YEBEH-106833; No of Pages 6

Epilepsy & Behavior xxx (xxxx) xxx

Contents lists available at ScienceDirect

Epilepsy & Behavior

journal homepage: www.elsevier.com/locate/yebeh

Indonesian Public Attitudes Toward Epilepsy (PATE) scale: Translation

and psychometric evaluation
Zhi Jien Chia a, Seilly Yunita Jehosua b, Kheng Seang Lim a,⁎, Herlyani Khosama b, Donny Hamdani Hamid c,
Si Lei Fong a, Chong Tin Tan a
a
Division of Neurology, Department of Medicine, Faculty of Medicine, University of Malaya, Malaysia
b
Department of Neurology, Faculty of Medicine, University of Sam Ratulangi/Prof RD Kandou Hospital, Manado, Indonesia
c
Division of Neurology, Pasar Rebo Hospital, Jakarta, Indonesia

a r t i c l e i n f o a b s t r a c t

Article history: Introduction: Epilepsy stigma has been associated with poor quality of life among people with epilepsy (PWE). It
Received 17 October 2019 is important to understand the variation and degree of epilepsy stigma in one of the most populous and culturally
Revised 25 November 2019 diverse nations in the world, Indonesia. Hence, this study aimed to test the validity and reliability of the
Accepted 25 November 2019 Indonesian version of the Public Attitudes Toward Epilepsy (PATE) scale.
Available online xxxx
Method: The translation was performed according to standard principles and tested in 200 native Indonesian
speakers who were aged above 18-year-old for psychometric validation.
Keywords:
Epilepsy
Results: The items in each domain had similar means and standard deviations (equal item variance), means rang-
Attitudes ing from 2.17 to 2.86 in general domain and 2.75 to 3.56 in personal domain and, standard deviations ranging
Indonesia from 0.87 to 1.05 and 0.88 to 1.01 in general and personal domain, respectively. Item-domain correlations
Stigma were more than 0.5 for all items, and they correlate higher within their own domain compare with the other
domain (convergent and divergent validity). Multitrait analysis showed similar variance, floor, and ceiling
patterns to a great extent compared with the initial study. The Indonesian PATE scale also showed mostly similar
correlation with demographic characteristics except monthly income. Principle axis analysis revealed strong
factor loading (N 0.3) in their hypothesized domain, except item 14. The Cronbach's α values for general and
personal domains were 0.836 and 0.765, which were within the accepted range of 0.7 to 0.9.
Conclusion: The Indonesian PATE scale is a validated and reliable translation for measuring public attitudes
toward epilepsy.
© 2019 Elsevier Inc. All rights reserved.

1. Introduction
Epilepsy is one of the most common neurological disorder character-
ized by enduring predisposition to epileptic seizures [1]. In the past, sei-
zures and epilepsy were often associated with sins, demonic possession,
insanity, and considered contagious, especially in Asian countries [2].
Although this has improved with better understanding of the biological
basis of the disease over the years, the remnant of adverse attitudes
remained disturbing to patients with epilepsy. This reflected in various
aspects of life, such as difficulties in finding jobs, lesser opportunity to
receive formal education, marriage, and friendships [2]. Oftentimes,
stigma in epilepsy is also associated with poorer quality of life and
other psychopathologies [3]. Demographic characteristics associated
with epilepsy stigma are low socioeconomic status, low education
⁎ Corresponding author at: Neurology Laboratory, University Malaya Medical Center,
6th Floor Menara Selatan, 59100 Kuala Lumpur, Malaysia.
E-mail address: kslimum@gmail.com (K.S. Lim).
level, older age group, single, and those who have no contact with peo-
ple with epilepsy [4].
The earliest epilepsy stigma study was done in 1949 by Caveness [5].
Subsequently, the publications increased dramatically since the 90s, es-
pecially in Asia, followed by Africa and Europe [4]. Many quantitative
scales were developed in attempt to quantify the magnitude of stigma,
such as the Attitudes and Beliefs about Living with Epilepsy (ABLE)
scale [6], Attitudes Toward Epilepsy and Persons with Epilepsy (ATPE)
[7], Epilepsy Attitude Scale [8], and Elementary School Epilepsy Survey
(ESES) [9]. However, some items in these scales are culturally depen-
dent and may not be accurate in certain populations. The Public Atti-
tudes Toward Epilepsy (PATE) scale was first designed with cultural
crossability in mind [10]. It was first used in Malaysia [11–14], a multi-
ethnic country with great cultural diversity, and was then translated
into a number of other languages in Asia [15–19].
As the largest island country in the world with a population of 264
million, Indonesia is also an ethnically diverse country with different
cultural backgrounds, which make it important to understand the vari-
ability and degree of the attitudes toward epilepsy in this country. In

