Asian Nursing Research 14 (2020) 300e305

Contents lists available at ScienceDirect

Asian Nursing Research

journal homepage: www.asian-nursingresearch.com

Research Article

The Indonesian Version of the Exercise Self-Efficacy Scale: Cross-

cultural Adaptation and Psychometric Testing*
Arif R. Hakim,1 Shan-Tair Wang,2, 3 Fransiskus X. Widiantoro,4 Mujib Hannan,1
Chi-Jane Wang,5, * Suzan J. Fetzer6
1
Department of Nursing, University of Wiraraja, Sumenep, Indonesia
2
Ditmanson Medical Foundation, Chia-Yi Christian Hospital, Chiayi, Taiwan
3
Institute of Gerontology, College of Medicine, National Cheng Kung University, Tainan, Taiwan
4
Department of Nursing, School of Health Sciences Saint Borromeus, Bandung, Indonesia
5
Department of Nursing, College of Medicine, National Cheng Kung University, Tainan, Taiwan
6
Department of Nursing, University of New Hampshire, Durham, United States

a r t i c l e i n f o s u m m a r y

Article history: Purposes: The study aimed to translate the Exercise Self-Efficacy Scale (ESES) into Indonesian and test
Received 16 December 2019 the cultural equivalence, reliability, and validity of the new version for university students.
Received in revised form Methods: The cross-sectional study recruited 379 Indonesian university students using convenience
22 August 2020
sampling. Phase 1, a culturally appropriate version of the ESES was developed in the Indonesian lan-
Accepted 26 August 2020
guage. Phase 2, the psychometric properties were determined through exploratory factor analysis,
bootstrap factor analysis, and confirmatory factor analysis. The internal consistency reliability was tested
Keywords:
using Cronbach's a, whereas the stability using intraclass correlation coefficient to assess.
factor analysis, statistical
psychometrics
Results: The students' ages ranged from 17 to 39 years, and 65.0% were women. For translation equiv-
self efficacy alence, the mean item content validity indexes ranged from 3.5 to 4, and all items were understandable.
translating The 16-item scale exhibited cross-cultural appropriateness and readability, with a three-factor model
validation study explaining 62.3% of the variance in exercise self-efficacy. A bootstrap analysis using 100 resamples
further confirmed the three-factor model. The indices of the good-fit model that used the three-factor by
two-stage least squares method were satisfactory, with c2/df ¼ 3.3, goodness of fit index ¼ .88, and root
mean-square error of approximation ¼ .05 (p < .001). The Cronbach's a was .78, .80, and .92 for factors 1,
2, and 3, respectively. The test–retest reliability was demonstrated with an intraclass correlation coef-
ficient of .91, indicating adequate measurement stability.
Conclusion: The 16-item ESES-I has acceptable validity and reliability; however, a broader application of
the scale requires further testing in different populations to confirm its external validity.
© 2020 Korean Society of Nursing Science. Published by Elsevier BV. This is an open access article under
the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction

Promoting physical activity (PA) to maintain wellness and pre-

vent disease is a global health priority [1]. Nearly one-third of the
world's population fails to meet the recommended levels of PA to
Arif R. Hakim: https://orcid.org/0000-0003-1723-480X; Shan-Tair Wang: https:// stay healthy [2]. In Indonesia, more than 80.0% of adolescents and
orcid.org/0000-0003-0061-9295; Fransiskus X. Widiantoro: https://orcid.org/ almost 25.0% of adults do not meet the recommended levels of PA
0000-0002-9106-3394; Mujib Hannan: https://orcid.org/0000-0002-9405-6819; [3]. These age groups include university students, who are
Chi-Jane Wang: https://orcid.org/0000-0001-8204-8574 vulnerable to physical inactivity due to academic pressure and the
*
PS: Readers who are interested in obtaining the Indonesian version of ESES,
habit of sitting. Promoting PA among Indonesian young people,
please email the corresponding author
* Corresponding author. Chi-Jane Wang, PhD, MPH, RN, Department of Nursing, especially university students, is crucial as the largest group [4],
College of Medicine, National Cheng Kung University, No.1, University Road, Tainan, and they are in an opportune period for establishing healthy life-
701, Taiwan. styles. One strategy to promote PA is to increase an individual's
E-mail address: [email protected]

