SWB is defined as the affective and cognitive evaluation people make about how happy and satisfied they are with their lives [9] [10]. Within psychology, there are cumulative studies supporting the view of SWB as relatively stable and moderately positive (e.g. [11] [12]). One of the first studies to indicate this stable positivity [13] combined the population means from 16 life satisfaction studies conducted in Western nations. Data were standardized to a statistic called the percentage of scale maximum (%SM) which converts scale scores into a range from 0 to 100. Using the mean values from each survey as data yielded a mean of 75 points and a standard deviation (SD) of 2.5. Hence, the range 70% - 80% SM included all of the means from these disparate studies. However, the subsequent inclusion of survey means from non-Western nations revealed that the mean of SWB was 70 (SD = 5), thus causing the normative range to expand downward, to 60% - 80% SM [14]. This downward expansion is due to two different influences. One is the effect of poverty and disadvantage in economically underdeveloped countries. The second is the downward influence of cultural response bias, acting particularly within Confucian-based cultures [15].

In order to explain the stable positivity of SWB as described, a revised homeostatic model of SWB is used (see Figure 1 below):

The affective factor of homeostatically protected mood (HPMood) is a construct that evolved from Russell’s [16] conception of core affect. Similar to core affect, HPMood exists without reference to objects or events [17]. It is proposed as a genetically determined, constant positive affective background, that pervades many thought processes but most especially those that are evaluative of personal and general characteristics. The archetypical form of such evaluative is “How satisfied are you with your life as a whole?” It is found that the response to this item to be heavily saturated with HPMood [18]. Based on the above conceptualization and use of structural equation modelling, Davern et al. [18] found HPMood to comprise three affects as: content, happy and excited. They also proposed that HPMood is the basis of SWB set-point and positive mood that is defended by homeostasis. The domination of SWB by HPMood has also been confirmed by [19] [20].

Regarding the three cognitive buffers of self-esteem, optimism and perceived control, they are not only strongly perfused with HPMood but also have a cognitive component which is responsible for adjusting each buffer in order to defend HPMood [21]. That is, they are in intimate interaction with momentary experience and assist in the process of defending HPMood against life experiences [22]. It is proposed that each of the cognitive buffers; namely, self-esteem, optimism and perceived control assists in the process of defending HPMood against life experiences [21]. Thus, people with high self-esteem sustain their wellbeing more effectively than those with low self-esteem [23] [24] and optimists buffer against adversity by holding the global expectation that current difficulties will not last and things will get better with time [25] [26]. Additionally, when facing threatening events, people with high perceived control tend to hold positive belief about their ability to change a situation and achieve a desirable outcome [27] [28]. In short, the homeostatic model proposes that SWB is the output of interaction between HPMood, cognitive buffers and experiential factors. In this study, CSE will be added as a cognitive buffer in the homeostatic model and it is predicted that CSE will contribute unique SWB variance beyond the variance accounted for by the homeostatic factors.

Self-esteem emphasizes that feelings of self-worth and self-respect are derived from, or related to, an individual’s personal attributes, competencies and standing, relative to other individuals [6]. Contrarily, CSE refers to the feelings and evaluations of self-worthiness based on the social groups, such as ethnic or work groups, of which one is a member. Hence, CSE is the value people place on themselves as members of their social groups [29]. The distinctiveness between these two concepts can be illustrated by their relative influence, as the dominant source of worthiness; in different cultures i.e. individualism and collectivism.

In the individualist societies, people’s identity is derived from individual attributes. This kind of personal identity makes individuals view themselves as separate and autonomous entities. People are therefore emotionally independent from groups and the guiding principle for conduct is their individual interest [30] [31]. However, in the collectivist societies, individuals derive their identity from their social groups. Their social identity, so derived, lets them view themselves as interconnected and embedded in interdependent social relationships. Hence, people are emotionally dependent on collectivities and priority is given to the collective interest [30] [31].

In light of the differential characteristics of individualist and collectivist societies, it seems that self-esteem should be the dominant source of felt worthiness for the individuals in individualist societies, as people emphasize individual self and the self-concept mainly relates to personal identity. Indeed, this claim has been empirically supported [6] [7]. Contrarily, CSE should have much higher influence in collectivist societies than individualist societies as the source of worthiness, since these people are more concerned about the collective self and their self-concept is primarily in association with social identity. In support of this, it has been demonstrated that for those people who live within the collectivist cultures, they are more likely to derive their sense of worthiness from the collective nature of social identity [5].

