ,453** ,344** ,485** ,623** ,569** 1
**p<.001, *p<.01
Marathon, Self-efficacy and Positive Orientation 7
All pairwise correlations between scores for a given variable at different
time points were significant and positive (Pearson’s r: .54 to .64, p< .001)
indicating moderate stability of both PO and GSE scores. There was also a
moderate correlation between PO and GSE at baseline (r= .50, p < .001). At
subsequent time points the correlation between GSE and PO was similar to
that observed at baseline; t1: r = .48, p < .001; t2: r= .49, p< .001.
Latent Trajectory Model
In the first step in the analysis we examined the trajectories of PO and
GSE over time, both for the whole sample and for men and women sepa-
rately (see Figures 1 and 2). Qualitative analysis of the trajectories suggested
that PO and GSE were higher immediately after the marathon than at the
baseline and decreased slowly during the follow-up period; it also suggested
gender differences in trajectories.
Fig. 1. - Separate trajectories for positive orientation over three time points.
To verify these observations we built latent trajectory models including
six observable variables (PO and GSE scores at each of the three time
points), and four latent variables representing PO intercept, PO slope, GSE
intercept, and GSE slope. In the LTM paradigm, intercepts represent the
estimated means of variables at baseline, and slopes represent the rate of
change in variables over time. We fixed the values of three path coefficients
leading from the latent variables representing the PO and GSE intercepts to
observable variables at each time point at 1. We also fixed the path coeffi-
cient for the path from latent slope variable to observable variable at t0to 0,
and the path coefficient for the path from the latent slope variable to the
observable variable at the t1to 1. As we expected that both PO and GSE
would have a non-linear trajectory, more specifically that the rate of change
would be greater between t0and t1than between t1and t2, we did not fix the
path parameters for the paths from the slope variables to the observable vari-
ables at t2; instead we allowed these parameters to vary freely during analysis
8A. Gorczyca, T. Jankowski, P. Oles
Fig. 2. - Separate trajectories for general self-efficacy over three time points.
Marathon, Self-efficacy and Positive Orientation 9
(Bolen & Curran, 2006). In the next step we added gender into the model
(women were coded 0 and men 1) as a predictor of both intercepts and
slopes. We also allowed residuals representing the variance of intercepts and
slopes which was not explained by gender to correlate with each other. To
minimise the problems associated with the small sample size, we adjusted the
model parameters using a bootstrapping procedure (Bollen & Stine, 1990).
Bootstrapping is used to estimate model stability, particularly when the data
are not normally distributed. For this purpose, we generated 2,000 bootstrap
samples for further analyses. Figure 3 represents the final model. The path
coefficients in the figure are presented in the standardised form while, in the
text, we present their unstandardised form.
The model was an excellent fit to the data, c2(8) = 5.21, p = .74; c2ratio
= .65; CFI = 1; IFI = 1.01; RMSEA = .001, 90%CI: 0 - 096. To verify the
hypothesis that both PO and GSE are higher after completion of a marathon,
Fig. 3. - Latent trajectory model of positive orientation and general self-efficacy
change. Factor loadings for latent intercepts are fixed to 1. Factor loadings for latent
slopes are fixed to 0 and 1 for time 1 and time 2 respectively. All other path coeffi-
cients are presented in a standardized form.
