A framework for identifying perceived Quality of Life indicators for the elderly in the neighbourhood context: a case study of Kolkata, India

Review of existing research reveals sporadic explorations of the vast concept of perceived Quality of Life (QoL) of elderly. Assessment of QoL of elderly in South Asian countries have been so far based on measurement models developed on western ideologies. The distinctively dissimilar perspectives and philosophies of the elderly living in South Asian countries seeks for development of QoL models from their perspective. The importance of the neighbourhood environment in the lives of the elderly and the preference of the elderly to age-in-place, also necessitates research on improving the QoL of the elderly in the context of their residential environment. This paper, exploring the perspectives of a group of Indian elderly, describes the stages of identification of a set of factors (components) and a broad set of domains (latent constructs which these factors cater to), influencing perceived QoL of Indian elderly to promote ageing-in-place. The methodology combines qualitative and quantitative research techniques, which includes literature review, followed by focus group discussions and close-ended interviews with Indian elderly from varied socio-demographic backgrounds. The present study, using Principal Component Analysis, identifies seven broad domains and 37 QoL factors influencing perceived QoL of Indian and elderly, which can be further validated on a larger sample for the development of a QoL index. The structure of domains and factors, identified for the Indian elderly varies considerably from that identified from literature. The study concludes with an exploration of the prioritisation of the identified domains among different socio-demographic groups.


Introduction
The elderly in South Asian countries, including India, have traditionally, preferred to live with their family and receiving home or family based care. Urbanisation has led to changes in the traditional family structure and has shifted youth to work based locations away from home. As a consequence of this, the traditional multi-generational Indian family structure vice to elders and the need for respect for elders and the older generation are emphasized in the existing religious and cultural systems in India (North and Fiske 2015).
In the South Asian context, despite the growing need for research on various aspects influencing the well-being of their elderly, studies on their perceived QoL is still a comparatively less explored area of research. In context of the South Asian countries, the limited existing studies on the perceived QoL of elderly have primarily used existing established instruments of QoL, developed in the US or European contexts. There is a significant dearth of original studies investigating the perceived needs and perspectives of the elderly of these countries and the heterogeneity in the perception among the myriad groups of these elderly. Some general tendencies observed amongst the elderly of these countries reveal their preference for lesser involvement in intensive activities, acceptance of death in general, and adherence to the system of inter-dependence instead of independence. These tendencies indicate a difference in their perspective from that of some of prevalent models of ageing in the western countries, which are primarily based on productivity and independence. Since, the senior citizens in these South Asian countries differ from the elderly in the western countries on their perspective to ageing (Brijnath 2012;Lamb 2014), the western models might not prove to be very effective in the assessment of their perceived QoL.
Kolkata, being one of the oldest South Asian cities, housing a heterogeneous set of population, has been selected as a study area for this research. The selection of Kolkata as a representative of an Indian city for the study of the differences in the perceptions of older adults in western and south-east Asian nations was also observed in Giles et al. (2010). Kolkata, having experienced various waves of in and out migration over the past decades, is presently, home to substantially large groups of a heterogeneous mix of older adults, some of whom live alone and some who live with their adult children and families (Gangopadhyay 2019). Besides, the conventional acceptance of "disengagement' and inter-dependence in old age among the elderly in south east Asian countries, recent studies on Indian elderly in Kolkata, have also discussed about the emergence of a new group of Indian elderly who believe in leading active, independent lives (Gangopadhyay 2019). Since, Kolkata, presently houses both the groups of elderly who adhere to the traditional system of inter-dependence and also a new emerging group of those who strive for activity and independence, it would be interesting to explore how these differences in attitudes influence the perception of the dimensions of well-being. However, the study accepts the smaller sample size, as a limitation in the applicability of the generalizability of the findings of the present study, in the context of the entire south east Asian countries. The study framework, however, can be further explored on a larger context and on a larger sample to arrive at a more generalized set of findings.
The Maintenance and Welfare of Parents and Senior Citizens (MWPSC) Act, 2007, published by the Ministry of Law and Justice, India, defines senior citizens as any person who is a citizen of India and has attained the age of '60 years and above'. In this study, we have referred to 'senior citizens' as elderly, older adults and older persons and our study respondents belong to the age group of 60 years and above. The population of India was expected to increase from 121.1 crores to 152.2 crores during the period 2011-2036 -an increase of 25.7% in twenty-five years at the rate of 1.0% (Ministry of Health and Family Welfare, Govt. of India, 2020). The number of older persons in India was expected to increase by more than double from 2011 to 23 crores in 2036 -an increase in their share to the total population from 8.4 to 14.9% annually (Ministry of Health and Family Welfare, Govt. of India, 2020).
A dearth in studies focusing specifically on the understanding of QoL from the perspective of the Indian elderly and the growing requirement for provision of home and neighborhood based care of the elderly, establish the need for a study. Considering the need to develop an assessment model specifically for the needs of the Indian elderly, the present study has formulated the following two objectives: a) First, identification of the broad domains and their respective set of indicators or subdomains or factors, which can be used for the development of a QoL index for assessment of QoL of Indian elderly. Considering the importance of aging-in-place, the present study attempts to identify only the factors of QoL which can be catered to by the neighborhood environment.
b) Second, understanding how the prioritisation of QoL domains varies across different socio-demographic groups. The understanding of the variation in prioritisation can be useful in the development of policies targeted for specific groups. The complete methodology of this paper can be divided into five steps.
i. Compiling an initial list of broad domains associated with the QoL of elderly, identified from the study of the items used for the assessment of QoL of elderly in existing scales or instruments. (discussed in Section 2.1).
ii. Identification of an initial list of factors associated with each of these domains from the study of QoL instruments, theories on ageing, research on older adults, and studies on Indian elderly. (discussed in Section 2.2).
iii. A focus group using the broad domains identified in the previous step (Section 2.1) to update and validate the identified list of factors (identified in Section 2.2), based on the perspectives of the Indian elderly. (discussed in Section 2.3).
iv. Refinement and item reduction through discussion and identification of a final list of factors influencing the QoL of Indian elderly from the results obtained from the previous steps (Section 2.2 and 2.3). (discussed in Section 2.4).
v. The final round of survey conducted on 408 Indian elderly respondents across all socio-demographic groups to arrive at the final set of domains and their respective set of factors influencing the perceived QoL of Indian elderly (survey process and details discussed in Section 2.5). The technique adopted for the analysis of the responses for identification of the final list of QoL domains, obtained from the final round of survey has been discussed in Sect. 2.5.4. and that for prioritisation of domains according to socio-demographic groups have been discussed in Sect. 2.5.5. Figure 1 demonstrates the sequence of steps adopted for this study. The following subsections shall describe in detail the process (methodology) followed for collection of data or information, and analysis of the same for each step. Section 3 shall discuss the results and observations from the survey.

