Chew, J., & Haase, A. M. (2016).

Psychometric Properties of the

Family Resilience Assessment Scale: A Singaporean perspective.
Epilepsy and Behavior, 61, 112-119.
https://doi.org/10.1016/j.yebeh.2016.05.015

Peer reviewed version

Link to published version (if available):

10.1016/j.yebeh.2016.05.015

Link to publication record in Explore Bristol Research

PDF-document

This is the author accepted manuscript (AAM). The final published version (version of record) is available online
via Elsevier at http://www.sciencedirect.com/science/article/pii/S1525505016300981. Please refer to any
applicable terms of use of the publisher.

University of Bristol - Explore Bristol Research

General rights

This document is made available in accordance with publisher policies. Please cite only the
published version using the reference above. Full terms of use are available:
http://www.bristol.ac.uk/red/research-policy/pure/user-guides/ebr-terms/
Running head: FAMILY RESILIENCE ASSESSMENT SCALE 1

Title

Psychometric Properties of the Family Resilience Assessment Scale: A Singaporean perspective.

Author names and affiliations

Judith Chewa,b and Anne M Haase, PhDa

aSchool for Policy Studies, University of Bristol

8 Priory Road, Bristol BS8 1TZ

United Kingdom

bMedical Social Work Department, KK Women’s and Children’s Hospital

100 Bukit Timah Road, S(229899)

Singapore

Corresponding author

Correspondence for this article should be addressed to Judith Chew

Email: judith.chew@bristol.ac.uk or judith.chew.fh@kkh.com.sg

FAMILY RESILIENCE ASSESSMENT SCALE 2

Abstract

Families of young people with chronic illnesses are more likely to experience higher levels of stress.

In turn, their ability to cope with multiple demands is likely to affect young people’s adaptation. The

purpose of this study was to examine psychometric properties of the Family Resilience Assessment

Scale (FRAS), an assessment tool that measures the construct of family resilience. A total of 152

young people with epilepsy, aged 13 to 16 years old, from KK Women’s and Children’s Hospital,

Singapore, completed the FRAS along with Rosenberg’s Self-Esteem scale. Factor structure of the

FRAS was examined. Exploratory factor analysis resulted in a 7-factor solution – Meaning making

and Positive outlook; Transcendence and spirituality; Flexibility and Connectedness; Social and

economic resources (community); Social and economic resources (neighbors); Clarity and Open

emotional expression; Collaborative problem-solving – accounting for 83.0% of the variance. Internal

consistency of the scale was high (α=0.92). Family resilience was significantly correlated with higher

levels of self-esteem. Our study provides preliminary findings that suggest FRAS is a reliable and

valid scale for assessing the construct of family resilience among young people with epilepsy in

Singapore.

Key words

Family resilience; Young people; Adolescents; Epilepsy; Factor analysis, Instrument


FAMILY RESILIENCE ASSESSMENT SCALE 3

1. Introduction

Studies have shown that young people with epilepsy are three to nine times more likely to have

poorer outcomes when compared to healthy peers, young people with other medical conditions and/or

their siblings [1-3]. The impact of epilepsy is not restricted to individuals but is also extended to their

families. Having a child with epilepsy is likely to place additional stress and burden on families in

coping with unpredictable patterns of seizure occurrence, seizure severity, and complexities of

medical treatment. Thus, living with a family member affected by epilepsy is likely to have an impact

on family functioning.

Preventing and reducing psychosocial problems in young people with chronic illnesses have been of

interest to both researchers and practitioners [4]. As young people are situated within several

systems, such as families, peers and schools, it is necessary to consider these influences on young

people’s adaptation [5]. Family functioning, which plays a significant role in young people’s

adaptation, has been identified as one of the modifiable processes for intervention [6, 7]. Compared to

their peers, young people with epilepsy had poorer parent-child relationships, greater problems with

family functioning (e.g., poorer communication between family members, lower family cohesion),

higher levels of stress and conflict within their families [8]. Associations between family functioning

and a range of psychosocial and health outcomes in young people with epilepsy has also been

demonstrated. Poorer levels of family functioning have been shown to predict higher levels of

behavioral problems [9, 10], lower self-esteem [11], social competencies [10, 12], academic

achievement [13, 14], and treatment adherence [15, 16].

Overall, these findings demonstrate the influence of family functioning on young people’s outcomes.

However, the number of studies that examined family influences on young people’s outcomes is

lacking. Among quantitative studies that examined relationships between family factors and

psychosocial outcomes, most used parent reports to measure family functioning. In addition, these

studies often adopt a deficit perspective and utilize assessment measures that focus on family

pathology. This is in contrast with the proliferation of literature in areas of individual and family

resilience that emphasizes a strengths perspective. Alongside the proliferation of research in the area

of resilience, a range of scales is available for measuring this construct [17]. However, the majority

focus on identifying individual traits (e.g., personality) and intrapersonal factors (e.g., emotional


FAMILY RESILIENCE ASSESSMENT SCALE 4

regulation) and fail to consider the influence of higher level systemic factors, such as family

processes. Commonly used assessment measures, such as Family Adaptability and Cohesion Scales

(FACES-IV), Family Assessment Device (FAD), and Family Assessment Measure (FAM), may not be

suitable for examining resilience prompting processes as they focus on family dysfunction. There are

several measures for families with an explicit focus on strengths, such as Family Resource Scale and

Family Support Scale [18], yet these measures identify sources of support and do not focus on

specific family processes.

Therefore, with an increasing emphasis on resilience, there is a need for assessment measures to

reflect the construct of family resilience, instead of dysfunction. Sixbey [19] responded to this need by

developing the Family Resilience Assessment Scale (FRAS), which measures family resilience as

conceptualized by Walsh [20], to aid understanding in how families deal and cope with adversity.

