Development and Initial Validation of a Multidimensional Scale Assessing Subjective Well-Being: The Well-Being Scale (WeBS)

Abstract

Numerous scales currently exist that assess well-being, but research on measures of well-being is still advancing. Conceptualization and measurement of subjective well-being have emphasized intrapsychic over psychosocial domains of optimal functioning, and disparate research on hedonic, eudaimonic, and psychological well-being lacks a unifying theoretical model. Lack of systematic investigations on the impact of culture on subjective well-being has also limited advancement of this field. The goals of this investigation were to (1) develop and validate a self-report measure, the Well-Being Scale (WeBS), that simultaneously assesses overall well-being and physical, financial, social, hedonic, and eudaimonic domains of this construct; (2) evaluate factor structures that underlie subjective well-being; and (3) examine the measure’s psychometric properties. Three empirical studies were conducted to develop and validate the 29-item scale. The WeBS demonstrated an adequate five-factor structure in an exploratory structural equation model in Study 1. Confirmatory factor analyses showed that a bifactor structure best fit the WeBS data in Study 2 and Study 3. Overall WeBS scores and five domain-specific subscale scores demonstrated adequate to excellent internal consistency reliability and construct validity. Mean differences in overall well-being and its five subdomains are presented for different ethnic groups. The WeBS is a reliable and valid measure of multiple aspects of well-being that are considered important to different ethnocultural groups.

Keywords

Eudaimonia hedonia psychometric social well-being structural equation modeling

Introduction

Psychological research on mental health has moved from focusing on understanding psychopathology and alleviating distress to a more holistic examination of positive characteristics such as subjective well-being (SWB). Studies on individual differences in SWB have contributed to this empirical understanding of adaptive functioning among diverse groups of people (youth, older adults, college students, people acculturating to a new culture, and people with mental and physical disabilities; Arber, Fenn, & Meadows, 2014; Lane & Fink, 2015; Lehman, 1983). This body of work has helped us understand the associations between SWB and personality (Diener, Suh, Lucas, & Smith, 1999), motivation (Ryan & Deci, 2000), leisure (Kuykendall, Tay, & Ng, 2015), mortality (Wiest, Schüz, Webster, & Wurm, 2011), interpersonal relationships (Luhmann, Hofmann, Eid, & Lucas, 2012), and culture (Ahuvia, 2001; Chang et al., 2014).

Despite these important contributions to our knowledge about people’s optimal functioning, further advancement of research on SWB and positive psychology has been limited by the factors associated with existing assessment approaches and tools. First, present conceptualization and assessment have tended to focus on disparate components of SWB. Although research has demonstrated the value of considering hedonia, eudaimonia, and psychological well-being as domains of SWB, and has separately demonstrated the fundamental roles of financial satisfaction and physical wellness in an empirical understanding of SWB, no measures have considered them simultaneously using a coherent theoretical framework. Second, whereas the established definition of SWB emphasizes cognitive evaluation of life as a whole and/or in specific domains, and people’s emotional feelings about their circumstances (Diener et al., 2016), measures have generally omitted measuring aspects of well-being associated with satisfaction in social connectedness (Keyes, 1998). This tradition likely is because of a reliance on a cultural framework that is individualistic in orientation and one that values more intrapsychic elements of SWB, with less emphasis on domains of well-being that are distinctly salient to collectivistic cultures (Cooke, Melchert, & Connor, 2016). Third, the factor structure underlying SWB is still under debate. Some studies have demonstrated that SWB has a multifactorial structure (e.g., Ryff, 1989; Vella-Brodrick & Allen, 1995), whereas others have indicated that this construct can be represented by either a single-factor or a bifactor structure (e.g., Jovanović, 2015; Longo, Coyne, Joseph, & Gustavsson, 2016).

To address these research gaps, the aims of this investigation are to (a) construct a new measure of SWB that includes multiple key components of SWB considered important in diverse cultures, (b) explore the factor structure that would best represent the construct of SWB, and (c) validate this scale in ethnically diverse samples.

Theoretical consideration in the study of SWB

The Hierarchy of Needs theory proposed by Abraham Maslow (1954) is one that outlines five areas of basic human needs and motivations that contribute to well-being. These five areas include (1) physiological needs for food and shelter, (2) safety and security, (3) social needs in terms of love and belonging, (4) esteem needs that focus on feelings of respect and environmental mastery, and (5) self-actualizing needs that emphasize self-direction and reaching for one’s potential. Although there is a rich body of literature that illuminates the role of individual backgrounds and culture in the nature and composition of happiness (Diener et al., 2016; Diener & Oishi, 2005), emerging evidence suggests that these five basic needs are linked to SWB in similar ways across geographic and cultural regions (Tay & Diener, 2011). Analysis of the relative importance of these areas has highlighted that it is the fulfillment of basic needs—rather than an actual amount of resources—that accounts for the unique variance in SWB. A comprehensive conceptualization and measure of SWB, therefore, should systematically include satisfaction in, and emotional feelings toward, one’s life in each of these areas in order to capture the lived experiences of a flourishing and optimal life.

The Hierarchy of Needs theory is likely one of the few comprehensive theoretical frameworks that simultaneously considers many aspects of physical, material, social, and psychological domains of wellness. Whereas there are other theories that focus on single aspects of SWB (see Ryff & Singer, 2006 for psychological well-being; Ryan & Deci, 2000 for self-determination theory; Waterman, 2007 for eudaimonic well-being; Diener et al., 1999 for hedonic well-being), none of them systematically addresses other aspects of human experiences and conditions for flourishing, such as satisfaction with physical wellness and cognitive evaluations of basic and financial needs (Cooke et al., 2016). Additionally, there appears to be limited meaningful difference between psychological and eudaimonic well-being (Ryan & Deci, 2001). Thus, a comprehensive investigation of SWB should include subjective, physical, financial, social, and psychological domains in order to capture the broad range of this construct.

Multidimensional conceptualization and measurement of SWB

The present disparate, nonintersecting research on SWB likely is due to the divergent emphasis on what this construct represents. Original theoretical postulation and recent empirical evidence have indicated that physical needs are to be pursued and fulfilled before attending to needs associated with financial and material resources, social relationships, and the inner psyche (Maslow, 1954; Tay & Diener, 2011). In fact, traditional research on well-being often has operationalized this construct using health indicators such as mobility and physical activity. For instance, the 36-item Short-Form Health Survey (Ware, Gandek, & and IQOLA Project, 1994) assesses physical functions, bodily pain, role limitations, and emotional well-being. The relations among physical activity, self-evaluated health statuses, and life satisfaction have been found to be stronger among middle-aged and older adults than young adults (Maher, Pincus, Ram, & Conroy, 2015). These associations may be mediated by healthy self-image. Additionally, it is likely that obtaining resources for basic survival and adopting physically healthy behaviors enhances happiness (Breslin & McCay, 2012). Therefore, it is probably more useful to consider a person’s subjective evaluation of his/her physical health status rather than objective health (Diener et al., 1999; Tao et al., 2007) as a component of general SWB.

Beyond physical well-being, a large body of research has shown that perceived availability and accessibility of material resources are fundamental to an optimal level of SWB (Diener & Fujita, 1995). Although some empirical evidence does not support a reliable and strong relation between income and well-being within a particular income or national group (Diener & Biswas-Diener, 2009), other research has shown that people are happier and more able to pursue other psychological goals when safety and financial needs such as housing are fulfilled, and when they are satisfied with these life circumstances (e.g., Johnson & Krueger, 2006). Similar to physical well-being, it is likely that when people believe that they have greater efficacy in managing their financial resources and subjectively experience lower levels of financial strain, they are more likely to enjoy life and flourish (Arber et al., 2014). As a result, perceived financial well-being should be considered a critical domain of SWB.

