- Email: [email protected]
Assessing social support among South Asians: The multidimensional scale of perceived social support
Asian Journal of Psychiatry 5 (2012) 164–168
Contents lists available at SciVerse ScienceDirect
Asian Journal of Psychiatry journal homepage: www.elsevier.com/locate/ajp
Assessing social support among South Asians: The multidimensional scale of perceived social support Kareen Tonsing a,*, Gregory D. Zimet b, Samson Tse a a
The University of Hong Kong, 13/F, K.K. Leung Building, Department of Social Work and Social Administration, Pokfulam Road, Hong Kong Special Administrative Region Indiana University School of Medicine, Health Information & Translational Sciences, Pediatrics & Clinical Psychology, Section of Adolescent Medicine, 410 W. 10th Street, HS 1001, Indianapolis, IN 46202, USA
b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 9 November 2011 Received in revised form 20 February 2012 Accepted 21 February 2012
The psychometric properties of the Multidimensional Scale of Perceived Social Support (MSPSS) of the Urdu (MSPSS-U) and Nepali (MSPSS-N) versions were investigated among Pakistani (n = 148) and Nepalese (n = 153) respondents living in Hong Kong. The factor analysis of the MSPSS-N conducted in this study confirmed the three factors of the original MSPSS, namely, Family, Friend, and Significant Others, while only two factors were extracted from the MSPSS-U (i.e., the Family and Friends subscales). With regard to construct validity, both the MSPSS-U and MSPSS-N were inversely correlated with depression, anxiety and stress as assessed by the Depression Anxiety Stress Scale (DASS-21). The internal reliability and construct validity of the MSPSS-U and MSPSS-N were also established and provided support for the validity of both versions as tools for measuring perceived social support among South Asian migrants. ß 2012 Elsevier B.V. All rights reserved.
Keywords: Urdu Nepali Social support Factor analysis Psychometric properties
1. Introduction Over the past 20 years there has been great interest in the role of social support as a mechanism for the maintenance of psychological well-being under conditions of stress. Social support is a multifaceted construct and a review of the literature indicates that the construct of social support consists of both a structural and a functional component (Goebert and Loue, 2009). The structural component includes informal and formal support, such as an individual’s social network (Hirsh, 1981), the size of the social network, frequency of contact with members of the network, reciprocal support and quality of the support (Canty-Mitchell and Zimet, 2000). Functional support refers to the perceived level of the support received, such as emotional support, affirmative support and tangible support (House and Kahn, 1985; Kahn and Antonucci, 1980). These two concepts of social support can be broadly distinguished in terms of received and perceived support. In recent years there has been a growing interest in examining the role of social support during the acculturation process (Choi and Thomas, 2009; Miranda and Matheny, 2000; Stone-Feinstein and Ward, 1990). It has been observed that the presence of supportive family members, relatives and friends can serve as a buffer from acculturative stress, especially during the initial
* Corresponding author. Tel.: +852 61723206; fax: +852 28587604. E-mail address: [email protected] (K. Tonsing). 1876-2018/$ – see front matter ß 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.ajp.2012.02.012
periods of acculturation (Cortes, 1995; Miranda and Matheny, 2000). Various studies have shown that among the two types of support (i.e., received support and perceived support), perceived support is an important predictor for psychological well-being (Aranda et al., 2001; Patel, 1992). Previous studies on social support have utilized a number of measures that includes both open-ended questions and self-report surveys (e.g., Gavazzi, 1994; Lifrak et al., 1997; Norbeck et al., 1981), which are often lengthy and thus may preclude their use in more time-intensive research that also examines multiple constructs. The concept of social support may be viewed differently in different societies, and while it is important to understand how people appraise their social relationships, a lengthy assessment tool is often a barrier to gathering this information. Hence, a short, reliable, and psychometrically acceptable social support assessment tool such as the Multidimensional Scale of Perceived Social Support (MSPSS) is useful. The 12-item MSPSS developed by Zimet et al. (1988) measures perceived social support from three sources: Family, Friends and Significant Others. The inclusion of significant others is unique to the MSPSS. The MSPSS is simple, easy to understand, and can be administered in a short time and is psychometrically sound. The brevity of the scale also makes it suitable for use in research that requires assessing a number of variables within a limited time. The MSPSS as a measure of social support has been used among varied populations (e.g., Canty-Mitchell and Zimet, 2000; Kazarian and McCabe, 1991). However, with a few exceptions (e.g., Akhtar et al., 2010; Cheng and Chan, 2004; Chou, 2000; Ng et al., 2010;
K. Tonsing et al. / Asian Journal of Psychiatry 5 (2012) 164–168
Rizwan and Aftab, 2009), these studies were conducted using western samples. While most of these studies reported good internal reliability and construct validity of the translated MSPSS, only a few were able to replicate the three factor solution of the MSPSS. In the study conducted by Akhtar et al. (2010) among Pakistani women, the Urdu translation of the MSPSS yielded only a single-factor solution, indicating that the participants in the study viewed social support as a single construct. In the study among Chinese adolescents (Chou, 2000), only two factors was identified, that of Family and Friends. The items on the Significant Others and Friends subscales loaded on a single factor, which was combined into a single Friends factor. It is probable that many of the participants may have considered their friends/peers as their significant others (Chou, 2000). The aim of the current study is to examine the psychometric properties of the MSPSS among South Asians using the Urdu (MSPSS-U) and Nepali (MSPSS-N) translated versions of the MSPSS. This article will specifically focus on examining the factorial structure, reliability, and construct validity of the translated versions of the MSPSS.
165
translated from the original language to the targeted languages (i.e., Urdu and Nepali) following the translation back-translation method (Brislin, 1970). The translated versions were then back-translated to the source language by two independent bilingual translators. Translators were asked to take effort to ensure that the translated version were culturally adapted to the local language so that they could be easily understood by the participants. To this end, the translators were from the same community as the research participants and so were able to understand the cultural terms and meanings of the words, taking effort to ensure that the translated meanings remained as close as possible to the original meanings. For instance, the item ‘‘I felt downhearted and blue’’ (item # 11, DASS-21) was translated as ‘‘I felt depressed and sad’’ because there was no meaningful translation for the word ‘‘blue’’ in either of the targeted languages. A pilot study was then conducted with six participants (two Pakistani and four Nepalese) in May 2010 using both the Urdu and Nepali versions of the MSPSS. The participants in the pilot study indicated that the translated versions were clear and easily understandable. 2.4. Measures
2. Methods 2.1. Study design This study involved a cross-sectional survey conducted during the period from July to December 2010. Employing the convenience sampling method, data were collected through survey questionnaires from social service agencies, community centers and religious organizations. Participants were asked to complete a set of questionnaires which included background information, the MSPSS, and the Depression Anxiety Stress Scale (DASS-21). It took approximately 20–25 min to administer the questionnaires. 2.2. Participants The present study was conducted among two sub-groups of South Asian migrants living in Hong Kong, namely, Pakistani and Nepalese. According to the 2006 by-Census data, there were about 55,356 South Asians living in Hong Kong, consisting of people from India, Pakistan, and Nepal. The Indians and Pakistani initially came to Hong Kong as part of Britain’s overall expansion of its colonies in the mid 19th century (Vaid, 1972). While many of the present-day Pakistanis in Hong Kong are the descendants of early 19th century migrants, there are also a significant number of recent migrants. According to the ‘Place of Birth’ information from the 2006 byCensus, nearly 60% of the total Pakistani population in Hong Kong had recently migrated to Hong Kong (Census and Statistics Department, 2007, p. 35). The Nepalese, on the other hand, initially came to the Territory in 1948 as part of the Gurkha Brigade of the British Army (FEONA, 2000). The Sino-British Joint Declaration of 1984 granted right of abode (ROA) to the Gurkhas and to all Nepalese children born in Hong Kong prior to 1983. Since the early 1990s, many Nepalese have exercised this right and migrated to Hong Kong. The total sample was 301: 153 Nepalese (47.7% males) and 148 Pakistanis (39.2% males). The mean age was 33.2 (SD = 8.90) for the Nepalese and 32.4 years (SD = 8.09) for the Pakistani sample.
