Sage Journals: Discover world-class research

Abstract

COVID-19 and the actions to curtail its spread have elevated the value of mental health as well as public health crises and pandemics. This calls for increased research in this area and the use of consistent and valid instruments to measure diverse aspects of mental health in different populations. This study presents preliminary psychometric properties (i.e., factor structure, internal consistency, convergent and discriminant validity) of the WHO-5 index as compared to other mental health instruments used in three countries (Botswana, Zimbabwe, and Malaysia). Data was collected from three countries during the first 5 months after the onset of COVID-19 in the Southern African and Southeast Asia regions: Botswana (N = 422; Mean Wellbeing = 9.8; SD = 5.4), Zimbabwe (N = 325; Mean = 9.4; SD = 5.3), and Malaysia (N = 425; Mean = 13; SD = 5.3). After data quality and scaling properties were evaluated, factor structures were assessed using principal component analysis and internal consistency of the extracted components were examined using Cronbach’s alpha (α). Construct validity was examined using Pearson’s correlations to establish both convergent validity and discriminant validity among the three mental health constructs (i.e., well-being, generalized anxiety, and loneliness). With Cronbach’s alpha of the total WHO-5 Wellbeing index of .86 (Botswana sample), α = .85 (Zimbabwean data) and α = .88 (Malaysian data), as well as (α > .7) for the selected demographic subgroups, the findings broadly suggest that WHO-5 is a unique, reliable, and valid instrument for measuring subjective well-being, and by extension mental health among diverse subgroups—in Botswana, Zimbabwe, Malaysia, and possibly, other similar settings.

Keywords

psychometrics evaluation wellbeing anxiety COVID-19 Botswana Zimbabwe Malaysia loneliness measurement

A wide variety of researchers (e.g., O’Neil et al., 2020; Peng et al., 2020; Qiu et al., 2020) and development partners such as the World Health Organization (WHO) and the International Labour Organization (ILO) have raised concerns that go beyond the widely reported physical health challenges. These include a wide variety of economic, social, mental, and psychosocial challenges emanating from the coronavirus disease (COVID-19) and the measures used to curtail its spread (e.g., national lockdowns, quarantines, isolations, and social distancing). Indeed, COVID-19 has demonstrated that physical health is not only influenced by biological factors, as the psychological, social, and environmental factors interact in multifaceted ways to influence physical health, and vice versa. The extant literature thus far suggests that COVID-19 is likely worsening people’s mental health resulting in short to long-term negative impacts for individuals, communities, and healthcare systems. Such concerns highlight the need to present well-researched personal and psychological resources to prevent and combat mental health challenges as well as valid instruments to measure such. The current study focuses on the measurement of a specific mental health construct (i.e., well-being).

Understanding Well-Being

Well-being is considered an accurate estimate of individuals’ ability to cope with daily stressors and to make positive contributions to society at large (World Health Organisation [WHO], 2013). It has also been discussed as a predictor of both mental health and overall health, including the quality of life (QOL) (Sischka et al., 2020) of an individual and society as a whole. It is therefore important to evaluate, promote and protect the well-being of people and societies because as the WHO suggests, there is no health without well-being. Establishing the definition of well-being as a construct has been challenging (Taylor, 2015) despite it being a well-studied concept (Dodge et al., 2012). There has been a debate about whether well-being is an objective or a subjective construct (La Placa et al., 2013; Taylor, 2015). The dissonance has resulted in two schools of thought concerning the source and predictors of well-being and consequently, two main conceptualizations of well-being: (1) objective well-being (OWB) and (2) subjective well-being (SWB). OWB is said to be determined by socioeconomic factors (Diener & Seligman, 2004; Kingdon & Knight, 2006), autonomy, and health factors (Taylor, 2015). One of the most prominent determinants of OWB is income—guided by the assertion that money secures a better QOL and ultimately, well-being (Diener & Seligman, 2004; Kingdon & Knight, 2006), especially in developing countries.

In contrast, supporters of SWB argue that the conditions surrounding one’s life cannot determine well-being on their own—why else would two people going through the same challenge respond differently (one positively and one negatively) if not for the difference in subjective well-being—inquired Diener et al. (1999). Similar to well-being in general, there are two main approaches to SWB (see Disabato et al., 2016). The first approach, called the hedonic approach, is indicated as positive psychological or affective responses to life and its challenges (Diener & Seligman, 2004; Taylor, 2015), resulting in life satisfaction (Taylor, 2015) and happiness (Cummins et al., 2009). The second approach is the eudaimonic approach—it emphasizes having a meaningful life and focuses on dimensions such as acceptance of self and personal growth (Ryff & Singer, 2008).

There are, however, debates about the unique differences between the two approaches (Disabato et al., 2016). Sischka et al. (2020) assert that well-being is an aspect of perceived QOL, thus supporting the subjective nature of well-being. Dodge et al. (2012) asserted that well-being is a multi-dimensional construct that is present when “individuals have the psychological, social and physical resources they need to meet a particular psychological, social, and/or physical challenge” (p. 230). Thus, combining both objective and subjective perspectives of well-being makes it important to distinguish well-being (OWB and SWB) from the other mental health constructs. These definitions support the definition of well-being held by the WHO (2013) which has been used in the current paper.

Correlates of Well-Being

Some have argued that well-being encompasses both negative and positive aspects of affect as one response to life events (Diener, 2006; La Placa et al., 2013). The former includes the presence of depression and anxiety (anxiety is an important construct in the current study), while the latter includes satisfaction, engagement, happiness, and contentment (Diener, 2006; Diener & Seligman, 2004). In line with the WHO (2013) making the association between well-being and mental health, the WHO-5 Well-Being Index has been used to screen for several psychological health challenges. For instance, some researchers have evaluated the psychometric soundness of WHO-5 to screen for depression. In a review of over 200 articles, Topp et al. (2015) evaluated the WHO-5 Well-Being Index as an effective screening tool for depression. Similar results were reported by Henkel et al. (2003). While we did not measure depression in the current study, we found it necessary to evaluate well-being against two other aspects of mental health (i.e., loneliness and generalized anxiety) as such conceptual distinctiveness from other mental health constructs can aid further research, policy, and practice.

Besides being used to measure depression, the WHO-5 has been adopted and used in other research areas related to cancer, stroke, personality disorders, and alcohol abuse (Sischka et al., 2020). Like the current study, the WHO-5 has recently been used to investigate subjective well-being in various studies during the current COVID-19 crisis (Eurofound, 2020). Social isolation is another important associate of well-being as it has been the most associated with the increase in the rates of loneliness in modern times (Jeste et al., 2020). Similarly, it has been projected that social isolation resulting from the lockdown measures implemented in countries to reduce the rates of COVID-19 infections, could be the leading cause of loneliness during the pandemic (Holmes et al., as cited in Groarke et al., 2020). All of these outcomes could be expected to have implications for people’s levels of SWB including those living in Botswana, Zimbabwe, and Malaysia since the governments of the three countries put lockdown measures in place during the early stages of COVID-19.