we evaluated the significance of means for the slope factors. As we allowed
the path parameters for the paths from the slopes to the observable variables
at follow-up to vary, the slope factors can only be interpreted as a rate of
change for the period between t0 and t1, after controlling for the effects of
gender. Both means for the slope factors were significant: PO slope mean =
2.98 (bootstrap 90% CI:[1.79 to 4.33]), p < .001; GSE slope mean = 1.8
(bootstrap 90% CI:[.68 to 3.08]), p = .004. However, because the paths from
gender to the latent variables were also significant we had to consider rate of
change separately for men and women. At the baseline, men had higher PO
and GSE than women by a factor of 2.45 (bootstrap 90% CI:[.60 to 4.48]),
p = .02 and 2.44 (bootstrap 90% CI:[.83 to 4.02]), p = .009, respectively. The
rates of change in PO and GSE were significantly lower for men than for
women, by a factor of 1.83 (bootstrap 90% CI:[-3.41 to -.52]), p = .029 and
1.94 (bootstrap 90% CI:[-3.36 to -.74]), p = .008, respectively. Since men
were coded 1 and women 0, simple arithmetic showed that, for men, the rate
of change in PO was only 1.15 (2.98 - 1.83) and in GSE it was -0.11 (1.83 -
1.94). This suggests that only women experienced a significant change in
GSE and PO as a result of completing the marathon; there appears to have
been little change in men’s initially higher levels.
We also observed strong negative correlations between latent residuals
of intercepts and slopes related to a given construct (PO: r = -.89; GSE: r = -
.86). These correlations follow the well-known law of initial values: the lower
the initial score, the greater the expected change over time (Jin, 1992). More
interestingly, there was also a strong correlation between PO and GSE latent
slope variables, r =.74, which suggests that an increase in GSE is accompa-
nied by an increase in PO. LTM does not, however, allow one to determine
the causal nature of relationships. LTM tells us that there were changes in
both GSE and PO between the baseline and post-marathon assessments and
that these changes were more marked in women; however we do not know
whether baseline GSE facilitates change in PO or whether it is baseline PO
which causes an increase in GSE. We considered the former mechanism
more likely and therefore formulated an autoregressive model of change
based on this hypothesis.
Autoregressive Model
To determine whether GSE had a positive influence on PO over time we
formulated a bivariate autoregressive model based on a three-point series of
two observable variables. Because of the small sample we decided to reduce
10 A. Gorczyca, T. Jankowski, P. Oles
Marathon, Self-efficacy and Positive Orientation 11
the number of variables in the model. In this analysis we were not interested
in gender effects; autoregressive models do not allow for extraction of true
change latent factor, so we could not directly check how gender modifies
changes in PO and GSES scores, that is, why we did not include gender in
the model. The baseline levels of PO and GSE were exogenous variables and
levels at t1and t2were regressed on t0and t1, respectively. We also specified
cross-lagged paths in both directions in order to determine which path para-
meters were significant predictors of the PO-GSE relationship. We also
allowed error terms at t1and t2to be correlated. Similarly, as in the LTM, we
used the bootstrapping procedure with 2,000 samples for further analyses.
The final autoregressive model is displayed in Figure 4.
The model was a very good fit to the data: c2(8) = 1.16, p = .56; c2/df =
.58; CFI = 1; IFI = 1; RMSEA = .001, 90%CI: 0 -.19; p (RMSEA<0.05) = .62.
All autoregressive coefficients were significant (p < .001); however, stan-
dardised coefficients of paths between t1and t2(PO: 1.14, bootstrap 90%
CI:[.83 to 2.09]; GSE: 1.11, bootstrap 90% CI:[.79 to 1.83]) were larger than
the corresponding values for the paths from t0to t1(PO: .62, bootstrap 90%
CI:[.43 to .77]; GSE: .45, bootstrap 90% CI:[.28 to .57]). This suggests that
post-marathon levels of GSE and PO were less stable relative to the follow-
up assessment than baseline levels were, relative to post-marathon levels. The
only significant cross-lagged path was from PO at t0to GSE at t1(.27, boot-
strap 90% CI:[.14 to .40], p = .002); the remaining cross-lagged coefficients
were insignificant, which suggests that, contrary to our hypothesis, baseline
Fig. 4. - Autoregressive model of mutual, crosslegged effects of positive orientation
and general self-efficacy on each other. Path coefficients are presented in a standard-
ized form.
PO facilitates residual change in GSE when baseline levels are compared
with post-marathon levels. There was no such pattern observable at follow-
up: neither post-marathon PO nor post-marathon GSE predicted the level of
other constructs at the follow-up assessment several weeks later.