Two scales were administered to the participants: the MSPSS (Zimet et al., 1988) and the DASS-21 (Lovibond and Lovibond, 1995). The MSPSS is a 12-item instrument which measures perceived support from three sources: Family, Friends and Significant Others (Zimet et al., 1988). Items are scored on a 7-point rating scale ranging from 1 (very strongly disagree) to 7 (very strongly agree) with possible scores ranging from 12 to 84. The MSPSS has shown high internal reliability (Cronbach’s alpha = .87, .85, and .91 respectively for the Family, Friends and Significant Others subscales). The 21-item DASS was administered to assess emotional distress in the past week. The DASS-21 consists of three subscales which are designed to measure the severity and frequency of experiencing symptoms of depression, anxiety and stress (Lovibond and Lovibond, 1995). It is important to note that the DASS-21 is used as a measure to assess the dimensional nature of emotional distress and to report participants’ experiences of related symptoms rather than as a diagnostic tool for categorizing different areas of mental health. Scores on the DASS-21 range from 0 to 63, with higher scores indicating a higher level of emotional distress. The DASS-21 has been used in various research studies, reporting meaningful discriminant validity in a variety of settings (Anthony et al., 1998; Crawford and Henry, 2003; Norton, 1998). The Cronbach’s coefficient alphas are: Depression .81; Anxiety .73; and Stress .81 (Lovibond and Lovibond, 1995).
3. Results 3.1. Descriptive statistics The mean and standard deviations of the subscales of the MSPSS and the total scale for the two groups of participants are provided in Table 1. An overview of the mean scores indicated that participants reported relatively high mean scores on support.
2.3. Procedures
3.2. Analyses of age and gender differences
Ethical approval was obtained from the Human Research Ethics Committee for Non-clinical Faculties, the University of Hong Kong, to conduct the study. Participants were assured of the confidentiality of their responses and the voluntary nature of their participation. Signed consent forms were obtained from all of those who volunteered to participate. All of the scales used in this study were
To assess for the potential effects of age and gender, a multivariate analysis of variance (MANOVA) was computed for the two subscales of the MSPSS-U and for the three subscales of the MSPSS-N, separately for each group. The MSPSS-U subscales of Family and Friends and the three subscales of the MSPSS-N were entered as the dependent variables and age and gender as the
K. Tonsing et al. / Asian Journal of Psychiatry 5 (2012) 164–168
166
Table 1 MSPSS means and standard deviations by race/gender sub-groups. Nepalese
Pakistani
Family Friends Significant Others Total scale
Male (n = 58)
Female (n = 80)
Male (n = 73)
Female (n = 80)
5.69 5.65 5.50 5.62
5.72 5.33 5.40 5.48
5.44 4.87 4.49 4.93
5.70 4.58 4.84 5.04
(1.12) (1.07) (1.22) (1.02)
(1.27) (1.35) (1.49) (1.23)
(1.44) (1.33) (1.77) (1.30)
(1.14) (1.27) (1.60) (1.10)
MSPSS, multidimensional scale of perceived social support.
between-subject factors. Results did not indicate any significant age or gender effect or an overall Gender Age group interaction. 3.2.1. Factor analysis Factor analysis was performed for both the MSPSS-N and MSPSS-U using Statistical Package for Social Sciences (SPSS 18), with principal axis factoring, and subject to oblique rotation, which is highly recommended when factors are intercorrelated (Floyd and Widaman, 1995). Although prior studies have documented the factor structures of the MSPSS, exploratory studies among varied samples have also yielded alternative factor structures of the scale. For instance, the Urdu version of the MSPSS reported a three-factor structure of the MSPSS in one study (Rizwan and Aftab, 2009), while another study reported only a single factor (Akhtar et al., 2010). Secondly, the factor structures of the Nepali version of the MSPSS have not been investigated in any prior studies. Therefore, it was deemed more appropriate to conduct an exploratory factor analysis (EFA) in the current study in order to determine whether the factor structures of the MSPSS could be replicated with the current data. EFA is conceptually a ‘multi-tailed’ test, as opposed to confirmatory factor analysis (CFA), which is conceptually a ‘onetailed’ test and as such can be used for confirmatory purposes (Gorsuch, 1997, p. 536). The stringent requirement imposed by CFA often requires making adjustment to the structural equation model, which may result in inaccurate probabilities due to capitalization on chance. Therefore, EFA is an appropriate alternative to attempting to adjust the confirmatory model (Gorsuch, 1997). Moreover, the procedure of the principal axis factoring approach, as used in the current study, usually results in the identification of similar number of factors as CFA (Gorsuch, 1997). Thus, an EFA was conducted for both the MSPSS-N and MSPSS-U, with eigenvalues set at 1.00 as the criterion and primary loadings of .40 for all the indices. The results are discussed below. 3.2.2. MSPSS-N For the MSPSS-N, the analysis yielded a three factor solution. The oblique rotation pattern matrix indicated that all of the items
loaded highly (.52 and above) on each of the respective subscales (see Table 2). Factor 1 (Friends) accounted for 46.73% of the variance (eigenvalues = 5.96), Factor 2 (Family) accounted for 9.05% (eigenvalues 1.41) of the variance, and Factor 3 (Significant Others) accounted for 7.82% of the variance (eigenvalues = 1.29), and in combination accounted for 63.61% of the total variance. The intercorrelation among the three subscales was found to be significantly correlated: Family with Friends, r = .530; p < .01; Family with Significant Others, r = .540, p < .01; and Significant Others with Friends, r = .575, p < .01. In short, the factorial structure of the MSPSS-N is similar to that of the original MSPSS. 3.2.3. MSPSS-U For the MSPSS-U, the analysis yielded only two factors as opposed to the three-factor solution in the original MSPSS (see Table 3). The Friends subscale was retained, while the items on Family and Significant Others loaded on a single factor, suggesting a more broad-based Family subscale. Factor 1 (eigenvalues = 7.11) accounted for 59.32% of the variance and included items relating to support from Family and Significant Others, and Factor 2 (eigenvalues = 1.14) accounted for 9.57% of the variance and included items that reflect perceived social support from friends, and in combination accounted for 68.89% of the total variance. The subscales of Family and Friends were found to be significantly correlated (Family with Friends, r = .722; p < .01), confirming the factor structure of the MSPSS-U despite the fact that the Family and Significant Others subscales of the original MSPSS were combined in the MSPSS-U. 3.3. Reliability To assess the internal reliability of the MSPSS-N and MSPSS-U, Cronbach’s coefficient alpha was computed for the total MSPSS and for each subscale. The Cronbach’s alpha for the total MSPSS-N was .90. The Family, Friends, and Significant Others subscales demonstrated coefficient alphas of .86, .84 and .80, respectively. The coefficient alpha for the total MSPSS-U was .93, and .91 and .90
Table 2 Summary of factor loading from principal axis factor analysis of MSPSS-N. Factor loading Friends 7 9 12 6 3 4 11 8 2 1 5 10
I can count on my friends when things go wrong I have friends with whom I can share my joys and sorrows I can talk about my problems with my friends My friends really try to help me My family really tries to help me I get the emotional help and support I need from my family My family is willing to help me make decisions I can talk about my problems with my family There is a special person with whom I can share my joys and sorrows There is a special person around when I am in need I have a special person who is a source of comfort to me There is a special person in my life who cares about my feelings
MSPSS-N, multidimensional scale of perceived social support of the Nepali.