Measuring Well-Being

On the whole, measures of well-being are determined by whether the researchers are interested in OWB or SWB. OWB measures focus on collecting information on factors such as socioeconomic status and opportunities for advancement (Diener & Seligman, 2004; Kingdon & Knight, 2006; Taylor, 2015) as well as social relationships (Cummins et al., 2009). Contrastingly, measures of SWB are usually self-report measures investigating levels of positive affect and life satisfaction, and/or happiness (Cummins et al., 2009; Diener et al., 1999). As a result, measures of SWB are differentiated by how they operationalize SWB—as affect, happiness, quality of life, life satisfaction, or a combination of these. For example, Cummins et al. (2009) compare and contrast three measures of SWB that focus on life satisfaction. One is the single question, “How satisfied are you with your life as a whole?” (Andrews & Withey, as cited in Cummins et al., 2009). The others use the Satisfaction with Life Scale (Diener, Emmons, Larsen, & Griffin, as cited in Cummins et al., 2009) and the Personal Well-being Scale (International Well-being Group [IWBG], as cited in Cummins et al., 2009).

Comprising only five items, the WHO-5 is favored for its brief, simple, and non-intrusive design (Topp et al., 2015) making it one of the most utilized measures of subjective well-being (Sischka et al., 2020; Topp et al., 2015). WHO-5 was initially presented as part of the DepCare Project at a WHO meeting in 1998 (Topp et al., 2015). The WHO-5 items were extracted from the WHO-10 scale, which was based on a scale containing 28-items (Bech et al., as cited in Topp et al., 2015; Sischka et al., 2020). The said scale comprised items from the Zung scales measuring depression and distress among others as well as items from the Psychological General Well-being Scale and the General Health Questionnaire (Bech, as cited in Sischka et al., 2020; Topp et al., 2015). However, in line with the WHO’s positive view of SWB, all of the items in the WHO-5 are positively phrased (Topp et al., 2015).

Aims and Goals of the Study

The present study assessed the psychometric soundness of the widely used World Health Organization Well-Being Index (WHO-5) by analyzing data collected from Botswana, Zimbabwe, and Malaysia. Data came from a large study that used a battery of psychosocial measures administered to some of the global south populations amid the COVID-19 pandemic. Evaluating the psychometric soundness of measuring instruments such as WHO-5 is important; since the instruments used should be valid and reliable for researchers and consumers of research to have confidence in research data and findings (see Clark & Watson, 2016; Drost, 2011).

The current study is thus important because one challenge in measuring well-being has been in the conceptualization of the construct and this has possibly resulted in a lack of clarity regarding what is being measured (i.e., looking at the different aspects of well-being as well as the construct’s relationship with associated constructs). For example, the WHO-5 considers positive well-being as another term for mental health (Topp et al., 2015)—while other researchers speak of different aspects of well-being. Even before the onset of COVID-19, mental health and well-being had gained attention as predictors of health (WHO, 2005). Besides, according to a review of the WHO-5 (see Topp et al., 2015), the WHO-5 has high clinimetric validity, can be used as a quality outcome measure that balances treatment effects on the intended and unintended, is a sensitive and specific screening tool for depression, as well as being highly applicable to a wide variety of study fields. It is unclear from the literature whether well-being is equivalent to mental health or a separate indicator. Furthermore, a few studies have evaluated the psychometric soundness of the WHO-5 as an instrument to assess subjective well-being in a multi-national, cross-cultural setting and during the COVID-19 pandemic like this current study.

Method

An online cross-sectional convenience sampling method was used in the three countries. Similar to other studies attempting to measure mental processes and human behavior, this study was guided by a positivist view and the use of empirical-analytic approaches (see Smallbone & Quinton, 2004) as measurement instruments were used to observe human behavior for analysis (see Drost, 2011). Generally, this process may require either measurement development or adaptation of existing instruments and the establishment of reliability and validity of measuring instruments. In the present study, the pre-existing measures described below were used. Piedmont et al. (2000) identify the validity of individual test protocols as a great and legitimate concern—and the current authors concur that validity and reliability are important when considering measurement instruments. While reliability is concerned with the consistency of measures, validity assessments seek to ensure that instruments are indeed measuring what they were intended to measure (Drost, 2011). We attempted to test the construct validity of WHO-5 by examining the relationship between this scale with loneliness and anxiety scales. We expected WHO-5 to correlate significantly and negatively with these two scales. Having noted the wide use of WHO-5 in different studies, we still deemed it important to investigate both the reliability and validity of the scale in different populations—and during a pandemic like COVID-19.

Sampling and Data Collection

A cross-sectional study using an online survey design was used to collect data 2 weeks before COVID-19 was considered a major concern in the three countries and the 2 weeks before taking the survey. Zimbabwe, Botswana, and Malaysia had instituted a national lockdown during these 2 weeks. The study was part of an international parent study titled, “Personal and Family Coping with Covid-19 in the Global South” also administered to adults aged 18 years and above in several other countries. Because of national lockdown measures put in place by these countries, data were collected using a combination of convenient and snowballing sampling techniques. To gain access to the study, a link was sent using different digital media platforms to allow participants to complete the survey using an internet-based device, at their own time and in their own space. Furthermore, participants, who had completed the questionnaire were asked to distribute the link to others in their networks.

Instruments

The WHO-5 Well-Being Index

WHO-5 (WHO, 1998) was used to measure subjective well-being over 2 weeks. The scale has five items that required participants to indicate how they have felt in the past 2 weeks (Topp et al., 2015; WHO, 1998). In the current study, the original items were amended to: “I feel cheerful and in good spirits,”“I feel calm and relaxed,”“I feel active and vigorous,”“I wake up feeling fresh and rested,” and “My daily life is filled with things that interest me.” The scale uses a six-point Likert scale ranging from 0 (At no time) to 5 (All of the time). Other researchers have revealed good internal consistency. For example, α = .91 (Löwe et al., 2004) and α = .89 (Newnham et al., 2010). Additionally, the WHO-5 is a valid measure of subjective well-being in several samples (Newnham et al., 2010).

Two Scales Used to Test Construct Validity

The Three-Item Loneliness Scale

This scale was adapted from the UCLA Loneliness Scale (Russell et al., 1978). The scale is considered a relatively short scale to measure general loneliness that could progress empirical research on the phenomenon (Russell et al., 1978). The UCLA Loneliness Scale comprised 20 items and a four-item response scale ranging from “I often feel this way,” to “I never feel this way” with a high internal consistency (α = .96) and a moderate test-retest reliability of 0.73 (Russell et al., 1978). However, Hughes et al. (2004) felt that the UCLA Loneliness Scale was too long for use in a large survey thus they developed the Three-Item Loneliness Scale. It comprises the following items adapted from the UCLA Loneliness Scale “How often do you feel that you lack companionship?,”“How often do you feel left out?” and “How often do you feel isolated from others? (Hughes et al., 2004). The response items included: “Hardly ever” (scored 1), “Some of the time” (scored 2), and “Often” (scored 3), (Hughes et al., 2004), with higher total scores indicating a higher degree of loneliness or social isolation. The Three-Item Loneliness Scale is reliable with an alpha of .72 (Hughes et al., 2004) and .83 (Groarke et al., 2020).