QoL scales or measures or instruments.
The multi-dimensionality of the concept of QoL has led to a lack of consensus on a fixed list of domains that could be considered to influence the same in old age. The significance of health related aspects like health status, and functional ability in contributing to the QoL of elderly has however been accepted widely among researchers (Rojo-Pérez, Fernández-Mayoralas, Rodríguez-Rodríguez, & Rojo-Abuín, 2007). Recent studies have also explored the concept in the context of the physical environment, and architectural design as the physical environment have been observed to influence the QoL of people, especially that of the elderly (Mandzuk and McMillan 2005).
Due to the lack of conceptual clarity in defining QoL, some issues, which have been observed to be associated with regard to the measurement and selection of QoL, are: (i) subjectivity versus objectivity, (ii) generic versus specific, (iii) self-report versus proxy report, and (iv) reliability and validity (Lin et al. 2013;Verdugo et al. 2005).
Subjective indicators for assessment of QoL have been widely used owing to the extensive body of research available on the level of satisfaction across populations and the ability to assess the relative importance of specific QOL domains (Verdugo et al. 2005). However, with respect to some types of QoL, like HRQOL (health-related Quality of Life), Lin et al. (2013) suggests combining both subjective and objective components for measurement in order to assess both the measure of target symptoms and the impact of diseases and symptoms on all the aspects of life. The choice of the subjectivity and objectivity of the indicators in a scale shall therefore depend on the use for which the scale has been designed. In this study, considering the objective of understanding perceived well-being, we have used the subjective approach to defining QoL. The different QoL measurement models or scales or instruments have conceptualized the dimensions contributing to QoL based on varied purposes, scopes, and the context of study. Existing scales can be observed to be of two types: (a) scales which measure several features or aspects of an individual's life (generic) and (b) scales which specifically measure one or more aspects contributing to the QoL of an individual (specific). Selection of generic or specific measures based on depends on the research objectives, and its applications in practice, or policy analysis (Lin et al. 2013). Studies assessing QoL of elderly have also been observed to use both QoL models developed for all groups of adults, and those developed specifically for the elderly. This study has adopted a generic approach for specifically addressing the QoL of Indian elderly as the study attempts to understand the QoL of the Indian elderly from a holistic perspective.
Sometimes due to the limitations of respondents experiencing difficulty in functioning, or for those who are too weak to respond to the questions completely and efficiently or have difficulty in understanding of questions, researchers use proxy respondents, which includes their caregivers or relatives to answer questions on their behalf. However, as observed by Lin et al. (2013), the relation between the self-and proxy reports varies depending on the domains of measurement and the relationships between the proxy and the respondent. Since proxy data can never be considered as an absolute substitute for self-reported data, Verdugo et al. (2005) suggests combining the use of both the types of data, followed by their respective analysis, and test the degree to which there is an agreement between the two types of data. The present study has used only self-reported data.
Reliability and validity of measures refer to the accuracy and ability of the indicators in a measure or the questionnaire to efficiently and accurately measure that which it is designed for.

Components of QoL scales or measures or instruments.
Depending upon the type of questions in existing QoL scales or measures or instruments, they can be broadly classified to be of two types: unidimensional or multi-dimensional. In a unidimensional scale (example, Geriatric Depression Scale-Short Version), questions, also referred to as items (example: 'Are you basically satisfied with your life?') are asked with respect to the overall status of QoL or with respect to the specific aspect for which it is designed. In a multi-dimensional QoL scale (example: WHOQOL-OLD), questions are asked with respect to varied aspects of life, referred to as domains or dimensions (6 domains: Sensory Abilities, Autonomy, Past, Present and Future Activities, Social Participation, Death and Dying and Intimacy). In case of the latter, the purpose of the measure is more holistic (assessment of overall well-being) and can be assumed to consist of a set of dimensions or domains which measure specific aspects related to QoL or well-being, like interpersonal relations, personal development, physical well-being, emotional well-being, etc. Each of these domains are then measured through a set of questions, also referred to as items or indicators or factors (in WHOQOL-OLD, each of the 6 domains consist of four or five questions corresponding to the respective domain, e.g.: 'Opportunities to love' under the domain of 'Intimacy' or 'Afraid of not being able to control death' under the domain of 'Death and dying'). In this study, we have used the term 'factors' for referring to QoL indicators and 'domains' to refer to the broader set that these factors cater to.