Family
Significant Others
.883 .737 .631 .528 .869 .784 .740 .608 .873 .823 .718 .652
K. Tonsing et al. / Asian Journal of Psychiatry 5 (2012) 164–168
167
Table 3 Summary of factor loading from principal axis factor analysis of MSPSS-U. Factor loading Family I get the emotional support I need from my family My family really tries to help me My family is willing to help me make decision I can talk my problems with my family There is a special person in my life who cares about my feelings I have a special person who is a real source of comfort to me There is a special person with whom I can share my joys and sorrows There is a special person who is around when I am in need I can talk about my problems with my friends My friends really try to help me I can count on my friends when things go wrong I have friends with whom I can share my joys and sorrows
4 3 11 8 10 5 2 1 12 6 7 9
Friends
.882 .836 .822 .626 .748 .630 .587 .491 .865 .860 .792 .675
MSPSS-U, multidimensional scale of perceived social support of the Urdu. Table 4 Intercorrelation of Family, Friends and Significant Others subscales of the MSPSS-N, Depression, Anxiety and Stress subscales of the DASS-21: Nepalese group (N = 153).
1 2 3 4 5 6
MSPSS-Family MSPSS-Friends MSPSS-Significant Others DASS-Depression DASS-Anxiety DASS-Stress
1
2
3
4
5
6
1 .530** .540** .233** .320** .297**
1 .575** .108 .150 .192*
1 .046 .162* .123
1 .741** .788**
1 .744**
1
MSPSS-N, multidimensional scale of perceived social support of the Nepali; DASS-21, depression anxiety stress scale. * p < .05. ** p < .01.
for the Family and Friends subscales, respectively. The results indicate that the overall reliabilities of both the MSPSS-U and MSPSS-N were quite good and comparable with the original scale. 3.4. Construct validity An underlying assumption of the development of the MSPSS was that it would be negatively correlated to reported anxiety and depression symptoms (Zimet et al., 1988). This hypothesis was supported by the correlations observed between the MSPSS subscales and the Depression Anxiety and Stress subscales of the DASS-21. The correlation matrices of the scales for the two ethnic groups are provided in Tables 4 and 5. For the Nepalese, the Family dimension of the MSPSS-N subscale (Table 4) was significantly inversely correlated with depression, anxiety, and stress (p < .01), while perceived support from Friends was negatively correlated with stress (p < .05) and Significant Others was negatively correlated with anxiety (p < .05). As can be observed in Table 5, for the Pakistanis, perceived support from family and friends was inversely correlated with depression and stress but not with anxiety.
Table 5 Intercorrelation of Family and Friends subscales of the MSPSS-U, Depression, Anxiety and Stress subscales of the DASS-21: Pakistani group (N = 148).
1 2 3 4 5
MSPSS-Family MSPSS-Friends DASS-Depression DASS-Anxiety DASS-Stress
1
2
1 .722** .296** .127 .335**
1 .215** .083 .324**
3
4
5
1 .756**
1
1 .776** .786**
MSPSS-U, multidimensional scale of perceived social support of the Urdu; DASS-21, depression anxiety stress scale. ** p < .01.
4. Discussion The goal of this study was to assess the psychometric properties of the Urdu and Nepalese versions of the MSPSS, which were administered to two sub-groups of South Asians living in Hong Kong. The factor structure proposed in the original study (Zimet et al., 1988) has been confirmed in this study by the results from the Nepalese sample, and participants could clearly differentiate among the three sources of support, (i.e., Friends, Family, and Significant Others). Perceived social support from family was more strongly inversely correlated with anxiety, stress, and depression than support from Friends and Significant Others (Table 4). This set of findings indicates that for the Nepalese, the sources of perceived support from Family, Friends and Significant Others are considered to be independent but internally consistent. On the other hand, the MSPSS-U yielded only two factors. While the Friends subscale of the original three factors of the MSPSS was retained, the subscales of Family and Significant Others were combined into a single factor of the Family subscale. Although the MSPSS-U reported high internal consistency, the factor analysis of the scale did not support the three-factor solution reported in previous studies (Ng et al., 2010; Rizwan and Aftab, 2009; Zimet et al., 1988). Possible explanations for this could be the translation method or cultural factors or a combination of these factors which may have accounted for the discrepancy in the MSPSS-U. In this study, 81.1% of the participants were married. It is probable that the ‘special person’ in the Significant Others subscale may have been interpreted as referring to their spouse as the items making up the Significant Others subscale were constructed to leave interpretation up to participants (Canty-Mitchell and Zimet, 2000). Others studies have also identified problems with the Significant Others subscale in which items on this subscale were found to load on either the Friends subscale (e.g., Cheng and Chan, 2004; Chou, 2000), or the Family subscale (Stanley et al., 1998).