The Generalized Anxiety Disorder Seven-Item (GAD-7) Scale

The scale was originally developed and used in primary care settings to assess generalized anxiety disorder (Spitzer et al., 2006). The scale has seven items measuring (1) feelings of anxiety, nervousness, or edginess, (2) ability to stop or control worrying, (3) worrying excessively about different things, (4) trouble relaxing, (5) being restless, (6) being easily irritated and annoyed, and (7) feeling afraid as if something awful might happen (Johnson et al., 2019). The scale is a reliable and valid measure of generalized anxiety disorder in both the psychiatric populations and general populations (Kertz et al., 2012; Löwe et al., 2008; Rutter & Brown, 2016). The scale has also shown excellent reliability with an alpha of .91 (Monteiro et al., 2022), .88 (Johnson et al., 2019), and .75 (Terrill et al., 2015). For example, it was found to have high internal consistencies of α = .89 and α = .91 by Löwe et al. (2008) and Kertz et al. (2012) respectively. However, it has been found to have only adequate internal consistency in some African countries. Barthel et al. (2014) reported an alpha of .69 in Côte d’Ivoire and of .67 in Ghana. Adjorlolo (2019) found the scale to have an alpha of .69 in a Ghanaian sample. The scale has also shown excellent reliability with an alpha of .91 (Monteiro et al., 2022) and Cronbach’s alpha of .88 (Omani-Samani et al., 2018).

Statistical Analyses

Statistical analyses were performed using IBM SPSS 25.0. Participants with incomplete cases were excluded from the different analyses. Firstly, to explore whether WHO-5, GAD-7, and the Three-Item Loneliness Scale (TILS) can measure different aspects of mental health as different constructs. Principal Component Analysis (PCA) was used to investigate the factor structure of the three scales. The scales were expected to manifest a multidimensional set of items (manifest variables) reflecting variance in the three dormant variables. Secondly, to explore the internal consistency of the three scales, a series of reliability analyses were carried out using the three data sets. More detailed reliability analyses were carried out for the WHO-5 Well-Being Index. Thirdly, to establish the discriminant and convergent validity of the scale, correlation analyses were carried out. Results of the three analyses (i.e., PCA, Cronbach alpha, and correlation analyses) are presented in the proceeding section.

Results

Demographic Characteristics of Respondents

The Botswana sample consisted of 422 participants, the majority of whom were aged between 18 and 24 years (57.8%), and were female (n = 296: 70.1%). Many of the participants identified as single (never married) (n = 268; 63.5%) and without children (69.7%) (See Table 1). Most of them fell within the low to lower-middle-income range (76.7%) and lived in an urban area (80.4%). Additionally, the sample comprised students (46.3%) while 37.4% reported that they were engaged in some form of formal employment.

Table 1.

Characteristics of Participants From Botswana, Zimbabwe, and Malaysia.

Total	Botswana		Zimbabwe		Malaysia
	N = 422		N = 332		N = 543
	n	%	n	%	N	%
Gender
Male	120	28.4	123	37.0	113	26.6
Female	296	70.1	205	61.7	310	72.9
Age (years)
18–24	242	57.8	124	37.3	136	32.0
25–34	55	13.1	87	26.2	102	24.0
35–44	59	14.1	70	21.1	67	15.8
45–54	28	6.7	31	9.3	47	11.1
55–64	27	6.4	16	4.8	62	14.6
65 above	8	1.9	4	1.2	9	2.1
Residence
Urban	336	80.4	310	93.7	353	83.1
Rural	67	16.0	18	5.4	55	12.9
Prefer not to say	15	3.6	3	0.9	17	4.0
Relationship status
Married	87	20.6	122	36.7	158	37.2
Widowed	2	0.5	9	2.7	5	1.2
Divorced/separated	11	2.6	14	4.2	6	1.4
Single but cohabiting	31	7.3	16	4.8	26	6.1
Single never married	268	63.5	154	46.4	215	50.6
Have multiple partners	4	0.9	2	0.6	0	0
Prefer not to say	19	4.5	15	4.5	15	34.5
Children
No children	294	69.7	186	56.2	278	65.4
One	36	8.5	40	12.1	27	6.4
Two	36	8.5	46	13.9	54	12.7
Three	33	7.8	35	10.6	43	10.1
Four	13	3.1	18	5.4	14	3.3
More than four	7	1.7	6	1.8	6	1.4
Prefer not to say	3	0.7	0	0	3	0.7
Education
I did not attend school	1	0.2	0	0	0	0
Primary	0	0	2	0.6	0	0
Secondary	111	26.4	48	14.5	27	6.4
Post-secondary school diploma/certificate/degree	107	25.4	72	21.8	104	24.5
Bachelor	120	28.5	124	37.5	143	33.6
Masters	42	10.0	70	21.1	91	21.41
Doctorate/PhD	32	7.6	10	3.0	57	13.4
Prefer not to say/don’t say	8	1.9	5	1.5
Employment
No employment	57	13.5	61	18.5	49	11.5
Agriculture/farming	2	0.5	7	2.1	2	0.5
Daily workers/small income job	13	3.1	8	2.4	8	1.9
Private employee	18	4.3	37	11.2	56	13.2
Employee	66	15.7	49	14.8	41	9.6
Government employee	42	10.0	42	12.7	102	24.0
Business owner	17	4.0	14	4.2	17	4.0
Student	195	46.3	99	30.0	130	30.6
Taking care of the household	1	0.2	10	3.0	9	2.1
Prefer not to say	10	2.4	3	0.9	11	2.6
Socioeconomic status
Low income	136	32.4	154	46.5	51	12.0
Lower middle income	186	44.3	131	39.6	204	48.0
Higher middle income	96	22.9	43	13.0	161	37.9
High income	2	0.5	3	0.9	9	2.1

Most of the participants in the Zimbabwe sample totaling 332 participants identified as female (n = 205: 61.7%) and were between the ages of 18 and 34 (63.6%). Many of them lived in an urban area (93.7%) and were of low to lower middle income (86.1). Most of them did not have children (56.2%) and they identified as single (never married) (46.4%) followed by married (36.7%). They were engaged in some form of employment (47.4%), students (30%), and unemployed (18.5%). Further details of the participants are listed in Table 1.

There were 543 Malaysian participants (male = 113: 26.6%, female = 310: 72.9%) who took part in the present study. More than half of the participants were in the age range from 18 to 34 years (56%). There were 56.7% identified as single and 37.2% of them were married and most of them did not have children (65.4%). In terms of income, 48% of the participants were in the lower to the middle-income range, followed by 37.9% in the high-middle-income bracket, 12% in the low-income range, and 2.1% of the participants were in the high-income range. Most of the participants lived in an urban area (83.1%). There were 53.2% of participants engaged in some form of formal employment, while 30.6% were students and 11.5% were unemployed (see Table 1).

Well-Being as a Separate Construct From Other Mental Health Constructs

Two separate PCAs without a restricted total number of factors were performed. As displayed in Table 2 below, both PCA yielded a three-factor solution with most of the items correlating at least 0.3 and with at least one other item, suggesting reasonable factorability (see Pallant, 2016).

Table 2.

