The WeD-QoL aims to operationalise this definition
by directly assessing the relationship between people’s “goals,
expectations, standards and concerns”. It was also influenced
by the ‘gap’ theories popular in QoL and Health-related
QoL research (for example, Calman , Michalos , Ruta
et al ), which argue that the main determinant of people’s
perceived quality of life is the gap between their expectations
and their experiences. However, their expectations are not static,
but are influenced by prevailing ideologies, social comparison,
adaptation, and values. These influences cause people to evaluate
their situation more or less favourably than might be expected,
accounting for the discrepancies between objective and subjective
measures, which have previously caused researchers to undervalue
subjective accounts (e.g. Sen 1984:309).
3. How it contributes to WeD
The process of developing the WeD-QoL has already
enabled us to:
i. Understand the categories and components of QoL
in the specific localities we are working in and, within these,
the relative influence of the subjective, inter-subjective, and
material dimensions of people’s lives
ii. Distinguish between components of QoL at the universal level
and those of the locality and the culture-group, and explore the
relationship between them
iii. Assess the level at which it is possible to produce a measure
of QoL that is meaningful to respondents, analysts and development
practitioners, and develop methods that will best achieve this
This culminated in the development of the WeD-QoL
Phase 3 instrument, which provides data on people’s happiness
and satisfaction with ‘life as a whole’ (subjective
wellbeing), and the interaction of goals, resources, and values
in producing these states, and will ultimately be used to produce
the WeD-QoL measure itself.
At this stage (prior to the
psychometric analyses of Phase 4) we cannot specify the content
of the ultimate WeD-QoL measure.
At Phase 3, The WeD-QoL instrument from which data
were collected had two components:
1. A battery of four ‘native’ scales
that measure Goals, Resource Availability, Goal Achievement, and
‘personal’ and social Values.
The native scales consist of the following:
‘core’ of cross-cultural ‘items’
items (for example, ‘feeling metta-karuna [universal love-compassion]
for others’ in Thailand)
scales, either three or five point
Peru and Ethiopia also developed items specific
to the site or section of the site to capture important differences
between the communities.
2. Two adaptations of internationally validated
scales that measure the following:
· Satisfaction with life as a whole (Satisfaction
with Life Scale, Diener et al 1985)
· Presence of positive and negative affect (Positive &
Negative Affect Scale, Watson et al 1988)
5. How WeD-Qol Phase 3 was developed
The four scales that formed the core of the WeD-QoL
instrument (Goals, Resource Availability, Goal Achievement, and
Values) were developed using data from multiple sites within Ethiopia,
Bangladesh, Thailand, and Peru, which was collected during the
first phase of the QoL research.
The initial exploratory phase (Phase 1)
The key questions for the exploratory phase were
goals and experiences (‘havings’, doings, beings,
feelings, and meanings) contribute to people’s QoL in the
context of their particular community, cultural grouping, and
are considered necessary to achieve or maintain QoL.
do values play in this process
The process of establishing the conditions for QoL
in each country began with the WeD Inaugural Workshop in 2003
and continued with ‘expert’ workshops on QoL during
the summer of 2003.
It also involved field research with a wider range
of informants, which produced both data and methodological reflections.
The WeD QoL Phase 1 fieldwork took place in rural, peri-urban, and
urban sites in Bangladesh, Ethiopia, Thailand and Peru. The average
sample size for the countries was 360 (range 314-419) and age and
gender were used as the key breaking variables, followed by religion
or ethnicity. The fieldwork used locally appropriate qualitative
and quantitative methods, including semi-structured interviews , focus groups,
the Person Generated Index (PGI)[Ruta et al 1994, 1998], and the Satisfaction with
Life Scale (SWLS) [Diener et al 1985].
The Phase 1 data were supplemented by information
from the Community Profiling, some of which asked questions relating
to quality of life and wellbeing (see particularly the Wellbeing
and Illbeing Dynamics in Ethiopia study, WIDE 2). It also
incorporated findings from the Resources and Needs Questionnaire
(RANQ) (see RANQ toolbox).
The second conceptual phase (Phase 2)
The exploratory phase was followed by a second conceptual
phase of analysis and theoretical integration. At an international
workshop in the UK country representatives reviewed the findings
of the first phase and discussed whether, and how, we might develop
a generic measure of QoL specific to developing countries.
It was agreed to extend the psychometric methodology
used by WeD-Peru to all the countries and develop and test a common
suite of measures (the WeD-QoL instrument used in Phase 3) which
could then be subjected to psychometric analysis. This decision
was ratified through face-to-face and email consultations with
our QoL Steering Group (comprising psychologists and psychometricians).
The draft structure of the WeD-QoL Phase 3 instrument and the
Terms of Reference for developing and administering it were then
agreed at a further international workshop in Thailand.
Once the country versions of the WeD-QoL Phase 3
instrument had been agreed and translated (a process that involved
all members of the research team), they were tested in a selection
of research sites with older and less-educated respondents who
were likely to find the WeD-QoL questions more difficult to respond
to (‘pre-piloting’). The draft WeD-QoL Phase 3 instrument
was then revised to take the interviewers’ feedback into
account and tested again in a similar selection of research sites
with similar respondents (‘piloting’). The instrument
was revised again following feedback from the piloting and was
then ready for the final administration.
This iterative development process ensured first
that all items were relevant to the locations in which the instrument
would be used, and the second that the majority could be analysed
6. How WeD-Qol
Phase 3 was implemented
The WeD-QoL Phase 3 instrument was administered
to 370 men and women in each country (approximately 60 per research
site) in the local language by a team of interviewers selected
by each of the country teams. Before this fieldwork commenced,
interviewers underwent intensive training from country researchers,
supported by researchers from WeD-Peru and WeD-Central who had
both extensive psychometric experience and experience of the QoL
fieldwork of another WeD country. The interviewers’ understanding
was furthered by their participation in the pre-piloting and piloting
phases of the WeD-QoL instrument.