168
K. Tonsing et al. / Asian Journal of Psychiatry 5 (2012) 164–168
The initial results of the present study show good construct validity for the MSPSS-N and MSPSS-U as evinced by the inverse correlation of the subscales with the measures of depression, anxiety, and stress (DASS-21). This is consistent with previous studies which have also reported the negative correlation between perceived social support and psychological distress (e.g., Chou, 2000; Zimet et al., 1988). The results of this study also indicate that the MSPSS-N and the MSPSS-U demonstrate adequate psychometric properties in relation to internal consistency and content validity, lending support to prior research (e.g., Canty-Mitchell and Zimet, 2000; Zimet et al., 1990) and suggesting that the MSPSS can be used among varied groups of populations with confidence. Although the findings of this study show that the MSPSS-N and the MSPSS-U are psychometrically sound and valid instruments, the study has certain limitations. Firstly, the samples were drawn using convenience sampling, and hence the findings of this study cannot be assumed to be generalizable to the larger population. Secondly, the findings from this correlational study do not determine the causal direction of the relationship of social support to emotional distress. The focus of this study was to assess the factor structure of the MSPSS across two groups, and as such the buffering effects of social support was not tested. The identification of only two factor structures in the MSPSS-U suggests that there is still a need to explore the subjective constitution of the ‘special person’ in the Significant Others subscale which could help to avoid the ambiguity of the items in this subscale. References Akhtar, A., Rahman, A., Husain, M., Chaudhry, I.B., Duddu, V., 2010. Multidimensional scale of perceived social support: psychometric properties in a South Asian population. Journal of Obstetrics and Gynaecology Research 6, 845–851. Anthony, M.M., Bieling, P.J., Cox, B.J., Enns, M.W., Swinson, R.P., 1998. Psychometric properties of the 42-item and 21-item versions of the Depression Anxiety Stress Scale in clinical groups and a community sample. Psychological Assessment 10, 176–181. Aranda, M.P., Castaneda, I., Lee, P.-J., Sobel, E., 2001. Stress, social support, and coping as predictors of depressive symptoms: gender differences among Mexican Americans. Social Work Research 25, 37–49. Brislin, R.W., 1970. Back-translation for crosscultural research. Journal of CrossCultural Psychology 1, 185–216. Canty-Mitchell, J., Zimet, G.D., 2000. Psychometric properties of the multidimensional scale of perceived social support in urban adolescents. American Journal of Community Psychology 28, 391–400. Census and Statistics Department, 2007. 2006 Population By-Census. Thematic Report – Ethnic Minorities. Census and Statistics Department, Hong Kong. Cheng, S.T., Chan, C.M., 2004. The multidimensional scale of perceived social support: dimensionality and age and gender differences in adolescents. Personality and Individual Differences 37, 1359–1369. Choi, J.B, Thomas, M., 2009. Predictive factors of acculturation attitudes and social support among Asian immigrants in the USA. International Journal of Social Welfare 18, 76–84.
Chou, K.L., 2000. Assessing Chinese adolescents’ social support: the multidimensional scale of perceived social support. Personality and Individual Differences 28, 299–307. Cortes, D.E, 1995. Variations in familism in two generations of Puerto Ricans. Hispanic Journal of Behavioral Sciences 17, 249–255. Crawford, J.R., Henry, J.D., 2003. The Depression Anxiety Stress Scale (DASS): normative data and latent structure in a large non-clinical sample. British Journal of Clinical Psychology 42, 111–131. Feona, 2000. Equality for All Races and Sexes. Far East Overseas Nepalese Association, Hong Kong. Floyd, F.J., Widaman, K.F., 1995. Factor analysis in the development and refinement of clinical assessment instruments. Psychological Assessment 62, 465–471. Gavazzi, S.M., 1994. Perceived social support from family and friends in a clinical sample of adolescents. Journal of Personality Assessment 62, 465–471. Goebert, D., Loue, S., 2009. Social support, mental health, minorities and acculturative stress. In: Loue, S., Sajatovic, M. (Eds.), Determinants of Minority Mental Health and Wellness. Gorsuch, R.L., 1997. Exploratory factor analysis: its role in item analysis. Journal of Personality Assessment 68, 532–560. Hirsh, B.J., 1981. Social networks and social support in community mental health. In: Gottlieb, B. (Ed.), Social Networks and Social Support. Sage, Beverly Hills, CA. House, J.S., Kahn, R.L., 1985. In: Cohen, S., Syme, S.L. (Eds.), Measures and Concepts of Social Support. Academic Press, New York. Kahn, R.L., Antonucci, T.C., 1980. Convoys over the life course: attachment, roles, and social support. Life-span Development and Behavior 3, 253–286. Kazarian, S.S., Mccabe, S.B., 1991. Dimensions of social support in the MSPSS: factorial structure, reliability, and theoretical implications. Journal of Community Psychology 19, 150–160. Lifrak, P.D, Mckay, J.R., Rostain, A., Alterman, A.I., O’ Brien, C.P., 1997. Relationship of perceived competencies, perceived social support, and gender to substance use in young adolescents. Journal of the American Academy of Child and Psychiatry 36, 933–940. Lovibond, S.H., Lovibond, P.F., 1995. Manual for the Depression Anxiety Stress Scales. Psychology Foundation, Sydney. Miranda, A.O., Matheny, K.B., 2000. Socio-psychological predictors of acculturative stress among Latino adults. Journal of Mental Health Counseling 22, 306–317. Ng, C.G., Siddiq, A.a.N., Aida, S.A., Zainal, N.Z., Koh, O.H., 2010. Validation of the Malay version of the Multidimensional scale of perceived social support (MSPSS-M) among a group of medical students in faculty of medicine, University Malaya. Asian Journal of Psychiatry 3, 3–6. Norbeck, J.S., Lindsey, A.M., Carrieri, V.L., 1981. The development of an instrument to measure social support. Nursing Research 30, 264–269. Norton, P.J., 1998. Depression Anxiety and Stress Scales (DASS-21): psychometric analysis across four racial groups. Anxiety, Stress, & Coping 20, 253–265. Patel, N., 1992. Psychological disturbance, social support and stressors: a community survey of immigrant Asian women and the indigenous population. Counseling Psychology Quarterly 5, 263–277. Rizwan, M., Aftab, S., 2009. Psychometric properties of the multidimensional scale of perceived social support in Pakistani young adults. Pakistan Journal of Psychology 40, 51–65. Stanley, M.A., Beck, J.G., Zebb, B.J., 1998. Psychometric properties of the MSPSS in older adults. Aging and Mental Health 2, 186–193. Stone-Feinstein, E., Ward, C., 1990. Loneliness and psychological adjustment of sojourners: new perspectives on culture shock. In: Keats, D.M., Munro, D., Mann, L. (Eds.), Heterogeneity in Cross-cultural Psychology Lisse. Swets & Zeitlinger, The Netherlands. Vaid, K.N., 1972. The Overseas Indian Community in Hong Kong, Hong Kong. Centre of Asian Studies, The University of Hong Kong. Zimet, G.D., Dahlem, N.W., Zimet, S.G., Farley, G.K., 1988. The multidimensional scale of perceived social support. Journal of Personality Assessment 52, 30–41. Zimet, G.D., Powell, S.S., Farley, G.K., Werkman, S., Berkoff, K.A., 1990. Psychometric properties of the multidimensional scale of perceived social support. Journal of Personality Assessment 55, 610–617.