Identification of initial list of broad domains of QoL for the present study.
For the present research, QoL instruments were identified from two studies by Halvorsrud and Kalfoss (2007), and Haywood et al. (2005) on QoL measures for older people, a study by Linton et al. (2016) on review of well-being measures on adults and the wellbeing scales such as Flourish Index, Secure Flourish Index, Psychological Well-being Scale and Social wellbeing scale (Saha, Basu & Pandit, 2022). Few other QoL measures were identified from online searches on elderly specific QoL instruments. Approximated 1400 items were extracted from 54 scales. Each of these scales consist of questions that aim at assessment of items or issues pertaining to the specific aspects of QoL the scales investigate. The detailed questionnaires of the instruments were studied to first identify the aspects of life each of the items in the questionnaire dealt with. Next, using the method of I-CVI (Item Content Validity-Index), a total of eight domains, namely, (1) a) Health (physical), b) Health (mental), (2) Social activities and relationships, (3) Leisure activities, (4) Level of independence, (5) Life and Self perceptions affecting emotional well-being, (6) Religious/ spiritual beliefs, (7) Financial stability and (8) Support from neighbourhood or surrounding built environment, were identified (Saha, Basu & Pandit, 2022). Saha, Basu & Pandit (2022) discusses in detail the process for identification of QoL domains from QoL scales using the technique of I-CVI. In this paper, since we are focusing on identifying the health based and socio-psychological factors which can be influenced by the neighborhood environment, we shall discuss in detail about the factors for the QoL domains 1 to 7 and exclude the identification of factors belonging to the domain 'Support from neighbourhood or surrounding built environment', which deals mainly with the infrastructural components of the built environment.

Research on ageing and theories of ageing
Ageing has been viewed from varied perspectives: both positive and negative. An understanding of the underlying concepts of these perspectives shall help us in identifying the characteristics unique to the older adults. This section shall discuss the research and theories on ageing based on the varied perspectives, observed majorly from the western literature.
An individual's perception of age and ageing depends upon their innate characteristics which includes their physical, mental, psychological and social capabilities. Researchers have viewed aging both from optimistic and pessimistic perspectives leading to different types of dualisms in the theories of aging. While few researchers have viewed aging from the perspective of onset of diseases and focused on the associated degradation in quality of physical and social capacities, few others have explored the sense of freedom that one experiences from being relieved of the duties and responsibilities of adulthood, enabling one to indulge in hobbies and other activities of interest (Andrew 2008).
The existing theories on ageing can be classified under the broad domains of biological and psycho-social theories. In this study, we have considered some of the psycho-social theories on ageing, because the focus of the present study is to understand perceived QoL which depends upon the inherent attitudes and perceptions of an individual, and are also influenced by their socio-cultural context.
The theory of Disengagement postulated in the year 1961 states it as natural and acceptable for the older adults to withdraw from the society and decrease interaction with the others in their social system. The Activity theory, on the other hand, is in direct contrast to this theory and states that life satisfaction in the later stages of one's life depends upon the active maintenance of social relationships and involvement in meaningful activities (Andrew 2008). According to Erikson (1959) finding a sense of meaning is very important in the later stages of one's life. The continuity theory presented by Havens and later developed by Atchley (Wadensten 2006) states that the habits and preferences developed by an individual during adulthood becomes a part of their personality and these traits are carried forward in their future stages of life (Wadensten 2006).
In old age, people often indulge in introspection and review their lives, trying to assess the gap between what their life goals and reality, i.e. what they had wished to accomplish and what they have actually been able to accomplish. This helps them to view their life with a deeper sense of meaning and significance (Krause 2009). Tornstam's (2005) theory of Gerotranscendence agrees with the theory of Erikson, and notes that as people age, their perceptions about the world undergoes a change, often shifting from materialism to more spiritual and deeper exploration of one's own inner self (Krause 2009). For contributing to the concept of Successful Aging, MacArthur, Rowe andKahn (1987, 1997) suggested three factors: (a) health (avoiding disease, and maintaining a healthy lifestyle), (b) maximizing intellectual and physical functioning by learning new things that help in improving their cerebral capacity, (c) increasing social engagement by continuing and maintaining their relationships with their family and friends (Zarit 2009).
Among the many theories focusing on the social groups and relations in old age, the convoy model, suggests that the characteristics of social support required by everyone may not be the same (Antonucci, Birditt, & Akiyama, 2009). The objective characteristics of social support alone cannot determine the degree of satisfaction or dissatisfaction with the support (Antonucci et al. 2009). For example, simply having a family cannot ensure that an individual receives the quality and amount of support that he/ she seeks for. An individual's assessment of the support is essential to understand the degree of satisfaction with the support being provided and received.
Other researches on old age include explorations of means to promote integrity in old age. Some of these methods include life review, reminiscence therapy, guided autobiography, etc. (Zarit 2009), which allows people to engage in recollection of their past lives, reminiscing and cherishing their achievements and happy moments.
These theories have provided us insight into the various aspects of ageing and the underlying concepts behind the changes brought about in late life. Researchers have also explored few other aspects influencing the QoL of old age. Health has been observed to be an essential influencing the experience of QoL in old age (Victor 2005). Among various factors associated with health of elderly, falls have been observed to be the most common problem, which is often associated with mortality, morbidity, and deteriorated functioning (Kenny 2005; Walker and Mollenkopf 2007) emphasizes the importance of mobility for maintaining an active late life. Another problem commonly associated with health in elderly is sensory impairment, which includes related to vision, hearing, taste, smell and touch (Margrain & Boulton, 2005). Cognitive skills, which are required for integration and processing of information, have also been observed to decline considerably in old age. Decline in cognitive skills is also associated with increase in levels of depression, and stress (Rabbitt 2005). Decline in memory, Dementia and Alzheimer's are also observed to be a common problem in old age (Maylor 2005). In the context of socialization, it has been observed that elderly maintain relationship with their neighbors through religious functions, and organized group participation (Wong and Waite 2016). It has been observed that the elderly compensate for changes or losses in their personal social networks by becoming more involved in community based social activities (Wong and Waite 2016). Research on leisure activities for elderly have observed that activities that are psychologically appealing and meaningful, contribute to wellbeing in late life (Kleiber 2016). Volunteering or activities which can help in contributing to the society can help in contributing to the improvement in the sense of self-esteem of the elderly (Liechty and Genoe 2007).
Similarly, researchers have explored various other aspects of the lives of the elderly and the factors that impact upon them. A compilation of all the factors observed to influence different aspects of the lives of the elderly, from existing literature, is provided in Table 1.