Despite the theorized “universal” human need for enjoying positive social relationships (Maslow, 1954), most existing research on SWB has not systematically examined people’s happiness and satisfaction in interpersonal relationships (Keyes, 1998). This omission of how individuals navigate social challenges and their cognitive and emotional experiences with social connections limits our comprehensive understanding of SWB across life domains. Consistent with the Hierarchy of Needs theory, research has suggested that close social relationships are essential components of optimal functioning (Diener & Oishi, 2005). Additionally, whereas satisfaction with social resources and relationships has been shown to be equally important to people across individualistic and collectivistic cultural orientations, and greater social support is related to better subjective mental health status (Rubenstein et al., 2016; Sarason, Sarason, & Gurung, 2001; Turner & Marino, 1994), people from collectivistic cultures may find social well-being distinctly central to their overall evaluation of functioning. Enjoying cohesive communities and feeling positive about the provision and receipt of social support to and from others may be fundamental elements of SWB for people of African American, Asian, and Latino backgrounds (e.g., Schwartz et al., 2010). Thus, systematic assessment of satisfaction in areas such as social cohesion, belongingness, and support, which researchers have identified as important to well-being (House, Umberson, & Landis, 1988; Slade, 2010) would capture one of the most important life tasks that comprise general well-being.

Psychological, intrapsychic experiences are proposed to reflect more advanced human needs (Maslow, 1954). These underscore the importance of feeling good about oneself and possessing optimal levels of aspiration to achieve one’s full potential. Individuals’ subjective evaluation of life circumstances and balance of positive and negative emotions are known as hedonic well-being (cf. Diener, 2006). Research has shown that hedonia is shaped by cultural and environmental contexts in which individuals are embedded (Ahuvia, 2001; Diener & Oishi, 2005; Diener et al., 1999). Together, cognitive appraisals of adjustment and affective experiences shape one’s self-respect and sense of efficacy, and are associated with greater SWB in the short term (Huta & Ryan, 2010). Common measures that focus on the cognitive evaluation aspect of well-being include the Satisfaction With Life Scale (SWLS; Diener, Emmons, Larsen, & Griffin, 1985) and Qualify of Life Inventory (Frisch, Cornell, Villanueva, & Retzlaff, 1992). There are also other well-validated, single-domain instruments that focus on emotional well-being, including the Bradburn Affect Balance Scale (Harding, 1982) and the Positive and Negative Affect Scales (Watson, Clark, & Tellegen, 1988).

In addition to the hedonic aspect, proponents of eudaimonic or psychological well-being posit that hedonic well-being represents only a proxy of the construct (Raibley, 2012). These researchers (Ryff, 1989; Seligman, 2002; Waterman, 2007) propose that personal growth, pursuit of meaning in life, and autonomy are most important in conceptualizing psychological wellness. Some measures of eudaimonic or psychological well-being capture personal satisfaction and positive experiences relating to “mind, body, and spirit.” Particularly, the Mental, Physical, and Spiritual Well-being Scale (WeBS) (Vella-Brodrick & Allen, 1995) and the Psychological WeBSs (Ryff & Singer, 1996) have shown adequate psychometric properties in capturing attitudes or emotions relating to meaning in life. Representing the apex of the hierarchy of needs, the self-actualizing needs to learn new things, develop competencies, gain self-direction, and contribute to society appear to underlie the commonalities among eudaimonic and psychological well-being. Research suggests that eudaimonic well-being may be less prone to fluctuations with life experiences and circumstances, and explains unique variances of general well-being independent of, and beyond, hedonic well-being (Huta & Ryan, 2010).

Taken together, the literature on SWB supports the Hierarchy of Needs theory and shows that SWB is a complex construct that likely involves multiple life domains that make for a global sense of wellness. A balance of various needs satisfaction will indicate optimal functioning and flourishing; therefore, any instrument that measures people’s SWB should tap into physical, financial, social, hedonic, and eudaimonic domains of well-being. Yet, these aspects of SWB also likely overlap in their abilities to indicate people’s general optimal functioning. An effective multidimensional measure of SWB, therefore, should assess overall well-being with items that share a common source of variance, while simultaneously demonstrating domain-specific areas of this construct including financial, physical, social, hedonic, and eudaimonic well-being. These five domains of SWB are theorized to be orthogonal to each other; for example, people who enjoy high levels of social well-being can have either high or low levels of eudaimonic well-being and hedonic well-being. Consistent with emerging evidence, this organization of SWB suggests a bifactor structure of this variable, in which a general factor of overall well-being as well as five domain-specific factors can be extracted (Chen, Bai, Lee, & Jing, 2016; de Bruin & du Plessis, 2015; Jovanović, 2015; Longo et al., 2016). Conceptually, such a complex representation of SWB implies that the construct can be conceptualized not only as a broad representation of people’s overall psychological functioning but that it also can manifest in several domains of people’s lives.¹

The present study

The first goal of this investigation was to develop a self-report measure of SWB that systematically assesses the overall construct and five distinct domains of well-being: physical well-being, financial well-being, social well-being, hedonic well-being, and eudaimonic well-being. The present conceptualization of SWB was expected to reflect satisfaction in the five elements of basic needs outlined in the Hierarchy of Needs theory. The second goal of this investigation was to validate the psychometric properties of this newly developed measure in multiple ethnically diverse samples. We aim to demonstrate reliability of the new measure, and to examine the factor structure that underlie SWB, and test the correlations between SWB and other related constructs.

Based on the theoretical framework and consistent with emerging research on the bifactor nature of SWB, we hypothesize that a bifactor model of the present measure would fit the scale data better than unifactorial and correlated factor structures. Hence, we expected that a general factor of SWB and five independent first-order factors for the SWB domains would be identified in exploratory (EFA) and confirmatory factor analyses (CFA). This would indicate that SWB could be understood best as comprising a broadband factor of well-being, while five orthogonal domain-specific factors would explain additional variance of this construct. We expected that domains of SWB would be significantly and positively correlated with existing measures of life satisfaction and negatively correlated with measures of depression and anxiety symptoms. We also aimed to explore differences in SWB across major ethnic groups in the United States. A systematic review has shown that 63% of published papers in positive psychology do not report the ethnic compositions of their samples, among which non-White participants were underrepresented (Rao & Donaldson, 2015). Therefore, we purposively employed ethnically diverse samples in the following studies.

In Study 1, we reported the procedures and results associated with item generation, reliability analysis, scale refinement, and initial exploratory factor analysis (EFA). We then reported in Study 2 the procedures and results of identifying a factor structure that best fit the well-being data. In Study 3, we tested the fit of various competing factor structure in a separate sample and examined the correlations between this new measure of SWB and other variables.

Study 1

The goal of the present study was to develop a new multidimensional scale that assesses five distinct domains of SWB, and to test the psychometric properties of this measure in an ethnically diverse sample.