https://doi.org/10.1016/j.yebeh.2019.106833
1525-5050/© 2019 Elsevier Inc. All rights reserved.

Please cite this article as: Z.J. Chia, S.Y. Jehosua, K.S. Lim, et al., Indonesian Public Attitudes Toward Epilepsy (PATE) scale: Translation and
psychometric evaluation, Epilepsy & Behavior, https://doi.org/10.1016/j.yebeh.2019.106833
2 Z.J. Chia et al. / Epilepsy & Behavior xxx (xxxx) xxx
Indonesia, Javanese (40%) is the largest ethnic groups in Indonesia,
followed by Sudanese (15%), Malays (3%), Sulawesians (3%), and others
[20]. However, they are distributed in different geographical locations,
e.g., Javanese in Java and Minahasan (Sulawesian) in Manado. Though
there are many languages spoken in Indonesia, Bahasa Indonesia
(Indonesian) has been adopted as the national language to facilitate
communication and national unity. It is taught in school and is now
widely spoken across the country. The aim of this study was to test
the validity and reliability of the Indonesian version of the PATE scale.
2. Methodology
2.1. PATE scale
The PATE scale is a 14-item scale that consists of 2 domains: general
and personal [10]. Each domain consists of 9 and 5 items, respectively.
General domain explores the general attitudes toward epilepsy without
personal involvement whereas personal domain requires the partici-
pant to consider personal commitment, such as dating, marriage, and
employment. Each item is scored using a 5-point Likert scale in which
1 indicates “strongly disagree” and 5 indicates “strongly agree”. The
scale is reversely scored in which higher score indicates poorer atti-
tudes. The PATE scale is a valid and reliable test instrument demonstrat-
ing appropriate content and construct validity. Cronbach's coefficient
was reported to be 0.633 and 0.868 for each respective dimension in
the original version [10].
2.2. Participants
Data were collected in 2 cities in Indonesia: Jakarta and Manado. Par-
ticipants from diverse socioeconomic background who were more than
18 years old were recruited via convenient sampling. Questionnaires
were distributed anonymously and were self-reported. Written consent
was obtained. Basic demographic characteristics were obtained, includ-
ing gender, age, marital status, education level, employment status, and
monthly income. This study was approved by the ethics committee of
Prof RD Kandou Hospital Medical Research Ethics Committee (MREC;
reference no.: PP 04.03/XIX.2/545) and University of Malaya Medical
Centre MREC (reference no.: 878.10).
2.3. Translation and cultural adaptation
The translation and cultural adaptation processes were based on the
Translation and Cultural Adaptation of Patient Reported Outcomes Mea-
sures—Principles of Good Practice (PGP) [21].
2.3.1. Translation
The original English PATE scale was transferred into a spreadsheet,
which was then forward-translated by two independent translators
into Indonesian language. The two forward translations were
reconciliated and were then back-translated into English by the other
two independent translators. Two backtranslations were reviewed
and reconciliated. The reconciliated backtranslation was compared
with the original PATE scale. Discrepancies were discussed by a panel
who consists of 3 neurologists; changes were made to ensure that the
original concepts were preserved. The final Indonesian PATE translation
was further examined for typo or grammatical error.
2.3.2. Cognitive debriefing
Pretest was carried out in 10 Indonesian native speakers from differ-
ent social backgrounds and education levels to ensure consistency in
comprehensibility. Difficulties in answering were also explored.
Please cite this article as: Z.J. Chia, S.Y. Jehosua, K.S. Lim, et al., Indonesian Public Attitudes Toward Epilepsy (PATE) scale: Translation and
psychometric evaluation, Epilepsy & Behavior, https://doi.org/10.1016/j.yebeh.2019.106833
2.3.3. Finalization
The face validity of the translated version was reviewed in an expert
meeting after considering the result of the pretesting and comparability
with the original PATE scale.
2.4. Statistical analysis
Sample size was based on the rule of 10 samples per variable in fac-
tor analysis, and the minimum sample size was 140 [22]. All data were
analyzed using Statistical Package for Social Sciences version 21 (SPSS
version 21). Demographic data were analyzed using descriptive statis-
tics and presented in frequencies and percentages. Means of each
group were compared using independent t-test.
2.5. Psychometric and validation testing
Validity of the translation was assessed using multitrait analysis and
principal axis factoring based on the following hypothesis:
1. Equal item variance: Items measuring the same concept should have
approximately equal variances (standard deviations).
2. Convergent validity: Variables should correlate well with the
hypothesized domain and roughly equal item-total correlation.
3. Discriminant validity: Each variable should correlate stronger within
their hypothesized domain compared with the other domain.
4. Construct validity: Confirmatory factor analysis using principal axis
factoring with direct oblimin rotation was used to test whether the
variables fit into its hypothesized domain as in the original study
[10]. The Kaiser–Meyer–Olkin (KMO) test was also examined to en-
sure sampling adequacy. In addition, it was hypothesized that
(a) the means and standard deviations and floor and ceiling effects
of this translation and (b) the relative relationship between the per-
sonal and general domains were comparable with the initial results
of the PATE scale in the Malaysian population [10]. The mean score
of each domain should also correlate with demographic characteris-
tics in a similar manner to the initial study [10].
2.6. Reliability
Internal consistency was evaluated, and Cronbach's α values of 0.7
to 0.9 were considered acceptable, whereas values of 0.6 to 0.7 were
considered satisfactory [23]. Mean inter-item correlation was used if α
values were below 0.7, with an acceptable range of 0.2 to 0.4 [24].
Item total correlation of more than 0.3 indicated good overall correla-
tion between each variable to the scale overall. Test–retest reliability
was not included as a measure of reliability as it may underestimate
the degree of reliability of measurements over time by interpreting
true change as measurement instability [25].
3. Results
Two hundred participants were recruited with a mean age of 35 ±
12.81 years, ranging from 17 to 64 years old, slightly younger than the
original study (41.4 ± 14.6 years). The number of male and female par-
ticipants were almost equal in this sample. Racial distribution was de-
pendent on geographical location, where in Jakarta consists of mostly
Javanese people and in Manado, the Sulawesians. This racial distribution
of the sample is representative of the population in their geographical
location, i.e., Javanese in Java and Sulawesian in Manado. Majority of
the participants were married (62%), had secondary education level or
lower (62.0%), were employed (58.5%), and belong to the low-income
group, with either no income or less than Indonesian Rupiah (INR) 1.5
million (United States Dollar [USD] 106)/month (52.5%) (Table 1).
 Z.J. Chia et al. / Epilepsy & Behavior xxx (xxxx) xxx 3

