Volume 74, No.4: 2022 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

217

Original Article

SMJ

Pat Rojmahamongkol, M.D., Supaluk Tangvalelerd, M.D.

Department of Pediatrics, Faculty of Medicine, Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.

The Reliability of the Thai Version of the Toddlers’

Temperament Questionnaire

ABSTRACT

Objective: To evaluate the reliability of the Early Childhood Behavior Questionnaire (ECBQ)-very short form, ai

version; and to investigate ai toddlers’ temperaments

Materials and Methods: e English version of the very short form ECBQ was translated into ai language. e

primary caregivers of 360 healthy, 18-36 months old children, were asked to ll the questionnaire from the period

of April, 2018 to June, 2019. e scores were calculated to determine the child’s temperament. e reliability of

the very short form ECBQ- ai version was assessed by Cronbach’s alpha coecient for internal consistency and

Intraclass correlation coecient (ICC) for inter-rater reliability, two-week and six-month test-retest reliability.

Results: e Cronbach’s alpha coecients were 0.627-0.692, which indicated questionable internal consistency,

but almost reach those of the original ECBQ. e inter-rater ICCs were 0.463-0.670, which were comparable to

those in the original English version. e two-week ICCs were 0.602-0.750, which indicated moderate reliability,

whereas the six-month ICCs decreased to 0.459-0.602. However, these values almost reached those of the original

English version. Most toddlers were reported to have a surgency temperament. Boys signicantly had more surgency

temperament and had higher mean surgency scores, whereas girls exhibited a more eortful-control temperament.

Conclusion: e very short-form ECBQ, ai version can be implemented in assessing ai toddlers’ temperament.

e majority of ai toddlers in this study demonstrated surgency temperament. Boys tended to be more surgency,

while girls tended to be more eortful control.

Keywords: temperament, ECBQ questionnaire, ai toddlers (Siriraj Med J 2022; 74: 217-224)

Corresponding author: Pat Rojmahamongkol

E-mail: patojung@yahoo.com

Received 14 January 2022 Revised 24 January 2022 Accepted 25 January 2022

ORCID ID: https://orcid.org/0000-0003-3853-0026

http://dx.doi.org/10.33192/Smj.2022.27

INTRODUCTION

Temperament is dened as an inborn dierence

in reactivity and self-regulation when a child interacts

with his or her environment. Reactivity is the child’s

reaction to changes in the environment. Self-regulation

is a process to modulate one’s reactivity. Although one’s

temperament persists throughout life, heredity, maturation,

and experiences inuence it over time

, for example,

parental discipline and responses to a child’s behavior

can aect the child’s expressions and habits. e most

popular temperament concept among pediatricians

has been omas and Chess’s approach, which was

introduced in 1977.

is concept has included nine

dimensions of temperament, which were activity level,

rhythmicity or regularity of physical functions, approach

to or withdrawal from a new situation, adaptability, one’s

sensory threshold to external stimuli, intensity of reactions,

quality of mood, distractibility, and attention span or

persistent focus on tasks. Children are categorized from

the nine dimensions into three groups, which are easy,

dicult, and slow-to-warm up ones. e “easy child” is

easy to raise and care for, exhibits regular sleep, waking

Volume 74, No.4: 2022 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

218

and feeding times, good moods, and high adaptability.

e “dicult child” has characteristics that are opposite

to those in the rst category. e “slow-to-warm-up

child” demonstrates initial withdrawal from new stimuli

and needs more time to adapt to those new situations,

though eventually he or she can adjust to such stimuli.

Some behavioral problems have been claimed to be

associated with temperament. e child who has a very

high activity level and is very distractible might mimic

the one with attention decit/ hyperactivity disorder

(ADHD). Toddlers with autism exhibited more sensitivity

to stimuli and have more negative moods, compared to

their typically developing peers.

Moreover, the child’s

temperament may have bidirectional interactions with

the parents’ responses and psychopathology. Parents

of negative-mood and low-impulse-control children

tended to report a history of anxiety.