Exploratory Factor Analysis of the WHO-5 and the Other Two Mental Health Scales Used in the Botswana, Zimbabwe, and Malaysia Population.

Scale items	Component—Botswana			Component—Zimbabwe			Component—Malaysia
Scale items	GAD-7 items	WHO-5 items	3ILS items	GAD-7 items	WHO-5 items	3ILS items	GAD-7 items	WHO-5 items	3ILS items
Not being able to stop or control worrying: The past 2 weeks (ANX2B 2)	0.827			0.857			0.830
Worrying too much about different things: The past 2 weeks (ANX2B 3)	0.826			0.854			0.824
Being so restless that it’s hard to sit still: The past 2 weeks (ANX2B 5)	0.811			0.852			0.822
Feeling afraid as if something awful might happen: The past 2 weeks (ANX2B 7)	0.804			0.848			0.809
Feeling nervous, anxious, or on edge: The past 2 weeks (ANX2B 1)	0.790			0.842			0.798
Trouble relaxing: The past 2 weeks (ANX2B 4)	0.788			0.837			0.763
Becoming easily annoyed or irritable: The past 2 weeks (ANX2B 6)	0.567			0.740			0.761
I feel cheerful and in good spirits: The past 2 weeks (WHOV1B 1)		0.855			0.865			0.842
I feel active and vigorous: The past 2 weeks (WHOV1B 3)		0.842			0.822			0.832
I feel calm and relaxed: The past 2 weeks (WHOV1B 2)		0.821			0.808			0.817
I wake up feeling fresh and rested: The past 2 weeks (WHOV1B 4)		0.725			0.728			0.815
My daily life is filled with things that interest me: The past 2 weeks (WHOV1B 5)		0.712			0.715			0.806
How often do you feel left out?: The past 2 weeks (LON2TB 2)			0.850			0.882			0.908
How often do you feel that you lack companionship: The past 2 weeks (LON2TB 1)			0.835			0.834			0.903
How often do you feel isolated from others?: The past 2 weeks (LON2TB 3)			0.826			0.828			0.841

Note. Extraction method: Principal Component Analysis. Rotation method: Oblimin with Kaiser Normalization. Rotation converged in four iterations. Factor loadings <0.2 are suppressed. WHO-5 = WHO Well-Being Index (WHO, 1998); GAD-7 = Generalized Anxiety Disorder Seven-Item Scale (Spitzer et al., 2006); 3ILS = Three-Item Loneliness Scale (Hughes et al., 2004).

After determining item correlations, to further verify that the data set was suitable for factor analysis, the Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) and Bartlett’s Test of Sphericity was assessed. Kaiser-Meyer-Olkin’s measure of sampling adequacy was 0.90 for the Botswana population, 0.91 for the Zimbabwean population, and 0.90 for the Malaysian population. These were considered sufficient as both numbers were above the recommended value of 0.6. Bartlett’s test of sphericity was also significant (χ² (105) = 3,190.40 p < .05) for the Botswana population and (χ² (105) = 2,758.44 p < .05) for the Zimbabwean population and (χ² (120) = 6,105.73 p < .05) for the Malaysian population. The aforementioned also supports the factorability of the correlation matrix in both data sets (see Bartlett, 1954; Kaiser, 1970, 1974).

In the Botswana sample, PCA revealed the presence of three components with eigenvalues of more than 1. The three components explained 42.26%, 13.60%, and 10.91% of the variance. Similar results were established in the Zimbabwean sample with the three components explaining 43.68%, 14.05%, and 11.04% of the variance, and in the Malaysian sample with the three components explaining 42.49%, 12.84%, and 11.04% of the variance. Collectively, the three factors-factor solutions explained 66.77% of variance in the Botswana sample, 68.77% of the variance in the Zimbabwean data, and 66.36% in the Malaysian sample. An inspection of the scree plot revealed a clear break after the second component—indicating a leveling-off of eigenvalues on the scree plot after the three factors (see Cattell’s, 1966 explanation of the scree test). To aid in the interpretation of the three components, oblimin rotation was performed. The results indicate that there was little difference between the three-factor varimax and oblimin solutions, thus both solutions were examined in subsequent analyses before deciding to use an oblimin rotation for the final solution.

The rotated solution revealed the presence of a simple structure—as suggested by Thurstone (1947), with all the three components showing several strong loadings on each component (see Table 2). All the five well-being items were loaded on the same factor and all had reasonably high loadings (ranging from 0.71 to 0.85 in the Botswana sample, ranging from 0.71 to 0.87 in the Zimbabwean sample, and ranging from 0.81 to 0.84 in the Malaysian sample). All of the seven generalized anxiety items with factor loadings ranging from 0.57 to 0.83 in the Botswana sample, 0.74 to 0.86 in the Zimbabwean sample, and 0.76 to 0.83 in the Malaysian sample also loaded in a different factor. The three loneliness items also loaded on one factor with loadings ranging from 0.82 to 85 in the Botswana population, with similar high factor loadings ranging from 0.82 to 88 in the Zimbabwean population and loadings ranging from 0.84 to 91 in the Malaysia Population.

Internal Consistency Reliability of All the Measuring Instruments

Having distinguished the WHO-5 Well-Being Index from loneliness and generalized anxiety using factor analysis, the next step entailed establishing the Cronbach’s alpha coefficients for the three scales. Cronbach’s alpha coefficients for the total WHO-5 Well-Being Index in the Botswana sample score was .86; the GAD-7 and the Three-Item Loneliness Scale also had good Alphas (.89, and .83). For the Zimbabwean data, the total WHO-5 Well-Being Index’s α = .85; GAD-7 α = .93; and the TILS α = .82. For Malaysian data, the total WHO-5 Well-being index α = .88; GAD-7 α = .91; and TILS α = .88. To further establish the reliability of the WHO-5 Well-Being Index in the three samples, we also split the datasets by gender, age, relationship status, and levels of education. Tables 3 to 5 below reveal internal consistency reliability coefficients according to a list of selected demographic characteristics.

Table 3.

Overall and Subgroup Alpha Coefficients of WHO-5 (WHO, 1988): Botswana.

			Cronbach’s alpha if item deleted the past 2 weeks
	No. of cases	Cronbach’s alpha	I feel cheerful and in good spirits	I feel calm and relaxed	I feel active and vigorous	I wake up feeling fresh and rested	Life filled with things that interest me
Total	406	.86	.828	.821	.841	.857	.841
Gender
Male	114	.85	.811	.803	.797	.856	.817
Female	286	.87	.826	.821	.853	.851	.843
Age (years)
18–24	233	.833	.782	.782	.812	.818	.806
25–34	55	.837	.816	.765	.783	.861	.786
35–44	55	.911	.887	.876	.898	.902	.894
45–54	27	.959	.948	.949	.948	.951	.947
55–64	28	.926	.896	.897	.908	.899	.949
65 plus	8	.816	.747	.793	.703	.841	.786
Relationship status
Married	83	.897	.948	.993	.944	.944	.948
Widowed	2	.967	.953	.937	.952	.941	.941
Divorced/separated	9	.955	.568	.468	.579	.709	.589
Single but cohabiting with a significant other	31	.641	.821	.815	.836	.850	.830
Single never married	259	.860	.800	.793	.793	.992	.793
Have multiple partners	4	.885	.860	.897	.865	.859	.912
Prefer not to say/don’t know	18	.902	.948	.993	.944	.944	.948
Education
Secondary	109	.823	.767	.775	.804	.798	.794
Post-secondary school diploma/certificate/degree	99	.857	.810	.811	.824	.854	.835
Bachelor	117	.872	.852	.813	.843	.869	.846
Masters	42	.882	.851	.847	.852	.875	.861
Doctorate/PhD	31	.904	.870	.876	.898	.887	.882

Table 4.