Contents lists available at SciVerse ScienceDirect
Asian Journal of Psychiatry journal homepage: www.elsevier.com/locate/ajp
Assessing social support among South Asians: The multidimensional scale of perceived social support Kareen Tonsing a,*, Gregory D. Zimet b, Samson Tse a a
The University of Hong Kong, 13/F, K.K. Leung Building, Department of Social Work and Social Administration, Pokfulam Road, Hong Kong Special Administrative Region Indiana University School of Medicine, Health Information & Translational Sciences, Pediatrics & Clinical Psychology, Section of Adolescent Medicine, 410 W. 10th Street, HS 1001, Indianapolis, IN 46202, USA
b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 9 November 2011 Received in revised form 20 February 2012 Accepted 21 February 2012
The psychometric properties of the Multidimensional Scale of Perceived Social Support (MSPSS) of the Urdu (MSPSS-U) and Nepali (MSPSS-N) versions were investigated among Pakistani (n = 148) and Nepalese (n = 153) respondents living in Hong Kong. The factor analysis of the MSPSS-N conducted in this study confirmed the three factors of the original MSPSS, namely, Family, Friend, and Significant Others, while only two factors were extracted from the MSPSS-U (i.e., the Family and Friends subscales). With regard to construct validity, both the MSPSS-U and MSPSS-N were inversely correlated with depression, anxiety and stress as assessed by the Depression Anxiety Stress Scale (DASS-21). The internal reliability and construct validity of the MSPSS-U and MSPSS-N were also established and provided support for the validity of both versions as tools for measuring perceived social support among South Asian migrants. ß 2012 Elsevier B.V. All rights reserved.
Keywords: Urdu Nepali Social support Factor analysis Psychometric properties
1. Introduction Over the past 20 years there has been great interest in the role of social support as a mechanism for the maintenance of psychological well-being under conditions of stress. Social support is a multifaceted construct and a review of the literature indicates that the construct of social support consists of both a structural and a functional component (Goebert and Loue, 2009). The structural component includes informal and formal support, such as an individual’s social network (Hirsh, 1981), the size of the social network, frequency of contact with members of the network, reciprocal support and quality of the support (Canty-Mitchell and Zimet, 2000). Functional support refers to the perceived level of the support received, such as emotional support, affirmative support and tangible support (House and Kahn, 1985; Kahn and Antonucci, 1980). These two concepts of social support can be broadly distinguished in terms of received and perceived support. In recent years there has been a growing interest in examining the role of social support during the acculturation process (Choi and Thomas, 2009; Miranda and Matheny, 2000; Stone-Feinstein and Ward, 1990). It has been observed that the presence of supportive family members, relatives and friends can serve as a buffer from acculturative stress, especially during the initial
* Corresponding author. Tel.: +852 61723206; fax: +852 28587604. E-mail address: [email protected] (K. Tonsing). 1876-2018/$ – see front matter ß 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.ajp.2012.02.012
periods of acculturation (Cortes, 1995; Miranda and Matheny, 2000). Various studies have shown that among the two types of support (i.e., received support and perceived support), perceived support is an important predictor for psychological well-being (Aranda et al., 2001; Patel, 1992). Previous studies on social support have utilized a number of measures that includes both open-ended questions and self-report surveys (e.g., Gavazzi, 1994; Lifrak et al., 1997; Norbeck et al., 1981), which are often lengthy and thus may preclude their use in more time-intensive research that also examines multiple constructs. The concept of social support may be viewed differently in different societies, and while it is important to understand how people appraise their social relationships, a lengthy assessment tool is often a barrier to gathering this information. Hence, a short, reliable, and psychometrically acceptable social support assessment tool such as the Multidimensional Scale of Perceived Social Support (MSPSS) is useful. The 12-item MSPSS developed by Zimet et al. (1988) measures perceived social support from three sources: Family, Friends and Significant Others. The inclusion of significant others is unique to the MSPSS. The MSPSS is simple, easy to understand, and can be administered in a short time and is psychometrically sound. The brevity of the scale also makes it suitable for use in research that requires assessing a number of variables within a limited time. The MSPSS as a measure of social support has been used among varied populations (e.g., Canty-Mitchell and Zimet, 2000; Kazarian and McCabe, 1991). However, with a few exceptions (e.g., Akhtar et al., 2010; Cheng and Chan, 2004; Chou, 2000; Ng et al., 2010;
K. Tonsing et al. / Asian Journal of Psychiatry 5 (2012) 164–168
Rizwan and Aftab, 2009), these studies were conducted using western samples. While most of these studies reported good internal reliability and construct validity of the translated MSPSS, only a few were able to replicate the three factor solution of the MSPSS. In the study conducted by Akhtar et al. (2010) among Pakistani women, the Urdu translation of the MSPSS yielded only a single-factor solution, indicating that the participants in the study viewed social support as a single construct. In the study among Chinese adolescents (Chou, 2000), only two factors was identified, that of Family and Friends. The items on the Significant Others and Friends subscales loaded on a single factor, which was combined into a single Friends factor. It is probable that many of the participants may have considered their friends/peers as their significant others (Chou, 2000). The aim of the current study is to examine the psychometric properties of the MSPSS among South Asians using the Urdu (MSPSS-U) and Nepali (MSPSS-N) translated versions of the MSPSS. This article will specifically focus on examining the factorial structure, reliability, and construct validity of the translated versions of the MSPSS.
165
translated from the original language to the targeted languages (i.e., Urdu and Nepali) following the translation back-translation method (Brislin, 1970). The translated versions were then back-translated to the source language by two independent bilingual translators. Translators were asked to take effort to ensure that the translated version were culturally adapted to the local language so that they could be easily understood by the participants. To this end, the translators were from the same community as the research participants and so were able to understand the cultural terms and meanings of the words, taking effort to ensure that the translated meanings remained as close as possible to the original meanings. For instance, the item ‘‘I felt downhearted and blue’’ (item # 11, DASS-21) was translated as ‘‘I felt depressed and sad’’ because there was no meaningful translation for the word ‘‘blue’’ in either of the targeted languages. A pilot study was then conducted with six participants (two Pakistani and four Nepalese) in May 2010 using both the Urdu and Nepali versions of the MSPSS. The participants in the pilot study indicated that the translated versions were clear and easily understandable. 2.4. Measures
2. Methods 2.1. Study design This study involved a cross-sectional survey conducted during the period from July to December 2010. Employing the convenience sampling method, data were collected through survey questionnaires from social service agencies, community centers and religious organizations. Participants were asked to complete a set of questionnaires which included background information, the MSPSS, and the Depression Anxiety Stress Scale (DASS-21). It took approximately 20–25 min to administer the questionnaires. 2.2. Participants The present study was conducted among two sub-groups of South Asian migrants living in Hong Kong, namely, Pakistani and Nepalese. According to the 2006 by-Census data, there were about 55,356 South Asians living in Hong Kong, consisting of people from India, Pakistan, and Nepal. The Indians and Pakistani initially came to Hong Kong as part of Britain’s overall expansion of its colonies in the mid 19th century (Vaid, 1972). While many of the present-day Pakistanis in Hong Kong are the descendants of early 19th century migrants, there are also a significant number of recent migrants. According to the ‘Place of Birth’ information from the 2006 byCensus, nearly 60% of the total Pakistani population in Hong Kong had recently migrated to Hong Kong (Census and Statistics Department, 2007, p. 35). The Nepalese, on the other hand, initially came to the Territory in 1948 as part of the Gurkha Brigade of the British Army (FEONA, 2000). The Sino-British Joint Declaration of 1984 granted right of abode (ROA) to the Gurkhas and to all Nepalese children born in Hong Kong prior to 1983. Since the early 1990s, many Nepalese have exercised this right and migrated to Hong Kong. The total sample was 301: 153 Nepalese (47.7% males) and 148 Pakistanis (39.2% males). The mean age was 33.2 (SD = 8.90) for the Nepalese and 32.4 years (SD = 8.09) for the Pakistani sample.