One study found

that the authoritative parenting style - when parents were

always sensitive to the child’s needs and give him or her

freedom to make decisions in general situations while

limiting the child’s decision-making in more serious

ones, especially potentially dangerous activities - was

correlated with more patience and less irritability among

their children.

is highlights the fact that a successful

management strategy for behavioral problems based on

temperamental origin should harmonize with the child’s

temperament. us, a correct interpretation of the child’s

temperament helps pediatricians and other medical

personnel to assist parents in better understanding their

children’s characteristics and to guide parents in dealing

with his or her problems.

Although omas and Chess’s temperament concept

is clinically practical for the evaluation of children’s

behavior and in terms of offering recommendation

to parents, however, the Toddler Temperament Scale

questionnaire developed by Fullard et al, in 1984

measure the nine dimensions of temperament contains

97 question items which is not clinically practical for ai

context. Most of ai-version questionnaires developed

for time-constraint clinical use consists of 15-40 question

items.

7-9

ere is also the concept of temperament developed

by Rothbart in 1981. Rothbart’s temperament is dened

as biologically-rooted individual dierences in reactivity

and self-regulation in one’s emotional, activational,

and attentional processes.

In their eort to measure

children’s temperaments, Putnam et al.

formulated 18

scales of temperamental components based on Rothbart’s

temperamental concept, which are shown in the appendix.

e children were classied into three groups: “surgency,”

which refers to the ones with a high activity level and

positive anticipation regarding new activities; “negative

aectivity,” referring to children who display fear and

frustration; and “eortful control,” which applies to those

with high impulse control and attentiveness to tasks. is

approach led to the development of the Early Childhood

Behavior Questionnaire (ECBQ),

which was designed

to assess temperament in children from 18-36 months

old. e psychometric properties have demonstrated

good internal consistency and reliability.

ere are three

versions of the ECBQ - the standard ECBQ (consisting

of 201 items), the short-form ECBQ (107 items), and

the very short-form ECBQ (36 items).

e ECBQ has been widely used in English and in

more than 18 languages, such as Chinese, Japanese, and

French. e very short-form version has been widely

implemented in English and ten non-English-speaking

countries.

To the best of our knowledge, there is no temperament

questionnaire for ai toddlers. As a result, research on

temperament in the ai population has been very limited.

e aim of this study was to translate the original English

very short form of ECBQ into ai and to evaluate the

reliability of the ai version. We also aimed to study ai

toddlers’ temperament by using this form. e English

version of the very short-form ECBQ was selected for

this research study because the number of the question

items is feasible for implementation in busy clinical

settings in ailand.

MATERIALS AND METHODS

e details of the original questionnaire

The very short version of the Early Childhood

Behavior Questionnaire (ECBQ) is used to evaluate

the temperament of 18-36-month-old children. e

questionnaire includes 36 items regarding children’s

behaviors, which are specic to each temperament type.

ere are 12 items for surgency, 12 for eortful control, and

another 12 for negative aectivity. Parents or caregivers

give a Likert scale score regarding the frequency of their

child’s behaviors. e scores range from 1 (never) to 7

(always), and “Does not apply” (if the question is not

relevant to the child). e mean score of each temperament

type is calculated, and the highest mean scores dene

the child’s temperament type. If there is more than one

equally highest mean score, the child is classied as

more than one - i.e. a mixed-temperament type. Internal

consistency was evaluated by using Cronbach’s alpha

coecient. e internal consistency of negative aectivity,

surgency, and eortful-control temperament were 0.70,

0.72, 0.72, respectively. e longitudinal stability over

six months was 0.65.

Rojmahamongkol et al.