Table 1 3.2. Validation testing

Demographic characteristics of participants (n = 200).

Characteristics 3.2.1. Equal item variance

Mean (SD)
The means and standard deviations of each item in the Indonesian
Age, years 35 (12.81) PATE scale were tabulated in Table 2. In general domain, the mean
and standard deviation are largely similar, ranging from 2.17 to 2.86
Frequency (%)
Gender
and 0.87 to 1.05, respectively. Similar results can be seen in the personal
Male 101 (50.5) domain, with the means ranging from 2.75 to 3.56 and standard devia-
Female 99 (49.5) tion ranging from 0.88 to 1.01. In addition, the mean of the personal do-
Race main (3.13) was higher than general domain (2.39), which indicated a
Javanese 109 (54.5)
poorer attitude from a personal perspective.
Sumatera 1 (0.5)
Sulawesian 84 (42.0)
Indonesian Chinese 3 (1.5) 3.2.2. Convergent and discriminant validity
Others 3 (1.5) Variables within the same hypothesized domain have item-domain
Marital status
Married 124 (62.0)
correlation of more than 0.5 and correlate higher within the same hy-
Single 72 (36.0) pothesized domain than the other (Table 3).
Widowed 4 (2.0)
Education level
3.2.3. Construct validity
Secondary level or lower 124 (62.0)
Tertiary level (diploma and bachelor) 76 (38.0)
Employment status 3.2.3.1. Compatibility with the previous study. The means, standard devia-
Student 34 (17.0) tions, floor, and ceiling effects of each item were tabulated in Table 1 in
Housewife 26 (13.0)
comparison with the initial study in Malaysian general population [10].
Employed 117 (58.5)
Unemployed 16 (8.0)
Indonesian population had higher mean scores in both general and per-
Retired 7 (3.5) sonal domains compared with the initial study and specifically in items 3,
Individual monthly income 4, 5, 8, and 9 in the general domain and items 10, 11, 12, and 14 in personal
No income 74 (37.0) domain. Moreover, standard deviations of current study were higher than
bINR 1,500,000 (bUSD 106) 31 (15.5)
the initial study (range of 0.87–1.05 compared with 0.56–0.97). The floor
INR 1,500,000–INR 10,000,000 (USD 106–USD 705) 86 (43.0)
NINR 10,000,000 (NUSD 705) 9 (4.5) and ceiling effects possessed the same pattern compared with the original
Family member with epilepsy/seizures results in which the floor is higher than the ceiling. Two items (10 and
Yes 1 (0.5) 11) related to marital relationship in personal domain have higher ceiling
No 199 (99.5)
effects compared with floor, while the rest had almost equal percentages.

3.2.3.2. Correlation of mean score with demographic characteristics

3.1. Translation equivalence and acceptability
The backtranslation of the Indonesian PATE was comparable with
the original version. All 10 participants involved in the cognitive
debriefing were clear with the terminologies used, indicated that the
items were relevant and easy to understand and that they had no
issue in answering the questions. No change was made to the reconciled
Indonesian translation of the scale in the final expert panel meeting.
(Table 4). The mean scores of both domains were significantly lower
in those with higher education level. Married participants had signifi-
cantly higher mean scores compared with single or divorced counter-
parts in the personal domain but not in the general domain. On the
contrary, people from the higher monthly income group had signifi-
cantly lower mean scores in the general domain but not in personal do-
main. Age was positively correlated with mean scores in personal
domain. There was no significant difference in both domains among dif-
ferent genders and employment statuses.

Table 2
Mean, standard deviation, floor, and ceiling effects of each item.

Mean (std. dev) Floor/ceiling (%)

Indonesian PATE Initial study Indonesian Initial

(n = 200) (n = 130) PATE study

General domain 2.39 (0.63)⁎⁎⁎ 2.09 (0.59)

1. People with epilepsy should not study in college or university. 2.39 (1.05) 1.82 (0.70) 14.0/3.5 32.3/0.8
2. People with epilepsy have the same rights as all people. 2.17 (0.87) 2.05 (0.91) 16.5/1.0 26.2/2.3
3. People with epilepsy should be isolated from others. 2.15 (0.96)⁎ 1.94 (0.84) 21.5/3.0 32.3/0.8
4. People with epilepsy should not participate in social activities. 2.41 (0.93)⁎⁎⁎ 2.00 (0.78) 11.0/0.5 22.3/0.8
5. I will not mind to be seen in the company with someone known to have epilepsy. 2.29 (0.92)⁎ 2.06 (0.77) 12.0/3.0 16.9/1.5
6. People with epilepsy should not marry. 2.23 (0.95) 2.06 (0.78) 17.5/3.5 22.3/0.8
7. I would stay away from a friend if I know she/he has epilepsy. 2.16 (0.93) 2.08 (0.92) 19.5/2.0 21.5/3.8
8. People with epilepsy should study in a special school. 2.86 (1.00)⁎⁎⁎ 2.43 (0.97) 4.5/2.5 16.9/1.5
9. Schools should not place children with epilepsy into regular classrooms. 2.82 (0.98)⁎⁎⁎ 2.35 (0.90) 3.5/2.0 13.8/2.3
Personal domain 3.13 (0.69)⁎⁎⁎ 2.72 (0.56)
10. I would date someone even though he/she has epilepsy. 3.41 (0.95)⁎⁎⁎ 2.78 (0.84) 2.0/12.0 3.1/3.1
11. I will marry someone with epilepsy, even though he/she has epilepsy. 3.56 (1.00)⁎⁎⁎ 3.03 (0.85) 1.0/21.0 2.3/3.8
12. I feel uncomfortable working with someone who has epilepsy. 2.75 (0.88)⁎⁎⁎ 2.42 (0.87) 1.5/2.0 10.8/1.5
13. I will advise my family members against marrying someone with epilepsy. 3.08 (0.99) 2.95 (0.91) 4.5/4.5 6.9/2.3
14. If I am an employer, I would give equal employment opportunity to someone with epilepsy. 2.84 (1.01)⁎⁎⁎ 2.41 (0.96) 5.0/4.0 10.8/4.6