Two scales were administered to the participants: the MSPSS (Zimet et al., 1988) and the DASS-21 (Lovibond and Lovibond, 1995). The MSPSS is a 12-item instrument which measures perceived support from three sources: Family, Friends and Significant Others (Zimet et al., 1988). Items are scored on a 7-point rating scale ranging from 1 (very strongly disagree) to 7 (very strongly agree) with possible scores ranging from 12 to 84. The MSPSS has shown high internal reliability (Cronbach’s alpha = .87, .85, and .91 respectively for the Family, Friends and Significant Others subscales). The 21-item DASS was administered to assess emotional distress in the past week. The DASS-21 consists of three subscales which are designed to measure the severity and frequency of experiencing symptoms of depression, anxiety and stress (Lovibond and Lovibond, 1995). It is important to note that the DASS-21 is used as a measure to assess the dimensional nature of emotional distress and to report participants’ experiences of related symptoms rather than as a diagnostic tool for categorizing different areas of mental health. Scores on the DASS-21 range from 0 to 63, with higher scores indicating a higher level of emotional distress. The DASS-21 has been used in various research studies, reporting meaningful discriminant validity in a variety of settings (Anthony et al., 1998; Crawford and Henry, 2003; Norton, 1998). The Cronbach’s coefficient alphas are: Depression .81; Anxiety .73; and Stress .81 (Lovibond and Lovibond, 1995).
3. Results 3.1. Descriptive statistics The mean and standard deviations of the subscales of the MSPSS and the total scale for the two groups of participants are provided in Table 1. An overview of the mean scores indicated that participants reported relatively high mean scores on support.
2.3. Procedures
3.2. Analyses of age and gender differences
Ethical approval was obtained from the Human Research Ethics Committee for Non-clinical Faculties, the University of Hong Kong, to conduct the study. Participants were assured of the confidentiality of their responses and the voluntary nature of their participation. Signed consent forms were obtained from all of those who volunteered to participate. All of the scales used in this study were
To assess for the potential effects of age and gender, a multivariate analysis of variance (MANOVA) was computed for the two subscales of the MSPSS-U and for the three subscales of the MSPSS-N, separately for each group. The MSPSS-U subscales of Family and Friends and the three subscales of the MSPSS-N were entered as the dependent variables and age and gender as the
K. Tonsing et al. / Asian Journal of Psychiatry 5 (2012) 164–168
166
Table 1 MSPSS means and standard deviations by race/gender sub-groups. Nepalese
Pakistani
Family Friends Significant Others Total scale
Male (n = 58)
Female (n = 80)
Male (n = 73)
Female (n = 80)
5.69 5.65 5.50 5.62
5.72 5.33 5.40 5.48
5.44 4.87 4.49 4.93
5.70 4.58 4.84 5.04
(1.12) (1.07) (1.22) (1.02)
(1.27) (1.35) (1.49) (1.23)
(1.44) (1.33) (1.77) (1.30)
(1.14) (1.27) (1.60) (1.10)
MSPSS, multidimensional scale of perceived social support.
between-subject factors. Results did not indicate any significant age or gender effect or an overall Gender Age group interaction. 3.2.1. Factor analysis Factor analysis was performed for both the MSPSS-N and MSPSS-U using Statistical Package for Social Sciences (SPSS 18), with principal axis factoring, and subject to oblique rotation, which is highly recommended when factors are intercorrelated (Floyd and Widaman, 1995). Although prior studies have documented the factor structures of the MSPSS, exploratory studies among varied samples have also yielded alternative factor structures of the scale. For instance, the Urdu version of the MSPSS reported a three-factor structure of the MSPSS in one study (Rizwan and Aftab, 2009), while another study reported only a single factor (Akhtar et al., 2010). Secondly, the factor structures of the Nepali version of the MSPSS have not been investigated in any prior studies. Therefore, it was deemed more appropriate to conduct an exploratory factor analysis (EFA) in the current study in order to determine whether the factor structures of the MSPSS could be replicated with the current data. EFA is conceptually a ‘multi-tailed’ test, as opposed to confirmatory factor analysis (CFA), which is conceptually a ‘onetailed’ test and as such can be used for confirmatory purposes (Gorsuch, 1997, p. 536). The stringent requirement imposed by CFA often requires making adjustment to the structural equation model, which may result in inaccurate probabilities due to capitalization on chance. Therefore, EFA is an appropriate alternative to attempting to adjust the confirmatory model (Gorsuch, 1997). Moreover, the procedure of the principal axis factoring approach, as used in the current study, usually results in the identification of similar number of factors as CFA (Gorsuch, 1997). Thus, an EFA was conducted for both the MSPSS-N and MSPSS-U, with eigenvalues set at 1.00 as the criterion and primary loadings of .40 for all the indices. The results are discussed below. 3.2.2. MSPSS-N For the MSPSS-N, the analysis yielded a three factor solution. The oblique rotation pattern matrix indicated that all of the items
loaded highly (.52 and above) on each of the respective subscales (see Table 2). Factor 1 (Friends) accounted for 46.73% of the variance (eigenvalues = 5.96), Factor 2 (Family) accounted for 9.05% (eigenvalues 1.41) of the variance, and Factor 3 (Significant Others) accounted for 7.82% of the variance (eigenvalues = 1.29), and in combination accounted for 63.61% of the total variance. The intercorrelation among the three subscales was found to be significantly correlated: Family with Friends, r = .530; p < .01; Family with Significant Others, r = .540, p < .01; and Significant Others with Friends, r = .575, p < .01. In short, the factorial structure of the MSPSS-N is similar to that of the original MSPSS. 3.2.3. MSPSS-U For the MSPSS-U, the analysis yielded only two factors as opposed to the three-factor solution in the original MSPSS (see Table 3). The Friends subscale was retained, while the items on Family and Significant Others loaded on a single factor, suggesting a more broad-based Family subscale. Factor 1 (eigenvalues = 7.11) accounted for 59.32% of the variance and included items relating to support from Family and Significant Others, and Factor 2 (eigenvalues = 1.14) accounted for 9.57% of the variance and included items that reflect perceived social support from friends, and in combination accounted for 68.89% of the total variance. The subscales of Family and Friends were found to be significantly correlated (Family with Friends, r = .722; p < .01), confirming the factor structure of the MSPSS-U despite the fact that the Family and Significant Others subscales of the original MSPSS were combined in the MSPSS-U. 3.3. Reliability To assess the internal reliability of the MSPSS-N and MSPSS-U, Cronbach’s coefficient alpha was computed for the total MSPSS and for each subscale. The Cronbach’s alpha for the total MSPSS-N was .90. The Family, Friends, and Significant Others subscales demonstrated coefficient alphas of .86, .84 and .80, respectively. The coefficient alpha for the total MSPSS-U was .93, and .91 and .90
Table 2 Summary of factor loading from principal axis factor analysis of MSPSS-N. Factor loading Friends 7 9 12 6 3 4 11 8 2 1 5 10
I can count on my friends when things go wrong I have friends with whom I can share my joys and sorrows I can talk about my problems with my friends My friends really try to help me My family really tries to help me I get the emotional help and support I need from my family My family is willing to help me make decisions I can talk about my problems with my family There is a special person with whom I can share my joys and sorrows There is a special person around when I am in need I have a special person who is a source of comfort to me There is a special person in my life who cares about my feelings
MSPSS-N, multidimensional scale of perceived social support of the Nepali.