Volume 74, No.4: 2022 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

219

Original Article

SMJ

e translation process

e ai version of the very short-form ECBQ

was translated with the permission of the pioneer of

the original English version. e content validity was

measured by three ai developmental and behavioral

pediatricians. e acceptable content validity index was

0.67. Each item’s content validity was initially ranged

from 0.33 to 0.67. e items which had low content

validity were retranslated. Aer substantial agreement

from those three specialists, reected by a content validity

index of 1, the translation back into English was done to

conrm that the ai version adhered with the original

English version.

Population and procedure

is cross-sectional, questionnaire-based research

included caregivers of 18-to-36-month-old children

who attended the well child clinic and daycare at Siriraj

Hospital, or who were in pre-kindergarten and daycare

in Bangkok and the surrounding metropolitan area. e

recruitment period was from April, 2018 to June, 2019.

e exclusion criteria were those whose children had a

diagnosis of developmental delay or chronic diseases.

Caregivers who could not read ai were also excluded.

is study was approved by the Human Research Protection

Unit, Faculty of Medicine Siriraj Hospital, Mahidol

University.

Aer signing an informed consent, 360 caregivers

were asked to ll out the 20-minute ai very short-form

ECBQ. Of these, 20 were randomized for two-week (short-

term) test-retest reliability; another 20 were allocated for

six-month (long term) test-retest reliability. Eighty-three

pairs of primary and secondary caregivers were recruited

for inter-rater reliability testing. e number of the

participants in each group randomized for test-retest and

inter-rater reliabilities were based on the reliabilities from

the original English version (0.6 and 0.31, respectively),

estimated type I error of 0.05, and the power of 80%. e

scores obtained from the questionnaire were calculated

so as to interpret the children’s temperament. At the end

of the study, all caregivers received the information about

their child’s temperament, as well as the developmental

and behavioral pediatrician’s guidance in child-rearing

practices specic to each child’s temperament.

Statistical analysis

e data were prepared and analyzed using PASW

Statistics 18.0 (SPSS Inc., Chicago, IL, USA). Descriptive

statistics were used to describe the participants’ characteristics.

Number and percentage were used to describe qualitative

data, while mean and standard deviation (SD) were used

to describe normally distributed quantitative data. A

median (25

, 75

percentile) was used to report non-

normally distributed data. e Chi-Square test was used

to evaluate statistical signicance.

Cronbach’s alpha coecient was used to evaluate the

internal consistency of the questionnaire. e coecient,

which was < 0.5, 0.5-0.6, 0.6-0.7, and > 0.7, was dened

as unacceptable, poor, questionable, and good internal

consistency, respectively.

13,14

Test-retest reliability was

interpreted by using the intraclass correlation coecient

(ICC). Inter-rater reliability was evaluated using the two-

way random-eects model, while intra-rater reliability

was evaluated by using the two-way mixed-eects model.

In this study, an ICC which was < 0.5, 0.5-0.75, and

0.75-0.9 was classied as of poor, moderate, and good

reliability, consecutively.

RESULTS

e demographic data are shown in Table 1. Of the

360 caregivers, 323 (89.7%) answered the questionnaire

regarding the child’s gender. e majority of these caregivers

were mothers. e short-term (two weeks) test-retest

reliability was performed involving 20 fathers or mothers

of the children, which included nine (45%) boys and 11

(55%) girls. Long-term (six months) test-retest reliability

was performed involving 20 participants whose children

were 11 (55%) boys and nine (45%) girls. Sixteen out

of 20 caregivers, who were randomized for long- term

test-retest reliability, were fathers or mothers. Another 4

out of 20 were other family members. Of the 83 pairs of

primary and secondary caregivers recruited for inter-rater

reliability, 46 (55.4%) were the caregivers of boys, and

37 (45.6%) were the caregivers of girls. All the children

were ai and lived in Bangkok and the surrounding

metropolitan area. e mean age (SD) was 27.9 (6.2)

months old.