According to Walsh [20], there are nine key processes within three domains of family functioning that

promote family resilience. In the first domain of family functioning – family beliefs – processes that

promote resilience include making meaning of adversity, positive outlook, and transcendence and

spirituality. Processes that foster resilience in the second domain of family functioning –

organizational patterns – are flexibility in a family’s structure, connectedness among family

members and utilization of social and economic resources. The third dimension of family functioning –

communication – involves processes that have clarity, involve open emotional expression, and

facilitate collaborative problem-solving [20]. Sixbey’s family resilience measure (i.e., FRAS) has six

subscales, which measured these nine family processes [19].

The FRAS, which was developed in the United States, offers promising potential utility in measuring

family resilience. It provides researchers and practitioners with a tool to assess, plan and evaluate

interventions designed to promote family resilience and its influence on young people’s outcomes.

Therefore, it is essential for this measure to be reliable and valid when used with other populations

from different cultures. However, as meanings of constructs such as family resilience are likely to vary

across cultures, it begs the question of whether there is conceptual equivalence when using Western-

developed measurement scales instead of developing culturally specific instruments. Several studies

used FRAS as a measure of family resilience [21-27]. When reported, Cronbach’s alpha coefficients

for FRAS ranged between 0.76 and 0.93 [23, 25, 26]. Of these studies, only Kaya and Arici [23]


FAMILY RESILIENCE ASSESSMENT SCALE 5

examined the factor structure of FRAS and found a four-factor structure instead of the original six-

factor structure. In order to address concerns regarding FRAS’ factor structure, a more thorough

analysis of its psychometric properties is warranted. Hence, the aim of this study was to examine the

reliability and validity of the FRAS in Singapore, a multi-cultural population where the measure has yet

to be tested.

2. Methods

2.1 Participants

Between November 2013 and August 2014, young people who met the following criteria: (i)

diagnosed with epilepsy, (ii) aged between 13 and 16 years old, and (iii) attending mainstream school,

were recruited from the pediatric neurology services in KK Women’s and Children’s Hospital,

Singapore (KKH). KKH is an 830-bed academic healthcare institution that provides specialized

pediatric and women’s healthcare services. It is one of two public hospitals in Singapore with a

pediatric neurology unit providing inpatient and outpatient services, such as diagnosis and

management of young people with epilepsy [28].

2.2 Procedures

SingHealth Centralised Institutional Review Board approved this study. Consent was obtained from

young people and their parents. Young people completed the survey while waiting to see their

physicians at KKH.

2.3 Measures

Only young people completed self-reported measures of family resilience and self-esteem. They also

provided individual-level demographic data, while their parents provided family-level data, such as

household income and family structure. Physicians provided clinical information on number of

medications, seizure frequency, and their assessment of seizure control (i.e., whether seizures were

effectively controlled by AED).

2.3.1 Family Resilience Assessment Scale (FRAS)


FAMILY RESILIENCE ASSESSMENT SCALE 6

As mentioned, FRAS measures the construct of family resilience, specifically, processes that support

a family's ability to cope successfully with adversity [19]. Although the 54-item FRAS was developed

to measure nine distinct family processes as conceptualized by Walsh [20], Sixbey’s original study

demonstrated a six-factor solution instead [19]. These six subscales include; (i) family communication

and problem solving (e.g., ‘We consult with each other about decisions’), (ii) utilizing social and

economic resources (e.g., ‘We ask neighbors for help and assistance’), (iii) maintaining a positive

outlook (e.g., ‘We trust things will work out even in difficult times’), (iv) family connectedness (e.g.,

‘We show love and affection for family members’), (v) family spirituality (e.g., ‘We attend

prayers/services at temple/mosque/church/other places of worship’), and (vi) ability to make meaning

of adversity (e.g., ‘We accept that stressful events as part of life’). Respondents indicated on a 4-point

Likert scale, which ranged from 1 (Strongly disagree) to 4 (Strongly agree), rating their level of

agreement with statements that describe family processes. Four items were negatively phrased (33,

37, 45, and 50) and were reversed scored before summing all items to obtain a total score for family

resilience. The total score range for FRAS lies between 54 and 216, with higher scores indicative of

higher levels of family resilience. Similarly, subscale scores were obtained through the summation of

values for items in each subscale. Cronbach’s alpha coefficients for total and subscales are reported

in Table 2a.

2.3.2 Rosenberg Self-Esteem Scale (RSS)

Young people’s global self-esteem was measured with Rosenberg Self-Esteem Scale (RSS) [29].

This 10-item scale evaluates global self-esteem through positive and negative perceptions of self.

Examples of positive and negative worded items are, ‘On the whole, I am satisfied with myself’ and ‘At

times I think I am no good at all’, respectively. Respondents rated each item on a 4-point Likert scale

ranging from 1 (Strongly disagree) to 4 (Strongly agree). Items that reflected negative perceptions

were reverse scored (3, 5, 8, 9, and 10) and all 10 items were summed to provide a total score that

range between 10 and 40. Higher scores are indicative of higher levels of global self-esteem, i.e. a

positive sense of one’s value as a person [29]. The Cronbach’s alpha coefficient was 0.90.

2.3.3 Illness severity


FAMILY RESILIENCE ASSESSMENT SCALE 7

The severity of young people’s illness has been determined based on: (i) seizure types, (ii) seizure

frequency, and (iii) number of AED and its side effects [30-33]. Often, composite scores were derived

from these classifications. In this study, illness severity was operationalized as the extent to which

young people’s seizures were controlled by AED use: (i) No seizures, AED not required (Low); (ii)

Seizures controlled with AED (Moderate); and (iii) Seizures despite AED (High).