Family
Significant Others
.883 .737 .631 .528 .869 .784 .740 .608 .873 .823 .718 .652
K. Tonsing et al. / Asian Journal of Psychiatry 5 (2012) 164–168
167
Table 3 Summary of factor loading from principal axis factor analysis of MSPSS-U. Factor loading Family I get the emotional support I need from my family My family really tries to help me My family is willing to help me make decision I can talk my problems with my family There is a special person in my life who cares about my feelings I have a special person who is a real source of comfort to me There is a special person with whom I can share my joys and sorrows There is a special person who is around when I am in need I can talk about my problems with my friends My friends really try to help me I can count on my friends when things go wrong I have friends with whom I can share my joys and sorrows
4 3 11 8 10 5 2 1 12 6 7 9
Friends
.882 .836 .822 .626 .748 .630 .587 .491 .865 .860 .792 .675
MSPSS-U, multidimensional scale of perceived social support of the Urdu. Table 4 Intercorrelation of Family, Friends and Significant Others subscales of the MSPSS-N, Depression, Anxiety and Stress subscales of the DASS-21: Nepalese group (N = 153).
1 2 3 4 5 6
MSPSS-Family MSPSS-Friends MSPSS-Significant Others DASS-Depression DASS-Anxiety DASS-Stress
1
2
3
4
5
6
1 .530** .540** .233** .320** .297**
1 .575** .108 .150 .192*
1 .046 .162* .123
1 .741** .788**
1 .744**
1
MSPSS-N, multidimensional scale of perceived social support of the Nepali; DASS-21, depression anxiety stress scale. * p < .05. ** p < .01.
for the Family and Friends subscales, respectively. The results indicate that the overall reliabilities of both the MSPSS-U and MSPSS-N were quite good and comparable with the original scale. 3.4. Construct validity An underlying assumption of the development of the MSPSS was that it would be negatively correlated to reported anxiety and depression symptoms (Zimet et al., 1988). This hypothesis was supported by the correlations observed between the MSPSS subscales and the Depression Anxiety and Stress subscales of the DASS-21. The correlation matrices of the scales for the two ethnic groups are provided in Tables 4 and 5. For the Nepalese, the Family dimension of the MSPSS-N subscale (Table 4) was significantly inversely correlated with depression, anxiety, and stress (p < .01), while perceived support from Friends was negatively correlated with stress (p < .05) and Significant Others was negatively correlated with anxiety (p < .05). As can be observed in Table 5, for the Pakistanis, perceived support from family and friends was inversely correlated with depression and stress but not with anxiety.
Table 5 Intercorrelation of Family and Friends subscales of the MSPSS-U, Depression, Anxiety and Stress subscales of the DASS-21: Pakistani group (N = 148).
1 2 3 4 5
MSPSS-Family MSPSS-Friends DASS-Depression DASS-Anxiety DASS-Stress
1
2
1 .722** .296** .127 .335**
1 .215** .083 .324**
3
4
5
1 .756**
1
1 .776** .786**
MSPSS-U, multidimensional scale of perceived social support of the Urdu; DASS-21, depression anxiety stress scale. ** p < .01.
4. Discussion The goal of this study was to assess the psychometric properties of the Urdu and Nepalese versions of the MSPSS, which were administered to two sub-groups of South Asians living in Hong Kong. The factor structure proposed in the original study (Zimet et al., 1988) has been confirmed in this study by the results from the Nepalese sample, and participants could clearly differentiate among the three sources of support, (i.e., Friends, Family, and Significant Others). Perceived social support from family was more strongly inversely correlated with anxiety, stress, and depression than support from Friends and Significant Others (Table 4). This set of findings indicates that for the Nepalese, the sources of perceived support from Family, Friends and Significant Others are considered to be independent but internally consistent. On the other hand, the MSPSS-U yielded only two factors. While the Friends subscale of the original three factors of the MSPSS was retained, the subscales of Family and Significant Others were combined into a single factor of the Family subscale. Although the MSPSS-U reported high internal consistency, the factor analysis of the scale did not support the three-factor solution reported in previous studies (Ng et al., 2010; Rizwan and Aftab, 2009; Zimet et al., 1988). Possible explanations for this could be the translation method or cultural factors or a combination of these factors which may have accounted for the discrepancy in the MSPSS-U. In this study, 81.1% of the participants were married. It is probable that the ‘special person’ in the Significant Others subscale may have been interpreted as referring to their spouse as the items making up the Significant Others subscale were constructed to leave interpretation up to participants (Canty-Mitchell and Zimet, 2000). Others studies have also identified problems with the Significant Others subscale in which items on this subscale were found to load on either the Friends subscale (e.g., Cheng and Chan, 2004; Chou, 2000), or the Family subscale (Stanley et al., 1998).