In short, this study intends to examine the relative contribution of self-esteem and CSE to SWB by embedding in the homeostatic model, for the Chinese in Hong Kong (HK) as well as the 1^st generation Chinese immigrants (AU1) and 2^nd generation Chinese immigrants (AU2) in Australia. Given Hong Kong is predominantly a Chinese society while Australia is a Western one; it is likely that CSE will be more important than self-esteem in contributing to SWB of Hong Kong Chinese. Additionally, compared with AU1 who migrated to Australia in the later part of their life, CSE will be less important than self-esteem in contributing to SWB for AU2, as this group is likely to be exposed to higher influence of Western culture during their growth in Australia. Hence the hypothesis to be tested is that: CSE will explain unique SWB variance beyond the other homeostatic model factors, and the proportion of unique SWB variance contributed by CSE will be the highest for HK, intermediate for AU1 and lowest for AU2. However, the proportion of independent contribution made by self-esteem will be in reverse order for these three groups.

Through convenience sampling, three groups of samples were drawn as: HK (N = 716), AU1 (N = 204) and AU2 (N = 134). The composition of participants is summarized below.

In terms of gender distribution, except for AU2 which was slightly dominated by males (male: 53.44%), the other groups were more female dominant [HK (female: 59.68%); AU1 (female: 64.47%)]. In terms of age, the majority of HK (57.81%) and AU1 (44.78%) were in middle adulthood (36 - 55 years), while over half of AU2 (53.38%) were aged 17 or below. Three income levels were used for both the Australian participants (Low: AU$30,999 or less; Medium: AU$31,000 - $60,000; High: AU$60,001 and above) [32] and Hong Kong participants (Low: HK$14,999 or less; Medium: HK$15,000 - $20,000; High: HK$20,001 and above) [33]. The income categories correspond with the standard income levels of each country. Regarding income distribution, nearly half of HK (49.62%), AU1 (43.11%), AU2 (42.99%) belonged to the high income group.

The Chinese participants in both Hong Kong and Australia completed a questionnaire using an 11-point end-de- fined scale for all items.

Subjective Wellbeing (SWB): This was measured by the Personal Wellbeing Index [34]. The scale comprises seven items measuring satisfaction with domains as: standard of living, health, achieving in life, relationships, safety, community-connectedness and future security. The reliability coefficient for this study was 0.88.

HPMood: The three affective predictors-content, happy and excited were measured by asking participants to indicate how each of them described their feelings when they thought about their life in general. The coefficient alpha of 0.85 was obtained in this study.

Self-Esteem: The ten-item Rosenberg Self-Esteem Scale [1] was used and the coefficient alpha in this study was 0.78.

Optimism: The Life Orientation Test-Revised (LOT-R) [35] comprises six questions framed either in an optimistic or pessimistic fashion. This survey only used the three optimistically framed questions. In this study, the reliability coefficient reported for this three-item scale was 0.79.

Perceived Control: A six-item scale was developed as a measure of primary and secondary control, which was extracted from [36] and originally consists of nine items, including a three-item measure of relinquished control. In this study, using the six-item version, an alpha of 0.83 was obtained.

Experiential Input: Life events were measured by asking participants whether anything had happened recently that caused them to feel happier or sadder than normal. Participants were asked to respond to three categories of response: “yes, happier” = 3, “no” = 2 and “yes, sadder” = 1.

Collective Self-esteem (CSE): The sixteen-item CSE scale [4] was used which is comprised of four subscales. In this study, the coefficient alphas for the subscales were: the Membership Esteem subscale (alpha = 0.66), the Private CSE subscale (alpha = 0.73), the Public CSE subscale (alpha = 0.72), and the Importance to Identity subscale (alpha = 0.60). The total scale alpha was 0.84.

Individualism-Collectivism: The eight-item Individualism-Collectivism Scale [37] was used, with a high score indicates collectivistic belief and a low score reflects individualistic belief. In this study, an alpha of 0.68 was obtained.