Method

Instrument: Development and pre-test of the Well-Being Scale (WeBS)

The construction and validation of this new WeBS were guided by the Hierarchy of Needs theory as well as the rich empirical evidence on different domains of well-being. We followed the steps recommended by Hinkin (1998) to develop the current measure. After conducting a comprehensive review of the literature, common items used to assess the various domains of this construct, including hedonic well-being, eudaimonia, and perceived physical well-being were selected and modified. Additional items—particularly those that capture social well-being and financial well-being were then developed and included in the measure. Sixty initial items were generated, which tapped into physical well-being (e.g., “I am satisfied with my physical appearance”), financial well-being (e.g., “I feel in control of my finances”), social well-being (e.g., “I know I can count on my friends and/or family in a time of crisis”), hedonic well-being (e.g., “I am satisfied with my life”), and eudaimonic well-being (e.g., “Life has meaning for me”). Each domain of SWB was assessed by at least six items (cf. Hinkin, 1998). More items were generated than necessary; it was expected that better functioning items would be retained through qualitative evaluation and statistical analyses. Items were placed on a Likert-type scale ranging from 1 (strongly disagree) to 6 (strongly agree), with some negatively worded items being reverse-scored. This initial set of items in the new measure was named the WeBS.

An ethnically diverse group of students from a state university in southern California (N = 54, 85% women) was recruited for the study, and served as the pilot sample. These individuals’ age ranged from 18 to 43 (M_age = 23.1, SD_age = 4.9). The majority of the sample self-identified as Latinx/Hispanic American (55.6%); another 18.5% self-identified as Asian American, 7.4% as Black/African American, 5.6% as Euro/White American; and 12.9% as mixed or “other.” Participants completed the WeBS and a brief demographic survey in small groups, following which they were asked to provide written feedback on the readability and possible redundancy of items within the initial 60-item scale. Students identified as Latinx/Hispanic and Asian Americans were oversampled to ensure that the initial item pool adequately represented SWB in more collectivistic cultures. The 60-item scale data demonstrated adequate internal consistency reliability. Eighteen items were deleted based on poor item-total and inter-item correlations, redundancy, and respondents’ feedback. Analysis on the readability of this 42-item pool indicated that the scale was comprehensible to individuals with a third- or fourth-grade English reading ability (Flesch–Kincaid grade-level index = 3.7).

Participants

Participants were undergraduate students from two large state universities in the Midwest and California (N = 359, 62.4% women, M_age = 20.31). About one-third (36.5%) of the participants identified as Euro/White Americans, 33.7% as Asian Americans, 20.6% as Latinx/Hispanic Americans, 3.9% as Black/African Americans, 1.1% as Native Americans, 2.5% as mixed race, and 1.4% as “other” ethnicity. Many of the participants were either foreign-born first-generation immigrants to the U.S. (22.0%) or second-generation Americans (21.5%). Consistent with the growing trend of demographics in the U.S., a vast majority of these individuals were of Asian and Hispanic backgrounds (Pew Research Center, 2012). Most of the Euro Americans and Black Americans identified as third- or fourth-generation immigrant in this sample. Thirty-one percent of the sample reported a relatively high annual household income (over $10,000), and there was a relatively even share of other income levels: 10.6% under $20,000, 11.5% $20,001–40,000, 15.9% 60,001–80,000, and 18.1% 80,001–100,000.

Procedures

Participants were recruited from the psychology subject pools at their respective universities. Participation was voluntary and anonymous. Participants completed a demographic survey and the WeBS via a secured online survey-hosting website or in a paper-and-pencil format.

A series of EFAs and exploratory structural equation models (ESEMs) were conducted to identify latent factors underlying the WeBS items and evaluate the model fit. A principal-axis factoring (PAF) analysis with oblique, promax rotation was conducted using the Statistical Package for Social Sciences (SPSS) version 20. Eigenvalues, scree plots, parallel analysis, and the minimum average partial (MAP) test (Velicer, Eaton, & Fava, 2000) were used to determine the number of factors to be extracted. An oblique rotation was chosen because prior research has demonstrated that domains of the SWB are correlated with, but distinct, from each other. EFAs with parallel analysis were also conducted using Mplus 7.11 (Muthén & Muthén, 1998–2014). Once the number of factors to be extracted was determined, we examined the model fit and loadings of all items on these factors in an ESEM. The model fit of ESEMs was evaluated with comparative fit index (CFI), Tucker–Lewis fit index (TLI), root mean square error of approximation (RMSEA), and maximum likelihood (ML)-based standardized root mean squared residual (SRMR). Although generally considered a poor indicator of overall model fit in samples of this size (Marsh, Balla, & McDonald, 1988), the model χ² value was reported as a common practice but it was not used to assess the fit of the present model. Guidelines for evaluating model fit suggest that CFI and TLI values close to or greater than .90 (Bentler, 1990), RMSEA values close to or smaller than .08, and SRMR values close to or smaller than .10 indicate adequate fit, and CFI and TLI ≥ .95, RMSEA ≤ .06, and SRMR ≤ .08 indicate closet fit (Hu & Bentler, 1999).

Results and discussion

Data from this study demonstrated a normal distribution with no univariate and multivariate outliers or curvilinear relations among items; therefore, factor analyses with ML were deemed appropriate. Three items were removed because the initial communalities were lower than .30 and did not significantly load on to any factors (see Comrey & Lee, 1992). These items assessed subjective experiences with minor bodily pain, being alone, and general curiosity. The remaining 39 items were subjected to a PAF with promax rotation. Although eight factors with eigenvalues ≥1.0 could be extracted, the scree test suggested that a five-factor structure would be optimal. Parallel analyses with SPSS macros (O’Conner, 2000) and in Mplus, as well as a MAP test all indicated that five factors should be retained. Consistent with the theoretical framework adopted for the measure and supported by empirical analyses, the WeBS data should be best understood as five distinct factors. Using an iterative process, we identified and eliminated 10 items that did not load on any of the five factors (loadings <.30). This resulted in a 29-item scale. Results from the ESEM for the reduced 29-item WeBS showed that the five-factor structure was an acceptable fit to the present data, χ²(271) = 649.02, CFI = .91, TLI = .87, RMSEA = .062 (95% CI = .056–.068), SRMR = .037.

Table 1 summarizes the means, standard deviations, and item loadings for each factor. Findings from this study showed that the 29 items in the newly constructed measure of SWB-tapped financial well-being, physical well-being, social well-being, eudaimonic well-being, and hedonic well-being. Results supported our hypothesis and demonstrated empirical support for the five domains of well-being based on satisfaction with needs fulfillment. Each of these factors had more than three item indicators. Because ESEM permitted each item to load onto multiple factors, a few items cross-loaded on two items. For example, “Life has meaning for me” simultaneously loaded on the factors labeled hedonic well-being and eudaimonic well-being. This suggested that these items may reflect happiness and eudaimonia, both aspects of SWB relating to psychological needs. This may explain why some research has found overlap in these two domains, while also being distinct subdomains of overall SWB (Huta & Ryan, 2010). Overall, the results of this study indicated that the WeBS is a reliable measure of SWB.

Table 1.

Means, standard deviations, and item loadings of the 29-item WeBS from exploratory structural equation modeling in Sample 1 (N = 359).