Table 2 demonstrates Cronbach’s alpha coecient

for each temperament and the range of Cronbach’s

alpha coecient when one item was deleted. e overall

Cronbach’s alpha coecients were 0.682, 0.627, and

0.692 for surgency, negative aectivity, and eortful-

control temperament, respectively. e maximum of

the coecient when one item was deleted was 0.707.

Regarding the inter-rater reliability, the ICCs were

0.638, 0.670, and 0.463 for surgency, negative aectivity,

and eortful-control temperament, respectively. e

ICCs for short-term intra-rater reliability were 0.614,

0.750, and 0.602 for surgency, negative aectivity, and

eortful-control temperament, respectively. For long-

term one, the ICCs decreased to 0.602, 0.459, and 0.476,

respectively. (Table 3)

Volume 74, No.4: 2022 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

220

TABLE 1. Demographic data of the children and the caregivers

TABLE 2. Cronbach’s alpha coecient for each temperament.

Characteristics Number (%)

Child’sgender

 Boy 174(53.9)

 Girl 149(46.1)

 Missingdata 37

Child’sAge(N=357)

 18-23months 98(27.5)

 24-30months 110(30.8)

 31-36months 149(41.7)

 Missingdata 3

Caregivers

 Parents 346(96.1)

 Grandparents 12(3.3)

 Others 2(0.5)

Caregivers’maritalStatus

 Single 32(8.9)

 Married 317(88.1)

 Divorced 11(3.1)

Caregivers’education

 UnderBachelordegree 42(11.7)

 Bachelordegree 194(53.9)

 AboveBachelordegree 124(34.4)

Family’smonthlyincome

 ≤30,000Bahts 111(31.7)

 30,001-50,000Bahts 107(30.6)

 50,001-100,000Bahts 82(23.4)

 ≥100,001Bahts 50(14.3)

 Missing 10

 Temperament Cronbach’salphacoefcient Cronbach’salphacoefcient

when item is deleted

 Surgency 0.682 0.645to0.692

 NegativeAffectivity 0.627 0.576to0.636

 EffortfulControl 0.692 0.649to0.707

Rojmahamongkol et al.

Volume 74, No.4: 2022 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

221

Original Article

SMJ

TABLE 3. Inter-rater reliability, short-term and long-term intra-rater reliabilities.

Temperament ICC

(95% CI) ICC

(95% CI)

 Surgency 0.638(0.495-0.747) 0.614(0.228-0.832) 0.602(0.244-0.819)

 NegativeAffectivity 0.670(0.536-0.772) 0.750(0.461-0.896) 0.459(0.021-0.746)

 EffortfulControl 0.463(0.280-0.614) 0.602(0.231-0.824) 0.476(0.040-0.756)

ICC: Intraclass correlation coecient

Intraclass correlation coecient between primary and secondary caregivers

Intraclass correlation coecient, two weeks apart (short-term)

Intraclass correlation coecient, six months apart (long-term)

e temperament classications of 360 ai children

in this study were 205 (56.9%) for surgency, 135 (37.5%)

for eortful control, and 3 (0.8%) for negative aectivity.

ere were 16 (4.4%) children whose caregivers rated

their temperaments as surgency and eortful control

equally; one (0.3%) child was classied as manifesting

surgency and negative aectivity. Boys signicantly had

more surgency temperament and a higher mean surgency

scores than girls. Girls exhibited more eortful-control

than boys. ere was no statistical signicance of gender

dierence in terms of negative aectivity and mixed

temperament. (Table 4)

TABLE 4. Descriptive statistics for the scale scores of each temperament in boys and girls.