Research on ageing and QoL of elderly in India
The earliest studies on ageing in India dates back to 'Ayurveda' an ancient Indian study on medicine systems and Manu's 'Dharmasatra' (Ashok & Ali, 2003;Ramamurti 2003), which dealt with medicinal plant based treatments for age related disorders (Ramamurti 2003) (Ashok & Ali, 2003), and the four stages of a man's life and their associated activities, respectively. Dharmasastra suggests disengagement during old age, as one gradually moves into the stage of 'Vanaprastha', disengaging oneself from family and community, renouncing materialistic pursuits (Ashok & Ali, 2003;Ramamurti 2003). Presently, research on ageing in India has progressed into the fields of medical and biological sciences, socio-psychological studies, behavioral sciences, and studies on culture and economic independence and financial assistance for the older adults.
Studies on medical and biological sciences have identified the different ailments associated with old age and issues related to access to health care. Mental health disorders associated with the Indian elderly have been identified as depression, dementia, alzheimer's, etc. (Krishnaswamy et al. 2008;Prakash 1999).
Lack of awareness about health related disorders, passive acceptance of ill-health in old age, extreme financial, and physical dependence on family for access to health care, etc. (Krishnaswamy et al. 2008) and limitations in existing economic resources in catering to expenses for treatments have been observed to be some of the major problems associated with health care of the Indian elderly.
The perception of the elderly in India have been observed to change in modern societies. The traditional Indian family, revered the elderly and consulted them for their wisdom and important family decisions (Krishnaswamy et al. 2008). However, in recent times, they are in some instances, now considered as a burden due to their increased financial and activity based dependence (Prakash 1999). Research on Indian elderly in context of the socio-cultural influences include studies on successful aging, disability, coping, nature and characteristics of centenarians, etc. (Panruti, Liebig, & Duvvuru, 2015). Krishnaswamy et al. (2008) notes that limited financial security is a major problem of the Indian elderly, where a significant percentage of the elderly are observed to be dependent on their families. The primary reason behind the major focus of existing research on the Indian elderly on demographic statistics instead of the socio-cultural influences on the perceptions of QoL (Ladusingh and Ngangbam 2016), can be owed partly to the lack of empirical data on domains influencing subjective well-being and partly to the unawareness about the importance of subjective well-being components in the development of public policy for the older adults.
Demographic influences on QoL of Indian elderly have been explored by various studies (Roy and Das 2011). Age, marital status, gender, educational background, living arrangement, have been observed to influence the QoL of Indian elderly. Gupta, Mohan, Tiwari, Singh, and Singh (2014) observes that, marital status and health status significantly influence the perceived QoL of the elderly. Ghosh (2015) observes in her study that gender plays a significant role in determining perceived quality of life of elderly as females, who are perceived to have a comparatively lower position in the society have lesser expectation and hence a higher perceived QoL. Education status, type of family and marital status have been observed to influence the QoL in a study conducted by (Kumar, Majumdar, & Pavithra, 2014). The study also observes that living arrangements, like living with a spouse, or living in joint families or a nuclear family significantly influences the overall well-being. Respondents living in joint families were observed to score better QoL compared to those living in nuclear families (Ganesh Kumar et al. 2014). Another study by Varma et al. (2010) observed that, age of the respondents influenced the domain of physical functioning in overall QoL, and educational status had a significant impact on many QoL domains, as respondents having education level greater than the primary level reported higher QoL scores. All these studies have used existing QoL scales like WHOQOL-BREF ( As stated before, there is a dearth in studies on development of a holistic scale or instrument for the measurement of QoL of Indian elderly. Due to existence of varied diverse cultural groups of people in India, the choice of life styles, and the experiences of people in India can also be assumed to be varied. Therefore, in order to develop policies catering to the well-being of the Indian elderly, it is required to identify factors influencing perceived QoL of the different groups of the Indian elderly from their own perspective. The methods of focus group survey and content analysis were adopted for this purpose.