https://doi.org/10.1016/j.anr.2020.08.008
p1976-1317 e2093-7482/© 2020 Korean Society of Nursing Science. Published by Elsevier BV. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).
 A.R. Hakim et al. / Asian Nursing Research 14 (2020) 300e305 301

exercise self-efficacy [5]. Exercise self-efficacy refers to an in- participating in the study, the students were informed of the pur-
dividual's belief in his or her ability to perform and succeed in pose of the research and were assured of their right to refuse to
challenging exercise situations [6]. Identifying exercise self-efficacy participate in the study or withdraw their consent at any stage. The
is fundamental to designing interventions to promote PA; however, data were collected from July to August 2018.
there is no available instrument that measures exercise self-efficacy
for the Indonesian population. Measurements/Instruments
The most important consideration in using a self-efficacy scale is
that it must be tailored to specific activity domains [7]. In addition, The original 18-item ESES describes situations in which it may
the scale assesses the multifaceted ways in which efficacy beliefs be difficult to adhere to an exercise routine. The items are
operate within the selected activity domain [8]. The 18-item Exer- responded using an 11-point scale (0e10), with the higher scores
cise Self-Efficacy Scale (ESES), developed by Bandura, is a useful indicating higher confidence in one's ability to regularly exercise
measure of exercise beliefs. It describes situations in which it may [6]. This scale was used with Bandura's permission.
be difficult to adhere to an exercise routine [8].
To date, the ESES has been adapted and validated in Korean [9], Procedure
which reported a one-factor structure. Similarly, a study using a
sample of elderly Australians undergoing cardiac rehabilitation In Phase 1, the translation and cross-cultural adaptation used
found a single factor [10]. Conversely, studies in the Netherlands the five-step process delineated by Beaton et al. [17], including
[11] and Iran [12] reported a three-factor structure. Note that the forward translation, synthesis, back translation, expert committee
adaptation and validation of the ESES in several countries have review, and pretesting (Table 1). Six bilingual experts of exercise
been conducted among unhealthy populations [9e12]. In a study of self-efficacy (three nursing education specialists, a psychologist, a
healthy undergraduate students from the United States, a two- health promotion specialist, and a nursing teacher) formed an
factor model emerged [13]. A recent study on Malaysian univer- expert review panel to evaluate the cross-cultural validity and
sity students simplified the ESES to 12 items and reported a three- content validity index at the item-level (I-CVI). During pretesting,
factor structure [14]. Although the original ESES was published as a 33 university students rated the instructions and scale items as
single dimension, the varied nature of each item (environment, clear or unclear. They were also asked to suggest modification or
health, and situation) contributes to a multidimensional structure. elimination of items. In Phase 2, the psychometric assessment of
These studies illustrate possible for cultural differences when the construct validity was conducted through exploratory factor
measuring exercise self-efficacy. analysis (EFA) and confirmatory factor analysis (CFA).
In addition, the reliability of ESES-I was assessed through
Purpose of the Study Cronbach's a and the intraclass correlation coefficient (ICC). SPSS
AMOS 18.0 (IMP Corp., Armonk, NY, USA) was used to conduct the
This study aimed to translate the ESES into Bahasa Indonesia EFA and applied an oblique promax rotation to the factors, whereas
(hereinafter, the ESES-I) and psychometrically test the scale for the CFA and reliability test were computed using SAS version 9.2
cross-cultural application among a younger adult population. (SAS Institute Inc, Cary, NC).