Significant at ⁎p b 0.05 and ⁎⁎⁎p b 0.001 compared with the results of the initial study performed among a Malaysian population [10].

Please cite this article as: Z.J. Chia, S.Y. Jehosua, K.S. Lim, et al., Indonesian Public Attitudes Toward Epilepsy (PATE) scale: Translation and
psychometric evaluation, Epilepsy & Behavior, https://doi.org/10.1016/j.yebeh.2019.106833
4 Z.J. Chia et al. / Epilepsy & Behavior xxx (xxxx) xxx

Table 3
Item-domain correlation and item-total correlation.

General Personal Corrected Cronbach's α

domain domain item-total if item deleted
correlation

General domain
1. People with epilepsy should not study in college or university. 0.55⁎ 0.13 0.45 0.83
2. People with epilepsy have the same rights as all people.a 0.51⁎ 0.19⁎ 0.50 0.82
3. People with epilepsy should be isolated from others. 0.66⁎ 0.23⁎ 0.58 0.82
4. People with epilepsy should not participate in social activities. 0.68⁎ 0.19⁎ 0.62 0.81
5. I will not mind to be seen in the company with someone known to have epilepsy.a 0.58⁎ 0.19⁎ 0.51 0.82
6. People with epilepsy should not marry. 0.63⁎ 0.32⁎ 0.59 0.81
7. I would stay away from a friend if I know she/he has epilepsy. 0.69⁎ 0.28⁎ 0.67 0.81
8. People with epilepsy should study in a special school. 0.67⁎ 0.35⁎ 0.48 0.83
9. Schools should not place children with epilepsy into regular classrooms. 0.66⁎ 0.30⁎ 0.52 0.82
Personal domain
10. I would date someone even though he/she has epilepsy.a 0.25⁎ 0.80⁎ 0.66 0.68
11. I will marry someone with epilepsy, even though he/she has epilepsy.a 0.34⁎ 0.85⁎ 0.74 0.64
12. I feel uncomfortable working with someone who has epilepsy. 0.42⁎ 0.64⁎ 0.51 0.73
13. I will advise my family members against marrying someone with epilepsy. 0.25⁎ 0.69⁎ 0.50 0.73
14. If I am an employer, I would give equal employment opportunity to someone with epilepsy.a 0.23⁎ 0.57⁎ 0.30 0.80

Numbers in bold indicate that the correlation of the item is higher with its hypothesized scale than with scales measuring other concepts.
a
These items were scored in reverse.
⁎ Spearman's correlation b0.01 (2-tailed).