Temperament Boys Girls p-value

Surgency  

N(%)  107(61.5) 72(48.3) 0.019*

Meanscores±SD 5.36±0.62 5.16±0.69 0.008*

NegativeAffectivity  

N(%)  0 3(2) 0.097

Meanscores±SD 2.86±0.67 3.00±0.73 0.056

EffortfulControl  

N(%)  57(32.8) 68(45.6) 0.022*

Meanscores±SD 5.11±0.65 5.09±0.68 0.740

Surgency-EffortfulControlequally  

N(%)  10(5.7) 5(3.4) 0.428

Meanscores±SD N/A N/A N/A

Surgency-NegativeAffectivityequally  

N(%)  0 1(0.3) 0.461

Meanscores±SD N/A N/A N/A

p-value < 0.05 is considered statistically signicant

* p-value < 0.05

Volume 74, No.4: 2022 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

222

DISCUSSION

is is the rst study in ailand that has used the

translated very short form of Early Childhood Behavior

Questionnaire (ECBQ), which is the clinically practical

evaluative tool for assessing ai toddlers’ temperament.

is study is also the rst preliminary report on ai

toddlers’ temperament. e majority of ai toddlers in

this study demonstrated surgency temperament. Boys

exhibited more surgency, while girls manifested more

eortful control.

The very short form of ECBQ, Thai version demonstrated

an internal consistency coecient at 0.682 for surgency

temperament, 0.627 for negative-aectivity temperament,

and 0.692 for eortful-control temperament. Although these

coecients suggested questionable internal consistency,

they were slightly lower than those of the original English

version (0.72 for surgency, 0.7 for negative aectivity,

and 0.72 for eortful control).

e maximum of the

Cronbach’s alpha coecient of the ai version of the

questionnaire, when any item was deleted, was 0.707,

which was slightly higher than the overall Cronbach’s alpha

coecient. is suggests that no item disproportionately

aected the overall internal consistency (i.e. No specic

item should be removed from the ai version of the

very short-form ECBQ). A recent study translating the

very short-form ECBQ into Czech, which recruited 709

children ages 18-36 months, also revealed an internal

consistency by Cronbach’s alpha coecients of 0.65, 0.70,

and 0.71 for surgency, negative aectivity, and eortful

control, consecutively. However, there was no report

regarding the inter-rater and test-retest reliabilities.

e intraclass correlation coecients exhibited

moderate to good short-term intra-rater reliability

(0.614, 0.602, and 0.750 for surgency, eortful control,

and negative aectivity, respectively). In contrast, the

long-term one had poor to moderate reliability (0.602

for surgency, 0.476 for eortful control, and 0.459 for

negative-aectivity temperament). We postulated that the

decline in the reliability might be due to chronological

fading. However, the long-term ICCs almost reached

those of the original English version, which was ranged

from 0.55-0.83.

e inter-rater reliability was classied

as being of poor to moderate reliability - i.e. 0.463,

0.638, and 0.670 for eortful control, surgency, and

negative aectivity, respectively, which were slightly

more favorable than in the original English version,

which reported an overall inter-rater reliability of 0.31

(0.32, 0.24, and 0.36 for eortful control, surgency, and

negative aectivity, respectively).

All of the properties

demonstrated that very short-form ECBQ, ai version can

be fruitfully used in research on temperament among ai

children.

Our study found that the majority of our participants

were reported as having a surgency temperament. Although

no previous studies reported the most common temperament

type in their populations, there has been plentiful evidence

that culture aects temperament. American children

get higher scores for surgency and lower scores for

negative aectivity than Japanese children.

One possible

explanation is that self-reliance, autonomy, and creativity

are highly valued and promoted in the Western cultures,

whereas such characteristics tend not to be in the Asian

ones. However, the American children got higher scores

for negative aectivity,

and tended to demonstrate

more eortful control than Chilean and Italian toddlers,

respectively.

erefore, how ai children’s temperament

diers from that of children’s from other regions should

be the subject of future research. Our ndings regarding

gender dierences are comparable to those of previous

international studies.

16,20

Possible explanations involve

how boys and girls are raised to behave dierently.

ere were some limitations to this study. e data

collection about parental temperament scoring did not

be specied that the parent was a father or a mother.