Focus group survey 2
A brief overview of the survey process is provided in this section. Semi-structured interviews were conducted in ten different neighbourhoods (including housing complexes and organically developed neighbourhoods) in the city of Kolkata, India on 83 elderly respondents. Before starting the survey, the survey purpose and type of questions to be asked were explained to the respondents. Only respondents who agreed to willingly participate in the survey after the introduction session were included. Besides, the respondents were also given the choice to not respond to any question against their will. To maintain anonymity of respondents, their personal details like name, address, etc. was not documented. The interview questions focused on the experience of respondents on the initial list of identified QoL domains, and how the experiences had changed with the onset of old age. The responses were analysed with content analysis; a descriptive approach to analyse qualitative data. The detailed process of focus group discussion, the technique used for extraction of QoL factors from focus group discussions through content analysis and the discussion of the results is elaborated in Saha, Basu & Pandit (2022). The focus group survey resulted in the identification of 47 QoL factors for all QoL domains (1 to 7, as mentioned in Section 2.1). The summary of the responses has been provided in Table 1.

Determination of final list of QoL factors 3
The final list of QoL factors identified for each QoL domain, from focus group survey, study of QoL scales and theories on ageing and older adults, were first listed together, followed by a stage of refinement of these factors. The factors were framed in order to enable the questionnaire or scale to have questions that could be asked with respect to how satisfied the respondents were with each of the factors. The assessment of the level of satisfaction with each of these factors can be assumed to give an assessment of the overall QoL of the respondents.
While refining and clubbing of items identified from QoL scales, items or factors that could not be catered to by the neighborhood infrastructure questions, were not included in their original form. For example, factors pertaining to questions like "I believe God has a plan for me", "I feel forgiving of those who have harmed me", "I am aware of inner peace", etc. were not used directly in our questionnaire. While considering these items, we have instead analysed the aspect that these questions or items cater to. For example, "I believe God has a plan for me" has been clubbed under "Comfort from faith in spiritual/ religious systems", or, "I feel forgiving of those who have harmed me" and "I am aware of inner peace" have been clubbed under "Mental peace (forgiveness and letting go)". Similarly, all other comparable or analogous items were clubbed together to arrive at the final list of 37 QoL factors, which could be catered to by the neighborhood infrastructure and attributes. Table 1 shows the list of factors identified for each domain from this two stage process of categorization. Table 1 also provides the definition for each of the QoL factors (meanings associated with the factors).
Factors catering to similar aspects of life were combined together to arrive at a final list of 37 factors. The factors were framed in a way to enable satisfaction based questions to be asked with respect to each of these factors.

Identification of final structure of QoL domains and their respective list of factors from interviews with 408 elderly respondents 4
Convenience sampling technique was adopted for this study considering the length of the questionnaire, and the type of questions which required the respondents to invest a significant amount of time to reflect upon their life conditions. Convenience sampling is a type of non-probabilistic sampling where respondents are selected on the basis of certain requirements of the researcher, such as availability, accessibility, geographical proximity, or the willingness to participate in the study (Etikan et al. 2016).

Study area selection for final survey with 408 respondents
The study area selected for this research is Kolkata the capital of the state of West Bengal in India. Kolkata has been included as a member of the Global Network of Age-friendly Cities and Communities of World Health Organization (WHO, 2012). The study was conducted in different neighborhoods in Kolkata Municipal Corporation and the adjoining Bidhannagar Municipal Corporation, located in the state of West Bengal in India. The selection of study areas for this study was determined on the basis of the availability of respondents for the survey. The administrative or ward boundaries were used to demarcate the neigborhoods, and the boundaries of the housing complexes were used to demarcate the neighborhood boundaries of the housing complexes.

Survey question format and process of data collection
The researcher was accompanied by survey team of five members, who were hired to conduct the survey. The purpose of the questionnaire and each of the questions, were explained in advance to each of the surveyors, in order to avoid any kind of mis-interpretations of the questions. Since the survey consisted of questions which required the respondents to introspect and reflect upon their lives, the survey required the surveyors to first get acquainted with the respondents. The survey process started with the surveyors first explaining the purpose of the survey, and the type of questions that will be asked during the survey. Any respondent who was not comfortable with the entire survey process were not included in the survey. Only respondents who agreed to respond to the survey, after the introduction session, and were mentally and physically capable to complete the entire survey process, which spanned for approximately 1 h for each respondent, were included in the survey. Respondents were also informed that they could choose to not respond to any question they were uncomfortable with. To maintain anonymity of the respondents, personal details like names, address, etc. of the respondents were not documented. Each questionnaire was translated to the local language, Bengali, which was the mother tongue of majority of the respondents. In some cases, questions had to be asked in Hindi. After the survey, the answers were read out/ shown to the respondents. The questions consisted of satisfaction rating of QoL factors (identified in Table 1, Section 2.4), in a scale of 1-5, where for the satisfaction scale, 1 meant least satisfied and 5 meant most satisfied.

Survey respondent characteristics
The socio-demographic characteristics of the respondents in the final user perception survey have been shown in Table 2. Majority of the respondents belonged to Age group 1 (60-69), and only 11% respondents belonged to Age group 3 (80 and above), which was because the survey required the elderly respondents who were healthy enough to sit through the long 1.5 hours' survey, and respond to all the questions efficiently. Only 28.9% of the respondents were female, which could be due to the fact that Indian female elderly are not comfortable to interact with strangers. 52.7% of the respondents were retired employees. 85% of the respondents were married, among which 66.4% of the respondents lived with their spouse and children. Only 1 elderly respondent lived with a non-family based care-giver (ayah or nurse or maid). Worry of future of children

Sense of burden on others ( related to financial dependence on family members or others) Sense of burden on others ( related to
financial dependence on family members or others).