Methods
Data Analysis

Design
In Phase 1, the cross-cultural validity was assessed based on the
consideration of semantic, idiomatic, conceptual, and experiential
This study has a cross-sectional design and was conducted in
equivalence. Each item was rated from 1 (not equivalent) to 4
two phases. Phase 1 consisted of the translation and cross-cultural
(highly equivalent) by each expert. If any item received a score
adaptation of the ESES, whereas Phase 2 assessed the psychometric
below 3, it was discussed by the group. The mean score of each item
properties of the translated instrument.
was calculated. Furthermore, the I-CVI was calculated by the
number of scores that were three or four, divided by the total
Setting and Samples
number of experts [18].
In Phase 2, the EFA was conducted by applying the following:
In total, 379 students from an Indonesian university in Yogya-
Kaiser's stopping rule, scree plot examination to determine the
karta were recruited. They were divided into three groups for
number of factors, a priori criterion using eigenvalue 1 and a
different purposes to test the validity and reliability of the ESES-I.
Group 1 included 33 students who pretested the translated in-
strument. Group 2 included 312 students who completed the Table 1 Cross-Cultural Translation and Adaptation Processes of The ESES into Bahasa
translated instrument to determine the factor structure, validity, Indonesia.
and internal consistency. A rule of thumb for factor analysis states
Process Step and Roles
that a minimum of 300 cases is considered adequate [15], thus 312
students were recruited to meet that standard. Group 3 included 34 Forward  The original language (English) translate into two Bahasa
translation Indonesian versions independently (T1 & T2)
students who provided information for the test–retest reliability by Translators aware, but naïve to the outcome measure
completing the ESES-I twice, 2 days apart. This sample size exceeds Synthesis  Bahasa Indonesian versions T1 & T2 synthesized into T12
the estimated minimum (30 students) needed for a power of 90.0% Research team and two translators
and an alpha of .05 [16]. Back translation  Independent translation of the Indonesian T12 back into
English versions (BT1 & BT2)
Two native English native bilingual professionals
Ethical Considerations Expert committee  Analysis of semantic, idiomatic, experiential, and
review conceptual equivalence
The permission to conduct this study was received from the Six content experts
institutional review board of university of Muhammadiyah Pretesting  Complete the scale to analyze understanding of items
34 Indonesian university students
Yogyakarta (Approval no. 317/EP-FKIKUMY/VI/2018). Before
302 A.R. Hakim et al. / Asian Nursing Research 14 (2020) 300e305

Table 2 Factor Loading and t Statistic of 3- and 2-Factor Model of The ESES-I (N ¼ 312).

Item 3-Factor model 2-Factor model

Factor 1 Factor 2 Factor 3 Factor 1 Factor 2

T FL t FL T FL t FL t FL

1 0.54 .09 1.59 .18 16.79 .71 1.18 .04 29.74 .79
2 0.32 .06 0.54 .15 27.39 .93 0.13 .20 66.06 .89
3 0.9 .12 2.13 .25 17.33 .72 1.46 .03 2.40 .83
4 0.06 .14 5.10 .89 0.40 .04 2.86 .37 1.31 .28
5 2.14 .28 0.47 .21 9.32 .72 1.95 .12 8.57 .65
6 5.72 .52 1.40 .15 1.67 .19 17.07 .59 1.27 .22
7 0.72 .16 1.06 .31 2.37 .36 3.19 .31 3.42 .47
8 0.88 .18 1.24 .42 1.38 .19 4.79 .41 1.92 .34
9 0.89 .04 8.39 .68 2.44 .15 4.66 .41 2.22 .39
10 1.19 .11 6.84 .74 0.83 .00 7.64 .52 1.40 .26
11 1.54 .23 3.99 .68 0.77 .16 11.99 .62 0.25 .06
12 21.83 .76 0.45 .16 2.01 .23 19.10 .66 0.31 .14
13 47.55 .88 0.21 .06 0.40 .01 21.15 .86 1.33 .05
14 7.24 .59 0.30 -.01 1.71 .22 9.36 .57 0.72 .19
15 32.76 .89 2.28 .15 3.78 .18 36.21 .99 2.81 .17
16 44.91 .94 0.75 .00 2.09 .09 21.54 .96 1.98 .14
17 27.54 .89 1.18 .05 1.53 .08 25.73 .94 1.81 .11
18 6.92 .61 1.01 .12 0.04 .00 32.34 .69 0.65 .02

Note. FL ¼ factor loading.