Discussion
The hypotheses about changes in GSE and PO during marathon train-
ing were partially confirmed. We observed a significant increase in GSE and
PO after the marathon relative to the baseline; levels had decreased slightly
two months later but both GSE and PO were still higher than at the baseline.
Assuming that a first marathon is an important life event and perceived as a
personal success, we can suggest that our result indirectly replicates previous
findings that taking up a great challenge in sport is correlated with high self-
esteem and self-efficacy (Barkhoff & Heiby, 2010). Note that self-esteem is a
core component of PO, and self-efficacy is correlated with both PO and self-
efficacy (Caprara et al., 2010; Oles´ et al., 2013).
As we expected, the effects were only found in women. Why was this?
Our preliminary suggestion is that gender moderates the relationships
between outstanding sports performance and self-efficacy and PO. Men ben-
efit from their first marathon because it counts as another success, whereas
women benefit twice, in the same direct way as men and indirectly through
improvements in self-efficacy and PO. Does running a marathon have differ-
ent meanings for men and women? While we did not ask the participants
about the personal meanings connected to their first marathon, we can only
suggest that it is a more important life event for women than for men. We
cannot explore this possibility as we do not have relevant qualitative data,
however, this explanation is consistent with some previous results. For exam-
ple, Ziegler (1991) found that women investigated within two weeks after
marathon reported that running had a positive effect on self-image and that
their lives were richer because of running, while men treated participating in
marathon more instrumentally; they reported increased knowledge of physi-
cal capabilities, improved perseverance or increased energy after run. Our
results are also congruent with the results of Loughren’s study (2010), who
showed that women are motivated to run more by goals related to self-esteem
enhancement, while men are driven by personal goal achievement and com-
petition. Similar results were obtained by Ogles, Masters and Richardson
(1995). In their study, women endorsed self-esteem, psychological coping,
and life meaning, among others, as more important motives for running than
12 A. Gorczyca, T. Jankowski, P. Oles
Marathon, Self-efficacy and Positive Orientation 13
men did, who were, in turn, more motivated by physical well-being than
women. In our study, we focused on PO and GSE, therefore it is not sur-
prising that we could detect changes only in women, who value the psycho-
logical benefits of running marathons, such as self-esteem and satisfaction in
life, more than men.
A more detailed inspection of the results reveals that the sample means
for both GSES scores and PO were at least half a standard deviation above
population means throughout the study period. The mean GSES score for
the men in our sample was one standard deviation higher than the popula-
tion mean, whilst the mean GSES score for the women in our sample was
about half a standard deviation higher than the population mean; a similar
pattern applied to PO. In other words, the men and women in our sample
had relatively high levels of GSE and PO before they ran their first marathon.
It is possible that people who start training for a marathon have relatively
high GSE and PO compared with the general population and that, as a con-
sequence, our study was subject to a ceiling effect, particularly in men. Indi-
viduals who have positive or very positive beliefs about themselves, their
lives, their potential, and their future are more likely to seek to retain their
adaptive life attitude or look for new experiences adequate to their beliefs,
than to engage in an activity which can improve their beliefs about them-
selves.
Our results provided a rather clear and somewhat challenging answer to
our question about the relationship between PO and self-efficacy. There was
already good previous evidence of a strong association between PO and GSE
(e.g., Caprara, Alessandri, & Barbaranelli, 2010; Oles´ et al., 2013), so this
result was not new. Other research, however, suggested that marathon train-
ing produced an unexpected effect, namely a simultaneous increase in self-
efficacy and decrease in positive affect (negative affect does not change)
(Samson, Solmon, & Stewart, 2014). The inverse relationship between posi-
tive affect and self-efficacy poses a challenge worthy of further investigation,
as positive affect is very closely related to PO (Caprara, 2009, 2010; Caprara
et al., 2010).