Overall and Subgroup Cronbach’s Alpha Coefficients of WHO-5 (WHO, 1988): Zimbabwe.

			Cronbach’s alpha if item deleted
	No. of cases	Cronbach’s alpha	I feel cheerful and in good spirits: The past 2 weeks	I feel calm and relaxed: The past 2 weeks	I feel active and vigorous: The past 2 weeks	I wake up feeling fresh and rested: Past 2 weeks	My daily life is filled with things that interest me: The past 2 weeks
Total	319	.85	.806	.799	.821	.836	.833
Gender
Male	119	.87	.828	.820	.844	.870	.845
Female	196	.82	.773	.768	.789	.799	.813
Age (years)
18–24	121	.827	.782	.755	.806	.797	.818
25–34	84	.812	.758	.768	.771	.808	.770
35–44	66	.864	.811	.805	.828	.869	.858
45–54	28	.941	.918	.923	.928	.931	.936
55–64	16	.904	.899	.912	.861	.859	.876
65 plus	4	−.428	.782	.755	.806	.797	.818
Relationship status
Married	117	.897	.874	.858	.883	.883	.875
Widowed	9	.905	.859	.922	.879	.860	.894
Divorced/separated	12	.852	.796	.830	.781	.815	.873
Single but cohabiting with a significant other	14	.883	.844	.823	.827	.892	.893
Single never married	151	.775	.706	.698	.750	.755	.757
Prefer not to say/don’t know	15	.896	.875	.862	.875	.882	.873
Education
Secondary	45	.746	.691	.706	.613	.747	.731
Postsecondary school diploma/certificate/degree	71	.774	.799	.709	.679	.707	.750
Bachelor	121	.854	.828	.810	.810	.838	.834
Masters	67	.904	.891	.875	.877	.876	.896
Doctorate/PhD	9	.925	.906	.890	.920	.909	.914

Table 5.

Overall and Subgroup Cronbach’s Alpha Coefficients of WHO-5 (WHO, 1988): Malaysia.

			Cronbach’s alpha if item deleted
	No. of cases	Cronbach’s alpha	I feel cheerful and in good spirits: The past 2 weeks	I feel calm and relaxed: The past 2 weeks	I feel active and vigorous: The past 2 weeks	I wake up feeling fresh and rested: The past 2 weeks	My daily life is filled with things that interest me: The past 2 weeks
Total	416	.883	.856	.850	.863	.860	.861
Gender
Male	109	.890	.862	.859	.872	.869	.871
Female	306	.880	.853	.846	.859	.856	.857
Age (years)
18–24	135	.902	.881	.871	.874	.890	.887
25–34	99	.818	.761	.769	.818	.771	.789
35–44	66	.860	.843	.834	.819	.847	.810
45–54	46	.894	.860	.856	.885	.859	.891
55–64	59	.916	.895	.886	.913	.904	.890
65 plus	9	.877	.814	.848	.777	.846	.937
Relationship status
Married	227	.899	.868	.870	.879	.881	.883
Widowed	6	.913	.891	.911	.935	.879	.854
Divorced/separated	11	.926	.907	.900	.893	.896	.944
Single but cohabiting with a significant other	27	.828	.834	.781	.750	.788	.809
Single never married	244	.876	.849	.837	.860	.849	.854
Education
Secondary	27	.852	.832	.766	.776	.885	.826
Post-secondary-school diploma/certificate/degree	103	.856	.822	.819	.823	.835	.832
Bachelor	140	.887	.855	.858	.884	.864	.879
Masters	89	.886	.874	.849	.850	.857	.872
Doctorate/PhD	56	.919	.892	.899	.913	.905	.896

Further Tests of Divergent and Discriminant Validity

For examination of construct validity, correlations between WHO-5 Well-Being Index, the Three-Item Loneliness Scale, and the GAD-7 were established. Results are presented in Table 6 below. In the Botswana, Zimbabwean and Malaysian samples, loneliness moderately and negatively correlated with well-being score (r = −.350), (r = −.305) and (r = −.380) respectively. In addition, generalized anxiety moderately and negatively correlated with well-being (r = −.468) in the Botswana sample, (r = −.444) in the Zimbabwean sample, and (r = −.464) in the Malaysian sample.

Table 6.

Pearson Correlations of Well-being, Loneliness and Generalized Anxiety: Botswana, Zimbabwe, and Malaysia.

Botswana					Zimbabwe					Malaysia
N	M	SD	L	GA	N	M	SD	L	GA	N	M	SD	L	GA
395	5.6	2.1			322	6.0	2.0			535	4.6	1.7
389	8.3	6.1	0.43**		301	9.2	6.7	0.40**		538	6.9	6.7	0.41**
406	9.8	5.4	−0.35**	−0.47**	319	9.4	5.3	−0.30**	−0.44**	541	14.8	5.1	−0.46**	−0.38**

Note. L = loneliness; GA = generalized anxiety; WB = well-being.

Correlation is significant at the .01 level (two-tailed).

Discussion

This study examined the psychometric properties (i.e., factor structure, internal consistency reliability, discriminant validity, and convergent validity) of the WHO-5 Well-Being Index as compared to other mental health measuring instruments. PCA was used to establish both discriminant validity and convergent validity of WHO-5 by trying to distinguish it from generalized anxiety and loneliness. In the second analysis, the reliabilities of the three measuring instruments were established. Negative relationships between anxiety and well-being, and loneliness and well-being were presumed, thus we established the construct validity of the WHO-5 by investigating its correlation with loneliness and generalized anxiety.

Validity and Reliability Analyses

Factor analysis is a commonly used procedure in the development and evaluation of psychological measurements (Floyd & Widaman, 1995). In the present study, PCA was used to establish factor structure before conducting reliability and correlation analyses. This decision was based on Kaufman and Kaufman’s (1993) suggestions that “factor structure is the most important evidence of theory-based multi-scale test construct validity” (pp. 90–95). The decision was not unique to the current study since factor analysis has been widely used to develop theories and evaluate the construct validity of measures (Kieffer, 1999), and for revealing how theoretical constructs underlie latent variables in a given data set (Clark & Watson, 2016).

With factor analysis, the authors were able to confirm the multidimensionality of the three mental health scales used (i.e., well-being, anxiety, and loneliness) as well as the unidimensionality and validity of the WHO-5 Well-Being Index (see Nunnally, 1978). Literature has shown a lack of conceptual and measurement clarity regarding whether well-being is equivalent to mental health or a separate indicator of mental health. Such conceptual and measurement clarity is needed since the findings of a study such as the current one, have the potential to inform national policy, practice, and future research in this area. Beyond distinguishing well-being from anxiety and loneliness, the findings also showed that the WHO-5 is unidimensional—as all the manifest variables measuring well-being had high loadings on the same factor—in all the three samples. There were no cross-loadings between measures of well-being and the other two measurement instruments—confirming the unidimensionality of the WHO-5 Well-Being Index.