All respondents to the WeD-QoL Phase 3 instrument
had also completed RANQ, and as many as possible had been involved
in other research, for example, the Income
and Expenditure survey or the Core Household diaries. The remainder was sampled proportionately
according to age, socio-economic status, ethnicity, and religion.
Upon completion, the responses were entered by the
country teams into a specially designed Microsoft Access database
in English. In the case of Peru and Thailand, the data were ‘double
entered’ by two staff members into two copies of the database,
which were then compared to identify any errors in data entry.
In all cases, the databases were checked by the project data manager
‘by eye’ and through a series of queries, for further
errors and anomalies.
Once the four countries’ data were approved
as ‘clean’, they were exported into SPSS in preparation
for the psychometric scale development
process (Phase 4).
A further data entry check was conducted at this
stage, by back-checking a random sample of the original protocols
against the SPSS database, and calculating the percentage data
When the psychometric phase
has been completed, the resultant WeD-QoL measure will be ‘virtually’
returned to the field. This means that only those variables which
form part of the resulting internally coherent scales will be included
in the WeD-QoL measure. For the purpose of reliable and efficient
measurement of QoL, other items may be deleted from the scales.
A post-phase 4 dataset, containing data only from the variables
included in the final WeD-Qol measure, will also be produced as
a result of this ‘virtual fieldwork’. The data , the measure and the methodology
will be made available for use by other researchers or practitioners.
After its completion, the psychometric
development process will be documented comprehensively.
7. How can it
The WeD-QoL data will permit analysis a) within
a site, b) across sites of the same country, eg. by rural and
urban areas, and c) across the four countries.
In analyses, it will be possible to make use of
the whole range of variables included at Phase 3, as well as the
psychometrically developed Phase 4 WeD-Qol measure.
It will thus be possible to explore the following
for different individuals in different sites across the four countries:
1. What goals they perceive as important or relevant
2. How satisfied they are with their achievement of the goals
they care about
3. Their perception of the resources they have to pursue these
4. Their values
5. Their perception of the values of people in their community
6. Their mood
7. Their satisfaction with life as a whole
8. Their level of subjective quality of life
Below are some of the research questions the earlier
participatory work and the data from the measure will enable us
what extent do ‘felt’ needs and ‘objective’
needs satisfiers differ between sites and individuals within those
sites? How is perceived need satisfaction or goal achievement
related to life satisfaction and mood? Does this relationship
differ between sites and individuals within those sites?
perceived access to different types of resources differ between
sites and individuals within those sites?
socio-demographic variables (like education and religion) relate
to the personal values people endorse? Is there any relationship
between particular value clusters and goal achievement, life satisfaction,
is the relationship between perceived and objective resources?
(recorded in the RANQ and Income and Expenditure surveys) Which
category has the greater effect on life satisfaction and mood?
conceptions of the good life and sources of happiness and unhappiness
differ between sites and between individuals within those sites?
the connections between the exploratory quality of life research
and the results of the WeD-QoL? This mostly compares qualitative
and quantitative findings. For example, where people have said
in focus groups that having a strong family is the most important
aspect of a good life, how does this reflect in the number of
WeD-QoL Phase 3 respondents endorsing the relevant item as ‘very
necessary’, and, on a quantitative level, how does achieving
a strong family relate to greater happiness and life satisfaction?
8.Links to other
WeD research tools
The WeD-QoL data enables us to explore the relationship
between perceived and objective resources (recorded in the RANQ
and Income and Expenditure surveys), and between both types of resources
and feelings of positive emotion and satisfaction with life. It
complements the more descriptive components of RANQ,
the community profiles, and research into processes and structures by providing information on people’s subjective perceptions
of wellbeing. This allows exploration of the relationship between
objective and subjective dimensions of wellbeing.
Ethiopia (Phase 3) - draft version see Laura Camfield firstname.lastname@example.org
Generated Index (PGI) (Ruta et al)
with Life Scale (SWLS) (Diener et al)
10. Phase 1 Short summaries
- One page summary
- One page summary
- One page summary -
- ECB - Peru
11. Further Reading
Camfield, L., McGregor, J.A., & Yamamoto,
J. (2006) Quality of Life and its relationship to wellbeing (forthcoming)
Calman, K. C. (1984). QOL in cancer patients a hypothesis.
Journal of Medical Ethics, 10, 124-7.
Diener, E., Emmons, R., Larsen, J., Griffin, S. (1985).
The Satisfaction With Life Scale. J Personality Assessment, 49(1),
Michalos, A. C. (1985). Multiple discrepancies theory
MDT. Social Indicators Research, 16, 347-413.
Ruta, D. A., Garratt, A. M., Leng, M., Russell, I.T.,
& MacDonald, L.M. (1994). A new approach to the measurement
of QOL. The Patient Generated Index. Med Care, 32(11), 1109-1126.
Ruta, D. A. (1998). Patient generated assessment:
The next generation. MAPI Quality of Life Newsletter, 20, 461–489.
Sen, A. (1984). Resources, values and development.
Oxford, UK: Blackwell.
Watson, D., Clark, L.A., & Tellegen, A. (1988).
Development and validation of brief measures of positive and negative
affect: The PANAS scales. Journal of Personality and Social Psychology,
WHOQOL Group. (1995). The World Health Organisation
Quality of Life assessment (WHOQOL): Position paper from the World
Health Organisation. Social Science & Medicine, 41, 1403-1409.