This lack of information hindered the comparisons

between fathers’ and mothers’ perspectives that may

lead to low inter-rater reliability, especially regarding

eortful-control temperament. Although our sample

size was adequate for evaluating the reliability of the

questionnaire, it was obviously too low to represent all

ai toddlers. Future research should be done using a

considerably larger sample size and should include the

participants from every region of ailand.

CONCLUSION

e very short form Early Childhood Behavior

Questionnaire, ai version demonstrated acceptable

internal consistency and reliability. e majority of the

ai toddlers revealed a surgency temperament. Boys

exhibited more surgency, while girls manifested more

eortful control.

ACKNOWLEDGEMENTS

e authors gratefully acknowledge Professor Samuel

P. Putnam for giving us permission to translate the

Early Childhood Behavior Questionnaire, very short

form into ai, and to Dr. Sasima Tongsai and Ms.

Kanokwan Sommai from the Division of Research,

Faculty of Medicine Siriraj Hospital, Mahidol University

for help with the statistical analysis.

Conict of interest declaration

e authors declare no conicts of interest relating

to any aspect of this study.

Rojmahamongkol et al.

Volume 74, No.4: 2022 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

223

Original Article

SMJ

Funding disclosure

ere was no funding for this study.

APPENDIX

The ECBQ assesses the following 18 scales of

temperament:

Activity Level/Energy: Level (rate and intensity) of gross

motoractivity, including rate and extent of locomotion

Attentional Focusing: Sustained duration of orienting

on an object ofattention; resisting distraction

Attentional Shiing: e ability to transfer attentional

focus fromone activity/task to another

Cuddliness: A child’s expression of enjoyment in and

moldingof the body to being held by a caregiver

Discomfort: Amount of negative aect related to sensory

qualitiesof stimulation, including intensity, rate, or

complexity oflight, sound, texture.

Fear: Negative aect,including unease, worry, or nervousness

related to anticipated pain or distress and/or potentially

threatening situations;being startled by sudden events

Frustration: Negative aect related to interruption of

ongoingtasks or goal blocking

High-intensity Pleasure: Pleasure or enjoyment related

to situations involving high stimulus intensity, rate,

complexity, novelty and incongruity

Impulsivity: Speed of response initiation

Inhibitory Control: e capacity to stop, moderate,or

refrain from a behavior under instruction

Low-intensity Pleasure: Pleasure or enjoyment related

to situations involving low stimulus intensity, rate,

complexity,novelty and incongruity

Motor Activation: Repetitive small-motor movements;

dgeting

Perceptual Sensitivity: Detection of slight, low-intensity

stimuli from the external environment

Positive Anticipation: Excitement about expected

pleasurableactivities

Sadness: Tearfulness or lowered mood related to exposure

to personal suering, disappointment, object loss, loss

of approval,or response to other’s suering

Shyness: Slow or inhibited approach and/or discomfort

in social situations involving novelty or uncertainty

Sociability: Seeking and taking pleasure in interactions

with others

Soothability: Rate of recovery from peak distress,

excitement, or general arousal

Children’s temperament divided into three factors:

Surgency/Extraversion is characterized by high

positive loadings for Activity Level, High-Intensity

Pleasure, and the Impulsivity scale and strong negative

loadings on the Shyness scales.

Negative Affectivity is characterized by high

positive loadings for Anger/Frustration, Sadness, Fear,

and Discomfort and negative loading for the Soothability

scales.

Eortful Control is characterized by high positive

loadings for Inhibitory Control, Attention Control, and

the Perceptual Sensitivity scales.

REFERENCES

1. Rothbart MK, Derryberry D. Development of individual

dierences in temperament. In: M.E. Lamb ALB, editors.

Advances in developmental psychology. Vol. 1. Hillsdale, NJ:

Lawrence Erlbaum Associations; 1981. p. 33-86.

2. Chess S, omas A. Temperamental individuality from childhood

to adolescence. J Am Acad Child Psychiatry. 1977;16(2):218-

26.