In terms of reliability, some studies have confirmed, using Cronbach’s alpha, that the WHO-5 has good internal consistency (El-Den et al., 2018; Newnham et al., 2010). Some have raised concerns that certain psychometric issues such as the reliability of the WHO-5 and the measurement invariance across countries of this scale are unclear. The current study, therefore, filled some gaps in research by analyzing data from the Zimbabwean, Botswana, and Malaysia contexts, showing good internal consistency (α > .70). In the present study, the reliability of the WHO-5 instrument was also expressed in terms of coefficient alpha (see Tables 3 –5). The three tables also show additional internal consistency reliability coefficients according to a selected list of several demographic characteristics. These findings provided further evidence for the internal consistency of the WHO-5 Well-Being Index.

Since all the test scores were collected at the same time, correlation analysis was used to further establish discriminant and convergent validity. The findings suggest that while the three scales are used to measure mental health, well-being is different from generalized anxiety and loneliness. The correlations between well-being and the other two constructs were negative. Furthermore, the results broadly suggest that the WHO-5 Well-Being Index, GAD-7, and the Three-Item Loneliness Scale measures are unique, reliable, and valid instruments for measuring mental health among diverse populations in Botswana, Zimbabwe, and Malaysia—and possibly—in similar African and Asian settings.

These results also confirmed the findings and proposed links between anxiety, loneliness, and well-being (Andersson, 1998; Hughes et al., 2004; Mhaka-Mutepfa et al., in press). The findings that negative relationships exist between well-being and loneliness as well as relationships between generalized anxiety and well-being infer that well-being is particularly important in times of crisis (International Baccalaureate, 2020). Due to the impact of the COVID-19 outbreak, people are currently living in what could be considered a time of crisis as the pandemic has altered many people’s lives in ways that are yet to be understood (Mhaka-Mutepfa et al., in press). These preliminary findings broadly suggest that a sense of subjective well-being can buffer against feelings of loneliness and generalized anxieties, especially during this COVID-19 era.

Limitations

Some of the limitations of the current study are related to the design as well as sampling techniques used (convenience and snowballing sampling techniques) as these did not allow for the calculation of the consent rate and exclusion rate. As in any online survey, such limitations could reduce the generalizability of our findings. In addition, the online survey method used in this study excluded members of the population who did not have access to the internet or those who are computer illiterate and/or not internet savvy. Future research could be administered using in-depth interviews, and paper and pencil to include a wider variety of participants who would otherwise be valuable in the psychometric assessment of these instruments (e.g., the different ethnic groups living in different rural areas of the three countries where there is no internet access).

Implications for Research and Practice

COVID-19 is not the first public health crisis, epidemic, or pandemic to hit the world. In 2018, the WHO reported an estimate of1,307 epidemic events including diseases such as shigellosis, chikungunya, typhoid fever, cholera, zika virus disease, meningitis, West Nile, and fever between 2011 and 2017 (WHO, 2018). All these public health crises, epidemics, or pandemics are bound to have similar negative impacts—albeit with different severities. In relation to COVID-19, in October 2020, a joint statement by the ILO, FAO, IFAD, and WHO warned that the COVID-19 pandemic has not only led to the dramatic loss of human life worldwide but has put millions of enterprises at an existential threat, exposed billions of global workforce at risk of losing their jobs, and millions of individuals at the risk of becoming extremely poor (ILO, FAO, IFAD and WHO, 2020). The foregoing consequences of COVID-19 have deleterious effects on the mental health of people. Some of the mental health and psychosocial challenges that have been reported elsewhere as a result of COVID-19 include insomnia, loneliness, depression, anxiety, loss of social support, bereavement, uncertainty, and worry (O’Neil et al., 2020; Peng et al., 2020; Qiu et al., 2020). These findings highlight the need to study and measure these diverse aspects of mental health even in countries like Botswana, Zimbabwe, and Malaysia which have side-lined concerns regarding mental health.

While the WHO-5 Well-Being Index has been used frequently in large-scale multi-national studies (Sischka et al., 2020), only a few studies have evaluated the WHO-5 Well-Being Index as a valid cross-cultural measure of subjective well-being (e.g., El-Den et al., 2018; Sischka et al., 2020) and just a limited number have done so in specific countries with the WHO-5 being translated into the different countries’ respective languages (e.g., Floyd & Widaman, 1995). Even fewer researchers have reported the validity and reliability of the WHO-5 in COVID-19-related studies (e.g., White & Van Der Boor, 2020). Like most psychological phenomena, well-being cannot be accurately studied if it is unclear what is being measured, and if the instruments used to measure different aspects of well-being are not valid and reliable. Thus more research is a prerequisite in accurately investigating the prevalence, predictors, mediators, moderators, and outcomes of mental health in settings like Botswana and Zimbabwe. The findings of the present study are therefore important as they fill some gaps by confirming internal consistency, and construct validity (i.e., convergent validity and discriminant validity) of the WHO-5.

With regards to policy and practice, health is not just the absence of illness but “a state of complete physical, mental, and social well-being” (WHO, 2007, p. 1). Therefore, looking at our findings and the revelations about public health crises, epidemics, or pandemics and their wide negative impacts, the authors suggest that policymakers, practitioners, researchers, and individuals should (1) anticipate and prepare for a wide variety of pandemics, (2) identify and use all resources at their disposal (including personal, and psychological), that could help protect individuals and communities from the grave impact, as well as (3) work collaboratively to develop evidence-based policies, programs, interventions and improve national infrastructures. This holistic approach is beneficial because stakeholders have learnt that for better responses to similar public health crises and pandemics, solutions require cooperation among multiple professionals—including medical personnel, psychologists, and anthropologists.

Footnotes

Acknowledgements

The authors express gratitude to the Principal Investigators from the University of Rwanda (Dr Stefan Jansen and Dr Epaphrodite Nsabimana), and all the other members of the consortium who were co-investigators in the study “Coping with COVID-19 in Global South” for their diligence and willingness to share best practices with each other to make the study a success. In addition, all the participants who agreed to fill in the long questionnaire while grappling to cope with the pandemic.

Author Contributions

All authors conceptualized and edited the paper. MP wrote the results sections. LM and KM wrote the literature review. MP and MMM wrote the method section and MMM added the Zimbabwe data. WYL and BSC added the Malaysia data.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical Approval

Ethics approval was obtained from the IRB University of Botswana, Medical Research Council of Zimbabwe, and the Universit Malaya Research Ethics Committee, Kuala Lumpur. All aspects of ethics; confidentiality, anonymity, and consent were considered and met.