Factor loading  0.5 and t statistic > 0.2 in bold.

maximum of 25 rotation iterations, the percent of cumulative Results

variance to identify the model, and a parallel analysis to provide a
visual comparison of the results. A bootstrap factor analysis was Participant Demographics
conducted to test the stability of the two- and three-factor models
using a nonparametric approach. Resamples (n ¼ 100) were drawn A total of 379 Indonesian university students provided the data
with replacement; each resample had the same size as the original for the factor analysis and testeretest procedures. Their ages
[19]. When testing the stability across resamples, the expectation is ranged from 17 to 39 years, with a greater number of women
that the t-value exceeds 2.0 [20]. participating (66.5%, n ¼ 252).
The first-order CFA was analyzed using maximum likelihood
estimation (MLE), the robust procedure of two-stage least squares Phase 1: Translation and Cross-Cultural Appropriateness
(2SLS), and five-fold cross-validation. The five-fold cross-validation
was performed to evaluate the model fit through 100 repetitions. The translational equivalence and cross-cultural appropriate-
The total sample was randomly divided into five subsets; then, the ness of the scale were assessed by the six experts’ ratings of the
model fit was tested for each subset. The average value of 500 level of agreement between the Indonesian and the original
goodness-of-fit indices (GFIs) was obtained. Good construct validity version. There were no deleted or modified items; the mean item
is achieved when the fitness indices reach the required level; CVIs ranged from 3.5 to 4.0. All items in the ESES-I were rated as
fitness indices indicate how well the items measure their respective understandable during pretesting by the 33 university students.
latent constructs [21]. The model is considered a good fit when the
probability level of GFI value is >.9, standardized root mean square Phase 2: Psychometric Assessment
residual (SRMR) < .05, and the comparative fit index (CFI)  .90. A
c2/df ratio of <5 indicates a good fit between the observed and The 18-item scale had a dimensional structure as indicated by
reproduced correlation matrices [22]. Akaike Information Criteria the inter-item correlation matrix. The weakest correlation (r  .3)
(AIC) was used to perform model comparisons, and the model with was between items 2 and 11, “When I am feeling pressure from
a lower AIC was considered to be better [23]. work” and “When I have too much work to do at home,” whereas
The scale reliability was evaluated by measures of internal the strongest correlation (r  .8) was between items 15 and 16,
consistency using Cronbach's a with acceptable score  .7 [24]. For “Without support from my family or friends” and “During a
ICC analysis, the testeretest reliability was determined 2 days after vacation.” For the inter-item correlation matrix, some were highly
the first assessment to get an acceptable compromise between correlated with each other and poorly correlated with items in
recollection bias and unwanted clinical change [25]. A two-way other groups. Thus, it was necessary to explore the construct
mixed-effects model was used because repeated measurements validity.
cannot be regarded as randomized samples [26]. The absolute The EFA yielded three factors with eigenvalues of 8.816, 1.490,
agreement definition was adopted because the measurements and 1.068. However, the screen test revealed only two factors above
would be meaningless if there was no agreement between repeated an eigenvalue of 1.00. The percent of cumulative variance tests
measurements. ICC values < .5, .5e.75, .75e.9, and >.9 indicate indicated that the three-factor model had higher cumulative vari-
poor, moderate, good, and excellent reliability, respectively [26]. ance (63.2%) compared with the two-factor model (57.3%). Thus,
 A.R. Hakim et al. / Asian Nursing Research 14 (2020) 300e305 303

Figure 1. Confirmatory factor analysis by Maximum Likelihood Estimation.

