Hostname: page-component-7c8c6479df-nwzlb Total loading time: 0 Render date: 2024-03-27T07:27:18.384Z Has data issue: false hasContentIssue false

Validation of the Bangla WHO-5 Well-being Index

Published online by Cambridge University Press:  19 July 2021

Md. Omar Faruk
Affiliation:
Centre for Disability in Development (CDD), Dhaka, Bangladesh
Farzana Alam
Affiliation:
Dhaka University, Dhaka, Bangladesh
Kamal Uddin Ahmed Chowdhury
Affiliation:
Dhaka University, Dhaka, Bangladesh
Tanjir Rashid Soron*
Affiliation:
Telepsychiatry Research and Innovation Network, Telepsychiatry Innovation Lab, Dhaka, Bangladesh University of Edinburgh, Edinburgh, UK
*
Author for correspondence: Tanjir Rashid Soron, E-mail: tanjirsoron@yahoo.com
Rights & Permissions [Opens in a new window]

Abstract

Background

Subjective wellbeing in terms of objective outcome can be useful to determine the level of progress in clinical practice as well as research studies in Bangladesh. Besides, cultural understanding of well-being for Bangladeshi population is also equally important to report. A valid Bangla version of the five-item WHO Well-being Index can be a suitable measure to achieve the purposes. Therefore, the present study aimed at validating the WHO-5 Well-being Index for general population in Bangladesh.

Methods

After following the standard procedures for translation, back-translation, and committee translation, the initial Bangla version of the scale was developed and pretested. Based on the feedback during pretesting, a slight modification was made and the final version was developed. This final version was administered to 269 participants of different socioeconomic backgrounds to find out the reliability and validity of the scale from March 2019 to May 2019. The data analysis was conducted using SPSS 24.

Results

The scale demonstrated acceptable internal consistency (α = 0.754) and test-retest reliability (r = 0.713), divergent validity (r = −0.443, p < 0.01 with the Bangla version of Perceived Stress Scale-10) and convergent validity (r = 0.542, p < 0.01 with the Bangla version of Warwick-Edinburgh Mental Well-Being Scale). The data also yielded one-factor structure for the scale in exploratory factor analysis explaining 38.68% of total variance. The factor-structure was further supported in the confirmatory factor analysis (χ2 = 295.852, χ2/df = 2.017, RMSEA = 0.062, CFI = 0.986, TLI = 0.964, and SRMR = 0.0255).

Conclusion

The findings suggested the Bangla version of the WHO-5 Well-being Index is a psychometrically valid and reliable tool for general adult population in Bangladeshi when it comes to measuring subjective well-being both in clinical practice and research studies.

Type
Original Research Paper
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
Copyright © The Author(s), 2021. Published by Cambridge University Press

Introduction

The concept of well-being has different connotations for different individuals, groups, and cultures (World Health Organization, 2017) making it difficult to reach a universally accepted definition of well-being. However, in general, well-being refers to the cognitive, emotional, and behavioral responses at a personal level. Psychological well-being can thus be interpreted in the sociocultural context of an individual and be seen in the form of a spectrum (World Health Organization, 2017). Such different connotations and the spectrum operating for an individual appear to be very subjective. Therefore, efforts to define well-being have led to a number of perspectives and measures. For instance, in a multi-disciplinary review, Dodge et al. (Reference Dodge, Daly, Huyton and Sanders2012) argued that many attempts at expressing the nature of well-being have focused purely on dimensions of well-being, rather than on definition. Similarly, after reviewing 75 scientific articles with more than 100 different scales or questionnaires, Gill and Feinstein (Reference Gill and Feinstein1994) demonstrated that a clinimetric definition of subjective well-being was lacking. Thus, Gill and Feinstein (Reference Gill and Feinstein1994) advocated for the development of short global rating scales of subjective well-being reflecting a single dimension with high clinical face validity. The WHO-5 Well-being Index can be one such well-qualified measure for measuring subjective well-being.

With the translation into more than 30 languages, the five-item WHO Well-being Index is one of the most widely used questionnaires to assess subjective psychological well-being (Topp et al., Reference Topp, Østergaard, Søndergaard and Bech2015). In their systematic review, Topp et al. (Reference Topp, Østergaard, Søndergaard and Bech2015) demonstrated four distinguished characteristics of the scale: the high clinimetric validity of the scale, the sensitivity of the scale in controlled clinical trials, potentiality to screen depression, and robustness of the scale across study fields. The WHO-5 Well-being Index has been reported as a pure generic instrument when it comes to measuring general well-being (Hall et al., Reference Hall, Krahn, Horner-Johnson and Lamb2011). The construct validity of the scale demonstrated a unidimensional scale with each item adding unique information about the level of well-being. The predictive validity of the scale has also been shown higher in many studies (Topp et al., Reference Topp, Østergaard, Søndergaard and Bech2015).