2.4 Data analysis

2.4.1 Exploratory factor analysis

Exploratory factor analysis with principal axis factoring was conducted to examine the factor structure

of FRAS. Based on existing recommendations, a reasonable absolute minimum sample size of 50

was required to yield reliable results from an exploratory factor analysis [34, 35]. Additionally,

simulation studies demostrated that sample size adequacy is partly determined by the nature of the

data [35-37], thus, factor-to-variable ratio (over-determination) and communality of variables were

examined to determine whether the current sample size was sufficient.

Prior to conducting an EFA, Kaiser-Meyer-Olkin's (KMO) measure of sampling adequacy and

Bartlett's test of sphericity were used to determine if the data was suitable for factor analysis [36].

Kaiser’s criterion (eigenvalues≥1.0), percentage of variance accounted by the number of factors, and

scree plots were used to determine the number of factors to be retained [36]. In addition to orthogonal

(varimax) rotation, oblique (direct oblimin) rotation method was used as family processes were

hypothesized to be interrelated. Individual items were retained if its factor loading on a single factor

was above 0.4, and had at least a 0.2 difference from other factors. Missing variables (n=7) were

excluded listwise and the final sample used for EFA was 145.

2.4.2 Reliability and validity

Cronbach’s alpha coefficient was used to assess the internal consistency of FRAS and its subscales.

A high Cronbach coefficient value (α>0.70) was indicative of a reliable measure [38]. To evaluate

validity of FRAS scores, we examined associations between FRAS and theoretically relevant

variables such as self-esteem and illness severity. Based on existing evidence, we hypothesized that

young people with higher self-esteem report correspondingly higher levels of family resilience [39-41].


FAMILY RESILIENCE ASSESSMENT SCALE 8

In contrast, young people who experienced greater illness severity would have significantly lower

levels of family resilience [8, 10, 42-45]. Correlational analyses were performed to establish the

statistical significance of relationships between measures of family resilience and young people’s self-

esteem. One-way analysis of variance (ANOVA) with post-hoc comparisons were conducted to test

the hypothesis that young people with higher illness severity had lower levels of family resilience.

Hochberg’s GT2 post-hoc test was used, as group sizes were different. All statistical analyses were

performed using SPSS version 21.0.

3. Results

A total of 176 young people were invited and 156 participated in this study (response rate of 88.6%).

No further information is available on the twenty young people who declined participation. Scores

from 152 young people (79 males, 73 females) were included in the analyses, as four questionnaires

were incomplete. Clinical and demographic characteristics of this sample of young people are

presented in Table 1.

3.1 Preliminary analysis

Kaiser-Meyer-Olkin (KMO) statistic was 0.85, which is above the minimum criterion of 0.5, indicating

that the sample size is adequate for factor analysis. In addition, KMO values for individual items,

which were greater than 0.63, were above the minimum acceptable limit of 0.5 [38]. Bartlett’s test of

sphericity was significant (χ2=11021.51, p<0.001), indicating that FRAS items were adequately

correlated for a factor analysis to be performed.

3.2 Exploratory factor analysis

Exploratory factor analyses using principal axis factoring were conducted, and similar results were

obtained from varimax and oblimin rotations. Both rotations yielded seven factors, accounting for

80.56% of the variance. There was no difference in patterns of item loadings for each rotation, i.e.,

individual FRAS items loaded onto the same factors. However, the seven-factor solution produced a

factor with only two items (Factor 7). When allowed to correlate, through the use of direct oblimin

rotation, correlation between factors ranged between -0.57 (Factor 2 and Factor 6) and 0.36 (Factor 4

and Factor 7). This provides evidence that the constructs are interrelated, with each factor measuring


FAMILY RESILIENCE ASSESSMENT SCALE 9

a unique aspect of family resilience. As recommended by Henson and Roberts (2006), both pattern

and structure matrices derived from the EFA through use of an oblique rotation method, are

presented in Table 2(a) and (b).

The results from the EFA did not support Sixbey’s six-factor structure of the FRAS. Instead, a seven-

factor solution emerged from the analyses. Upon examination, it was noted that these factors and its

corresponding items had closer approximation to Walsh’s family resilience framework. On this basis, it

was concluded that the current seven-factor solution provided a better representation of family

resilience. A summary of FRAS item classifications according to Walsh’s conceptual framework, the

six-factor and seven-factor solutions yielded from Sixbey’s and this current study are presented in

Table 3.

3.3 Reliability and validity

Internal consistency for the total FRAS scale was high with Cronbach’s alpha value of 0.92. As all 54

items had factor loadings greater than 0.40, they were summated according to their respective factors

to form FRAS sub-scales. Cronbach’s alpha coefficients of these subscales, which ranged between

0.93 and 0.97, are reported in Table 2a.

As hypothesized, there was a significant positive relationship between family resilience (i.e., FRAS

total scale score) and self-esteem, r=0.58, p<0.001. Young people who reported higher levels of

family resilience also had higher levels of self-esteem. One-way ANOVA revealed significant

differences in young people’s family resilience across illness severity conditions, F(2,142)=4.84,

p<0.01. Hochberg’s GT2 post-hoc comparisons indicated that young people who had seizures despite

medication (high illness severity) had significantly lower levels of family resilience when compared to

those who did not have seizures (low or moderate severity). However, there was no significant

difference in average FRAS scores between young people with low and those with moderate illness

severity.

4. Discussion

4.1 Factor structure of FRAS


FAMILY RESILIENCE ASSESSMENT SCALE 10

The objective of this study was to examine the psychometric properties of FRAS. Sixbey’s original

FRAS six-factor structure was not replicated. Instead, a seven-factor solution emerged from the

exploratory factor analysis and it reflected dimensions of family resilience put forward by Walsh’s

conceptual framework [20]. These seven factors – Meaning-making and Positive outlook;

Transcendence and spirituality; Flexibility and Connectedness; Resources – Community; Resources –

Neighbors; Clarity and Open emotional expression; Collaborative problem-solving – accounted for

approximately 83% of the total variance with factor loadings ranging from 0.40 to 0.91. The total

FRAS scale also demonstrated good internal consistency, suggesting that the 54-items functioned

collectively to characterize the concept of family resilience.

