The Subjective Well-Being of the Homeless, and Lessons for Happiness

Abstract

The current study assessed the subjective well-being of a broad spectrum of homeless people. One-hundred-and-eighty-six homeless
people from the streets of Calcutta (India), California, and a tent camp in Portland (Oregon) were interviewed, and responded
to measures of subjective well-being. They answered questions about life satisfaction, satisfaction with various life domains,
and their experience of positive and negative emotions. The mean rating of life satisfaction was slightly negative for both
American samples but positive for the pavement dwellers in Calcutta. Satisfaction with self-related domains was positive,
whereas satisfaction with material related domains was generally negative. Satisfaction with social domains appears to be
the area of largest variation among the groups. We discuss the importance of social factors and basic material needs as they
relate to overall subjective well-being of the homeless.

Full text

ROBERT BISWAS-DIENER and ED DIENER
THE SUBJECTIVE WELL-BEING OF THE HOMELESS,
AND LESSONS FOR HAPPINESS
(Accepted 9 June 2005)
ABSTRACT. The current study assessed the subjective well-being of a broad
spectrum of homeless people. One-hundred-and-eighty-six homeless people from the
streets of Calcutta (India), California, and a tent camp in Portland (Oregon) were
interviewed, and responded to measures of subjective well-being. They answered
questions about life satisfaction, satisfaction with various life domains, and their
experience of positive and negative emotions. The mean rating of life satisfaction was
slightly negative for both American samples but positive for the pavement dwellers in
Calcutta. Satisfaction with self-related domains was positive, whereas satisfaction
with material related domains was generally negative. Satisfaction with social do-
mains appears to be the area of largest variation among the groups. We discuss the
importance of social factors and basic material needs as they relate to overall sub-
jective well-being of the homeless.
KEY WORDS: homeless, quality of life, subjective well-being, well-being
SUBJECTIVE WELL-BEING OF THE HOMELESS
Poverty is one of the most pressing social concerns in the world
today. In 1999, according to the United Nations Human Develop-
ment Report (2002), nearly half the people in sub-Saharan Africa and
more than a third of those in South East Asia lived on less than a
dollar a day. Of the poor, the most visible are the homeless. Whether
they are gangs of street youth or panhandling drifters, no society is
unaffected by the social problem of homelessness. There is little
agreement about the possible causes and solutions to this social ill.
Although past research has focused on psychopathology, incidence of
trauma, and the demography of the homeless, little attention has
been paid in the psychological literature to the overall quality of life
of the homeless. Instead, the bulk of the research literature has been
confined to clinical aspects of homelessness or to the effects of
Social Indicators Research (2006) 76: 185–205 Ó Springer 2006
DOI 10.1007/s11205-005-8671-9

homelessness on children (e.g., Aptekar, 1994). It is often assumed
that life on the street is fraught with difficulty, but little information
has been collected to suggest which specific life domains might be the
least problematic, and in which areas, if any, the homeless might
actually be flourishing. Unfortunately, overlooking the possible re-
sources and strengths of the homeless limits our ability to create
effective interventions.
The existing literature on homelessness strongly suggests that there
are many problems associated with life without a home. Studies have
shown that homelessness is associated with problem behaviors in
children (Edleman and Mihaly, 1989; DiBiase and Waddell, 1995),
strained family relationships (Vostanis et al., 1996; Nyamathi et al.,
1999), higher exposure to trauma (Hien and Bukszpan, 1999; Buhrich
et al., 2000), increased anger and depression (Marshall et al., 1996),
and the negative psychological impact of social stigma (Lankenau,
1999). Because of the methodological difficulties related to studying
homelessness, it is unclear whether factors such as depression and
alcohol abuse are causes or effects of homelessness. While prior his-
tory of mental illness is undoubtedly responsible for homelessness in
at least some cases, there is evidence to suggest that the experience of
homelessness causes or exacerbates many psychological problems. In
a study by Shlay (1994), for example, homeless people were found to
report greater emotional well-being and fewer behavioral problems in
their children after positive changes in their economic and social
status. The need for research on personal resources and successes on
the street becomes more pressing to the extent that people can over-
come the psychological ills that accompany homelessness.
Studies on the harmful effects of homelessness are consistent with
a larger body of literature examining the relation between income and
subjective well-being. In large national surveys, for example, income
has been shown to be moderately correlated with life satisfaction,
especially at the lower economic levels and in the poorest countries
(Diener et al., 1999; Diener and Lucas, 2000; Diener and Biswas-
Diener, 2002). Higher income has been shown to be related to in-
creased longevity (Wilkenson, 1996), better health (Salovy et al.,
2000), and greater life satisfaction (Diener et al., 1985; Diener and
Oishi, 2000). Scholars appear to agree that although correlations
between income and subjective well-being are often modest, there
appears to be a curvilinear relationship in which money has the
ROBERT BISWAS-DIENER AND ED DIENER
186