the three-factor model was preferred. Using plot displays, a parallel According to the EFA, both the two- and three-factor models
analysis identified two retained factors whose actual eigenvalues were plausible to be included in the CFA. Initially, the MLE pro-
were above or along the lines representing the randomly generated cedure was used, which revealed that the two-factor model
eigenvalues. The bootstrap analysis testing the stability of the two- without Item 7 showed a poor fit (c2/df ¼ 3.9, SRMR ¼ .06,
and three-factor models using 100 resamples further confirmed GFI ¼ .83, CFI ¼ .88, AIC ¼ 535.05). As shown in Figure 1, the three-
that the three-factor model was appropriate (Table 2). The result of factor model without Items 7 and 8 resulted in a good model fit (c2/
bootstrap analysis for the two-factor model showed that Item 7 was df ¼ 3.01, SRMR ¼ .06, GFI ¼ .89, CFI ¼ .93, AIC ¼ 304.06). Item 7
found to be cross-loading, and many items had a factor loading (“When I am feeling anxious”) and Item 8 (“After recovering from
below 0.50 (i.e., Items 4, 7, 8, and 9); the three-factor model showed an illness that caused me to stop exercising”) were removed from
that Items 7 and 8 had low factor loadings and t-statistic values. further analysis because of their low factor loadings.
304 A.R. Hakim et al. / Asian Nursing Research 14 (2020) 300e305

Table 3 Original and Bahasa Indonesia Versions of ESES.

Number of Original Version Number of Bahasa Indonesia Version

item Item

1 When I am feeling tired Situational/interpersonal

2 When I am feeling under pressure from work 6 Ketika saya sedang merasa tertekan
3 During bad weather 12 Ketika sedang ada tamu
4 After recovering from an injury that caused me to stop exercising 13 Ketika ada hal menarik lainnya untuk
Dikerjakan
5 During or after experiencing personal problems 14 Jika saya tidak mampu mencapai target olahraga saya
6 When I am feeling depressed 15 Tanpa dukungan dari keluarga atau teman saya
7 When I am feeling anxious 16 Selama liburan
8 After recovering from an illness that caused me to stop exercising 17 Ketika saya memiliki janji yang lain
9 When I feel physical discomfort when I exercise 18 Setelah mengalami masalah keluarga
10 After a vacation Competing demands
11 When I have too much work to do at home 4 Setelah sembuh dari cedera yang menyebabkan saya berhenti berolahraga
12 When visitors are present 9 Ketika saya merasa tidak nyaman dengan badan saya saat berolahraga
13 When there are other interesting things to do 10 Setelah liburan
14 If I do not reach my exercise goals 11 Ketika saya memiliki banyak pekerjaan dirumah
15 Without support from my family or friends Internal feelings
16 During a vacation 1 Ketika saya sedang merasa lelah
17 When I have other time commitments 2 Ketika saya sedang mengalami tekanan dalam pekerjaan
18 After experiencing family problems 3 Selama cuaca buruk
5 Selama atau setelah mengalami masalah pribadi