3.2.3.3. Correlation between the items and rotated principal components
(Table 5). The suitability for factor analysis was examined. Kaiser–
Meyer–Olkin test showed a result of 0.8, which was above the accepted
range of 0.6, suggesting a good sampling adequacy [26,27]. Bartlett's
Test of Sphericity was statistically significant [28]. Principle axis factor
analysis revealed most items having strong factor loading (≥ 0.3) in
line with their hypothesized domain, except item 14 with loading
b0.3. The two-factor model explained a total of 41.2% of the variance,
with factor 1 contributing 29.37% and factor 2 contributing 11.80%.
3.3. Reliability
The Cronbach's α values for general and personal domains were
0.836 and 0.765, respectively, and 0.840 for the overall PATE scale,
which were within the accepted range of 0.7 to 0.9. The mean inter-
item correlations were 0.364 and 0.396 for general and personal do-
mains, respectively. Both domains correlated strongly with the total
mean score, with correlation coefficients of 0.887 and 0.747 (p b 0.01).
Likewise, each individual item in the Indonesian PATE scale correlated
well with the scale in general, with all item-total correlation of more
than 0.3 (Table 3).
4. Discussion
This study showed that the Indonesian PATE scale is a reliable and
validated translated version. It was translated according to standard
principles [21] and tested with multitrait analysis and principal axis fac-
tor analysis, with an acceptable internal consistency in both general and
personal domains.
Though convenient sampling was used in this study, we managed to
recruit participants from various socioeconomic backgrounds with a
wide range of age (17–64 years) with equal gender representation.
The mean age of participants was compatible to the initial study (35.0
vs. 41.4 years, respectively) [10]. This enables generalization of the re-
sults to the general Indonesian population.
Multitrait analysis showed a similar variance, floor, and ceiling ef-
fects (Table 2) and relative relationship between the domains as the ini-
tial study. It also fulfilled the criteria for equal item variance, convergent,
and discriminant validity. Current study had smaller floor effects in all
items compared with the initial study and higher ceiling effects in
most of the items in personal domain. This was expected because of
the overall more negative attitudes toward epilepsy in Indonesian pop-
ulation than the initial study.

Table 4
Mean scores by domains and demographic characteristics (n = 200).

Variable General domain Personal domain Total

Age a
0.073 0.167⁎,b 0.135

Variable Group (n) Mean (SD) Mean (SD) Mean (SD)

Gender Male (101) 2.37 (0.59) 3.12 (0.69) 2.64 (0.53)

Female (99) 2.40 (0.67) 3.13 (0.70) 2.66 (0.57)
Marital status Married (124) 2.44 (0.65) 3.23 (0.69) 2.73 (0.56)
Others (76) 2.29 (0.59) 2.95 (0.66)⁎⁎ 2.52 (0.51)⁎
Education level Tertiary (76) 2.16 (0.51)⁎⁎⁎,b 2.88 (0.61)⁎⁎⁎ 2.42 (0.45)⁎⁎⁎
Secondary or lower (124) 2.52 (0.65) 3.28 (0.70) 2.79 (0.55)
Employment status Employed (117) 2.34 (0.57) 3.12 (0.74) 2.62 (0.53)
Not employed (82) 2.44 (0.70) 3.13 (0.62) 2.69 (0.57)
Monthly income Low: bINR 1.5 million or USD 106 (105) 2.53 (0.71)⁎⁎⁎ 3.15 (0.61) 2.75 (0.58)⁎⁎
High: ≥INR 1.5 million or USD 106 (95) 2.23 (0.48) 3.10 (0.78) 2.54 (0.49)

INR, Indonesian Rupiah.

⁎ Significant level at p b 0.05.
⁎⁎ Significant level at p b 0.01.
⁎⁎⁎ Significant level at p b 0.001.
a
Based-on Pearson correlation.
b
This difference was comparable with the initial PATE study.

Please cite this article as: Z.J. Chia, S.Y. Jehosua, K.S. Lim, et al., Indonesian Public Attitudes Toward Epilepsy (PATE) scale: Translation and
psychometric evaluation, Epilepsy & Behavior, https://doi.org/10.1016/j.yebeh.2019.106833
 Z.J. Chia et al. / Epilepsy & Behavior xxx (xxxx) xxx 5

Table 5
Correlation between the hypothesized domain and the rotated principal components.

Item Hypothesized domain Factor

1 2

1. People with epilepsy should not study in college or university. General 0.49
2. People with epilepsy have the same rights as all people. General 0.54
3. People with epilepsy should be isolated from others. General 0.68
4. People with epilepsy should not participate in social activities. General 0.69
5. I will not mind to be seen in the company with someone known to have epilepsy. General 0.61
6. People with epilepsy should not marry. General 0.65
7. I would stay away from a friend if I know she/he has epilepsy. General 0.78
8. People with epilepsy should study in a special school. General 0.40
9. Schools should not place children with epilepsy into regular classrooms. General 0.46
10. I would date someone even though he/she has epilepsy. Personal 0.84
11. I will marry someone with epilepsy, even though he/she has epilepsy. Personal 0.95
12. I feel uncomfortable working with someone who has epilepsy. Personal 0.43
13. I will advise my family members against marrying someone with epilepsy. Personal 0.60
14. If I am an employer, I would give equal employment opportunity to someone with epilepsy. Personal 0.29

Extraction method: Principal axis factoring.