In our study, PO did not only correlate with GSE at all three time points,
changes in GSE were also correlated with changes in PO. The results suggest
that increasing self-efficacy is conditioned on the baseline PO but not other-
wise as we hypothesised. The autoregressive model (Figure 4) illustrating this
is consistent with an interpretation of the influence of changes in PO on GSE
that contrasts with Caprara et al.’s (2010) results suggesting that GSE influ-
ences PO, although it is at least partially consistent with the inherited and
stable nature of PO (Caprara et al., 2009). It is important to state clearly that
we have not concluded from our results that PO influences self-efficacy; we
are only pointing out that data are consistent with such a model.
One obvious limitation of this study is typical of correlational research,
namely that we assume that participants respond honestly and accurately.
The second limitation is the lack of a control group; this would have helped
to determine how participating in marathons influenced PO and self-efficacy.
Another limitation relates to the indices of self-efficacy and PO. In this study,
self-efficacy was measured using the GSES, which is a valid index of an indi-
vidual’s beliefs about their ability to cope effectively with new and challeng-
ing life circumstances (Scholz, Dona, Sud, & Schwarzer, 2002); this contrasts
with Bandura’s (1989) claim that self-efficacy is related to specific activities.
It is also important to note that participants who dropped out had a higher
baseline PO than participants who went on to complete all the assessments;
it is difficult to interpret this finding, but it may have a bearing on the con-
clusions to be drawn from the study. More detailed inspection reveals that
the majority of participants who dropped out were men; only seven of 51
were women. It seems unlikely that including these participants in the analy-
sis would change the pattern of results. A final limitation, but not the least
important, is that PO was measured by using a short but valid scale, namely
the PS, rather than using separate scales to measure each component: self-
esteem, satisfaction with life, and optimism.
To improve our understanding of how outstanding physical perfor-
mance can modify beliefs about the self, it will be necessary to use a larger
sample and determine whether there are differences in effects which are
related to baseline differences in PO and self-efficacy. It would also be valu-
able to ask participants to comment on what running their first marathon
meant to them.
REFERENCES
Bandura, A. (1989). Human agency in social cognitive theory. American Psychologist,44,
1175-1184.
Barkhoff, H., & Heiby, E.M. (2010). The relationship between self-efficacy, competition suc-
cess, delight, self-esteem, and body-care among elite artistic roller skaters. Journal of
Behavioral Health and Medicine, 1, 43-51.
Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin,
107, 238-246. doi: 10.1037/0033-2909.107.2.238
Bezjak, J. E., & Langga-Sharifi, E. (1991). Validation of the Physical Fitness Opinion Question-
naire against marathon performance. Perceptual and Motor Skills, 73(3, Pt. 1), 993-994.
Bollen, K. A. (1990). Overall fit in covariance structure models: two types of sample size
effects. Psychological Bulletin, 107, 256–259. doi: 10.1037/0033-2909.107. 2.256
14 A. Gorczyca, T. Jankowski, P. Oles
Marathon, Self-efficacy and Positive Orientation 15
Bollen, K. A. & Curran, P. J. (2004). Autoregressive Latent Trajectory (ALT) models: A Syn-
thesis of two traditions. Sociological Methods & Research, 32, 336-383.
Bollen, K.A., & Curran, P.J. (2006). Latent curve models: A structural equation perspective.
New York: Wiley.
Bollen, K. A., & Stine, R. (1990). Direct and indirect effects: Classical and bootstrap estimates
of variability. Sociological Methodology, 20, 115-140.
Boudreau, A. L. (2010). How long distance trail running affects the lived experience of
women employees’ productivity at work. Dissertation Abstract International: Section B:
The Sciences and Engineering, 71(2-B), 1390.
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In: K. A. Bollen
& J. S. Long (Eds.), Testing Structural Equation Models (pp. 136–162). Newbury Park,
CA: SAGE Focus Editions.