greatest impact on psychological health at the lowest economic levels
(Inglehart and Klingemann, 2000).
Two theories are often advanced to explain these findings: basic
needs and adaptation. In the former theory, it is assumed that basic
physical needs such as food, water, and shelter, must be satisfied
before a person can attain higher order psychological fulfillment
(Maslow, 1954). By this reasoning, homeless individuals, who face
impediments to fulfilling basic needs, should exhibit lower levels of
subjective well-being. As homeless people gain better access to food
and shelter there ought to be a corresponding increase in psycho-
logical health. The theory of adaptation is also helpful in under-
standing the relation between income and subjective well-being.
Research on adaptation suggests that diminished responsiveness to
repeated stimuli allows people to adjust to life circumstances,
including adverse circumstances (Silver, 1982; Loewenstein and
Frederick, 1999). But while people can often adapt relatively well to
discrete instances of trauma, there are conditions to which it is more
difficult to adapt. Stroebe et al. (1996), for instance, found that
widows show higher average levels of depression than their non-be-
reaved counterparts, even 2 years after the death of their spouse. A
review of income and national happiness data by Diener and Diener
(1995) showed lower levels of SWB in poor nations, suggesting that
extreme poverty is a condition difficult to adapt to, even in the long
run.
The current study is, in part, an exploration of the limits of
adaptation. We hope to begin to answer the questions of which life
domains, including physical needs such as food and social needs such
as friendship, are necessary for long-term subjective well-being. By
examining the emotional quality of life of groups living in extreme
deprivation, we have the opportunity to assess both potential adap-
tation and the fulfillment of basic needs as they influence subjective
well-being. Only by turning research attention on tragic but common
cases in which people do not have enough, or have barely enough,
basic needs fulfillment will we be able to articulate how ‘‘basic’’ and
how ‘‘needed’’ are food, shelter, and social relationships.
Our prior research with homeless people (Biswas-Diener and Di-
ener, 2001) suggests that social factors may be of more fundamental
importance than Maslow (1954) believed. According to the findings
of our earlier study, social networks play a key role in overall sub-
SWB OF THE HAPPINESS
187

jective well-being. This is consistent with recent research by Diener
and Seligman (2002) suggesting that good social relationships are
necessary for high subjective well-being. Similarly, some theorists
argue that value judgments could be as important to emotional well-
being as basic needs (Brown, 1996). A sense of personal failure in the
eyes of society, for example, can be psychologically and socially
debilitating. A study of homeless people who later found housing
showed that improved quality of life depended on ‘‘what kind of
housing was obtained,’’ with people transitioning to independent
living situation making much larger gains in satisfaction than those
transitioning to dependent living situations (Wolf et al., 2001). To the
extent that social and self-actualization needs are fundamentally
important to subjective well-being, an examination of specific life
domains might help shed light on which aspects of the homeless
experience are most beneficial, and which are most detrimental, to the
emotional quality of life of people on the streets.
One of the primary difficulties in understanding the specific
experience of homelessness is the fact that homeless people are a
heterogeneous group (Brown, 1996). In the United States there is
wide variation in age, gender, and ethnicity of homeless people, as
well as causes of homelessness that include unemployment, disability,
mental illness, domestic violence, and poor social networks (Hagen,
1987). In other countries there are distinct categories of homeless
people ranging from ‘‘street children’’ (Aptekar, 1994) to sex workers
(Biswas-Diener and Diener, 2001). Likely, there are substantial dif-
ferences in the experience of these disparate groups, as well as their
collective resources for coping with homelessness. A complete
understanding of the homeless experience is beyond the scope of any
single study, but would be greatly aided by large surveys of homeless
populations. In an early example of one such study we attempted to
capture the similarities and differences in the experience of the
poorest citizens of Calcutta, including slum dwellers, pavement
dwellers, and sex workers (Biswas-Diener and Diener, 2001). While
certain elements of life were consistent among the three groups, such
as concerns related to income, there were also marked differences in
experience, such as the perception of social stigma specifically related
to prostitution.
The current study attempts to examine a broad spectrum of the
homeless experience by sampling three diverse groups: pavement
ROBERT BISWAS-DIENER AND ED DIENER
188