Abridged item		M	SD	F1	F2	F3	F4	F5
1	Physical health status	4.74	1.16	.02	.60	.07	.02	.00
2	Financial resources to meet needs	4.46	1.45	.88	.02	.02	−.04	−.04
3	Financial resources to have fun	4.25	1.44	.93	−.05	.02	−.04	−.02
4	Satisfaction with housing	4.68	1.36	.43	.04	.05	−.02	.11
5	Control of finances	4.24	1.39	.61	.01	−.15	.06	.21
6	Control over physical health	4.73	1.31	.18	.57	−.01	.09	.01
7	Satisfaction with weight	3.55	1.60	−.01	.81	.03	−.20	−.06
8	Energy to do necessary things	4.45	1.20	.28	.47	−.04	.07	.09
9	Care of physical health	4.38	1.21	−.07	.76	−.02	.04	.07
10	Plan for the future.	4.97	1.06	−.03	.05	.08	.45	.08
11	People to talk to when have problems	4.93	1.43	−.04	.09	.64	−.02	.17
12	Count on friends/family in crisis	5.19	1.17	.12	−.02	.81	.00	.02
13	People who love me	5.46	1.05	−.02	−.07	.54	.12	.20
14	Solve problems I face	4.84	.95	.13	.08	.01	.60	.10
15	Life at home	4.91	1.12	.31	.04	.15	.06	.23
16	Satisfaction with physical appearance	4.14	1.33	.06	.63	.03	−.04	.10
17	Getting along with people	5.08	.89	.03	.18	.26	.35	−.03
18	Spending time with friends/relatives	5.31	.96	.09	.12	.48	.23	−.02
19	Doing fun and interesting things	4.85	.97	.31	.15	.05	.08	.28
20	Reaching goals	5.19	.92	−.01	−.06	−.12	.63	.28
21	Making a difference in other’s lives	4.98	1.03	−.12	−.05	.15	.40	.37
22	Meaningful life	5.20	1.02	.00	−.09	.09	.43	.43
23	Satisfaction with spirituality	4.75	1.14	.06	.06	−.09	.37	.22
24	Confidence in self	4.58	1.17	.01	.03	.00	.68	−.13
25	Creativity	4.47	1.23	−.13	.15	−.02	.45	−.08
26	Stimulating conversations	4.76	1.10	.02	.01	.06	.62	−.02
27	Make me happy	5.30	.82	.15	.06	.09	.33	.33
28	Feel happy often	4.86	1.06	.11	.11	.02	−.02	.73
29	Enjoyment in life	5.10	.93	.02	.06	−.01	−.01	.92

Note. Loadings ≥ .30 are in boldface. F1: financial well-being, F2: physical well-being, F3: social well-being, F4: eudaimonic well-being, F5: hedonic well-being.

Study 2

The goal of this study was to further examine the factor structure that underlies the WeBS items to represent both overall SWB and five domains. Specifically, we hypothesized that a bifactor model would be the best fitting structure for this multidimensional scale of SWB than single, correlated, or higher-order factor models.

Method

Participants

Similar to the sample in Study 1, an ethnically diverse sample of college students participated in this study (N = 371, 58.2% women, M_age = 20.46). Of the participants, 38.9% self-identified as Asian Americans, 28.1% as Euro/White Americans, 20.8% as Latinx/Hispanic Americans, as 4.9% Black/African Americans, and self-identified as 7.1% mixed race or “other” ethnicity. One-third of the participants reported a household income of over $10,000. The majority of students (67.6%) reported living independently, such as in an apartment or at a university residence hall, while 29.1% reported living with family members.

Procedures

Participants were recruited from the subject pool as a part of a larger study examining college experiences of students. Respondents completed a demographic survey questionnaire and the 29-item WeBS. Based on the ESEM results from Study 1, we conducted a series of CFA with five factors in Mplus using the ML estimator. To achieve a simple structure, only primary factor loadings were permitted in the CFA. Model fit indices were consulted to examine the nature of items assessing the five domains of SWB. We also examined the fit of alternate factor structures for the current WeBS data, including a single factor model, a higher-order factor model, and a bifactor model (see Lui & Rollock, 2017 and Reise et al., 2007 for description of the differences among higher-order, correlated factor, and bifactor models). In the single factor model, all items loaded onto one factor as the common source of shared variance. In the correlated five-factor model, five factors were specified based on results from Study 1, and each item was specified to load onto one of the five factors. These five factors were specified to correlate with each other. In the higher-order factor model, items loaded onto five lower-order factors, which shared a common source of variance as a function of the higher order factor. Finally, in the bifactor model, all items loaded onto a general factor, and residual variance was explained by five orthogonal first-order factors. Differences in CFI and RMSEA were used to directly compare the relative fit across nonnested models.

Results and discussion

Table 2 summarizes the model fit indices for the five correlated first-order factors structure (M1) from a CFA. The model fit indices for a single-factor model (M2) and a higher-order factor structure (M3) are also presented in Table 2. Finally, a bifactor structure (M4) was specified when all items loaded onto a first-order general factor, and then their shared variance was accounted for by five uncorrelated, first-order specific factors. In this bifactor model, the general factor was not permitted to correlate with the specific factors.

Table 2.

Summary of fit indices across four models for subjective well-being.

Model	χ²	df	CFI	TLI	RMSEA (90% CI)	SRMR
Sample 2
M1. Five-factor	1316.15	369	.80	.78	.083 (.078–.088)	.135
M2. Single factor	2184.25	377	.62	.59	.114 (.109–.118)	.095
M3. Higher-order factor	1362.79	374	.79	.78	.084 (.080–.089)	.139
M4. Bifactor	822.61	350	.90	.87	.062 (.056–.067)	.049
Sample 3
M1. Five-factor	1249.02	369	.79	.77	.082 (.077–.087)	.128
M2. Single factor	1804.30	377	.66	.64	.103 (.099–.108)	.081
M3. Higher-order factor	1268.80	374	.79	.77	.082 (.077– .087)	.130
M4. Bifactor	787.47	350	.90	.88	.061 (.055–.066)	.053

Note. N = 371 in Sample 2 and N = 353 in Sample 3. CFI: comparative fit index; RMSEA: root mean squared error of approximation; SRMR: standardized root mean squared residual; TLI: Tucker–Lewis fit index.

Results showed that the five-factor structure with no cross-loadings did not adequately fit the present data, and that the single factor structure also did not fit the WeBS data. Specifying a higher-order factor structure that consisted of five lower-order factors did not improve the model fit. These findings suggested that SWB as measured by the 29-item WeBS should not be considered as comprising a single general factor, contrary to preliminary results from a recent study (Longo et al., 2016). Rather, our results showed that the confirmatory bifactor structure was a more appropriate fit with the data. It appeared that the WeBS can be best understood as having a domain-general SWB factor, and five domain-specific factors that capture the unique aspects of physical, financial, social, hedonic, and eudaimonic well-being. Although the bifactor structure demonstrated the best model fit, the fit indices revealed adequate but less than excellent fit to the data.

Study 3

The goals of this study were to confirm the bifactor structure of SWB in an independent sample, examine the reliability and construct validity of the new measure of well-being, and explore differences in SWB across ethnocultural groups. We hypothesized that the WeBS and its subscales would be positively correlated with life satisfaction and negatively correlated with depression and anxiety symptoms. We expected that these effects would be larger when well-being was measured with overall WeBS scores than when well-being was indicated by each of the five domain scores, given that the WeBS captures overall well-being and the five subscales taps specific domains of SWB.

Method

Participants

Undergraduate students participated in this online survey study (N = 353, 56.2% women, M_age = 20.42). In this ethnically diverse sample, 34.1% self-identified as Euro/White Americans, 37.8% as Asians/Asian Americans, 19.0% as Latinx/Hispanic Americans, 4.3% as Black/African Americans, and 4.3% of the participants endorsed “mixed” or “other” ethnicity. Immigration generation status was as follows: 28.1% first-generation immigrants, 19.3% second-generation immigrants, 12.3% third-generation immigrants, and 39.9% fourth-generation immigrants. Participants also reported their annual household income: 15.0% under $40,000, 12.4% $40,001–60,000, 17.0% $60,001–100,000, and 20.6% over $100,001.