Considering the non-normal distribution of the data, a robust confidence to exercise. A previous study has found that students
2SLS was performed to analyze the fit of the three-factor model. who exercise regularly have less anxiety than those who do not
The fit indices of both the MLE and 2SLS procedures were similar, [29]. The low factor loading of Item 8 indicated that the students
indicating that the distribution of the data did not influence the responded inconsistently on this item. Healthy students may have
model fit analysis (c2/df ¼ 3.3, SRMR ¼ .05, GFI ¼ .88, CFI ¼ .93, struggled to comment on exercising after an injury if they had not
AIC ¼ 404.68). In addition, the five-fold cross-validation also experienced any injury.
showed that the three-factor model was appropriate for the 16- Although a study conducted in the United States reported a
item ESES-I (SRMR ¼ .05, GFI ¼ .88, CFI ¼ .92). Based on the re- single factor [13], most of the studies conducted in Asian countries
sults of the EFA and CFA, the first-order three-factor structure have found that the ESES is multidimensional. The differences in
(Figure 1) was selected as the final model applied to the ESES-I. psychological constructs between Eastern and Western cultures
According to the item-component of the three-factor model of [30] may have contributed to such differences in beliefs about ex-
ESES-I, the name and description of each factor are as follows ercise habits. Regarding the number of factors, the 16-item ESES-I is
(Table 3): Factor 1 (8 items), “situational/interpersonal,” explained similar to the other translated versions, but the number of items
48.9% of the variance and includes items that are “situational or differs from the 18-item Korean [7], the 17-item Iranian [10], and
interpersonal influences” on exercise self-efficacy; Factor 2 (4 the 12-item Malaysian [14] versions. Four items in the ESES-I (Items
items), “competing demands,” explained 8.3% of the variance and 6, 11, 14, and 15) loaded onto different factors compared with the
represents situations in which the participants have little control; Korean, Iranian, and Malaysian versions. Although the Korean [7]
Factor 3 (4 items), “internal feelings,” explained 5.9% of the variance and Iranian [10] versions include Item 6 in Factor 2, the ESES-I
and represents emotions that influence exercise. agrees with the Malaysian study [14] that belongs to Factor 1. Be-
As for the reliability, the 34 participating students demonstrated ing depressed (Item 6) was deemed a situational factor as students
clear test–retest reliability (ICC ¼ .91, 95% confidence face academic stressors, which might influence their exercise
interval ¼ 0.82e0.95). Cronbach's a was .78, .80, and .92 for Factors behavior. The placement of Items 14 and 15 in Factor 1 of the ESES-I
1, 2, and 3, respectively, based on the 312 respondents' data. is different from the other studies that include it in Factor 2
[7,10,14]. Failure to achieve one's exercise goals (Item 14) was more
of a situation than a competing demand that influences one's
Discussion confidence to exercise. Item 15 was considered as a situational/
interpersonal factor because the support from family or friends
The cross-cultural adaptation of ESES to Bahasa Indonesia requires interpersonal relationships, communications, or in-
resulted in a three-factor model with 16 items. The EFA and CVA teractions, which could influence exercise self-efficacy. The place-
analyses with MLE, 2SLS, and five-fold cross-validation showed that ment of Item 11 is inconsistent among various studies. Although the
the three-factor model had the best fit. Among the fit indices, only study population was similar, a study in Malaysia includes it in
the Chi-squared p-value in both the MLE and 2SLS procedures Factor 3, whereas the ESES-I retains it in Factor 2. The students
showed significance, indicating that there was a difference be- interpreted “work at home” as completing school assignments at
tween the model-implied population covariance and the observed home, which competed with their time and ability to exercise. The
sample covariance. However, the Chi-squared statistic is very sen- number and placement of items are different among the studies on
sitive to sample size and no longer relied on as a basis for accep- Asian populations. This study provides a more valid cross-cultural
tance or rejection. The larger the sample size, the more likely a adaptation because it was conducted from a comprehensive
model will fail to fit using the Chi-squared goodness-of-fit test [27]. approach and with recommended procedures rather than a simple
Although the GFI was below .9, it is also known to depend on the translation.
sample size [28]. This study found that Indonesians have different and specific
In the CFA with first-order, three-factor structure of the final views of engaging in exercise compared with other populations. As
model of ESES-I, two items (Items 7 and 8) were deleted. Feeling a multicultural country, the variety of cultures, religions, and
anxious (Item 7) does not influence Indonesian students’
 A.R. Hakim et al. / Asian Nursing Research 14 (2020) 300e305 305