The WHO-5 Well-being Index has been used in a variety of areas along with research studies. The scale has been used in six controlled clinical trials as an outcome measure in the fields of psychiatry, oncology, endocrinology, otolaryngology, and depression to name a few. The use of WHO-5 Well-being Index in the diagnosis of depression has been demonstrated in numerous research studies and was shown as a highly sensitive measure with a sensitivity score of 0.93 and specificity of 0.83. Besides depression, the WHO-5 Well-being Index can be used in research studies to assess well-being over time or to compare well-being between groups. The scale has been used in many study fields with namely stress, neurology, suicidology, pain, health economics, and cardiology with endocrinology being the top field studied using the scale (Topp et al., Reference Topp, Østergaard, Søndergaard and Bech2015).

Validated in many countries across cultures as well as among targeted population, no such effort has yet been put in validating the WHO-5 Well-being Index for general population in Bangladesh. Therefore, the present study aimed to validate the five-item WHO Well-being Index into Bangla so that the index can be used for both clinical and research purposes.

There are few scales available to measure the well-being in Bangladesh. For example, designed for measuring children's well-being, Haque and Imran (Reference Haque and Imran2016) adapted the Stirling Children's Well-being Scale in Bangladeshi context putting further emphasis on establishing sensitivity and discriminant validity with the emergence of diagnostic features. In addition to that, Rahman and Imran (Reference Rahman and Imran2013) also adapted the Warwick-Edinburgh Mental Well-Being Scale (WEMWBS) (Tennant et al., Reference Tennant, Hiller, Fishwick, Platt, Joseph, Weich, Parkinson, Secker and Stewart-Brown2007) into Bangla but with no assessment of factor-structure. Huque and Begum (Reference Huque and Begum2005) developed a scale for measuring psychological well-being which lacks the determination of norms. The length (66 items) of the scale may create fatigue among responders. Against this backdrop, the need for validating the WHO-5 Well-being Index among general adult population with a view to determining the subjective feeling would be of paramount importance. The index is likely to contribute to the clinical practice as well as research studies in Bangladesh.

Methods

Study participants

Analysis of factor structure was one of the aims of the study. The rule of thumb for factor analysis considers a sample size of 200 as fair and 50 participants per factor is adequate (Wilson Van Voorhis and Morgan, Reference Wilson Van Voorhis and Morgan2007). The present study recruited 270 participants through purposive sampling in Dhaka city. Two hundred and sixty-nine participants (43.9% female and 55.8% male) aged 18–55 years (M = 24.66, s.d. = 7.45) were included in the study while one participant was excluded due to the unwillingness for using the responses in the study. Students as participants outnumbered (68.8%) the remaining participants that included businessperson, job holder, housewife, and unemployed participants (collectively 31.2%). Participants from all religious backgrounds comprised the main sample of the study. The demographic properties of 269 participants are presented in Table 1. Another 70 participants were also recruited for assessing test-retest reliability.

Table 1. Demographic properties of participants

Study procedure

Validating the Bangla version of the index was approved by the authority of the five-item WHO Well-being Index. The translation and back-translation procedures were carried out in accordance with the steps suggested by Gjersing et al. (Reference Gjersing, Caplehorn and Clausen2010). The original scale was given to five professionals including two psychiatrists, a language expert, and two clinical psychologists to translate it into Bangla language. An expert panel of five members comprising language experts and mental health professionals reviewed the translation process and checked their suitability. The expert panel reviewed the items in terms of their understandability and clarity for general people in Bangladesh. The reviewed items were then synthesized in a new version with all the issues addressed. The synthesized Bangla version was then given to another five independent professionals (two clinical psychologists, two language experts, and a psychiatrist) to blindly back-translate the items in the original language. The professionals involved in the back-translation were neither aware nor notified of the concepts under study. Three professionals including a language expert formed another expert committee to check the back-translation. The items were checked in terms of their semantic clarity, idiomatic, and conceptual equivalence. Two words were changed based on the consensus of the expert committee to make them more relevant for general people in Bangladeshi. The draft version of the scale was pretested on a sample of 35 participants. The pretesting aimed at the understanding of the questions being asked, whether the questions conveyed similar meaning to the participants and whether the participants were able to answer each question. The pretesting resulted in no further revision and the scale was, therefore, finalized. The final version of the scale was sent to the original author via email to check the accuracy of the resultant version. The scale was finalized for data collection with no objection from the author. The data were collected in Dhaka city by three psychology graduates trained by the second author prior to the data collection.

Ethics

The research ethics committee of the Department of Clinical Psychology, University of Dhaka, Bangladesh approved the study (project ID #MS190602, approved on 10 July 2019). The ethical standards suggested for human participants were maintained throughout the research. Participants were provided with an informed consent form with their participation as voluntary. No monetary compensation for the participation was provided.

Assessments

The following scales were used as instruments for the validation of the scale.