There are various reasons that could account for the lack of distinction between processes within

Walsh’s conceptualization of specific family functioning domains, for example, items measuring family

beliefs of meaning-making and positive outlook loaded onto a single factor instead of two. Thus, it

may be possible that FRAS items measure a single construct instead of distinct family processes.

Another reason might be that these items may not be sufficiently distinct to differentiate various

concepts of family processes. For example, young people may have interpreted the statement, ‘We

can work through difficulties as a family’, as an indication of their families’ ability to resolve problems

instead of reflecting their family beliefs. Third, the relationship between processes belonging to the

same family functioning domain may have masked distinctions, resulting in extraction of a single

factor. For instance, it is possible that a positive relationship between key communication processes

such as ‘Clarity’ and ‘Open emotional expression’, exists. It is likely for families, which encourage

expression of emotions (e.g., ‘We can ventilate at home without upsetting someone’) would also tend

to adopt processes that encourage clarity in communication between family members (e.g., ‘We can

be honest and direct with each other in our family’). Concurrently, there may be a small number of

families with high levels of clarity in their communication, but were less open in their expression of

emotions or vice versa. This lack of heterogeneity among communication processes within families of

the current sample may be one reason why a single factor was extracted instead of two.

It is of interest to note that items describing ‘Social and economic resources’ loaded onto two distinct

but correlated factors. Based on further examination of these items, it is postulated that young people

made a distinction between the availability of community resources (Factor 4) and the extent to which


FAMILY RESILIENCE ASSESSMENT SCALE 11

their families actually sought and received help from their neighbors (Factor 7). Another possible

reason for the distinction between factors is Asian families, such as Chinese and Indians, tend to rely

either on themselves [46] or on extended family members [47], instead of their neighbors.

4.2 Reliability and validity

There was low to moderate correlation between two pairs of subscales, ‘Flexibility and

Connectedness’ and ‘Collaborative problem-solving’; ‘Resources – Community’ and ‘Resources –

Neighbors’. These correlations suggest young people’s perceptions of family processes were related

but also conceptually distinct. Furthermore, it indicates that these subscales measure different

aspects of family resilience and supports the theoretical understanding of resilience as a

multidimensional construct [20, 48].

It appears the dimensionality of FRAS differed across countries in which its factor structure has been

examined. Kaya and Arici [23] conducted a confirmatory factor analysis (CFA) to examine the factor

structure of the Turkish version of FRAS. Results from their analysis did not support the original six-

factor structure but demonstrated a four-factor structure instead. One reason behind this diversity

could be differences in meanings of family resilience. Processes that foster resilience within families,

such as receiving aid from extended families versus neighbors and communities, may be dependent

on cultural contexts. Sample characteristics is another issue to consider when attempting to explain

differences in dimensions of family resilience. For example, Sixbey [19] recruited participants ranging

between 16 and 77 years old (mean=36.2 years). Kaya and Arici [23] recruited university students

with an average age of approximately 22 years old. In contrast, the average age of young people in

this study was 15 years. Participants’ age may reflect corresponding family life cycles and potential

variations in family processes during each period. In turn, these differences could be reflected in the

different FRAS structures.

Significant associations found between FRAS scores and measures of young people’s self-esteem

and illness severity, provide support for concurrent validity. As hypothesized, there was a strong

positive relationship between family resilience and self-esteem, where young people who reported

higher levels of self-esteem also perceived higher levels of resilience within their families. These

results are similar to findings in previous studies that examined the relationship between young


FAMILY RESILIENCE ASSESSMENT SCALE 12

people’s self-esteem and family functioning [40, 41, 49, 50]. It is possible that these family processes

supported young people’s efforts in managing illness-related demands and influenced how they

viewed themselves. The significant relationship between family resilience and young people’s self-

esteem underscores the importance of considering family factors when attempting to understand

factors influencing psychosocial adaptation to a chronic condition such as epilepsy. Therefore, future

research should continue to examine the influence of family factors on young people’s outcomes. In

particular, young people’s perspectives regarding their families and its processes, as there is a lack of

research in this area.

Family resilience was significantly lower among those who continued to have seizures despite AED,

compared to young people who achieved seizure control. This is similar to results from existing

studies that examined family functioning among young people with chronic illnesses [42-45]. For

example, greater neurological impairment was associated with higher levels of conflict and less

supportive family relationships within the family [45]. This contributes to the growing evidence that the

demands of epilepsy is likely to have a negative effect on young people and their families. Taken

together, this suggests that family processes are potential targets for interventions. Young people and

their families who exhibit moderate to high distress, particularly those who fail to achieve seizure

control despite medication could receive additional support services to promote positive outcomes.

Findings from this study have implications for practitioners who provide psychosocial interventions for

young people with epilepsy. With empirical evidence indicating that the FRAS is a reliable and valid

measure, practitioners could utilize this tool to measure and identify family processes, and in turn,

provide valuable information needed to develop interventions aimed at promoting resilience.

Additionally, it can be used to evaluate the effectiveness of these interventions.

4.3 Methodological considerations

The relatively small sample size (n=145) and low ratio of participants to number of variables (2.7:1)

may raise concerns about the EFA factor solution, as both do not meet traditional recommendations

regarding required sample sizes for factor analyses. However, there remain differing opinions on

adequate or acceptable sample sizes [35-37]. Early recommendations either emphasized minimum

sample size (e.g., at least 200) or a required ratio of participants to number of variables. Based on


FAMILY RESILIENCE ASSESSMENT SCALE 13

findings from simulation studies, several authors argued that greater emphasis on high factor over-

determination and communality of variables, instead [34, 35, 37, 51]. Although large samples are

beneficial, EFA should not be ruled out on the sole basis of a small sample size [34, 35]. Item

communalities of 54 items in FRAS ranged from 0.6 to 0.9 and these values were considered to be

high [52]. With the exception of one factor, the remaining six factors had at least 4 variables with

factor loadings of 0.8, indicative of high over-determination of factors. Despite the relatively small

sample size, conditions such as high communalities among variables and high factor over-

determination were met. Therefore, we have confidence that factor solutions in this study are reliable.