Measures

Participants were administered a demographic questionnaire, the 29-item WeBS, and the following self-report measures.

SWLS (Diener et al., 1985)

The SWLS is a 5-item measure of life satisfaction. Items are placed on a Likert-type scale from 1 (completely disagree) to 7 (completely agree). The scale has demonstrated adequate psychometric properties, including reliability and construct validity (Diener et al., 1985; Pavot, Diener, & Suh, 1998). Internal consistency for the current sample was adequate (Cronbach’s α = .88).

Center for Epidemiological Studies–Depression Scale (CES-D; Radloff, 1977)

The CES-D is a reliable 20-item checklist of depression symptoms used around the world. Respondents rated the frequency of experiencing depression symptoms in the previous week (e.g., “I felt that everything I did was an effort”) on a scale of 1 (rarely or none of the time) to 4 (most of or all of the time). Several items were reverse scored; higher scores indicated greater depressed mood. Internal reliability for this scale was adequate for the current sample (Cronbach’s α = .88).

State-Trait Anxiety Inventory (STAI; Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983)

The STAI is a well-validated 40-item scale measuring anxiety symptoms as either a reaction to current circumstances (STAI-State, 20 items) or an enduring pattern (STAI-Trait; 20 items). Only the 20-item STAI-State subscale was used for the present study. Respondents rated the degree to which each of the statements described their experiences (e.g., “I feel frightened”) on a scale from 1 (not at all) to 4 (very much so). Higher scores indicated greater state anxiety. Internal consistency of the STAI-S was excellent in the present data (Cronbach’s α = .92).

Procedures

The various factor structures were tested in this sample as in Study 2. Construct validity of the WeBS was examined by correlating the WeBS scores with the SWLS, CES-D, and STAI-S scores. We also examined mean level differences in the WeBS scales across ethnic groups.

Results and discussion

Similar to the results in Study 2, a bifactor CFA solution (M4) yielded better model fit than correlated five-factor (M1), single-factor (M2), and higher-order CFA solutions (M3; see Table 2 for summary of fit indices). Again, the model fit for the bifactor structure was adequate but not excellent with a sample of 353 individuals. Larger sample sizes likely will improve the fit statistics. Table 3 summarizes the item loadings in the bifactor CFA, and Figure 1 depicts the structure that underlies the present WeBS data.

Table 3.

Means, standard deviations, and item loadings of the final confirmatory bifactor model underlying the WeBS data in Sample 3 (N = 353).

Abridged item		M	SD	GF	SF1	SF2	SF3	SF4	SF5
1.	Physical health status	4.76	1.24	.41		.44
2.	Financial resources to meet needs	4.55	1.39	.37	.77
3.	Financial resources to have fun	4.31	1.46	.43	.80
4.	Satisfaction with housing	4.76	1.29	.37	.34
5.	Control of finances	4.18	1.35	.43	.47
6.	Control over physical health	4.77	1.19	.54		.47
7.	Satisfaction with weight	3.84	1.59	.25		.79
8.	Energy to do necessary things	4.37	1.26	.54		.38
9.	Care of physical health	4.47	1.17	.51		.50
10.	Plan for the future	4.98	1.10	.40
11.	People to talk to when have problems	5.13	1.24	.60			.59
12.	Count on friends/family in crisis	5.35	1.02	.56			.63
13.	People who love me	5.57	.87	.51			.46
14.	Solve problems I face	4.91	.96	.63
15.	Life at home	4.91	1.24	.59
16.	Satisfaction with physical appearance	4.34	1.25	.38		.64
17.	Getting along with people	5.13	.94	.66
18.	Spending time with friends/relatives	5.43	.82	.60			.20
19.	Doing fun and interesting things	4.80	1.09	.56
20.	Reaching goals	5.29	.87	.52				.54
21.	Making a difference in other’s lives	4.99	1.00	.46				.58
22.	Meaningful life	5.21	1.06	.45				.40
23.	Satisfaction with spirituality	4.75	1.16	.47				.23
24.	Confidence in self	4.61	1.07	.38				.32
25.	Creativity	4.45	1.15	.28				.43
26.	Stimulating conversations	4.94	1.04	.49				.22
27.	Make me happy	5.29	.94	.60					.25
28.	Feel happy often	4.94	1.08	.61					.64
29.	Enjoyment in life	5.15	1.03	.62					.65

Note. Item loadings >.20 are shown, per convention in bifactor modeling. GF: general factor of subjective well-being; SF1: financial well-being; SF2: physical well-being; SF3: social well-being, SF4: eudaimonic well-being; SF5: hedonic well-being; WeBS: Well-Being Scale.

Figure 1.

Confirmatory bifactor model representing subjective well-being (SWB), financial (FW), physical (PW), social (SW), hedonic (HW), and eudaimonic well-being (EW; specific factors) based on the 29-item Well-Being Scale data. See Table 3 for item loadings and the Appendix for all scale items.

We computed the internal consistency reliability of the overall WeBS scores and the five subscale scores. All of these showed adequate to excellent internal consistency in this sample (see Table 4). In addition, we examined the correlations between related constructs and both the overall WeBS scores and subscale scores because of a preliminary support for a bifactor model of well-being (in Studies 2 and 3) and adequate fit of the ESEM with five domain factors (in Study 1). Table 4 summarizes the zero-order correlations between WeBS scales and other related adjustment measures. As hypothesized, the overall WeBS scores and five subscale scores were significantly and positively correlated with life satisfaction self-reports, and significantly and negatively correlated with depression and anxiety symptoms. These results provided additional evidence for the construct validity of the WeBS. It was not surprising that compared to other domains, hedonic well-being was correlated with life satisfaction and yielded the largest effect size, because they both contained cognitive evaluation of overall quality of life. Additionally, eudaimonic well-being was strongly associated with life satisfaction, because having a sense of purpose in life and pursuing one’s potential likely overlap ped with general satisfaction with life circumstances. It is also possible that being content with past and present life events allows individuals to pursue meaning in life and explore ways to self-actualize, as proposed in the Hierarchy of Needs theory (Maslow, 1954; Tay & Diener, 2011).

Table 4.

Means, standard deviations, internal consistency reliability, and correlations among WeBS and related constructs (N = 353).

	M	SD	α	1.	2.	3.	4.	5.	6.	7.	8.
1. WeBS	4.82	.64	.92	–
2. FW	4.45	1.09	.80	.71	–
3. PW	4.42	.93	.81	.77	.46	–
4. SW	5.37	.78	.79	.70	.35	.36	–
5. EW	4.93	.66	.83	.87	.43	.51	.60	–
6. HW	5.13	.89	.85	.80	.48	.47	.62	.72	–
7. SWLS	5.13	1.20	.88	.67	.51	.47	.47	.56	.67	–
8. CES-D	1.87	.49	.88	−.49	−.24	−.33	−.43	−.48	−.50	−.46	–
9. STAI-S	1.99	.60	.92	−.50	−.25	−.33	−.38	−.50	−.51	−.47	.76

Note. All correlations p < .001. WeBS: Well-Being Scale; FW: financial well-being; PW: physical well-being; SW: social well-being; EW: eudaimonic well-being; HW: hedonic well-being; SWLS: Satisfaction With Life Scale; CES-D: Center for Epidemiological Study – Depression Scale; STAI-S: State-Trait Anxiety Inventory – State.