Perceived Stress Scale (PSS)

Originally developed by Cohen et al. (Reference Cohen, Kamarck and Mermelstein1983), the PSS 10 was used for its superior psychometric properties over the two other versions (PSS 14 and PSS 4). Moreover, PSS-10 demonstrated good internal consistency and test-retest reliability. Besides, it has been tested with diverse tools to determine the construct validity and resulted in moderate to strong correlation establishing it as a robust tool for measuring perceived stress (Mozumder, Reference Mozumder2017). The Bangla translated version of PSS-10 was used for the present study. No published data on the reliability and validity were found for the Bangla version of the PSS-10. The Bangla version is available at Sheldon Cohen's Laboratory for the study of stress (Mozumder, Reference Mozumder2017). The 10-item five-point Likert-type scale measures the degree to which an individual appraises his or her life as stressful. The scale combined both positive (4, 5, 7, 8) and negative items (1, 2, 3, 6, 9, 10). In case of negatively stated items, 0 signifies ‘never’, 1 ‘almost never’, 2 ‘sometimes’, 3 ‘fairly often’, and 4 ‘very often’ while the reverse can be seen in case of positive items.

Bangla version of Warwick-Edinburgh Mental Well-Being Scale (WEMWBS)

The WEMWBS was developed and validated in the UK by Tennant et al. (Reference Tennant, Hiller, Fishwick, Platt, Joseph, Weich, Parkinson, Secker and Stewart-Brown2007), and in 2013, the Bangla version of WEMWBS (Rahman and Imran, Reference Rahman and Imran2013) was validated. The 14 item five-point Likert-type scale measures mental well-being focusing entirely on positive aspects of mental health (Tennant et al., Reference Tennant, Hiller, Fishwick, Platt, Joseph, Weich, Parkinson, Secker and Stewart-Brown2007). The scale is scored by summing responses to each item answered from ‘none of the time’ up to ‘all of the time’. The higher the score the better the mental well-being. The Bangla version of the scale revealed a score of Cronbach's α 0.77, split-half reliability 0.87, and test-retest reliability 0.72. Convergent validity was found to be −0.53 (with regards to the General Health Questionnaire-12) (Rahman and Imran, Reference Rahman and Imran2013).

Statistical methods

The demographic data were analyzed by descriptive statistics. Internal consistency reliability was assessed using Cronbach's α suggested by Nunnally (Reference Nunnally1967, Reference Nunnally1978). Item analysis, test-retest reliability, divergent, and criterion validity were analyzed using Pearson correlations. Exploratory factor analysis was carried out with the maximum likelihood method. Multiple indices such as χ2, ratio of χ2 to df (χ2/df), root mean square error of approximation (RMSEA), and comparative fit index (CFI) were used to assess the adequacy of the model fit. The criteria for model fit were χ2 with p ⩾ 0.01, χ2/df ⩽ 2, RMSEA ⩽ 0.06, CFI ⩾ 0.95, SRMR ⩽ 0.08 (Mozumder, Reference Mozumder2017). The relationship between the scores of one item and the scores of all other items was examined by inter-item correlation. Most of the researchers considered the average inter-item correlation should be between 0.20 and 0.40 suggesting homogeneity of all items in relation to the same content domain (Piedmont, Reference Piedmont and Michalos2014). In order to analyze the data, SPSS 24 (IBM Corp, 2016) and AMOS 18 (Arbuckle, Reference Arbuckle2009) were used.

Results

The demographic properties of the participants are presented in Table 1.

The results included item analysis, reliability and validity assessment, and analysis of factor structure.

Item analysis

Item analysis of the scale revealed a correlation ranging from r = 0.445 to r = 0.561, all significant at p < 0.01 indicating good internal consistency with very good items (Ebel and Fresbie, Reference Ebel and Fresbie1991, p. 232). Items no. 5 and 2 demonstrated a lower inter-item correlation compared to others on the scale. However, the correlation between the items falls within the suggested range (Piedmont, Reference Piedmont and Michalos2014). Cristobal et al. (Reference Cristobal, Flavián and Guinalíu2007) suggested that corrected item-total correlation values lower than 0.30 are not acceptable. The corrected item-total correlation of all items in the current scale demonstrated a value of above 0.30.

The inter-item correlation and item-total statistics are presented in Table 2.

Table 2. Inter-item correlation and item-total statistics of the Bangla version of the WHO-5 well-being index

Convergent validity

Assessment of convergent validity was carried out using the Bangla validated version of WEMWBS (Rahman and Imran, Reference Rahman and Imran2013) and the Bangla version of the WHO-5 Well-being Index. The correlation between the two scales was found to be r = 0.542 (p < 0.01). Evidence suggests that a reliability coefficient ranging from 0.31 to 0.60 is adequate (Post, Reference Post2016). Convergent validity for the present scale was found adequate.

Divergent validity

Divergent validity was assessed correlating the Bangla version of the WHO-5 Well-being Index with the Bangla version of PSS (Mozumder, Reference Mozumder2017) as the study showed that stress can adversely impact subjective well-being (Ritchie et al., Reference Ritchie, Sedikides, Wildschut, Arndt and Gidron2011). The correlation between the two scales was r = −0.443 (p < 0.01) demonstrating a low negative correlation (Mukaka, Reference Mukaka2012).

Internal consistency reliability

The Bangla five-item WHO Well-being Index yielded a score of Cronbach's α 0.754 demonstrating a moderate value (Shrout, Reference Shrout1998).