Sense of burden due
to financial or family responsibilities

Analysis of responses for identification of QoL domains
The broad domains identified in this study was based on the assumption that people often interlink seemingly similar qualitative aspects of life during their assessment and evaluate them on the basis of concepts that are not directly observable (Bollen 2002). The satisfaction data on QoL factors and domains collected from the residents were analysed using Principal Component Analysis (PCA), a method of Exploratory Factor Analysis (EFA). The technique of PCA is most popularly used as a technique for variable reduction. However, the results from PCA, has been observed in many studies to help in the extraction of latent constructs or major components, to which the variables contribute to (Amérigo and Aragonés 1990;Amole 2009;Bezerra and Gomes 2015). This study uses Principal Component Analysis (PCA) with varimax rotation. The steps for PCA as followed in a study by Mazumdar and Paul (2016) was adopted in this research. PCA was conducted using the software SPSS. First, Bartlett's test of Sphericity was conducted and the correlation matrix was found to be an Identity Matrix, where all the diagonals are 1. Second, Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy was calculated, where the KMO value was observed to be 0.942 which is higher than the minimum acceptable value of 0.50 (shown in  Table 3). Third, components were extracted using varimax rotation in factor analysis. Factors having Eigenvalue greater than 1 were retained. The criteria for consideration of items or variables as significant was variables with factor loadings of 0.4 and above, as observed in studies by Bahari and Ling (2010); Chaplain (1995); Heung and Cheng (2000); Sidique et al. (2010).
The exploratory factor analysis yielded 7 components. The cumulative variance explained by the components is 54.929%. The variable groupings and the associated component is shown in Table 4. Each component was assigned a name on the basis of the variables that grouped together to form the same. The Cronbach's coefficient alpha for each component was then conducted to test the reliability of scales of the item variables. While there is no standard cut-off point for the alpha coefficient, and a value of 0.7 and above is generally considered as an acceptable lower limit (Hair et al., 1998 as cited by (Sidique et al. 2010)), values higher than 0.5 are also considered acceptable in exploratory research (Nunnally, 1978 as cited by (Sidique et al. 2010)). Cronbach's Alpha value of 0.6 as the lower cut off is also considered in studies by Fleming (2011); Setbon and Raude (2010) and the same has been used in this study. The findings have been discussed in Section 3.1.

Prioritization of QoL domains based on socio-demographic groups
In order to study the variation in priortisation among different socio-economic groups, we have first separated and segregated the responses according to the different categories in each socio-demographic characteristic, and conducted Ordinal Logistic Regression (OLR) for each of these categories. In each OLR for each category, the scores received for each of the domains from PCA (for the respondents from that category) were considered as the dependent variable and the score for the overall QoL for that category of respondents was considered as the independent variable.
The responses for the different categories under each socio-demographic group were then complied together to create a single comparative table. The results of prioritization of domains according to the different socio-demographic characteristics have been shown from Tables 5 and 6. The findings have been discussed in Section 3.2.

Domains of QoL
This section shall discuss the findings from Table 4. Each of the seven components derived in Tables 4, were assigned a title or name by the authors, based on the factors which clubbed together to form the same. To explain further with an example, the factors 'Companionship' KMO  iii. Ability to cope with your feeling of helplessness (due to self and worry over family's future)  and 'Quality of social interaction' were observed to club together to form Comp 6, and was assigned the name Quality of companionship' based on the aspects of life they catered to. Similarly, all the other components were assigned names or titles by the authors.
Comparison between the results of QoL domain and factor structure obtained in Tables 1 and 4, reveal that the groupings observed from Principal Component Analysis vary considerably from that which was developed from literature study and focus group discussion. The reason behind this difference might be due to the bigger sample size and the wider spectrum of individuals from varied socio-economic backgrounds considered for the final user perception study, in comparison to that used for the focus group survey.
The final list of domains obtained from the study are listed below: a) Health status (physical and mental) influencing performance of activities, b) Psychological state, c) General perceptions associated with old age, d) Financial status, e) Involvement in activities and social relationships, f) Quality of companionship and. g) Quality of sleep are the major domains.
The findings validate the assumption that the perception of the Indian elderly about the factors influencing their QoL is significantly dissimilar from those developed in the context of western countries. While, 'Quality of sleep' and 'Quality of companionship' are observed to be explicit domains in the QoL of Indian elderly, 'Religious/ spiritual activities' which are assumed to be separate domains contributing to QoL of the elderly (ref Table 1), have been clubbed with the factors influencing the domain of 'Psychological state' in the findings from the final user perception survey (ref Table 4). This finding can be corroborated to the influence of religious/ spiritual beliefs and practices on the psychological state of the elderly. The clubbing of factors which are most commonly associated with old age like change in  self-appearance, fear of death, dependence on health aids, etc. to form a separate domain might probably be linked to their awareness of the problems and insecurities faced due to the socio-physiological changes in old age.
The following section shall discuss in detail the observations from the findings ( Table 4) for each of these seven QoL domains.

Component or Domain1: Health status (physical and mental) influencing performance of activities.
The factors associated with physical and mental health, which directly influence the efficiency and ability to perform activities have been observed to group together to form this domain. Both the factors which influence the experience of physical health, as well as the factors which are affected by the physical health condition of an individual have grouped together. The fact that these factors impact upon the ability and efficiency to perform daily activities, can be further validated with the association of the factor 'Freedom to perform activities of choice' in this domain. Cognitive functioning, which was observed from literature study as an aspect of mental health status, has also been observed to associate with the factors in this domain, and this phenomenon can be ascribed to the fact that cognitive functioning influences one's efficiency in the performance of activities (Hall, Vo, Johnson, Barber, & O'Bryant, 2011).