4.4 Limitations of this study and future research

Existing limitations of this study should be taken into consideration. First, due to a small sample size,

only an EFA was performed. It is recognized that using subsets of the data for confirmatory factor

analyses would have provided additional evidence to either corroborate or contradict EFA findings.

However, this was not feasible due to the sample size of this study. Second, the FRAS factor

structure was derived from a clinical sample. Further research among the general population of young

people and adult population is necessary to determine if the current structure is invariant across

different populations. Third, the present study used a cross-sectional design and no assessments

were made to determine whether the seven-factor structure was constant over time. Assessments of

test-retest reliability in future studies could provide insight to the stability of this measure. Fourth, the

exclusive reliance on self-reports may give rise to common method variance, e.g., social desirability

and acquiescence. Future research could minimise such variances by obtaining data from various

Supplementing young people’s views by obtaining data from

other sources such as parents, siblings or significant others, is likely to be beneficial. The

convergence or divergence of data obtained from multiple perspectives provides valuable information

of different aspects of family processes. For instance, differences in family members’ perspective

regarding family processes could also suggest conflicting expectations and needs [53-55]. Left

unresolved, these differences could lead to increased stress and conflict within families.

5. Conclusion


FAMILY RESILIENCE ASSESSMENT SCALE 14

In conclusion, these findings provide preliminary evidence that FRAS is an adequate family resilience

measure for use among young people with epilepsy in Singapore. The seven-factor FRAS structure

reflects the construct of family resilience as theorized by Walsh and can be used to facilitate

practitioners' assessments of and supporting families in harnessing processes that foster resilience in

order to meet epilepsy-related challenges.

6. Acknowledgements

This exploratory factor analysis of Family Resilience Assessment Scale forms part of the first author’s

dissertation about young people’s experiences with epilepsy; this PhD study was supported by

SingHealth Talent Development Fund. We would like to thank the following individuals for their

assistance during the recruitment and data collection for this study: Gina Tan and Mavis Teo from the

hospital’s Medical Social Work Department; and Dr. Derrick Chan and his team from the Department

of Paediatrics, Neurology Service. The findings and conclusions of this report are those of the authors

and do not necessarily represent the official position of KK Women’s and Children’s Hospital or

School for Policy Studies, University of Bristol.

Reference

1. McDermott, S., S. Mani, and S. Krishnawami, A population-based analysis of specific

behavior problems associated with childhood seizures. Journal of Epilepsy, 1995. 8(2): p.

110-118.

2. Plioplys, S., D.W. Dunn, and R. Caplan, 10-year research update review: Psychiatric

problems in children with epilepsy. Journal of the American Academy of Child & Adolescent

Psychiatry, 2007. 46(11): p. 1389-1402.

3. Rodenburg, R., et al., Psychosocial issues for children with epilepsy. Epilepsy & Behavior,

2011. 22(1): p. 47-54.

4. Ekeland, E., et al., Exercise to improve self-esteem in children and young people. Cochrane

Database Systematic Reviews, 2004(1).

5. Carpenter, J. and R. McConkey, Disabled children's voices: The nature and role of future

empirical enquiry. Children & Society, 2012. 26(3): p. 251-261.


FAMILY RESILIENCE ASSESSMENT SCALE 15

6. Rodenburg, R., et al., Family predictors of psychopathology in children with epilepsy.

Epilepsia, 2006. 47(3): p. 601-614.

7. Wallander, J.L. and J.W. Varni, Effects of pediatric chronic physical disorders on child and

family adjustment. Journal of Child Psychology and Psychiatry, 1998. 39(1): p. 29-46.

8. Rodenburg, R., et al., Family factors and psychopathology in children with epilepsy: A

literature review. Epilepsy & Behavior, 2005. 6(4): p. 488-503.

9. Austin, J.K., et al., Behavioral issues involving children and adolescents with epilepsy and the

impact of their families: Recent research data. Epilepsy & Behavior, 2004. 5(s3): p. 33-41.

10. McCusker, C.G., et al., Adjustment in children with intractable epilepsy: Importance of seizure

duration and family factors. Developmental Medicine & Child Neurology, 2002. 44(10): p. 681-

687.

11. Austin, J.K., et al., Self‐ esteem and symptoms of depression in children with seizures:

Relationships with neuropsychological functioning and family variables over time. Epilepsia,

2010. 51(10): p. 2074-2083.

12. Tse, E., et al., Social skills problems in children with epilepsy: Prevalence, nature and

predictors. Epilepsy & Behavior, 2007. 11(4): p. 499-505.

13. Fastenau, P.S., et al., Neuropsychological predictors of academic underachievement in

pediatric epilepsy: Moderating roles of demographic, seizure, and psychosocial variables.

Epilepsia, 2004. 45(10): p. 1261-1272.

14. Mitchell, W.G., et al., Academic underachievement in children with epilepsy. Journal of Child

Neurology, 1991. 6(1): p. 65-72.

15. DiMatteo, M.R., Social support and patient adherence to medical treatment: A meta-analysis.

Health Psychology, 2004. 23(2): p. 207-218.

16. Mitchell, W.G., L.M. Scheier, and S.A. Baker, Adherence to treatment in children with

epilepsy: Who follows "doctor's orders"?. Epilepsia, 2000. 41(12): p. 1616-1625.