Caprara, G.V. (2009). Positive orientation: Turning potentials into optimal functioning. The
Bulletin of the European Health Psychologist, 11, 46-48.
Caprara, G.V. (2010). W kierunku integracji teorii cech i społeczno-poznawczej [Towards to
integration trait and social-cognitive theory]. In: S. Nowosad, B. ˙
Zurek (Eds.), Barwy
nauki. Nowoczesne technologie ICT w upowszechnianiu osia¸gnie¸c´ nauki (pp. 317-335).
Lublin: KUL.
Caprara, G.V., Alessandri, G., & Barbaranelli, C. (2010). Optimal functioning: Contribution
of self-efficacy beliefs to positive orientation. Psychotherapy and Psychosomatics, 79, 328-
330. DOI: 10.1159/000319532
Caprara, G.V., Fagnani, C., Alessandri, G., Steca, P., Gigantesco, A., Sforza, L.L.C., & Stazi,
M. A. (2009). Human optimal functioning: The genetics of positive orientation towards
self, life and the future. Behavioral Genetics,39, 277-284.
Caprara, G.V., Alessandri, G., Eisenberg, N., Kupfer, A., Steca, P., Caprara, M. G., Yam-
aguchi, S., Fukuzawa, A., & Abela, J. (2012). The Positivity Scale. Psychological Assess-
ment, 24, 701-712.doi: 10.1037/a0026681
Caprara, G.V., Steca, P., Alessandri, G., Abela, J.R.Z., & McWhinnie, C.M. (2010). Positive
orientation: Explorations on what is common to life satisfaction, self-esteem, and opti-
mism. Epidemiologia e Psichiatria Sociale,19, 63-71.
Diener, E. (1984). Subjective well-being. Psychological Bulletin, 95, 542-575.
Dzwonkowska, I., Lachowicz-Tabaczek, K., & Łaguna, M. (2008). Samoocena i jej pomiar.
Skala samooceny SES M. Rosenberga [Self-esteem and its measurement. Self-Esteem
Scale by M. Rosenberg]. Warszawa: Pracownia Testów Psychologicznych PTP.
Freund, W., Weber, F., Billich, C., Birklein, F., Breimhorts, M., & Schuetz, U. H. (2013). Ultra-
marathon runners are different: Investigations into pain tolerance and personality traits
of participants of the TransEurope FootRace 2009. Pain Practice, 13(7), 524-532.
Gayton, W. F., Matthews, G. R., & Burchstead G. N. (1986). An investigation of the validity
of the Physical Self-efficacy Scale in predicting marathon performance. Perceptual and
Motor Skills, 63 (2, Pt.2), 752-754.
Johnson, M. H., Stewart, J., Humphries, S. A., & Chamove, S. A. (2012). Marathon runners’
reactions to potassium iontophoretic experimental pain: Pain tolerance, pain threshold,
coping and self-efficacy. European Journal of Pain, 16, 767-774.
Laguna M., Ole P. K., & Filipiuk D. (2011). Orientacja pozytywna i jej pomiar: Polska adap-
tacja Skali Orientacji Pozytywnej. [Positive orientation and it’s measurement: Polish
adapt ation of the P ositivity Scale ]. Studia Psychologiczne, 49 (4), 47-54. doi:
10.2478/v10167-010-0035-7
Loughren, E.A. (2009). Motivation of First Time Marathoners to Adherence to Marathoning.
Disertation thesis. Temple University.
Martin, J. J. (1995). Competitive orientation, self-efficacy and goal importance in Filipino
marathoners. International Journal of Sport Psychology, 26, 348-358.
Masters, K.S., Lambert, M.J. (1989). The relation between cognitive coping strategies, reasons
for running, injury, and performance of marathon runners. Journal of Sport and Exercise
Psychology, 11, 161-170.
Masters, K.S. & Ogles, B.M. (1995). An investigation of the different motivations of marathon
runners with varying degrees of experience. Journal of Sport Behavior, 18, 69-79.