Test-retest reliability

With a gap of 2 weeks, the scale was administered twice on a group comprising 70 participants. The test-retest reliability for the Bangla five-item WHO Well-being Index was r = 0.713 (p < 0.01) suggesting an acceptable result (Benson and Clark, Reference Benson and Clark1982; Opacich, Reference Opacich, Christiansen and Baum1991).

Exploratory factor analysis

Exploratory factor analysis with maximum likelihood was performed. No rotation method was used as the scale yielded a single factor. The suitability of factor analysis was determined by the score of Kaiser-Meyer-Olkin (0.775) and Bartlett's test of sphericity (χ2 = 293.093, p < 0.001). With an Eigenvalue of 1, the exploratory factor analysis yielded one factor explaining 38.68% of the total variance. The KMO measure of sampling adequacy was found to be acceptable comparing with the rule of thumb (Spicer, Reference Spicer2005).

All items had a factor loading >0.60 except item no. 5 (factor loading 0.500). Elimination of factor loadings with a value <0.32 is generally recommended (Tabachnick and Fidell, Reference Tabachnick and Fidell2014). The scale illustrated factor loadings for each item above the recommended values.

The factor structure with loading score is presented in Table 3.

Table 3. Factor structure of the Bangla version of the WHO-5 Well-being Index

Anti-image correlations, communalities, and correlation matrices were also investigated. Anti-image correlations refer to the reflection of the pairwise correlation remaining after partialing out the effects of other variables. It is suggested that the diagonals of the anti-image correlation matrix should be over 0.5 (Hauben et al., Reference Hauben, Hung and Hsieh2017). The anti-image correlation matrix ranges from 0.762 to 0.779 (see Table 3). Removal of an item with a communality score <0.2 is suggested (Child, Reference Child2006). Except for item no. 5, the remaining items demonstrated acceptable value (see Table 3). The correlation matrix demonstrated a low correlation between items no. 2 and 5 with a value of 0.267. However, the value was found to be acceptable as the inter-item correlation should be between 0.20 and 0.40 (Piedmont, Reference Piedmont and Michalos2014).

Anti-image correlation and communalities are presented in Table 4.

Table 4. Anti-image matrices and communalities of the Bangla version of the WHO-5 Well-being Index

a Measures of sampling adequacy (MSA)

Confirmatory factor analysis

To test the goodness of fit for the one-factor structure, AMOS 18 (Arbuckle, Reference Arbuckle2009) was used to run the confirmatory factor analysis (CFA). It is recommended that χ2, RMSEA, CFI, and SRMR are the indices that at a minimum should be reported when testing the goodness of fit (Kline, Reference Kline2008). However, a number of fit indices were taken into consideration for the present study such as χ2, χ2/df, RMSEA, CFI, TLI, and SRMR as a good fit of the model. The χ2, χ2/df, RMSEA, CFI, TLI, and SRMR were found to be 295.852 (p > 0.01), 2.017, 0.062, 0.986, 0.964, and 0.0255, respectively. RMSEA and χ2/df were found to be slightly higher than the values considered (Mozumder, Reference Mozumder2017). Browne and Cudeck (Reference Browne, Cudeck, Bollen and Long1993) suggested an RMSEA value of 0.05 or less indicates a close fit. However, this value cannot be considered infallible due to its subjective judgment (Arbuckle, Reference Arbuckle2011) and often produces inappropriate decisions (Hu and Bentler, Reference Hu and Bentler1999). Therefore, an RMSEA value in the range of 0.05–0.08 has been considered a fair fit (Hu and Bentler, Reference Hu and Bentler1999; Hair et al., Reference Hair, Black, Babin and Anderson2010; Awang, Reference Awang2012). A value of χ2/df ranging from 2 to 1 or 3 to 1 suggests an acceptable fit (Carmines and McIver, Reference Carmines, McIver, Bohrnstedt and Borgatta1981, p. 80) while others have recommended using ratios lower than 2 or as high as 5 when it comes to indicating a reasonable fit (Marsh and Hocevar, Reference Marsh and Hocevar1985). Evidence suggests that a value of CFI above 0.87 indicates a marginal fit (Dagnall et al., Reference Dagnall, Denovan, Parker, Drinkwater and Walsh2018). A CFI >0.95 indicates a relatively good model-data fit (Xia and Yang, Reference Xia and Yang2018). TLI was also taken into consideration that produced a value of 0.964 indicating a good model fit (Xia and Yang, Reference Xia and Yang2018).

The one-factor initial analysis was found suggesting a correlation between error terms (items no. 1 and 5). Putting the error terms into the model, a better fit was found. The findings are presented in Table 5 and Fig. 1.

Fig. 1. The one-factor structure of the Bangla version of the WHO-5 Well-being Index.

Table 5. Goodness of fit indices for one-factor model of the Bangla version of the WHO-5 Well-being Index

Discussion

The concept of well-being is subject to change depending on a given cultural context. For example, personality patterns, life circumstances, and other cultural variables may shape the meaning of well-being (Diener et al., Reference Diener, Oishi and Lucas2003; Tov and Diener, Reference Tov and Diener2007). Therefore, any psychometric tool measuring well-being should be in relation to the particular context. Hence the present study was taken into consideration. Suggested measures were followed during the translation processes to ensure cultural representation.