ORCID iDs

Magen Mhaka-Mutepfa

Leonne M. Mfolwe

Bee Seok Chua

References

Adjorlolo

(2019). Generalised anxiety disorder in adolescents in Ghana: Examination of the psychometric properties of the Generalised Anxiety Disorder-7 Scale. African Journal of Psychological Assessment, 1, a10. https://doi.org/10.4102/ajopa.v1i0.10

Andersson

(1998). Loneliness research and interventions: A review of the literature. Aging & Mental Health, 2(4), 264–274. https://doi.org/10.1080/13607869856506

Barthel

Barkmann

Ehrhardt

Bindt

; International CDS Study Group. (2014). Psychometric properties of the 7-Item Generalized Anxiety Disorder Scale in antepartum women from Ghana and Côte d’Ivoire. Journal of Affective Disorders, 169, 203–211. https://doi.org/10.1016/j.jad.2014.08.004

Bartlett

M. S.

(1954). A note on the multiplying factors for various χ² approximations. Journal of The Royal Statistical Society, 16(2), 296–298.

Cattell

R. B.

(1966). The scree test for the number of factors. Multivariate Behavioral Research, 1(2), 245–276. https://doi.org/10.1207/s15327906mbr0102_10

Clark

Watson

(2016). Constructing validity: Basic issues in objective scale development. In Kazdin

A. E.

(Ed.), Methodological issues and strategies in clinical research (pp. 187–203). American Psychological Association.

Cummins

R. A.

Lau

A. A. L. D.

Mellor

Stokes

M. A.

(2009). Encouraging governments to enhance the happiness of their nation: Step 1: Understand subjective well-being. Social Indicators Research, 91, 23–36. https://doi.org/10.1007/s11205-008-9324-6

Diener

(2006). Guidelines for national indicators of subjective well-being and ill-being. Journal of Happiness Studies, 7, 397–404. https://doi.org/10.1007/s10902-006-9000-y

Diener

Seligman

M. E. P.

(2004). Beyond money: Toward an economy of well-being. Psychological Science in the Public Interest, 5(1), 1–31. https://doi.org/10.1111%2Fj.0963-7214.2004.00501001.x

10.

Diener

Suh

E. M.

Lucas

R. E.

Smith

H. L.

(1999). Subjective well-being: Three decades of progress. Psychological Bulletin, 125(2), 276–302. https://doi.org/10.1037/0033-2909.125.2.276

11.

Disabato

D. J.

Goodman

F. R.

Kashdan

T. B.

Short

J. L.

Jarden

(2016). Different types of well-being? A cross-cultural examination of hedonic and eudaimonic well-being. Psychological Assessment, 28, 471–482. https://doi.org/10.1037/pas0000209

12.

Dodge

Daly

Huyton

Sanders

(2012). The challenge of defining well-being. International Journal of Wellbeing, 2(3), 222–235. https://doi.org/10.5502/ijw.v2i3.4

13.

Drost

E. A.

(2011). Validity and reliability in social science research. Education Research and Perspectives, 38(1), 105–123.

14.

El-Den

Chen

T. F.

Gan

Y. L.

Wong

O’Reilly

C. L.

(2018). The psychometric properties of depression screening tools in primary healthcare settings: A systematic review. Journal of Affective Disorders, 225, 503–522. https://doi.org/10.1016/j.jad.2017.08.060

15.

Eurofound. (2020). Living, working and COVID-19: First findings. European Foundation for the Improvement of Living and Working Conditions. https://www.eurofound.europa.eu/topic/covid-19

16.

Floyd

F. J.

Widaman

K. F.

(1995). Factor analysis in the development and refinement of clinical assessment instruments. Psychological Assessment, 7(3), 286–299. https://doi.org/10.1037/1040-3590.7.3.286

17.

Groarke

J. M.

Berry

Graham-Wisener

McKenna-Plumley

P. E.

McGlinchey

Armour

(2020). Loneliness in the UK during the COVID-19 pandemic: Cross-sectional results from the COVID-19 psychological well-being study. PLoS One, 15(9), e0239698. https://doi.org/10.1371/journal.pone.0239698

18.

Henkel

Mergl

Kohnen

(2003). Identifying depression in primary care: A comparison of different methods in a prospective cohort study. BMJ Evidence Based Medicine, 326, 155200200–155200201.

19.

Hughes

M. E.

Waite

L. J.

Hawkley

L. C.

Cacioppo

J. T.

(2004). A short scale for measuring loneliness in large surveys: Results from two population-based studies. Research on Aging, 26(6), 655–672. https://doi.org/10.1177/0164027504268574

20.

ILO, FAO, IFAD and WHO. (2020). Impact of COVID-19 on people’s livelihoods, their health and our food systems. Joint statement by ILO, FAO, IFAD and WHO. who.int

21.

International Baccalaureate. (2020). Why well-being matters during a time of crisis: Well-being considerations for a successful post-Covid-19 educational transition. International Baccalaureate Organisation. https://www.ibo.org/globalassets/new-structure/research/images/why-wellbeing-matters-during-a-time-of-crisis-en.pdf

22.

Jeste

D. V.

Lee

E. E.

Cacioppo

(2020). Battling the modern behavioral epidemic of loneliness: Suggestions for research and interventions. Journal of American Medical Association Psychiatry, 77(6), E1–E2. https://doi.org/10.1001/jamapsychiatry.2020.0027

23.

Johnson

S. U.

Ulvenes

P. G.

Øktedalen

Hoffart

(2019). Psychometric properties of the General Anxiety Disorder 7-Item (GAD-7) Scale in a heterogeneous psychiatric sample. Frontiers in Psychology, 10, 1713. https://doi.org/10.3389/fpsyg.2019.01713

24.

Kaiser

H. F.

(1970). A second generation little jiffy. Psychometrika, 35(4), 401–415. https://doi.org/10.1007/bf02291817

25.

Kaiser

H. F.

(1974). An index of factorial simplicity. Psychometrika, 39, 31–36.

26.

Kaufman

N. L.

(1993). Kaufman adolescent and Adult Intelligence Test (KAIT). American Guidance Service.

27.

Kertz

Bigda-Peyton

Bjorgvinsson

(2012). Validity of the generalized anxiety disorder-7 scale in an acute psychiatric sample. Clinical Psychology & Psychotherapy, 20, 456–464. https://doi.org/10.1002/cpp.1802

28.

Kieffer

K. M.

(1999). An introductory primer on the appropriate use of exploratory and confirmatory factor analysis. Research in the Schools, 6(2), 75–92.

29.

Kingdon

G. G.

Knight

(2006). Subjective well-being poverty vs. income poverty and capabilities poverty? The Journal of Development Studies, 42(7), 1199–1224. https://doi.org/10.1080/00220380600884167

30.

La Placa

McNaught

Knight

(2013). Discourse on well-being in research and practice. International Journal of Wellbeing, 3(1), 116–125. https://doi.org/10.5502/ijw.v3i1.7

31.

Löwe

Decker

Müller

Brähler

Schellberg

Herzog

Herzberg

P. Y.

(2008). Validation and standardization of the Generalized Anxiety Disorder Screener (GAD-7) in the general population. Medical Care, 46(3), 266–274. https://doi.org/10.1097/MLR.0b013e318160d093

32.

Löwe

Spitzer

R. L.

Gräfe

Kroenke

Quenter

Zipfel

Buchholz

Witte

Herzog

(2004). Comparative validity of three screening questionnaires for DSM-IV depressive disorders and physicians’ diagnoses. Journal of Affective Disorders, 78, 131–140.