17. Windle, G., K.M. Bennett, and J. Noyes A methodological review of resilience measurement

scales. Health and Quality of Life Outcomes, 2011. 9, 1-18 DOI: 10.1186/1477-7525-9-8.

18. Early, T., Measures for practice with families from a strengths perspective. Families in

Society: The Journal of Contemporary Social Services, 2001. 82(3): p. 225-232.


FAMILY RESILIENCE ASSESSMENT SCALE 16

19. Sixbey, M.T., Development of the family resilience assessment scale to identify family

resilience constructs. 2005, University of Florida.

20. Walsh, F., Family resilience: A framework for clinical practice. Family Process, 2003. 42(1): p.

1-18.

21. Buchanan, T., Family resilience as a predictor of better adjustment among international

adoptees, in Social Work. 2008, The University of Texas at Arlington.

22. Cripe, C.T., Family resilience, parental resilience and stress mediation in families with autistic

children. 2013, Northcentral University.

23. Kaya, M. and N. Arici, Turkish version of shortened Family Resiliency Scale (FRAS): The

study of validity and reliability. Procedia - Social and Behavioral Sciences, 2012. 55: p. 512-

520.

24. Lum, C., The development of family resilience: Exploratory investigation of a resilience

program for families impacted by chemical dependency. 2008, San José State University.

25. Openshaw, K.P., The relationship between family functioning, family resilience, and quality of

life among vocational rehabilitation clients. 2011, Utah State University.

26. Plumb, J.C., The impacat of social support and family resilience on parental stress in families

with a child diagnosed with an autism spectrum disorder, in Social of Social Policy and

Practice. 2011, University of Pennsylvania.

27. Faqurudheen, H., S. Mathew, and T.M. Kumar, Exploring family resilience in a community

mental health setup in south India. Procedia Economics and Finance, 2014. 18: p. 391-399.

28. Chan, D., et al., Pediatric epilepsy and first afebrile seizure in Singapore: Epidemiology and

investigation yield at presentation. Journal of Child Neurology, 2010. 25(10): p. 1216-1222.

29. Rosenberg, M., Society and the adolescent self-image. 1965, Princeton, NJ: Princeton

University Press.

30. Austin, J.K., D.W. Dunn, and G.A. Huster, Childhood epilepsy and asthma: Changes in

behavior problems related to gender and change in condition severity. Epilepsia, 2000. 41(5):

p. 615-623.

31. Devinsky, O., et al., Risk factors for poor health‐ related quality of life in adolescents with

epilepsy. Epilepsia, 1999. 40(12): p. 1715-1720.


FAMILY RESILIENCE ASSESSMENT SCALE 17

32. Cramer, J.A., et al., Development of the Quality of Life in Epilepsy Inventory for Adolescents:

the QOLIE-AD-48. Epilepsia, 1999. 40(8): p. 1114-21.

33. Heimlich, T.E., et al., Brief report: Adolescents' attitudes toward epilepsy: Further validation of

the Child Attitude Toward Illness Scale (CATIS). Journal of Pediatric Psychology, 2000. 25(5):

p. 339-345.

34. de Winter, J.C.F., D. Dodou, and P.A. Wieringa, Exploratory factor analysis with small sample

sizes. Multivariate Behavioral Research, 2009. 44(2): p. 147-181.

35. Hogarty, K.Y., et al., The quality of factor solutions in exploratory factor analysis: The

influence of sample size, communality, and overdetermination. Educational and Psychological

Measurement, 2005. 65(2): p. 202-226.

36. Costello, A.B. and J.W. Osborne, Best practices in exploratory factor analysis: Four

recommendations for getting the most from your analysis. Practical Assessment Research &

Evaluation, 2005. 10(7): p. 1-9.

37. MacCallum, R.C., et al., Sample size in factor analysis. Psychological methods, 1999. 4(1): p.

84-99.

38. Field, A., Discovering statistics using SPSS: (and sex and drugs and rock 'n' roll). 3rd ed.

2009, London: SAGE Publications.

39. Drotar, D., Relating parent & family functioning to the psychological adjustment of children

with chronic health conditions: What have we learned? What do we need to know? Journal of

Pediatric Psychology, 1997. 22(2): p. 149-165.

40. Preechawong, S., et al., Relationship of family functioning, self-esteem, and resourceful

coping of Thai adolescents with asthma. Issues in Mental Health Nursing, 2007. 28(1): p. 21-

36.

41. Shek, D.T.L., Family functioning and psychological well-being, school adjustment, and

problem behavior in Chinese adolescents with and without economic disadvantage. The

Journal of Genetic Psychology, 2002. 163(4): p. 497-502.

42. Zashikhina, A. and B. Hagglof, Family functioning and juvenile chronic physical illness in

Northern Russia. Acta Pædiatrica, 2009. 98(2): p. 355-360.


FAMILY RESILIENCE ASSESSMENT SCALE 18

43. Iobst, E.A., et al., Pain, fatigue, family functioning, and attitude toward illness in children with

juvenile rheumatic diseases. Journal of Developmental and Physical Disabilities, 2007. 19(2):

p. 135-144.

44. Rodrigues, N. and J.M. Patterson, Impact of severity of a child's chronic condition on the

functioning of two-parent families. Journal of Pediatric Psychology, 2007. 32(4): p. 417-426.

45. Reiter-Purtill, J., et al., Parental distress, family functioning, and social support in families with

and without a child with Neurofibromatosis 1. Journal of Pediatric Psychology, 2008. 33(4): p.

422-434.

46. Shek, D.T.L., Help-seeking patterns of Chinese parents in Hong Kong. Asia Pacific Journal of

Social Work and Development, 1998. 8(1): p. 106-119.

47. Ow, R., Burden of care and childhood cancer: Experiences of parents in an Asian context.

Health & Social Work, 2003. 28(3): p. 232-240.