The scale was administered on a sample of 269 participants representing the general population. The scale demonstrated good internal consistency (Ebel and Fresbie, Reference Ebel and Fresbie1991, p. 232) for all items with scores ranging from 0.445 to 0.561. The inter-item correlation and the item-total statistics suggested further consistency of the scale.

Exploratory factor analysis (EFA) of the items yielded a one-factor structure for the Bangla version of the WHO-5 Well-being Index suggesting a consistent finding with the previous studies (Awata et al., Reference Awata, Bech, Yoshida, Hirai, Suzuki, Yamashita, Ohara, Hinokio, Matsuoka and Oka2007; De Wit et al., Reference De Wit, Pouwer, Gemke, Delemarre-van de Waal and Snoek2007; De Souza and Hidalgo, Reference De Souza and Hidalgo2011; Hajos et al., Reference Hajos, Pouwer, Skovlund, Den Oudsten, Geelhoed-Duijvestijn, Tack and Snoek2013; Mortazavi et al., Reference Mortazavi, Mousavi, Chaman and Khosravi2015; Chongwo et al., Reference Chongwo, Ssewanyana, Nasambu, Mwangala, Mwangi, Nyongesa, Newton and Abubakar2018; Dadfar et al., Reference Dadfar, Momeni, Asgharnejad, Nemati and Ghazie pour Abarghouie2018). Of the five items, the first four items showed a loading value above 0.60 while the remaining items showed a value of 0.500. Anti-image matrices and communalities were found to be acceptable (see Table 4).

CFA was used to test further the one-factor model obtained in the EFA. The study has shown that cross-cultural validation involving different data sets results in a lack of correspondence between EFA and CFA (Van Prooijen and van der Kloot, Reference Van Prooijen and van der Kloot2001). To rule out the problem, CFA was carried out on the same data set. Error items (items 1 and 5) were correlated in the same factor (see Fig. 1), as correlating error terms within factors are found to be a common phenomenon (Gerbing and Anderson, Reference Gerbing and Anderson1984). The resulting model demonstrated a good fit when multiple indices (χ2, χ2/df, RMSEA, CFI, TLI, and SRMR) were considered to avoid inappropriate findings (Lai and Green, Reference Lai and Green2016). The model fit suggested considerable support for the Bangla version of the WHO-5 Well-being Index.

The scale displayed a value of Cronbach's α 0.754 that falls within the ranges (0.70–0.95) recommended in general (Shrout, Reference Shrout1998; Tavakol and Dennick, Reference Tavakol and Dennick2011). The scale was administered twice on a group of 70 participants demonstrating an acceptable test-retest reliability score of 0.713 (Benson and Clark, Reference Benson and Clark1982; Opacich, Reference Opacich, Christiansen and Baum1991). Analysis of convergent validity revealed an adequate coefficient value (r = 0.542) (Post, Reference Post2016) when the Bangla version of the index was correlated with the Bangla version of WEMWBS (Rahman and Imran, Reference Rahman and Imran2013). A moderate correlation coefficient (r = −0.443) (Mukaka, Reference Mukaka2012) value between the index and the Bangla version of the PSS (Mozumder, Reference Mozumder2017) ensured the divergent validity.

The scale is the first of its kind to assess factor structures of well-being measures in Bangladesh. The psychometric measures validated or developed in Bangladesh did not assess the factor structure of the items or targeted only for children. Besides, the scale developed to measure psychological well-being contained a large number of items (Huque and Begum, Reference Huque and Begum2005). Studies demonstrated that psychometric measures short or medium in length can be useful in terms of performance compared to longer scales (Kearns et al., Reference Kearns, Cruickshank, McGuigan, Riley, Shaw and Snaith1982; Zigmond and Snaith, Reference Zigmond and Snaith1983; El-Rufie and Absood, Reference El-Rufie and Absood1995). The Bangla version of the WHO-5 Well-being Index can be a suitable tool to minimize the shortcomings mentioned above. The scale has been used as an outcome indicator in many research studies such as suicidology, alcohol abuse, diabetes, stroke, cancer, sleep problems, personality disorder, grief, and quality of life across the world (Topp et al., Reference Topp, Østergaard, Søndergaard and Bech2015). The Bangla version of the scale can also be used to assess the level of well-being in both clinical as well as research studies.

The study acknowledges a few limitations. As the study was conducted based on the Clinical Psychotherapy Unit of the University of Dhaka, an abundance of student participants was observed that skewed from the general adult population alongside other attributes. However, as the students came here from different backgrounds and every district of the country, this sample may be considered as a good representation of the country and Bangladeshi culture. Moreover, the scale used very simple and culturally neutral language, so it might have less influence on education and culture. The level of understanding in the Bangladeshi population of different backgrounds and ages was assessed during the pretesting. Determining the cut-off score in order for enabling the index as a screening tool was evident in a number of studies (Topp et al., Reference Topp, Østergaard, Søndergaard and Bech2015). However, as the present study attempted to validate the index among the general population without a clinical diagnosis, the cut-off score could not be determined marking it another limitation of the study. With a large sample size, it is also suggested to determine the diagnostic properties of the scale in future studies.