worship observances may be some factors affecting people's exer- 7. Pajares F. Self-efficacy beliefs in academic settings. Rev Educ Res. 1996;66(4):
543e78.
cise self-efficacy. This study found that exercise self-efficacy varied
8. Bandura A, Pajares F, Urdan T, editors. Guide for constructing self-efficacy
among Indonesian university students at the item level. The lowest scales. Greenwich, CT: Self-efficacy beliefs of adolescents; 2006. p. 307e37.
score belonged to item 12 (“When visitors are present”), which 9. Shin Y, Jang H, Pender NJ. Psychometric evaluation of the exercise self-efficacy
indicated that, for students, receiving visitors was more important scale among Korean adults with chronic diseases. Res Nurs Health. 2001;24(1):
68e76.
than exercising. In Indonesia, treating visitors is an obligation, and https://doi.org/10.1002/1098-240X(200102)24:1<68::AID-NUR1008>3.0.CO;2-C
consequently, other activities are abandoned for politeness. Ac- 10. Everett B, Salamonson Y, Davidson PM. Bandura's exercise self-efficacy scale:
cording to Muslim practices, people are encouraged to visit each validation in an Australian cardiac rehabilitation setting. Int J Nurs Stud.
2009;46(6):824e9. https://doi.org/10.1016/j.ijnurstu.2009.01.016
other and “glorify” their guests [31]. Item 4 (“After recovering from 11. Van der Heijden MM, Pouwer F, Pop VJ. Psychometric properties of the Exercise
an injury that caused me to stop exercising”) had the highest score, Self-efficacy Scale in Dutch Primary care patients with type 2 diabetes mellitus.
indicating that students were not deterred by an injury. Perhaps Int J Behav Med. 2014;21(2):394e401.
https://doi.org/10.1007/s12529-013-9308-z
they had been taught the importance of exercising when recov- 12. Noroozi A, Ghofranipour F, Heydarnia AR, Nabipour I, Tahmasebi R, Tavafian SS.
ering from an injury to restore health. The unstable results of the The Iranian version of the exercise self-efficacy scale (ESES) Factor structure,
various studies adapting the ESES suggest the need for a factor internal consistency and construct validity. Health Educ J. 2011;70(1):21e31.
https://doi.org/10.1177/0017896910374547
analysis when making a new version of the scale. 13. Cornick JE. Factor structure of the exercise self-efficacy scale. Meas Phys Educ
Exerc Sci. 2015;19(4):208e18.
Limitations https://doi.org/10.1080/1091367X.2015.1074579
14. Liu KT, Kuan G, Arifin WN, Kueh YC. Psychometric properties of the self-efficacy
scale among undergraduate students in Malaysia. Malays J Med Sci.
The present study examined the ESES-I in the context of Indo- 2019;26(3):119e28. https://doi.org/10.21315/mjms2019.26.3.10
nesian university students only; thus, the scale should be validated 15. Tabachnick BG, Fidell LS. Using multivariate statistics. 3rd ed. New York:
HarperCollins; 1996. p. 48.
when used with other populations. The EFA and CFA were con- 16. Shoukri MM, Asyali M, Donner A. Sample size requirements for the design of
ducted using the same sample; the scale needs to be validated reliability study: review and new results. Stat Methods Med Res. 2004;13(4):
using an independent sample. Moreover, the test–retest reliability 251e71. https://doi.org/10.1191/0962280204sm365ra
17. Beaton DE, Bombardier C, Guillemin F, Ferraz MB. Guidelines for the process of
was conducted over a short interval. A future study should examine cross-cultural adaption of self-report measures. Spine. 2001;25(24):3186e91.
the stability of the ESES-I over a longer period to ensure longitu- https://doi.org/10.1097/00007632-200012150-00014
dinal reliability. 18. Polit DF, Beck CT. The content validity index: are you sure you know what's
being reported? Critique and recommendations. Res Nurs Health. 2006;29(5):
489e97. https://doi.org/10.1002/nur.20147
Conclusions 19. Kline RB. Principles and practice of structural equation modeling. 3rd ed. New
York: Guilford; 2011. p. 91e230.