Component or Domain 2: Psychological state.
The factors that influence an individual's ability to maintain their overall psychological state or condition, despite all the adversities, that affect their daily life is observed to have the strongest influence in this domain. The other variables which have associated together, corroborate with our findings from literature, where, loneliness, helplessness, sense of responsibility have been commonly related to the emotional or psychological well-being of an elderly.

Component or Domain 3: General perceptions associated with old age.
A wrinkly faced (Ayalon 2015) bent body, with a supporting stick in hand, which is the most commonly linked to the image of old age is further validated by the association of the factors 'Change in self-appearance in old age', and 'Dependence on health aids' in the formation of this domain. Besides, the fear associated with the vulnerability of the elderly to crime and abuse and recent upsurge in crimes on elderly victims in the Indian context, is probably the reason why these variables have also associated together. Overall, this domain deals with the generic vulnerabilities and image-abilities associated with old age.

Component or Domain 4: Financial status.
The variables related to the ability of an individual to cater to their basic needs and desires based upon the economic resources available to them have associated together to form this domain. The sense of burden that is often imposed an individual due to the limitations in their financial capabilities is also a factor associated with this domain.

Component or Domain 5: Involvement in activities and social relationships.
According to sociometer theory, an individual's self-esteem is often strongly influenced by the reactions of people associated with him/ her because self-esteem is as a subjective measure of the degree to which an individual is being included or excluded or rejected by other people (Onoda et al. 2010). The factors which associated together to form this domain corroborate with the same theory. The sense of 'Ability to protect self from physical injury' influences the willingness to go outside one's home to participate in outdoor activities. The 'Ability of coping' and adjusting with other people along with the ability to accept and cope with the declining physical and mental capabilities in performing varied activities, is expected to influence an individual's 'Level of engagement in varied activities'. Participation in such social activities, in turn, help in building social relationships. Thus, each of these variables which have associated to form this domain, mutually influence each other, in regard to an individual's wish to participate in social activities and thereby enabling or unenabling them with the opportunity to build social relationships.
Component or Domain 6: Quality of companionship.
The factors which associated together to form this domain corroborate with the Convoy model. The respondents in this study also links both the objective characteristics (degree of companionship, i.e. if they have or do not have sufficient companionship) and the quality of relationship, as important factors influencing their companionship.

Component or Domain 7: Quality of sleep.
The variables which load on Component 7 relates to sleep related disorders like Nocturia, which affects the Quality of sleep in elderly. Based on our findings from literature study, this factor was observed to be included in domain of Health. The identification of this domain as a separate domain influencing QoL, from the study findings based on the direct feedback of the older adult population, highlights the importance of quality of sleep in affecting the perceived QoL of the respondents.

Discussion of results on prioritization of QoL domains based on sociodemographic groups
The exploration of the perception of prioritisation of domains across varied socio-demographic groups reveal that the social and economic background of an individual influences their prioritisation of the aspects catering to their well-being (here, QoL domains). Some of the findings also corroborate with the concept of "paradox of well-being" where old age is associated to be directly proportional to higher levels of SWB despite the substantial amount of physical and social losses that comes associated with it (Jung & Siedlecki, 2018).
Overall observations of Tables 5 and 6 reveal that health has been assigned in the high priority zone by most of the respondents. The aspect of 'quality of companionship' has been given either low priority or has not been assigned with any statistical significance. This finding is in contrast with the existing studies which assert on the importance of the quality of social relationships with increased age, especially the older old.
Owing to the limitations in sample size and sampling distribution in this study, this aspect can be further explored on a larger sample to arrive at a conclusive understanding of this domain on the QoL of the elderly.
Prioritisation based on age groups. Higher age groups (age groups: 70-79 and above 80 years) rated the Health domain with highest priority. The psychological state of mind was observed to be of highest importance for the respondents in age group 1. The lowest importance of financial status among respondents belonging to the highest age group (age group 3: above 80 years) can be owed to the complacence that one attains in old age with lesser materialistic demands.
Prioritisation based on gender groups. While health was observed to be of highest priority among male respondents, female respondents prioritized the involvement in activities and social relationships as highest. The higher need for social relationship corroborates with the findings of the study by Mukherjee (2013). The traditional gender based roles in the Indian society which restricts the freedom of elderly women to travel alone consequently leads to decreased contact with their friends and relatives, leading to solitude and loneliness (Mukherjee, 2013). Financial status is observed to be of higher priority among male respondents in comparison to female respondents.
Prioritisation based on marital status. The married respondents were revealed to prioritize health as most important, while the priority of health was comparatively lower for single respondents. The domains related to companionship, leisure and social relationships, were observed to be non-significant to the overall QoL of single respondents. The stated importance of factors like Companionship and Quality of social interaction were observed to be high for single respondents. The non-significance could be ascribed to the fact that unavailability of family based social relations might have led the single respondents to derive satisfaction in overall QoL from other aspects of life, specially their psychological state of mind. These findings also can be interpreted as unavailability of satisfactory social relationships in their present lives, which consequently highlights the need for creating opportunities to develop social relationships in the neighborhood.