Despite several advantages such as less time-intensive, easy, and cheaper, non-probability sampling (purposive sampling for the study) can be subjected to selection bias (Forster, Reference Forster2001; Galloway, Reference Galloway2005). Furthermore, self-reporting scales can produce inaccurate results led by social desirability bias (Kunda and Spencer, Reference Kunda and Spencer2003). Therefore, employing probability sampling with measures that reduce biases of all forms is recommended in future studies.

Conclusion

The Bangla version of the WHO-5 Well-being Index demonstrated reliable and valid psychometric properties in the Bangladesh context. It takes about 3–4 minutes to administer. This tool can help in tapping into the subjective well-being of the general population both in clinical as well as research studies in Bangladesh.

Acknowledgements

The authors are grateful to Muhammad Kamruzzaman Mozumder, Ph.D. for his generous help in the confirmatory factor analysis.

Author contributions

MOF and KUAC were responsible for the conceptualization and design of the study. MOF and FA contributed to the data collection training including data collection. MOF and FA were responsible for statistical data analysis and interpretation of the data. MOF drafted the manuscript which was reviewed and edited by all authors including TRS. MOF and FA had access to all the data and take responsibility for the integrity of the data and accuracy of the data analysis. All authors read and approve the final manuscript.

Financial support

The study received no financial grant.

Conflict of interest

None.

Ethical standards

The ethical review committee at the Department of Clinical Psychology, University of Dhaka, Bangladesh approved the study in 2019. The ethical standards outlined in the Helsinki declaration for human participants were maintained throughout the research. Participants were provided with an informed consent form with their participation as voluntary. No monetary compensation for the participation was provided.