Rotation method: Oblimin with Kaiser normalization.

The Indonesian PATE scale showed an almost similar correlation References

with demographic characteristics as in the initial Malaysian study [10]
whereby increasing age and married individuals were found to be cor-
related with more negative attitudes in the personal domain. In the ini-
tial study, higher education level showed more positive attitudes in
general domain but not in personal domain, contrary to the findings
from the current study which showed significantly more positive atti-
tudes in both domains among the highly educated. Current Indonesian
study also showed that those with higher monthly income had more
positive attitudes in the general domain, which were not found in the
initial study. These have shown that education level and socioeconomic
status were important factors affecting the attitudes toward epilepsy,
especially in Indonesia.
Although item 14 on employing people with epilepsy loaded stron-
ger on its hypothesized domain, it has shown a slightly weak factor
loading (b0.3) (Table 5). This may due to the nature of work in the re-
cruited cohort which are mostly self-employed (such as farmer or mer-
chant) or government servant who may not have a clear perception
about employing someone. In view of the significant correlation in
mean score between the item 14 and personal domain, no modification
has been made to the translated version to preserve the consistency of
comparison among various translations but will be tested again in fu-
ture studies.
The reliability of the Indonesian PATE scale was proven by
Cronbach's α values that were within an acceptable range in both gen-
eral and personal domains. In addition, the mean inter-item correlations
were within the acceptable range of 0.2 to 0.4, and all item-total corre-
lations were above 0.3 (Table 3). This indicated that the Indonesian
PATE scale translation is a reliable tool to measure public attitudes to-
ward epilepsy.
5. Conclusion
The Indonesian PATE scale is a validated and reliable translation for
measuring public attitudes toward epilepsy.
Declaration of competing interest
None of the authors has any competing interest to declare.
Acknowledgment
This study was sponsored by a University of Malaya Research Grant
from the university research fund (reference no.: RP052C-17HTM). We
would also like to acknowledge Prof. Dr. Choo Wan Yuen for her assis-
tance in statistics.
[1] Fisher RS, Acevedo C, Arzimanoglou A, Bogacz A, Cross JH, Elger CE, et al. ILAE official
report: a practical clinical definition of epilepsy. Epilepsia 2014;55:475–82.
[2] Mac TL, Tran DS, Quet F, Odermatt P, Preux PM, Tan CT. Epidemiology, aetiology, and
clinical management of epilepsy in Asia: a systematic review. Lancet Neurol 2007;6:
533–43.
[3] Lee SA, Yoo HJ, Lee BI, Group KQiES. Factors contributing to the stigma of epilepsy.
Seizure 2005;14:157–63.
[4] Lim KS, Lim CH, Tan CT. Attitudes toward epilepsy, a systematic review. Neurol Asia
2011;16:269–80.
[5] Caveness W. A survey of public attitudes toward epilepsy. Epilepsia 1949;4:
19–26.
[6] DiIorio CA, Kobau R, Holden EW, Berkowitz JM, Kamin SL, Antonak RF, et al. Devel-
oping a measure to assess attitudes toward epilepsy in the US population. Epilepsy
Behav 2004;5:965–75.
[7] Antonak RF. Psychometric analysis and validation of the scale of attitudes toward
persons with epilepsy. J Epilepsy 1990;3:11–6.
[8] Aydemir N. Familiarity with, knowledge of, and attitudes toward epilepsy in Turkey.
Epilepsy Behav 2011;20:286–90.
[9] Crystal LP, Camfield CS. Elementary school epilepsy survey (ESES): a new measure of
elementary school students' knowledge and attitudes about epilepsy. Epilepsy
Behav 2005;7:687–96.
[10] Lim KS, Wu C, Choo WY, Tan CT. Development and validation of a public attitudes
toward epilepsy (PATE) scale. Epilepsy Behav 2012;24:207–12.
[11] Lim KS, Choo WY, Wu C, Hills M, Tan CT. Chinese Public Attitudes Toward Epi-
lepsy (PATE) scale: translation and psychometric evaluation. Neurol Asia
2013;18:261–70.
[12] Lim KS, Choo WY, Wu C, Tan CT. Malay public attitudes toward epilepsy (PATE)
scale: translation and psychometric evaluation. Epilepsy Behav 2013;29:
395–9.
[13] Lim KS, Hills MD, Choo WY, Wong MH, Wu C, Tan CT. Attitudes toward epilepsy
among the primary and secondary school teachers in Malaysia, using the public at-
titudes toward epilepsy (PATE) scale. Epilepsy Res 2013;106:433–9.
[14] Lim KS, Hills MD, Choo WY, Wong MH, Wu C, Tan CT. A web-based survey of at-
titudes toward epilepsy in secondary and tertiary students in Malaysia, using
the Public Attitudes Toward Epilepsy (PATE) scale. Epilepsy Behav 2013;26:
158–61.
[15] Mohsen AK, Kolahi AA, Farsar AR, Kermaniranjbar S. Public awareness, attitudes, and
first-aid measures on epilepsy in Tehran. Iran J Child Neurol 2018;13:91–106.
[16] Lee SA, Choi EJ, Jeon JY, Paek JH. Attitudes toward epilepsy and perceptions of
epilepsy-related stigma in Korean evangelical Christians. Epilepsy Behav 2017;
74:99–103.
[17] Yang K, He Y, Xiao B, Wang J, Feng L. Knowledge, attitudes and practice towards ep-
ilepsy among medical staff in Southern China: does the level of hospitals make a dif-
ference? Seizure 2019;69:221–7.
[18] Yang K, Ma C, He Y, Wang J, Yue Z, Xiao B, et al. Attitudes toward epilepsy among
medical staffs in basic-level hospitals from southern China. Epilepsy Behav 2018;
89:23–9.
[19] Yang K, Yue Z, Zhou M, Wang J, Xiao B, Feng L. Investigation into attitudes toward
epilepsy among non-/neurological doctors and nurses in southern China. Epilepsy
Res 2019;154:79–85.
[20] Indonesia S. Nationality, ethnicity, religion, and daily language of Indonesian popu-
lation; 2010.
[21] Wild D, Grove A, Martin M, Eremenco S, McElroy S, Verjee-Lorenz A, et al. Principles
of good practice for the translation and cultural adaptation process for patient-
reported outcomes (PRO) measures: report of the ISPOR Task Force for Translation
and Cultural Adaptation. Value Health 2005;8:94–104.
[22] Pett MA, Lackey NR, Sullivan JJ. Designing and testing the instrument. In: Pett
MA, Lackey NR, Sullivan JJ, editors. Making sense of factor analysis: the use of