The correlations between WeBS and CES-D/STAI-S were smaller than those between WeBS and SWLS, which provided evidence for the scale’s discriminant validity. SWB scores were expected to share greater variance with life satisfaction than with either depression or anxiety. This is because both WeBS and SWLS captured variant breadth of positive adaptation and optimal functioning, whereas depression and anxiety were indicators of psychological distress, which might not be equivalent to simply diminished well-being. The present findings support the value of examining optimal functioning, wellness, and flourishing because a lack of dysphoria or distress may not reflect doing well or being happy (Lui, Rollock, Chang, Leong, & Zamboanga, 2016; World Health Organization, 1958).

Given the diverse ethnic representation in our sample, we explored differences in the levels of SWB across groups. Table 5 summarizes the means, standard deviations, and mean difference tests of these comparisons. Overall, Euro/White Americans tended to report significantly greater well-being than other ethnocultural groups, which is consistent with some existing research on well-being across ethnic groups (e.g., Rubenstein et al., 2016).

Table 5.

Means and Standard Deviations for WeBS scores across ethnic groups.

	Euro	Black	Latinx	Asian	Other	F
WeBS	5.01^a b (.65)	4.86 (.87)	4.70^a (.61)	4.95^b (.38)	4.82 (.64)	4.60***
FW	4.62^a (1.09)	4.33 (1.12)	4.04^a (1.13)	4.54 (1.00)	4.46 (1.08)	3.62**
PW	4.62^a b (.95)	4.68 (1.12)	4.27^a (1.03)	4.30^b (.84)	4.50 (.68)	2.72*
SW	5.59^c d e (.69)	4.90^c d (1.29)	5.41^d e (.73)	5.16^e (.78)	5.68 (.33)	7.29****
HW	5.31^b c (.81)	4.41^c (1.53)	5.02 (.93)	5.00^b (.87)	5.50 (.59)	3.76**
EW	5.08^b (.68)	5.21 (.69)	4.91 (.66)	4.75^b (.62)	5.05 (.43)	4.95***

Note. N = 353 (n = 120 for Euro/White Americans, 15 for Black/African Americans, 67 for Latinx/Hispanic Americans, 133 for Asian Americans, and 18 for individuals identified as “other” race). WeBS: Well-Being Scale; FW: financial well-being; PW: physical well-being; SW: social well-being; HW: hedonic well-being; EW: eudaimonic well-being.

Significant post hoc comparison between Euro Americans and Latinx Americans.

Between Euro Americans and Asian Americans.

Between Euro Americans and Black Americans.

Between Latinx Americans and Black Americans.

Between Latinx Americans and Asian Americans.

p < .05; **p < .01; ***p < .005; ****p < .001.

General discussion

The purpose of the current investigation was to develop and evaluate the psychometric properties of a new self-report measure that assesses multiple domains of SWB in three ethnically diverse samples of college students. To our knowledge, the WeBS is the first self-report measure that utilizes a theory of well-being and taps all five important domains of well-being and needs satisfaction. This brief measure of SWB demonstrated adequate construct validity and internal consistency reliability. We also showed preliminary findings that this construct could be conceptualized as comprising a broad, general factor, and five domain-specific factors. The exploratory comparisons of the levels of well-being of different ethnocultural groups highlighted the importance for further examination of how SWB may be understood, approached, and expressed in distinct ways among people of various ethnocultural backgrounds.

One of the unique contributions of this investigation is the inclusion of both personal emotions and cognitive evaluations of interpersonal relationships as core elements of SWB. Despite other research documenting the critical value of interpersonal relationships in the knowledge base of SWB (e.g., Cooke et al., 2016), this investigation is one of the few empirical studies that incorporates social well-being in a multidimensional assessment of this construct. Although social well-being may be a more fundamental part of life and meaningful to people from more collectivistic cultures, our findings show that social well-being is important to the definition of SWB among groups that may be more individualistic as well.

Model comparisons of the WeBS data suggest that SWB comprises a first-order general factor, and five first-order specific factors assessing physical, financial, social, hedonic, and eudaimonic well-being domains. On the one hand, SWB is typically experienced as a global experience of optimal functioning and flourishing. On the other hand, there are five distinct domains that contribute to and indicate SWB, including physical well-being (i.e., subjective experience of physical health and fulfillment of physiological needs), financial well-being (i.e., satisfaction with financial and material resources), social well-being (i.e., social connectedness and support), hedonic well-being (i.e., subjective feelings of balanced affect and evaluation of life), and eudaimonic well-being (i.e., a sense of meaningful life and potential to reach one’s goals). Researchers and healthcare professionals may find it helpful to differentiate between these domains of well-being and assess how they each contribute to a person’s overall functioning. Because the WeBS is designed to be a multidimensional, comprehensive self-report measure of SWB, people interested in assessing this construct could rely on this single measure rather than administer multiple unidimensional scales. This could reduce the burden on respondents and minimize bias in comparing levels of well-being domains and error due to measurement effects.

Future research should continue to cross-validate this bifactor structure of SWB in other samples of college students and nonstudents to confirm whether SWB would be best characterized as having a first-order breadth factor and items also reflect five first-order domain-specific factors (Jovanović, 2015). Given the tentative bifactor structure underlying the WeBS items, it remains unclear how to best make meaning of the domain subscale scores once the items’ shared variance is accounted for by the first-order general factor (see discussion in Lui & Rollock, 2017, Reise, Bonifay, & Haviland, 2013). Researchers, therefore, should be careful in their interpretion of the WeBS overall score and/or the domain subscale scores.

There are some limitations that should be considered in these studies. First, the cross-sectional nature of this study limits our understanding of how various domains of SWB change over time. Longitudinal research should examine how time and life circumstances affect SWB. Second, more work is needed to examine the psychometric properties of the scale in different samples and settings. Particularly, the scale should be cross-validated in older, noncollege community samples, and clinical samples. We expect the scale to remain reliable and valid in other community samples. Future research should aim to examine how various SWB domains are associated with other adjustment outcomes such as psychological distress, depression, and substance use. Third, our preliminary ethnic comparisons should be replicated in larger samples, particularly those including Black/African Americans.

Additionally, caution should be exercised in over-interpreting ethnic differences in the mean WeBS scores obtained in this investigation, because they may be related to response style differences across ethnic groups. Studies have consistently found that people of Euro/White American backgrounds are more likely to use extreme responses, whereas people from more collectivistic cultures such as Asian Americans tend to use midpoints on a Likert-type scale (cf. Rollock & Lui, 2016). Ethnic differences in SWB also may be attributable to socioeconomic differences and other health statuses (Chang et al., 2014). Furthermore, future cross-validation studies should examine whether the current bifactor structure underlying the WeBS data is equivalent across ethnocultural groups in multigroup analyses. Similarly, the present WeBS data showed promise in its ability to assess SWB across various domains; future studies should aim to further validate its factor structure (e.g., measurement invariance across gender) and examine its psychometric properties across sociocultural contexts. This investigation supports the important role of ethnicity in well-being research (Rao & Donaldson, 2015) by intentionally recruiting multiple ethnically diverse samples and exploring group differences in the level of SWB. As the U.S. population becomes more diverse in immigration status, ethnicity, and generation status, researchers should study how acculturation and cultural group membership affect the level of SWB and its correlations with other variables (Schwartz et al., 2012).

Keeping in mind the limitations of this group of studies, this present investigation nevertheless demonstrates that an effective measure of well-being can be developed utilizing a theoretical framework and being mindful of how SWB is conceptualized in diverse cultures. We believe that the WeBS has demonstrated its usefulness as a measure of well-being in both research and applied settings, and with people from different cultures. We hope that additional studies using the WeBS will be conducted so that there is broader, more comprehensive assessment of well-being when researchers attempt to study SWB.