References

Arbuckle, JL (2009) Amos 18, 18th Edn. Crawfordville, Florida: Amos Development Corporation.Google Scholar
Arbuckle, JL (2011) Amos 20.0 User's Guide. Chicago: IBM SPSS.Google Scholar
Awang, ZH (2012) A Handbook on SEM: Structural Equation Modeling, 4th Edn. Kualalumpur: Centre For Graduate Studies, University Teknologi MARA Kelantan.Google Scholar
Awata, S, Bech, P, Yoshida, S, Hirai, M, Suzuki, S, Yamashita, M, Ohara, A, Hinokio, Y, Matsuoka, H and Oka, Y (2007) Reliability and validity of the Japanese version of the World Health Organization-Five Well-Being Index in the context of detecting depression in diabetic patients. Psychiatry and Clinical Neurosciences 61, 112119.CrossRefGoogle ScholarPubMed
Benson, J and Clark, F (1982) A guide for instrument development and validation. The American Journal of Occupational Therapy 36, 789800.CrossRefGoogle ScholarPubMed
Browne, MW and Cudeck, R (1993) Alternative ways of assessing model fit. In Bollen, KA and Long, JS (eds.), Testing Structural Equation Models. Newbury Park, CA: Sage, pp. 136162.Google Scholar
Carmines, EG and McIver, JP (1981) Analyzing models with unobserved variables: analysis of covariance structures. In Bohrnstedt, GW and Borgatta, EF (eds.), Social Measurement: Current Issues. Beverly Hills: Sage Publications, Inc, pp. 65115.Google Scholar
Child, D (2006) The Essentials of Factor Analysis, 3rd Edn. New York: Continuum.Google Scholar
Chongwo, E, Ssewanyana, D, Nasambu, C, Mwangala, PN, Mwangi, PM, Nyongesa, MK, Newton, CR and Abubakar, A (2018) Validation of a Swahili version of the World Health Organization 5-item well-being index among adults living with HIV and epilepsy in rural coastal Kenya. Global Health Research and Policy 3, 17. doi: 10.1186/s41256-018-0081-z.CrossRefGoogle ScholarPubMed
Cohen, S, Kamarck, T and Mermelstein, R (1983) A global measure of perceived stress. Journal of Health and Social Behavior 24, 385.CrossRefGoogle ScholarPubMed
Cristobal, E, Flavián, C and Guinalíu, M (2007) Perceived e-service quality (PeSQ). Managing Service Quality: An International Journal 17, 317340.CrossRefGoogle Scholar
Dadfar, M, Momeni, SN, Asgharnejad, FAA, Nemati, M and Ghazie pour Abarghouie, F (2018) Reliability, validity, and factorial structure of the World Health Organization-5 Well-Being Index (WHO-5) in Iranian psychiatric outpatients. Trends in Psychiatry and Psychotherapy 40, 7984.CrossRefGoogle Scholar
Dagnall, N, Denovan, A, Parker, A, Drinkwater, K and Walsh, RS (2018) Confirmatory factor analysis of the inventory of personality organization-reality testing subscale. Frontiers in Psychology 9, 1116. doi: 10.3389/fpsyg.2018.01116.CrossRefGoogle ScholarPubMed
De Souza, CM and Hidalgo, MPL (2011) World Health Organization 5-item well-being index: validation of the Brazilian Portuguese version. European Archives of Psychiatry and Clinical Neuroscience 262, 239244.CrossRefGoogle ScholarPubMed
De Wit, M, Pouwer, F, Gemke, RJBJ, Delemarre-van de Waal, HA and Snoek, FJ (2007) Validation of the WHO-5 well-being index in adolescents with type 1 diabetes. Diabetes Care 30, 20032006.CrossRefGoogle ScholarPubMed
Diener, E, Oishi, S and Lucas, RE (2003) Personality, culture, and subjective well-being: emotional and cognitive evaluations of life. Annual Review of Psychology 54, 403425.CrossRefGoogle ScholarPubMed
Dodge, R, Daly, A, Huyton, J and Sanders, L (2012) The challenge of defining wellbeing. International Journal of Wellbeing 2, 222235.CrossRefGoogle Scholar
Ebel, RL and Fresbie, DA (1991) Essentials of Educational Measurement, 5th Edn. New Jersey: Prentice-Hall.Google Scholar
El-Rufie, OE and Absood, GH (1995) Retesting the validity of the Arabic version of the hospital anxiety and depression (HAD) scale in primary health care. Social Psychiatry and Psychiatric Epidemiology 30, 2631.CrossRefGoogle Scholar
Forster, JJ (2001) Sample surveys: nonprobability sampling. In, Smelser, N.J. and Baltes, pp. 1346713470. doi: 10.1016/b0-08-043076-7/00499-xGoogle Scholar
Galloway, A (2005) Non-probability sampling. Encyclopedia of Social Measurement, 859864. doi: 10.1016/b0-12-369398-5/00382-0CrossRefGoogle Scholar
Gerbing, D and Anderson, J (1984) On the meaning of within-factor correlated measurement errors. Journal of Consumer Research 11, 572580.CrossRefGoogle Scholar
Gill, TM and Feinstein, AR (1994) A critical appraisal of the quality of quality-of-life measurements. JAMA 272, 619626.CrossRefGoogle ScholarPubMed
Gjersing, L, Caplehorn, JR and Clausen, T (2010) Cross-cultural adaptation of research instruments: language, setting, time and statistical considerations. BMC Medical Research Methodology 10. doi:10.1186/1471-2288-10-13.CrossRefGoogle ScholarPubMed
Hair, J, Black, W, Babin, B and Anderson, RE (2010) Multivariate Data Analysis. New Jersey: Pearson Prentice Hall.Google Scholar
Hajos, TRS, Pouwer, F, Skovlund, SE, Den Oudsten, BL, Geelhoed-Duijvestijn, PHLM, Tack, CJ and Snoek, FJ (2013) Psychometric and screening properties of the WHO-5 well-being index in adult outpatients with type 1 or type 2 diabetes mellitus. Diabetic Medicine 30, 6369.CrossRefGoogle ScholarPubMed
Hall, T, Krahn, GL, Horner-Johnson, W, Lamb, G and Rehabilitation Research and Training Center Expert Panel on Health Measurement (2011) Examining functional content in widely used health-related quality of life scales. Rehabilitation Psychology 56, 9499.CrossRefGoogle ScholarPubMed
Haque, M and Imran, MA (2016) Adaptation of Stirling children's well-being scale (SCWBS) in Bangladesh context. The Dhaka University Journal of Biological Sciences 25, 161167.CrossRefGoogle Scholar
Hauben, M, Hung, E and Hsieh, WY (2017) An exploratory factor analysis of the spontaneous reporting of severe cutaneous adverse reactions. Therapeutic Advances in Drug Safety 8, 416.CrossRefGoogle ScholarPubMed