Please cite this article as: Z.J. Chia, S.Y. Jehosua, K.S. Lim, et al., Indonesian Public Attitudes Toward Epilepsy (PATE) scale: Translation and
psychometric evaluation, Epilepsy & Behavior, https://doi.org/10.1016/j.yebeh.2019.106833
6 Z.J. Chia et al. / Epilepsy & Behavior xxx (xxxx) xxx

factor analysis for instrument development in health care research. Sage; 2003.
p. 13–50.
[23] DeVellis RF. Scale development: theory and applications. Sage publications;
2016.
[24] Briggs SR, Cheek JM. The role of factor analysis in the development and evaluation of
personality scales. J Pers 1986;54:106–48.
[25] Carmines EG, Zeller RA. Reliability and validity assessment. Sage publications;
1979.
[26] Cerny BA, Kaiser HF. A study of a measure of sampling adequacy for factor-analytic
correlation matrices. Multivar Behav Res 1977;12:43–7.
[27] Kaiser HF. An index of factorial simplicity. Psychometrika 1974;39:31–6.
[28] Bartlett MS. A note on the multiplying factors for various χ 2 approximations. J R Stat
Soc B Methodol 1954:296–8.

Please cite this article as: Z.J. Chia, S.Y. Jehosua, K.S. Lim, et al., Indonesian Public Attitudes Toward Epilepsy (PATE) scale: Translation and
psychometric evaluation, Epilepsy & Behavior, https://doi.org/10.1016/j.yebeh.2019.106833