Footnotes

Appendix

Article Notes

References

Ahuvia

(2001) Well-being in cultures of choice: A cross-cultural perspective. American Psychologist 56: 77–78. doi:10.1037/0003-066X.56.1.77.

Arber

Fenn

Meadows

(2014) Subjective financial well-being, income and health inequalities in mid and later life in Britain. Social Science & Medicine 100: 12–20. doi:10.1016/j.socscimed.2013.10.016.

Bentler

P. M.

(1990) Comparative fit indexes in structural models. Psychological Bulletin 107: 238–246. doi:10.1037//0033-2909.107.2.238.

Breslin

McCay

(2012) Perceived control over physical and mental well-being: The effects of gender, age and social class. Journal of Health Psychology 18: 38–45. doi:10.1177/1359105312438600.

Chang

T. E.

Weiss

A. P.

Marques

Baer

Vogeli

Trinh

N. T.

Yeung

A. S.

(2014) Race/ethnicity and other social determinants of psychological well-being and functioning in mental health clinics. Journal of Health Care for the Poor and Underserved 25: 1418–1431. doi:10.1353/hpu.2014.0138.

Chen

F. F.

Bai

Lee

J. M.

Jing

(2016) Culture and the structure of affect: A bifactor modeling approach. Journal of Happiness Studies 17: 1801–1824. doi:10.1007/s10902-015-9671-3.

Chen

Jing

Hayes

Lee

(2013) Two concepts or two approaches? A bifactor analysis of psychological and subjective well-being. Journal of Happiness Studies 14: 1033–1068. doi: 10.1007/s10902-012-9367-x.

Chen

F. F.

Hayes

Carver

C. S.

Laurenceau

J. P.

Zhang

(2012) Modeling general and specific variance in multifaceted constructs: A comparison of the bifactor model to other approaches. Journal of Personality 80: 219–251. doi: 10.1111/j.1467-6494.2011.00739.x.

Comrey

A. L.

Lee

H. B.

(1992) A first course in factor analysis, Hillsdale, NJ: Lawrence Erlbaum Associates.

10.

Cooke

P. J.

Melchert

T. P.

Connor

(2016) Measuring well-being: A review of instruments. The Counseling Psychologist 445: 730–757. doi: 10.1177/0011000016633507.

11.

de Bruin

G. P.

du Plessis

G. A.

(2015) Bifactor analysis of the mental health continuum–Short form (MHC–SF). Psychological Reports 116: 438–446. doi:10.2466/03.02.PR0.116k20w6.

12.

Diener

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

13.

Diener

Biswas-Diener

(2009) Will money increase subjective well-being? A literature review and guide to needed research. In: Diener

(ed.) The science of well-being: The collected works of Ed Diener, New York, NY: Springer, pp. 119–154.

14.

Diener

Emmons

R. A.

Larsen

R. J.

Griffin

(1985) The Satisfaction With Life Scale. Journal of Personality Assessment 49: 71–75. doi:10.1207/s15327752jpa4901_13.

15.

Diener

Fujita

(1995) Resources, personal strivings and subjective well-being: A nomothetic and idiographic approach. Journal of Personality and Social Psychology 68: 926–935. doi:10.1037/0022-3514.68.5.926.

16.

Diener, E., Heintzelman, S. J., Kushlev, K., Tay, L., Wirtz, D., Lutes, L. D., & Oishi, S. (2016). Findings all psychologists should know from the new science on subjective well-being. Canadian Psychology. Advance online publication. doi:10.1037/cap0000063.

17.

Diener

Oishi

(2005) The nonobvious social psychology of happiness. Psychological Inquiry 16: 162–167. doi:10.1207/s15327965pli1604_04.

18.

Diener

Suh

Lucas

Smith

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

19.

Frisch

M. B.

Cornell

Villanueva

Retzlaff

P. J.

(1992) Clinical validation of the Quality of Life Inventory: A measure of life satisfaction for use in treatment planning and outcome assessment. Psychological Assessment 4: 92–101. doi:10.1037/1040-3590.4.1.92.

20.

Harding

S. D.

(1982) Psychological well-being in Great Britain: An evaluation of the Bradburn Affect Balance Scale. Personality and Individual Differences 3: 167–175. doi: 10.1016/0191-8869(82)90031-9.

21.

Hinkin

T. R.

(1998) A brief tutorial on the development of measures for use in survey questionnaires. Organizational Research Methods 1: 104–121. doi:10.1177/109442819800100106.

22.

House

J. S.

Umberson

Landis

K. R.

(1988) Structures and processes of social support. Annual Review of Sociology 14: 293–318. doi:10.1146/annurev.so.14.080188.001453.

23.

Bentler

P. M.

(1999) Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling 6: 1–55. doi:10.1080/10705519909540118.

24.

Huta

Ryan

R. M.

(2010) Pursuing pleasure or virtue: The differential and overlapping well-being benefits of hedonic and eudaimonic motives. Journal of Happiness Studies 11: 735–762. doi:10.1007/s10902-009-9171-4.

25.

Johnson, W., & Krueger, R. F. (2006). How money buys happiness: Genetic and environmental processes linking finances and life satisfaction. Journal of Personality and Social Psychology, 90, 680–691. doi:10.1037/0022-3514.90.4.680.

26.

Jovanović

(2015) A bifactor model of subjective well-being: A re-examination of the structure of subjective well-being. Personality and Individual Differences 87: 45–49. doi:10.1016/j.paid.2015.07.026.

27.

Keyes

C. L. M.

(1998) Social well-being. Social Psychology Quarterly 61: 121–140. doi: 10.2307/2787065.

28.

Kuykendall

Tay

(2015) Leisure engagement and subjective well-being: A meta-analysis. Psychological Bulletin 141: 364–403. doi:10.1037/a0038508.

29.

Lane

J. A.

Fink

R. S.

(2015) Attachment, social support satisfaction, and well-being during life transition in emerging adulthood. The Counseling Psychologist 43: 1034–1058. doi:10.1177/0011000015592184.

30.

Lehman

A. F.

(1983) The well-being of chronic mental patients: Assessing their quality of life. Archives of General Psychiatry 40: 369–373. doi: 10.1001/archpsyc.1983.01790040023003.

31.

Longo

Coyne

Joseph

Gustavsson

(2016) Support for a general factor of well-being. Personality and Individual Differences 100: 68–72. doi:10.1016/j.paid.2016.03.082.

32.

Luhmann

Hofmann

Eid

Lucas

R. E.

(2012) Subjective well-being and adaptation to life events: A meta-analysis on differences between cognitive and affective well-being. Journal of Personality and Social Psychology 102: 592–615. doi: 10.1037/a0025948.

33.

Lui, P. P., & Rollock, D. (2017). Greater than the sum of its parts: Development of a measure of collectivism among Asians. Cultural Diversity and Ethnic Minority Psychology. Advanced online publication. doi: 10.1037/cdp0000163.

34.

Lui

P. P.

Rollock

Chang

E. C.

Leong

F. T. L.

Zamboanga

B. L.

(2016) Big 5 personality and subjective well-being in Asian Americans: Testing optimism and pessimism as mediators. Asian American Journal of Psychology 7: 274–286. doi:10.1037/aap0000054.

35.

Maher

J. P.

Pincus

A. L.

Ram

Conroy

D. E.

(2015) Daily physical activity and life satisfaction across adulthood. Developmental Psychology 51: 1407–1419. doi:10.1037/dev0000037.