Hu, L and Bentler, PM (1999) Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Structural Equation Modeling 6, 155.CrossRefGoogle Scholar
Huque, P and Begum, HA (2005) Development of a scale for measuring psychological well-being for use in Bangladesh. Bangladesh Psychological Studies 15, 6374.Google Scholar
IBM Corp (2016) IBM SPSS Statistics for Windows. Armonk, NY: IBM Corp.Google Scholar
Kearns, NP, Cruickshank, CA, McGuigan, KJ, Riley, SA, Shaw, SP and Snaith, RP (1982) A comparison of depression rating scales. British Journal of Psychiatry 141, 4549.CrossRefGoogle ScholarPubMed
Kline, RB (2008) Principles and Practice of Structural Equation Modeling, 2nd Edn. New York: Guilford Press.Google Scholar
Kunda, Z and Spencer, SJ (2003) When do stereotypes come to mind and when do they color judgment? A goal-based theoretical framework for stereotype activation and application. Psychological Bulletin 129, 522544.CrossRefGoogle ScholarPubMed
Lai, K and Green, SB (2016) The problem with having two watches: assessment of fit when RMSEA and CFI disagree. Multivariate Behavioral Research 51, 220239.CrossRefGoogle ScholarPubMed
Marsh, HW and Hocevar, D (1985) Application of confirmatory factor analysis to the study of self-concept: first- and higher order factor models and their invariance across groups. Psychological Bulletin 97, 562582.CrossRefGoogle Scholar
Mortazavi, F, Mousavi, SA, Chaman, R and Khosravi, A (2015) Validation of the World Health Organization-5 well-being index; assessment of maternal well-being and its associated factors. Turkish Journal of Psychiatry 26, 4855.Google ScholarPubMed
Mozumder, MK (2017) Validation of Bengali perceived stress scale among LGBT population. BMC Psychiatry 17, 314. doi: 10.1186/s12888-017-1482-0.CrossRefGoogle ScholarPubMed
Mukaka, MM (2012) Statistics corner: a guide to appropriate use of correlation coefficient in medical research. Malawi Medical Journal 24, 6971.Google ScholarPubMed
Nunnally, JC (1967) McGraw-Hill Series in Psychology. Psychometric Theory. New York: McGraw-Hill.Google Scholar
Nunnally, JC (1978) Psychometric Theory, 2nd Edn. New York: McGraw-Hill.Google Scholar
Opacich, KJ (1991) Assessment and informed decision-making. In Christiansen, C and Baum, C (eds.), Occupational Therapy-Overcoming Human Performance Deficits. Thorofare, NJ: Slack, pp. 256372.Google Scholar
Piedmont, RL (2014) Inter-item correlations. In Michalos, AC (ed.), Encyclopedia of Quality of Life and Well-Being Research. Dordrecht: Springer, pp. 33033304. doi: 10.1007/978-94-007-0753-5_1493.CrossRefGoogle Scholar
Post, MW (2016) What to do with ‘moderate’ reliability and validity coefficients? Archives of Physical Medicine and Rehabilitation 97, 10511052.CrossRefGoogle Scholar
Rahman, ST and Imran, MA (2013) Bangladeshi adaptation of Warwick-Edinburgh mental well-being scale. The Dhaka University Journal of Psychology 37, 4960.Google Scholar
Ritchie, TD, Sedikides, C, Wildschut, T, Arndt, J and Gidron, Y (2011) Self-concept clarity mediates the relation between stress and subjective well-being. Self and Identity 10, 493508.CrossRefGoogle Scholar
Shrout, PE (1998) Measurement reliability and agreement in psychiatry. Statistical Methods in Medical Research 7, 301317.CrossRefGoogle Scholar
Spicer, J (2005) Making Sense of Multivariate Data Analysis: An Intuitive Approach. Thousand Oaks, CA: Sage, pp. 181202.CrossRefGoogle Scholar
Tabachnick, BG and Fidell, LS (2014) Using Multivariate Statistics, 6th Edn. Harlow: Pearson.Google Scholar
Tavakol, M and Dennick, R (2011) Making sense of Cronbach's alpha. International Journal of Medical Education 2, 5355.CrossRefGoogle ScholarPubMed
Tennant, R, Hiller, L, Fishwick, R, Platt, S, Joseph, S, Weich, S, Parkinson, J, Secker, J and Stewart-Brown, S (2007) The Warwick-Edinburgh mental well-being scale (WEMWBS): development and UK validation. Health and Quality of Life Outcomes 5, 63.CrossRefGoogle ScholarPubMed
Topp, CW, Østergaard, SD, Søndergaard, S and Bech, P (2015) The WHO-5 well-being index: a systematic review of the literature. Psychotherapy and Psychosomatics 84, 167176.CrossRefGoogle ScholarPubMed
Tov, W and Diener, E (2007) Culture and subjective well-being. Handbook of cultural Psychology.Google Scholar
Van Prooijen, JW and van der Kloot, WA (2001) Confirmatory analysis of exploratively obtained factor structures. Educational and Psychological Measurement 61, 777792.CrossRefGoogle Scholar
Wilson Van Voorhis, CR and Morgan, BL (2007) Understanding power and rules of thumb for determining sample sizes. Tutorials in Quantitative Methods for Psychology 3, 4350.CrossRefGoogle Scholar
World Health Organization, Promotion of Mental Well-being (2017) Retrieved from http://www.searo.who.int/entity/mental_health/promotion-of-mental-well-being/en/.Google Scholar
Xia, Y and Yang, Y (2018) RMSEA, CFI, and TLI in structural equation modeling with ordered categorical data: the story they tell depends on the estimation methods. Behavior Research Methods 51, 409428. doi: 10.3758/s13428-018-1055-2.CrossRefGoogle Scholar
Zigmond, AS and Snaith, RP (1983) The hospital anxiety and depression scale. Acta Psychiatrica Scandinavica 67, 361370.CrossRefGoogle ScholarPubMed
Figure 0

Table 1. Demographic properties of participants

Figure 1

Table 2. Inter-item correlation and item-total statistics of the Bangla version of the WHO-5 well-being index

Figure 2

Table 3. Factor structure of the Bangla version of the WHO-5 Well-being Index

Figure 3

Table 4. Anti-image matrices and communalities of the Bangla version of the WHO-5 Well-being Index

Figure 4

Fig. 1. The one-factor structure of the Bangla version of the WHO-5 Well-being Index.

Figure 5

Table 5. Goodness of fit indices for one-factor model of the Bangla version of the WHO-5 Well-being Index