Masters, K.S., Ogles, B.M., & Jolton, J.A. (1993). The Development of an Instrument to Mea-
sure Motivation for Marathon Running: The Motivation of Marathoners Scales
(MOMS). Research Quarterly for Exercise and Sport, 64(2), 134-143.
Nicolas, N., Banizette, M., & Millet, G.Y. (2011). Stress and recovery states after 24h ultra-
marathon race: A one-month follow-up study. Psychology of Sport and Exercise, 12, 368-374.
Ogles, B.M., Masters, K.S., Richardson, S.A. (1995). Obligatory running and gender: An
analysis of participative motives and training habits. International Journal of Sport Psy-
chology, 26, 233-248.
Okwumabua, T. M. (1985). Psychological and physical contribution to marathon perfor-
mance: An Exploratory investigation. Journal of Sport Behavior, 8, 163-171.
Ole, P. K., Alessandri, G., Ole, M., Bk, W., Jankowski, T., Laguna, M., & Caprara, G. V.
(2013). Positive orientation and generalized self-efficacy. Studia Psychologica, 55 (1), 47-59.
Preacher, K. J., Wichman, A. L., MacCallum, R. C., & Briggs, N. E. (2008). Latent growth
curve modeling. Thousand Oaks, CA: Sage Publications.
Jin, P. (1992). Toward a reconceptualization of the law of initial value. Psychological Bulletin,
111(1), 176-184.
Samson, A. (2014). Sources of self-efficacy during marathon training: A qualitative longitudi-
nal investigation. The Sport Psychologist, 28, 164-175, doi: 10.1037/t25467-000
Samson, A., Solmon, M., & Stewart, L. (2013). Sources of self-efficacy during marathon train-
ing: A qualitative longitudinal investigation. International Journal of Sport Psychology, 44,
55-68.
Scholz, U., Dona, B., Sud, S., & Schwarzer, R. (2002). Is general self-efficacy a universal con-
struct? European Journal of Psychological Assessment, 18, 242-151.
Scholz, U., Nagy, G. Schüz, B., Ziegelmann, J. P. (2008). The role of motivational and voli-
tional factors for self-regulated running training: Associations on the between- and
within- person level. British Journal of Social Psychology, 47(3), 421-439.
Schwarzer, R., & Jerusalem, M. (1995). Generalized Self-Efficacy scale. In J. Weinman, S.
Wright, & M. Johnston (Eds.), Measures in health psychology: A user’s portfolio. Causal
and control beliefs (pp. 35-37). Windsor, UK: Nfer-Nelson.
Selig, J. P. & Little, T. D. (2012). Autoregressive and cross-lagged panel analysis for longitudi-
nal data. In: B. Laursen, T. D. Little, N. A. Card, Handbook of developmental research
methods (pp. 265-278), NY: The Guilford Press.
Vertinsky, P. (2000). Eternally wounded women? Feminist perspectives on physical activity
and aging or a woman’s p(l)ace in the marathon of life. Journal of Aging Physical Activity,
8(4), 386-404.
Warner, L., Schulz, B., Wolff, J., Parschau, L., Wurm, S., & Schwarzer, R. (2014). Sources of
self-efficacy for phisical activity. Health Psychology. Advance online publication.
http://dx.doi.org/10.1037/hea0000085
16 A. Gorczyca, T. Jankowski, P. Oles
Marathon, Self-efficacy and Positive Orientation 17
Wicker, P., Hallmann, K. (2013). Estimating consumer’s willingness-to-pay for participation in
and traveling to marathon events. Event Management, 17(3), 271-282.
Wilson, V., Morley, N., & Bird, E. (1980). Mood profiles of marathon runners, joggers and
non-exercisers. Perceptual and Motor Skills, 50(1), 117-118.
Ziegler, S.G. (1991). Perceived benefits of marathon running in males and females. Sex Roles,
25, 119-127.
Manuscript submitted . Accepted for publication March 2016.