3. Macari SL, Koller J, Campbell DJ, Chawarska K. Temperamental

markers in toddlers with autism spectrum disorder. J Child

Psychol Psychiatry. 2017;58(7):819-28.

4. Kryski KR, Olino TM, Dyson MW, Durbin CE, Klein DN, Hayden

EP. Associations between observed temperament in preschoolers

and parent psychopathology. Personal Ment Health. 2018;12(2):

131-44.

5. Lee EH, Zhou Q, Eisenberg N, Wang Y. Bidirectional relations

between temperament and parenting styles in Chinese children.

Int J Behav Dev. 2012;37(1):57-67.

6. Fullard W, McDevitt SC, Carey WB. Assessing temperament in

one- to three-year-old children. J Pediatr Psychol. 1984;9(2):205-

17.

7. Laohaprasitiporn P, Monteerarat Y, Jaderojananont W,

Limthongthang R, Vathana T. Validity, reliability and

responsiveness of the ai version of patient-related wrist

evaluation. Siriraj Med J. 2021;73(4):275-81.

8. Janha P, Punyapas S, Ratta-apha W. Parent stress-coping

skills and resilience among parents of children with specic

learning disorders. Siriraj Med J. 2021;73(1):38-45.

9. Pinkaew B, Assansen P, Michel O, Talek K, Phonmanee T,

Kerdnoppakhun J. Impact assessment of smell and taste disorders

on quality of life in ais using the SF-36 health survey (ai

version). Siriraj Med J. 2019;71(2):102-9.

10. Putnam SP, Gartstein MA, Rothbart MK. Measurement of

ne-grained aspects of toddler temperament: the early childhood

behavior questionnaire. Infant Behav Dev. 2006;29(3):386-401.

11. Mary Rothbart’s temperament questionnaires. The Early

Childhood Behavior Questionnaire (ECBQ) [Internet]. [cited

2020 Nov 30]. Available from: https://research.bowdoin.edu/

rothbart-temperament-questionnaires/instrument-descriptions/

the-early-childhood-behavior-questionnaire/.

12. Putnam SP, Jacobs J, Gartstein MA, Rothbart MK. Development

and assessment of short and very short forms of the Early

Childhood Behavior Questionnaire. Poster presented at:

International Conference on Infant Studies; 2010 Mar; Baltimore,

MD.

13. Tavakol M, Dennick R. Making sense of Cronbach’s alpha.

Int J Med Educ. 2011;2:53-5.

14. Glen S. Cronbach’s alpha: simple denition, use and interpretation.

[Internet]; 2021 [cited 2020 30 Nov]. Availble from: https://

www.statisticshowto.com/cronbachs-alpha-spss/

15. Koo TK, Li MY. A guideline of selecting and reporting intraclass

Volume 74, No.4: 2022 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

224

correlation coecients for reliability research. J Chiropr Med.

2016;15(2):155-63.

16. Petra P, Potmesil M. e Early Childhood Behavior Questionnaire

Very Short Form (ECBQ VSF) and its adaptation to the population

of the Czech Republic. Psychiatr Psychol Klin. 2019;19:281-7.

17. Slobodskaya HR, Gartstein MA, Nakagawa A, Putnam SP.

Early temperament in Japan, the United States, and Russia: do

cross-cultural dierences decrease with age? J Cross Cult

Psychol. 2012;44(3):438-60.

18. Farkas C, Vallotton C. Dierences in infant temperament

between Chile and the US. Infant Behav Dev. 2016;44:208-18.

19. Cozzi P, Putnam SP, Menesini E, Gartstein MA, Aureli T,

Calussi P, et al. Studying cross-cultural dierences in temperament

in toddlerhood: United States of America (U.S.) and Italy.

Infant Behav Dev. 2013;36(3):480-3.

20. Else-Quest NM, Hyde JS, Goldsmith HH, Van Hulle CA.

Gender dierences in temperament: a meta-analysis. Psychol

Bull. 2006;132(1):33-72.

Rojmahamongkol et al.