20. Lu W, Miao J, McKyer EL. A primer on bootstrap factor analysis as applied to
The three-factor, 16-item ESES-Indonesian version demon- health studies research. Am J Health Educ. 2014;45(4):199e204.
strates cultural equivalence, reliability, and validity when used with https://doi.org/10.1080/19325037.2014.916639
university students. The differences in the number of factors and 21. Ahmad S, Zulkurnain NNA, Khairushalimi FI. Assessing the fitness of a mea-
surement model using confirmatory factor Analysis (CFA). IJIAS. 2016;17(1):
item placements, when compared with the scale's other language 159e68.
versions, may reflect the cultural differences among the countries. 22. Hooper D, Coughlan J, Mullen MR. Structural equation modelling: guidelines
for determining model fit. Electron J Bus Res Methods. 2008;6(1):53e60.
https://doi.org/10.21427/D7CF7R
Conflict of interest 23. Mohammed AA, Naugler C, Far BH. Emerging business intelligence framework
for a clinical laboratory through big data analytics. Emerging trends in
computational biology, bioinformatics, and systems biology: algorithms and
All authors declare no conflicts of interest in this study.
software tools. New York: Elsevier/Morgan Kaufmann; 2015. p. 577e602.
24. Cronbach LJ, Schoneman P, McKie D. Alpha coefficient for stratified-parallel
References tests. Educ Psychol Meas. 1965;25(2):291e312.
https://doi.org/10.1177/001316446502500201
25. Marx RG, Menezes A, Horovitz L, Jones EC, Warren RF. A comparison of two-
1. Vuori I. World health organization and physical activity. Prog Prev Med. time intervals for test-retest reliability of health status instruments. J Clin
2018;3(1):e0012. https://doi.org/10.1097/pp9.0000000000000012
Epidemiol. 2003;56(8):730e5.
2. Gonza lez K, Fuentes J, M
arquez JL. Physical inactivity, sedentary behavior and https://doi.org/10.1016/S0895-4356(03)00084-2
chronic diseases. Korean J Fam Med. 2017;38(3):111e5. 26. Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation
https://doi.org/10.4082/kjfm.2017.38.3.111
coefficients for reliability research. J Chiropr Med. 2016;15(2):155e63.
3. World Health Organization. Global recommendations on physical activity for https://doi.org/10.1016/j.jcm.2016.02.012
health [internet]. Geneva: World Health Organization; 2010 [cited 2018
27. Barrett P. Structural equation modelling: adjudging model fit. Pers Indiv Differ.
January 2]. Available from: https://www.who.int/dietphysicalactivity/ 2007;42(5):815e24. https://doi.org/10.1016/j.paid.2006.09.018
factsheet_recommendations/en/ 28. Mulaik SA, James LR, Van Alstine J, Bennett N, Lind S, Stilwell CD. Evaluation of
4. Badan Pusat Statistik. Penduduk menurut kelompok umur[internet]. Jakarta: goodness-of-fit indices for structural equation models. Psychol Bull.
Badan Pusat Statistik; 2019 [cited 2020 April 30]. Available from https://www. 1989;105(3):430e45. https://doi.org/10.1037/0033-2909.105.3.430
bps.go.id/statictable/2016/04/04/1904/penduduk-berumur-15-tahun-ke-atas-
29. Carruth E, Taylor N. The influence of aerobic exercise on state anxiety in college
menurut-golongan-umur-dan-jenis-kegiatan-selama-seminggu-yang-lalu- students. J Undergrad Res. 2009;7(2):1e11.
2008—2019.html. Bahasa Indonesia.
30. Markus HR, Kitayama S. Culture and the self: implications for cognition,
5. Farren GL, Zhang T, Martin SB, Thomas KT. Factors related to meeting physical emotion, and motivation. Psychol Rev. 1991;98(2):224e53.
activity guidelines in active college students: a social cognitive perspective. https://doi.org/10.1037/0033-295X.98.2.224
J Am Coll Health. 2017;65(1):10e21.
31. Tahir M, Zubairi SU. Towards the role of islamic education in promoting peace
https://doi.org/10.1080/07448481.2016.1229320 and harmony in a society: an analysis. Res J Islam Stud. 2017;4(1):25e36.
6. Bandura A. Self-efficacy mechanism in human agency. Am Psychol. 1982;37(2):
https://doi.org/10.2139/ssrn.2939061
122e47. https://doi.org/10.1037/0003-066X.37.2.122