48. Luthar, S.S. and D. Cicchetti, The construct of resilience: Implications for interventions and

social policies. Development and Psychopathology, 2000. 12(04): p. 857-885.

49. Burt, C., L. Cohen, and J. Bjorck, Perceived family environment as a moderator of young

adolescents' life stress adjustment. American Journal of Community Psychology, 1988. 16(1):

p. 101-122.

50. McNelis, A.M., et al., Factors associated with self-concept in children with asthma. Journal of

Child and Adolescent Psychiatric Nursing, 2000. 13(2): p. 55-68.

51. MacCallum, R.C., et al., Sample size in factor analysis: The role of model error. Multivariate

Behavioral Research, 2001. 36(4): p. 611-637.

52. Tabachnick, B.G. and L.S. Fidell, Using multivariate statistics. 6th ed. 2013, Boston, MA:

Pearson.

53. Van Roy, B., et al., Understanding discrepancies in parent-child reporting of emotional and

behavioural problems: Effects of relational and socio-demographic factors. BMC Psychiatry,

2010. 10(1): p. 56.

54. Silva, N., et al., Why the (dis)agreement? Family context and child–parent perspectives on

health-related quality of life and psychological problems in paediatric asthma. Child: Care,

Health and Development, 2015. 41(1): p. 112-121.


FAMILY RESILIENCE ASSESSMENT SCALE 19

55. Huygen, A.C.J., W. Kuis, and G. Sinnema, Psychological, behavioural, and social adjustment

in children and adolescents with juvenile chronic arthritis. Annals of the Rheumatic Diseases,

2000. 59(4): p. 276-282.


FAMILY RESILIENCE ASSESSMENT SCALE 20

Table 1. Characteristics of young people who participated in the survey (n=152).

n (%)1
Individual-level demographics
Age, mean ± SD 15.0 ± 1.13
Gender
Male 79 (52.0)
Female 73 (48.0)
Ethnicity1
Chinese 95 (62.5)
Malay 37 (24.3)
Indian 18 (11.8)
Others (Arab, Burmese) 2 (1.3)
Living arrangements
Parents and siblings 135 (88.8)
Parents, siblings and relatives 10 (6.6)
Single parent and siblings 4 (2.6)
Single parent, siblings and relatives 2 (1.3)
Step-family 1 (0.7)
Young person’s medical information
Age at which young person was diagnosed with epilepsy (mean ± SD, range) 8.79 ± 3.94
(≤1-16 years)
Number of years with epilepsy (mean ± SD, range) 6.21 ± 3.68
(≤1-15 years)
AED
Not on medication 18 (11.8)
Single AED 86 (56.6)
Multiple AED 48 (31.6)
Seizures
No seizures 95 (62.5)
At least once a month 35 (23.0)
Every three months 14 (9.2)
Single seizure episode within the past 3 months 8 (5.3)
Illness severity (n=152)
No seizures, AED not required 18 (11.8)
Seizures controlled with AED 77 (50.7)
Seizures despite AED 57 (37.5)
Family-level demographics
Respondents (n=148; 4 did not participate)
Father 48 (32.4)
Mother 100 (67.6)

Age of parent (mean ± SD, range) 48.3 ± 4.90

(37-59 years)
Employment (n=140)
Employed 97 (69.3)
Unemployed 43 (30.7)

Highest qualification attained (n=146)

Below secondary 13 (8.9)
Secondary (GCE ‘O’ or ‘N’ level) 57 (39.0)
GCE ‘A’ level / ITE 33 (22.6)
Polytechnic / other diplomas 23 (15.8)
University 20 (13.7)


FAMILY RESILIENCE ASSESSMENT SCALE 21

n (%)1
Household income (n=146)
No working person 2 (1.4)
Less than 1,999 SGD 16 (11.0)
2,000 – 4,999 SGD 60 (41.1)
5,000 – 9,999 SGD 33 (22.6)
10,000 SGD and above 35 (24.0)
1Percentages for ‘Ethnicity’ and ‘Household income’ do not add up to 100% due
n – study sample; SD – Standard deviation; NA – Not available; AED – Anti-epileptic drugs; GCE ‘O’,
‘N’ and ‘A’ levels refers to Singapore-Cambridge General Certificate of Education Ordinary, Normal
and Advance level, respectively; ITE – Institute of Technical Education; SGD – Singapore Dollars.



FAMILY RESILIENCE ASSESSMENT SCALE 22

Table 2a: Pattern matrix of exploratory factor analysis (direct oblimin rotation).
Item No. Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 Factor 7
Meaning making and Positive outlook
13 .909
40 .898
4 .898
18 .878
36 .869
7 .864
21 .861
22 .858
51 .853
34 .841
26 .837
5 .829
24 .814
Flexibility and Connectedness
47 .879
3 .872
8 .835
9 .832
30 .832
1 .828
54 .812
33 .807
45 .756
10 .730
Clarity and Open emotional expression
14 .940
15 .916
53 .913
48 .910
16 .909
29 .903
20 .897
23 .858
41 .855
37 .767
Resources - Community
31 .911
49 .889
39 .883
32 .876
2 .857
38 .833
19 .819
50 .627
Transcendence and spirituality
12 .916
42 .896
35 .881
44 .819
Collaborative problem-solving
17 -.890
27 -.872
25 -.869
28 -.867
52 -.849
6 -.757
46 -.672
Resources - Neighbors
43 .838
11 .753
Initial eigenvalues 12.92 12.53 8.87 4.41 2.65 2.28 1.17
% of variance explained 23.92 23.21 16.43 8.16 4.91 4.22 2.17
Cronbach’s alpha coefficients 0.97 0.97 0.97 0.96 0.93 0.96 0.90