33.

Monteiro

R. P.

Nascimento

B. S.

Monteiro

T. M. C.

da Silva

P. D. G.

Ferreira

A. J. C.

(2022). Psychometric evidence of the 7-Item Generalized Anxiety Disorder Questionnaire in Brazil. International Journal of Mental Health and Addiction, 20, 1023–1034.

34.

Mhaka-Mutepfa

Biswas

Mfolwe

Mathoote

K. D.

Hasan

M. T.

(in press). Wellbeing, loneliness, and quality of relationships before and during the COVID-19 pandemic in Botswana: Findings from a cross-sectional analysis. Sage Open.

35.

Newnham

E. A.

Hooke

G. R.

Page

A. C.

(2010). Monitoring treatment response and outcomes using the World Health Organization’s well-being index in psychiatric care. Journal of Affective Disorders, 122, 133–138. https://doi.org/10.1016/j.jad.2009.06.005

36.

Nunnally

J. C.

(1978). Psychometric theory (2nd ed.). McGraw-Hill.

37.

Omani-Samani

Maroufizadeh

Ghaheri

Navid

(2018). Generalized Anxiety Disorder-7 (GAD-7) in people with infertility: A reliability and validity study. Middle East Fertility Society Journal, 23(4), 446–449.

38.

O’Neil

Nicholls

S. J.

Redfern

Brown

Hare

D. L.

(2020). Mental health and psychosocial challenges in the COVID-19 pandemic: Food for thought for cardiovascular health care professionals. Heart, Lung & Circulation, 29(7), 960–963. https://doi.org/10.1016/j.hlc.2020.05.002

39.

Pallant

(2016). SPSS survival manual: A step by step guide to data analysis using SPSS program (6th ed.). McGraw-Hill Education.

40.

Peng

Wang

(2020). Challenges and opportunities in mental health services during the COVID-19 pandemic. General psychiatry, 33(5), e100275. https://doi.org/10.1136/gpsych-2020-100275

41.

Piedmont

R. L.

McCrae

R. R.

Riemann

Angleitner

(2000). On the invalidity of validity scales: Evidence from self-reports and observer ratings in volunteer samples. Journal of Personality and Social Psychology, 78(3), 582–593. https://doi.org/10.1037//0022-3514.78.3.582

42.

Qiu

Shen

Zhao

Wang

Xie

(2020). A nationwide survey of psychological distress among Chinese people in the COVID-19 epidemic: Implications and policy recommendations. General psychiatry, 33, e100213. https://doi.org/10.1136/gpsych2020-100213

43.

Russell

Peplau

L. A.

Ferguson

M. L.

(1978). Developing a measure of loneliness. Journal of Personality Assessment, 42(3), 290–294. https://doi.org/10.1207/s15327752jpa4203_11

44.

Rutter

L. A.

Brown

T. A.

(2016). Psychometric properties of the generalized anxiety disorder scale-7 (GAD-7) in outpatients with anxiety and mood disorders. Journal of Psychopathology and Behavioral Assessment, 39, 140–146. https://doi.org/10.1007/s10862-016-9571-9

45.

Ryff

C. D.

Singer

B. H.

(2008). Know thyself and become what you are: A eudaimonic approach to psychological well-being. Journal of Happiness Studies, 9, 13–39. https://doi.org/10.1007/s10902-006-9019-0

46.

Sischka

P. E.

Costa

A. P.

Steffgen

Schmidt

A. F.

(2020). The WHO-5 Well-Being Index – Validation based on item response theory and the analysis of measurement invariance across 35 countries. Journal of Affective Disorders Reports, 1(1), 100020. https://doi.org/10.1016/j.jadr.2020.100020

47.

Smallbone

Quinton

(2004). Increasing business students’ confidence in questioning the validity and reliability of their research. Electronic Journal of Business Research Methods, 2(2), 153–162.

48.

Spitzer

R. L.

Kroenke

Williams

J. B.

Löwe

(2006). A brief measure for assessing generalized anxiety disorder: The GAD-7. Archives of Internal Medicine, 166(10), 1092–1097. https://doi.org/10.1001/archinte.166.10.1092

49.

Taylor

T. E.

(2015). The markers of well-being: A basis for a theory-neutral approach. International Journal of Wellbeing, 5(2), 75–90. https://doi.org/10.5502/ijw.v5i2.5

50.

Terrill

A. L.

Hartoonian

Beier

Salem

Alschuler

(2015). The 7-Item Generalized Anxiety Disorder Scale as a tool for measuring generalized anxiety in multiple sclerosis. International Journal of MS Care, 17(2), 49–56. https://doi.org/10.7224/1537-2073.2014-008

51.

Thurstone

L. L.

(1947). Multiple factor analysis. University of Chicago Press.

52.

Topp

C. W.

Østergaard

S. D.

Søndergaard

Bech

(2015). The WHO-5 Well-being Index: A systematic review of the literature. Psychotherapy and Psychosomatics, 84, 167–176. https://doi.org/10.1159/000376585

53.

White

R. G.

Van Der Boor

(2020). Impact of the COVID-19 pandemic and initial period of lockdown on the mental health and well-being of adults in the UK. British Journal of Psychology 1904-1920, 6(5), 1–4. https://doi.org/10.1192/bjo.2020.79

54.

World Health Organisation. (1998). Well-being measures in primary health care/The Depcare Project. Author.

55.

World Health Organisation. (2005). Promoting mental health: Concepts, emerging evidence, practice. Author. https://www.who.int/mental_health/evidence/MH_Promotion_Book.pdf

56.

World Health Organisation. (2007). Basic documents (46th ed.). Author.

57.

World Health Organisation. (2013). Mental health action plan 2013-2020. Author. http://apps.who.int/iris/bitstream/handle/10665/78855/9789241506021_eng.pdf?sequence=1

58.

World Health Organisation. (2018). Managing epidemics: Key facts about major deadly diseases. Author. https://apps.who.int/iris/bitstream/handle/10665/272442/9789241565530-eng.pdf?sequence=1&isAllowed=y

Psychometrics Validation of the WHO-5 Wellbeing Index in Botswana,Zimbabwe,and Malaysia: Coping With COVID-19 Pandemic in the Global South Study

Abstract

Keywords

Understanding Well-Being

Correlates of Well-Being

Measuring Well-Being

Aims and Goals of the Study

Method

Sampling and Data Collection

Instruments

The WHO-5 Well-Being Index

Two Scales Used to Test Construct Validity

The Three-Item Loneliness Scale

The Generalized Anxiety Disorder Seven-Item (GAD-7) Scale

Statistical Analyses

Results

Demographic Characteristics of Respondents

Well-Being as a Separate Construct From Other Mental Health Constructs

Internal Consistency Reliability of All the Measuring Instruments

Further Tests of Divergent and Discriminant Validity

Discussion

Validity and Reliability Analyses

Limitations

Implications for Research and Practice

Footnotes

Acknowledgements

Author Contributions

Declaration of Conflicting Interests

Funding

Ethical Approval

ORCID iDs

References