36.

Marsh

H. W.

Balla

J. R.

McDonald

R. P.

(1988) Goodness-of-fit indices in confirmatory factor analysis: The effect of sample size. Psychological Bulletin 103: 391–410. doi:10.1037/0033-2909.103.3.391.

37.

Maslow

A. H.

(1954) Motivation and personality, New York, NY: Harper and Row.

38.

Muthén

L. K.

Muthén

B. O.

(1998–2014) Mplus user’s guide, Los Angeles, CA: Muthén & Muthén.

39.

O’Conner

B. P.

(2000) SPSS and SAS programs for determining the number of components using parallel analysis and Velicer’s MAP test. Behavioral Research Methods, Instrumentation, and Computers 32: 396–402. doi:10.3758/BF03200807.

40.

Pavot

Diener

Suh

(1998) The temporal satisfaction with life scale. Journal of Personality Assessment 70: 340–354. doi:10.1207/s15327752jpa7002_11.

41.

Pew Research Center (2012). Explaining why minority births now outnumber White births. Retrieved from http://www.pewsocialtrends.org/2012/05/17/explaining-why-minority-births-now-outnumber-white-births/.

42.

Radloff

L. S.

(1977) The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement 1: 385–401. doi:10.1177/014662167700100306.

43.

Raibley

J. R.

(2012) Happiness is not well-being. Journal of Happiness Studies 13: 1105–1129. doi:10.1007/s10902-011-9309-z.

44.

Rao

M. A.

Donaldson

S. I.

(2015) Expanding opportunities for diversity in positive psychology: An examination of gender, race, and ethnicity. Canadian Psychology 56: 271–282. doi:10.1037/cap0000036.

45.

Reise

S. P.

(2012) The rediscovery of bifactor measurement models. Multivariate Behavioral Research 47: 667–696. doi: 10.1080/00273171.2012.715555.

46.

Reise

S. P.

Bonifay

W. E.

Haviland

M. G.

(2013) Scoring and modeling psychological measures in the presence of multidimensionality. Journal of Personality Assessment 95: 129–140. doi: 10.1080/00223891.2012.725437.

47.

Reise

S. P.

Morizot

Hays

(2007) The role of the bifactor model in resolving dimensionality issues in health outcomes measures. Quality of Life Research 16: 19–31. doi:10.1007/s11136-007-9183-7.

48.

Rollock, D., & Lui, P. P. (2016). Measurement invariance and the five-factor model of personality: Asian international and Euro American cultural groups. Assessment, 23, 571–587. doi:10.1177/1073191115590854.

49.

Rubenstein

C. L.

Duff

Prilleltensky

Jin

Dietz

Myers

N. D.

Prilleltensky

(2016) Demographic group differences in domain-specific well-being. Journal of Community Psychology 44: 499–515. doi:10.1002/jcop.21784.

50.

Ryan

R. M.

Deci

E. L.

(2000) Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist 55: 68–78. doi:10.1037/0003-066X.55.1.68.

51.

Ryan

R. M.

Deci

E. L.

(2001) On happiness and human potentials: A review of research on hedonic and eudaimonic well-being. Annual Review of Psychology 52: 141–166. doi:10.1146/annurev.psych.52.1.141.

52.

Ryff

C. D.

(1989) Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of Personality and Social Psychology 57: 1069–1081. doi:10.1037/0022-3514.57.6.1069.

53.

Ryff

C. D.

Singer

B. H.

(1996) Psychological well-being: Meaning, measurement, and implications for psychotherapy research. Psychotherapy and Psychosomatics 65: 14–23. doi:10.1159/000289026.

54.

Ryff

C. D.

Singer

B. H.

(2006) Best news yet on the six-factor model of well-being. Social Science Research 35: 1103–1119. doi:10.1016/j.ssresearch.2006.01.002.

55.

Sarason

B. R.

Sarason

I. G.

Gurung

R. A. R.

(2001) Close personal relationships and health outcomes: A key to the role of social support. In: Sarason

Duck

(eds) Personal relationships: Implications for clinical and community psychology, Chichester, England: Wiley, pp. 15–42.

56.

Seligman

M. E. P.

(2002) Authentic happiness, New York, NY: Free Press.

57.

Slade

(2010) Mental illness and well-being: The central importance of positive psychology and recovery approaches. BMC Health Services Research 10: 26, Retrieved from http://www.biomedcentral.com/1472-6963/10/26/.

58.

Spielberger

C. D.

Gorsuch

R. L.

Lushene

P. R.

Vagg

P. R.

Jacobs

A. G.

(1983) Manual for the STAI (Form Y), Palo Alto, CA: Consulting Psychologists Press.

59.

Schwartz

S. J.

Weisskirch

R. S.

Hurley

E. A.

Zamboanga

B. L.

Park

I. J.

Kim

S. Y.

Greene

A. D.

(2010) Communalism, familism, and filial piety: Are they birds of a collectivist feather? Cultural Diversity and Ethnic Minority Psychology 16: 548–560. doi: 10.1037/a0021370.

60.

Schwartz

S. J.

Waterman

A. S.

Umana-Taylor

A. J.

Lee

R. M.

Kim

S. Y.

Vazsonyi

A. T.

Williams

M. K.

(2012) Acculturation and well-being among college students from immigrant families. Journal of Clinical Psychology 69: 298–318. doi:10.1002/jclp.21847.

61.

Tao

F. B.

M. L.

Kim

S. D.

Sun

P. Y.

Huang

(2007) Physical activity might not be the protective factor for health risk behaviours and psychopathological symptoms in adolescents. Journal of Paediatrics and Child Health 43: 762–767. doi:10.1111/j.1440-1754.2007.01217.x.

62.

Tay

Diener

(2011) Needs and subjective well-being around the world. Journal of Personality and Social Psychology 101: 354–365. doi:10.1037/a0023779.

63.

Turner

Marino

(1994) Social support and social structure: A descriptive epidemiology. Journal of Health and Social Behavior 35: 193–212. doi:10.2307/2137276.

64.

Velicer

W. F.

Eaton

C. A.

Fava

J. L.

(2000) Construct explication through factor or component analysis: A review and evaluation of alternative procedures for determining the number of factors or components. In: Goffin

R. D.

Helmes

(eds) Problems and solutions in human assessment, Boston, MA: Kluwer, pp. 41–71.

65.

Vella-Brodrick

D. A.

Allen

F. C. L.

(1995) Development and psychometric validation of the Mental, Physical, and Spiritual Well-Being Scale. Psychological Reports 77: 659–674. doi: 10.2466/pr0.1995.77.2.659.

66.

Ware

Gandek

IQOLA

Project

(1994) The SF-36 Health Survey: Development and use in mental health research and the IQOLA Project. International Journal of Mental Health 23: 49–73.

67.

Waterman

A. S.

(2007) Doing well: The relationship of identity status to three conceptions of well-being. Identity 7: 289–307. doi:10.1080/15283480701600769.

68.

Watson

Clark

L. A.

Tellegen

(1988) Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology 54: 1063–1070. doi:10.1037/0022-3514.54.6.1063.

69.

Wiest

Schüz

Webster

Wurm

(2011) Subjective well-being and mortality revisited: Differential effects of cognitive and emotional facets of well-being on mortality. Health Psychology 30: 728–735. doi:10.1037/a0023839.

70.

World Health Organization (1958) Constitution of the World Health Organization, Annex, Geneva, Switzerland: Author.