FAMILY RESILIENCE ASSESSMENT SCALE 23

Table 2b: Structure matrix of exploratory factor analysis (direct oblimin rotation).
Item No. Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 Factor 7
Meaning making and Positive outlook
13 .915
40 .907
4 .905
18 .902
36 .900
7 .897
21 .885
22 .871
51 .871
34 .869
26 .868
5 .845
24 .817
Flexibility and Connectedness
47 .943 -.607
3 .932 -.589
8 .915 -.581
9 .913 -.576
30 .894 -.589
1 .880 -.661
54 .875 -.555
33 .837 -.533
45 .774
10 .757 -.440
Clarity and Open emotional expression
14 .957
15 .937
53 .927
48 .924
16 .924
29 .912
20 .896
23 .892
41 .878
37 .712
Resources - Community
31 .936
49 .930
39 .910
32 .905 .444
2 .442 .905 .401
38 .855
19 .833
50 .733 .436
Transcendence and spirituality
12 .923
42 .896
35 .882
44 .427 .869
Collaborative problem-solving
17 .573 -.912
27 .583 -.909
25 .538 -.908
28 .541 -.902
52 .554 -.896
6 .669 -.833
46 .539 -.815
Resources - Neighbors
43 .477 .929
11 .452 .859


FAMILY RESILIENCE ASSESSMENT SCALE 24

Table 3: Family Resilience Assessment Scale (FRAS).

Walsh’s Sixbey’s Current
No. Items framework study study
(9 constructs)1 (6 factors)2 (7 factors)3
4 We accept stressful events as a part of life MM AMM MMPO
5 We accept that problems occur unexpectedly MM AMM MMPO
7 We are able to work through pain and come to an MM FCPS MMPO
understanding
18 We can deal with family differences in accepting a MM FCPS MMPO
loss
24 We can work through difficulties as a family MM FCPS MMPO
40 We learn from each other’s mistakes MM FCPS MMPO
13 We believe we can handle our problems PO MPO MMPO
21 We can solve major problems PO MPO MMPO
22 We can survive if another problem comes up PO MPO MMPO
26 We define problems positively to solve them PO FCPS MMPO
34 We feel we are strong in facing big problems PO MPO MMPO
36 We have the strength to solve our problems PO MPO MMPO
51 We trust things will work out even in difficult times PO MPO MMPO
12 We attend prayers/services at TS FS TS
temple/mosque/church/other places of worship
35 We have faith in Buddha/Deities/Allah/God/Hindu TS FS TS
gods/Others
42 We participate in temple/mosque/church/other TS FS TS
religious activities
44 We seek advice from religious advisors TS FS TS
1 Our family is flexible and can deal with unexpected Fl FCPS FLCO
events
8 We are adaptable to demands placed on us as a Fl FCPS FLCO
family
9 We are open to new ways of doing things in our family Fl FCPS FLCO
3 The things we do for each other make us feel a part of Co AMM FLCO
the family
10 We are understood by other family members Co FCPS FLCO
30 We feel good giving time and energy to our family Co FCPS FLCO
47 We show love and affection for family members Co FC FLCO
33 We feel taken for granted by family members* Co FC FLCO
45 We seldom listen to family members concerns or Co FC FLCO
problems*
2 Our friends value us and who we are SER FC R-C
11 We ask neighbors for help and assistance SER USER R-N
19 We can depend upon people in this community SER USER R-C
31 We feel people in this community are willing to help in SER USER R-C
an emergency
32 We feel secure living in this community SER USER R-C
38 We know there is community help if there is trouble SER USER R-C
39 We know we are important to our friends SER USER R-C
43 We receive gifts and favors from neighbors SER USER R-N
49 We think this is a good community to raise children SER USER R-C
50 We think we should not get too involved with people in SER FC R-C
this community*
14 We can ask for clarification if we do not understand Cl FCPS COEE
each other
15 We can be honest and direct with each other in our Cl FCPS COEE
family


FAMILY RESILIENCE ASSESSMENT SCALE 25

20 In our family, we can question or clarify when we do Cl FCPS COEE

not understand the communications between family
members
23 We can talk about the way we communicate in our Cl FCPS COEE
family
41 We mean what we say to each other in our family Cl FCPS COEE
53 We understand communication from other family Cl FCPS COEE
members
54 We work to make sure family members are not OEE FCPS FLCO
emotionally or physically hurt
16 We can ventilate at home without upsetting someone OEE FCPS COEE
29 We feel free to express our opinions OEE FCPS COEE
48 We tell each other how much we care for one another OEE FCPS COEE
37 We keep our feelings to ourselves* OEE FC COEE
6 We all have input into major family decisions CPS FCPS CPS
17 We can compromise when problems come up CPS FCPS CPS
25 We check with each other about decisions CPS FCPS CPS
27 We discuss problems and feel good about the CPS FCPS CPS
solutions
28 We discuss things until we reach a resolution CPS FCPS CPS
46 We share responsibility in the family CPS FCPS CPS
52 We try new ways of working with problems CPS FCPS CPS
1 MM=Making meaning of adversity; PO=Positive outlook; TS=Transcendence and spirituality;

Fl=Flexibility; Co=Connectedness; SER=Social and economic resources; Cl=Clarity; OEE=Open

emotional expression; CPS=Collaborative problem-solving.

2AMM=Ability to make meaning of adversity (3 items); FCPS=family communication and problem-

solving (27 items); MPO=Maintaining a positive outlook (6 items); FS=Family spirituality (4 items);
FC=Family connectedness (6 items); Utilizing social and economic resources (8 items).

3 MMPO=Meaning making and Positive outlook (13 items); TS=Transcendence and spirituality (4
items); FLCO=Flexibility and Connectedness (10 items); R-C =Resources – Community (8 items); R-
N=Resources - Neighbors (2 items); COEE=Clarity and Open emotional expression (10 items);
CPS=Collaborative problem-solving (7items). *Reverse scored.