168
K. Tonsing et al. / Asian Journal of Psychiatry 5 (2012) 164–168
The initial results of the present study show good construct validity for the MSPSS-N and MSPSS-U as evinced by the inverse correlation of the subscales with the measures of depression, anxiety, and stress (DASS-21). This is consistent with previous studies which have also reported the negative correlation between perceived social support and psychological distress (e.g., Chou, 2000; Zimet et al., 1988). The results of this study also indicate that the MSPSS-N and the MSPSS-U demonstrate adequate psychometric properties in relation to internal consistency and content validity, lending support to prior research (e.g., Canty-Mitchell and Zimet, 2000; Zimet et al., 1990) and suggesting that the MSPSS can be used among varied groups of populations with confidence. Although the findings of this study show that the MSPSS-N and the MSPSS-U are psychometrically sound and valid instruments, the study has certain limitations. Firstly, the samples were drawn using convenience sampling, and hence the findings of this study cannot be assumed to be generalizable to the larger population. Secondly, the findings from this correlational study do not determine the causal direction of the relationship of social support to emotional distress. The focus of this study was to assess the factor structure of the MSPSS across two groups, and as such the buffering effects of social support was not tested. The identification of only two factor structures in the MSPSS-U suggests that there is still a need to explore the subjective constitution of the ‘special person’ in the Significant Others subscale which could help to avoid the ambiguity of the items in this subscale. References Akhtar, A., Rahman, A., Husain, M., Chaudhry, I.B., Duddu, V., 2010. Multidimensional scale of perceived social support: psychometric properties in a South Asian population. Journal of Obstetrics and Gynaecology Research 6, 845–851. Anthony, M.M., Bieling, P.J., Cox, B.J., Enns, M.W., Swinson, R.P., 1998. Psychometric properties of the 42-item and 21-item versions of the Depression Anxiety Stress Scale in clinical groups and a community sample. Psychological Assessment 10, 176–181. Aranda, M.P., Castaneda, I., Lee, P.-J., Sobel, E., 2001. Stress, social support, and coping as predictors of depressive symptoms: gender differences among Mexican Americans. Social Work Research 25, 37–49. Brislin, R.W., 1970. Back-translation for crosscultural research. Journal of CrossCultural Psychology 1, 185–216. Canty-Mitchell, J., Zimet, G.D., 2000. Psychometric properties of the multidimensional scale of perceived social support in urban adolescents. American Journal of Community Psychology 28, 391–400. Census and Statistics Department, 2007. 2006 Population By-Census. Thematic Report – Ethnic Minorities. Census and Statistics Department, Hong Kong. Cheng, S.T., Chan, C.M., 2004. The multidimensional scale of perceived social support: dimensionality and age and gender differences in adolescents. Personality and Individual Differences 37, 1359–1369. Choi, J.B, Thomas, M., 2009. Predictive factors of acculturation attitudes and social support among Asian immigrants in the USA. International Journal of Social Welfare 18, 76–84.
Chou, K.L., 2000. Assessing Chinese adolescents’ social support: the multidimensional scale of perceived social support. Personality and Individual Differences 28, 299–307. Cortes, D.E, 1995. Variations in familism in two generations of Puerto Ricans. Hispanic Journal of Behavioral Sciences 17, 249–255. Crawford, J.R., Henry, J.D., 2003. The Depression Anxiety Stress Scale (DASS): normative data and latent structure in a large non-clinical sample. British Journal of Clinical Psychology 42, 111–131. Feona, 2000. Equality for All Races and Sexes. Far East Overseas Nepalese Association, Hong Kong. Floyd, F.J., Widaman, K.F., 1995. Factor analysis in the development and refinement of clinical assessment instruments. Psychological Assessment 62, 465–471. Gavazzi, S.M., 1994. Perceived social support from family and friends in a clinical sample of adolescents. Journal of Personality Assessment 62, 465–471. Goebert, D., Loue, S., 2009. Social support, mental health, minorities and acculturative stress. In: Loue, S., Sajatovic, M. (Eds.), Determinants of Minority Mental Health and Wellness. Gorsuch, R.L., 1997. Exploratory factor analysis: its role in item analysis. Journal of Personality Assessment 68, 532–560. Hirsh, B.J., 1981. Social networks and social support in community mental health. In: Gottlieb, B. (Ed.), Social Networks and Social Support. Sage, Beverly Hills, CA. House, J.S., Kahn, R.L., 1985. In: Cohen, S., Syme, S.L. (Eds.), Measures and Concepts of Social Support. Academic Press, New York. Kahn, R.L., Antonucci, T.C., 1980. Convoys over the life course: attachment, roles, and social support. Life-span Development and Behavior 3, 253–286. Kazarian, S.S., Mccabe, S.B., 1991. Dimensions of social support in the MSPSS: factorial structure, reliability, and theoretical implications. Journal of Community Psychology 19, 150–160. Lifrak, P.D, Mckay, J.R., Rostain, A., Alterman, A.I., O’ Brien, C.P., 1997. Relationship of perceived competencies, perceived social support, and gender to substance use in young adolescents. Journal of the American Academy of Child and Psychiatry 36, 933–940. Lovibond, S.H., Lovibond, P.F., 1995. Manual for the Depression Anxiety Stress Scales. Psychology Foundation, Sydney. Miranda, A.O., Matheny, K.B., 2000. Socio-psychological predictors of acculturative stress among Latino adults. Journal of Mental Health Counseling 22, 306–317. Ng, C.G., Siddiq, A.a.N., Aida, S.A., Zainal, N.Z., Koh, O.H., 2010. Validation of the Malay version of the Multidimensional scale of perceived social support (MSPSS-M) among a group of medical students in faculty of medicine, University Malaya. Asian Journal of Psychiatry 3, 3–6. Norbeck, J.S., Lindsey, A.M., Carrieri, V.L., 1981. The development of an instrument to measure social support. Nursing Research 30, 264–269. Norton, P.J., 1998. Depression Anxiety and Stress Scales (DASS-21): psychometric analysis across four racial groups. Anxiety, Stress, & Coping 20, 253–265. Patel, N., 1992. Psychological disturbance, social support and stressors: a community survey of immigrant Asian women and the indigenous population. Counseling Psychology Quarterly 5, 263–277. Rizwan, M., Aftab, S., 2009. Psychometric properties of the multidimensional scale of perceived social support in Pakistani young adults. Pakistan Journal of Psychology 40, 51–65. Stanley, M.A., Beck, J.G., Zebb, B.J., 1998. Psychometric properties of the MSPSS in older adults. Aging and Mental Health 2, 186–193. Stone-Feinstein, E., Ward, C., 1990. Loneliness and psychological adjustment of sojourners: new perspectives on culture shock. In: Keats, D.M., Munro, D., Mann, L. (Eds.), Heterogeneity in Cross-cultural Psychology Lisse. Swets & Zeitlinger, The Netherlands. Vaid, K.N., 1972. The Overseas Indian Community in Hong Kong, Hong Kong. Centre of Asian Studies, The University of Hong Kong. Zimet, G.D., Dahlem, N.W., Zimet, S.G., Farley, G.K., 1988. The multidimensional scale of perceived social support. Journal of Personality Assessment 52, 30–41. Zimet, G.D., Powell, S.S., Farley, G.K., Werkman, S., Berkoff, K.A., 1990. Psychometric properties of the multidimensional scale of perceived social support. Journal of Personality Assessment 55, 610–617.