Prioritisation based on living arrangement.
Analysis with respect to living arrangement also reveals a state of paradox. While, except for the domains of Health and Psychological state, the rest of the domains were observed to be non-significant among respondents living with children; the domains of involvement in activities and social relationships, quality of companionship and quality of sleep were also observed to be non-significant for respondents living alone. The elderly living alone might have adapted to their life of solitude and therefore have not associated any significance to companionship while rating their overall QoL. This paradox leads to the implication that both abundance of availability and extreme unavailability can lead to adaptation and acceptance of a situation as normal and can subsequently lead to the failure of consideration for the particular factor while assessing their QoL. Some studies have also indicated that people living alone often associate with higher level of well-being in comparison to those living with their children or others (Gee, 2000). Quality of companionship was also observed to be non-significant among respondents living with spouse and children. Financial status was observed to be non-significant among respondents living with their spouses only. Since Liv. Arr. 4 (residing only with non-family based care giver) had only 1 respondent, the response could not be considered for this stage of analysis.

Prioritisation based on educational backgrounds
The respondents in the category of Edu 1 have given the highest priority to involvement in activities and social relationships and assigned lower priority to financial status. Their life long financial crisis might have led them to learn to adapt to their lives with limited financial resources, and they have learnt to focus on drawing their satisfaction for their overall QoL from their social support and relationships. A study by Åberg, Sidenvall, Hepworth, O'Reilly, and Lithell (2005) suggests that satisfaction in life can be achieved by maintaining a balance between the realities of one's life and lowering one's expectations. Adapting to limited resources and decreasing one's desires can be ascribed to be a reason for the lower priority.

Prioritisation based on economic backgrounds.
Respondents belonging to Economically Weaker Section category (EWS) (Inc 1) were observed to give comparatively lower priority to their financial status. Older people's gradual acceptance of the realities and changes in perspective towards life, subsequently leads to the decrease in need for additional income, in comparison to people of other age groups (Foster, Tomlinson, & Walker, 2019Foster, Tomlinson, & Walker, 2019. This interpretation corroborates the 'relativist' theory, which suggests that older adults have the ability to adjust their expectations according to their existing financial situation (Foster, Tomlinson & Walker, 2018). These respondents from Inc 1 (EWS) along with the respondents belonging to the Lower Income Group (LIG) category (Inc 2) prioritised the domain 'General perceptions of old age'. This domain includes the factor of vulnerability to social abuse and crime, which highlights their vulnerability to the same.

Summary of results
Identification of latent constructs or the major domains influencing perceived QoL analysed using Principal Component Analysis (PCA) reveal that the groupings vary considerably from those developed from literature study and focus group discussion study. The findings validate the assumption that the perception of the Indian elderly about the factors influencing their QoL is significantly dissimilar from those belonging to the west. With respect to the domain of Health, higher (increasing) age groups were observed to assign higher priority to health. The lowest priority of financial status for the respondents belonging to the highest age group (age group 3) can be explained with the help of the Socioemotional selectivity theory, which refers to the complacence one attains in old age, consequently lessening the materialistic demands. The traditional gender based roles and restrictions of the Indian society have been affirmed with higher prioritisation of the need for involvement in activities and social relationships by the female respondents and higher prioritisation of financial status by the male respondents. The findings based on marital status reveal that both abundance of availability and extreme unavailability can lead to adaptation and acceptance of a situation as normal and can subsequently lead to the failure of consideration for the particular factor while assessing their QoL. The findings based on economic background of the respondents reveal that older people are more likely to accept the existing situation of their lives and adjust their needs accordingly, in contrast to the people from other age groups. This interpretation can be supported by the 'relativist' theory. It is important to highlight the higher priority assigned to the domain of 'General perceptions of old age' by the respondents belonging to EWS and LIG category. Since this domain comprises of the factors 'vulnerability to social abuse and crime', it can be assumed that the elderly belonging to the less privileged economic backgrounds either suffer from or are vulnerable to abuse and crime due to their limited economic access and resources.
The findings are however limited to the responses received from a limited number of respondents. The study framework can be further applied on a larger sample with a more heterogenic spread for a more generalized understanding of the variation in prioritisation among varied socio-demographic groups. Besides, the technique of structural equation modelling can be utilised on the findings of a larger sample size for obtaining a more robust domain-factor structure defining the determinants of perceived QoL.

Conclusions
The present study proposes a framework for combining quantitative and qualitative research techniques to identify a broad set QoL domains and their respective list of factors, which can be further utilised for the assessment of the perceived QoL of Indian elderly and for improving the same in the neighbourhood context. The present study has two major contributions: first, the incorporation of existing research on older adults, theories on ageing, items from existing QoL scales or models along with the direct perceptions of the Indian elderly and second, the identification of QoL factors which can be catered to in the neighbourhood context, through the provision of required infrastructure. The first contribution includes both the identification of a list of 37 QoL factors, and then with the use of exploratory factor analysis, the identification of a broader list of seven QoL domains, to which these factors contribute to. These findings validate the need for the study by establishing that the structure of domain and factors influencing perceived QoL of the Indian elderly, vary considerably from that of the models on QoL of elderly developed in the western context. The QoL domain and factor structure can be considered to be the third contribution of this study, which can be further explored on a larger sample size for the development of a QoL index specific to the needs of the Indian elderly, and which can be catered to in the neighbourhood context. The second contribution can be of immediate pragmatic use to policy makers who can involve planners to frame neighbourhood development guidelines to cater to the list of QoL factors, identified in this study. Further studies using the same model may be conducted on a larger sample size for better exploration of the variation in prioritisation among varied socio-demographic groups. Though the study attempts to ensure equal participation from all socio-demographic groups, due to certain existing socio-cultural limitations influencing willingness of participation of respondents, the sampling distribution is a limitation of this study. The limited sample size is another major limitation of this study. However, the proposed framework of this study can be further explored on a larger sample to arrive at conclusive findings on the factors and the prioritisation of the same in their influence on the QoL of Indian elderly.