dwellers in Calcutta (India), homeless people (pavement and shelter
dwellers) in Fresno, California (USA), and homeless people residing
in a self-governing tent camp in Portland, Oregon (USA). Although
these three groups do not represent the entire range of the homeless
experience, they are sufficiently diverse to enable a better under-
standing of which aspects of homelessness are universal and which
may be unique to specific demographic groups.
In addition, the current study is intended to provide important
information useful for policy decisions and applications of interven-
tions related to the homeless. Past research demonstrates that the
public is concerned about the problem of homelessness (Morgan
et al., 1997) but may have lost the compassion necessary to motivate
helping behaviors (Link et al., 1995). In survey data collected in
Fresno among a sample of people patronizing a filling station
(N = 29), we found that these members of the ‘‘normal’’ population
were widely misinformed about both the causes of homelessness, and
the ways in which homeless people gain income (Biswas-Diener and
Diener, 2000). Sixteen of our respondents identified drug abuse and
12 identified laziness as the major cause of homelessness, while
physical disability was mentioned by only one person (21 of our 78
homeless respondents received government disability payments).
Similarly, 21 of 29 pilot respondents mentioned ‘‘begging’’ as a source
of income for the homeless (only four of our 78 homeless respondents
reported panhandling, while 36 earned money from day labor). To
the extent that helping behaviors are guided by sympathy, empathy,
and a realistic understanding of the issues faced by homeless people,
an examination of homeless quality of life is an important addition to
the research literature. Our study was aimed both at furthering our
understanding of the homeless experience and at the more general
issues related to how conditions affect well-being.
METHODS
In the current study, we sampled from diverse groups of homeless
people. One hundred and eighty six homeless people were interviewed
using a variety of subjective well-being measures (described in detail
below) including those assessing general life satisfaction, satisfaction
with specific domains of life, and positive and negative affect. The
SWB OF THE HAPPINESS
189

measures were presented in written form for literate research partic-
ipants, and in interview format for illiterate participants. There were
scattered missing data for respondents, resulting in slightly lower N’s
for several of the analyses.
Research participants came from one of three groups: pavement
dwellers from Calcutta (India), homeless people from Fresno (Cali-
fornia), and people residing at ‘‘Dignity Village,’’ a self governing
tent camp for homeless people in Portland, Oregon (USA). Inter-
views with the respondents from both of the American samples were
conducted by the first author at or near the dwelling places of the
research participants. The Indian research participants were inter-
viewed at or near their dwellings by a trained, experienced research
assistant fluent in Bengali, Hindi, and English.
Respondents in Fresno were recruited by the first author at
Poverello House, an agency providing free meals and other services
for poor and homeless people. Female participants in Fresno were
recruited either from Poverello House or the nearby Holy Cross
Center for Women, a social service agency providing aid to poor
and homeless women and their children. Interviews were conducted
privately, and participants were offered minor compensation for
their participation. Five people refused to participate in the study,
but three refusals were due to an inability to speak English fluently.
Participants from Dignity Village were recruited on-site by the first
author, and offered a small amount of money for participation in
the study. Literate participants provided written answers to surveys,
while illiterate participants answered the same questions in inter-
view format. Only two people refused to participate at Dignity
Village.
In India respondents were recruited by word-of-mouth by a
trained research assistant at the Dum Dum railway platform and
from a footpath near the Marwari Hospital (both locations in Cal-
cutta). Twelve people refused to participate, citing either concerns
about what would be done with the research results (many pavement
dwellers live in illegal settlements), or an inability to understand the
research because of illiteracy. Interviews were conducted in relative
privacy, away from other homeless people.
Each homeless group is described in detail below:
1. Pavement dwellers in Calcutta. The term ‘‘pavement dweller’’
refers to individuals and families in Calcutta who reside on sidewalks,
ROBERT BISWAS-DIENER AND ED DIENER
190

railway platforms, parks, and street medians rather than in perma-
nent or semi-permanent shelters. By some estimates there are at least
100,000 homeless children and 200,000 refugee pavement dwellers in
Calcutta (Thomas, 1999). Pavement dwellers typically possess some
form of bedding and a tarpaulin or mosquito net for protection
against the elements. Those pavement dwellers residing together with
their family members often possess pots and other simple cooking
utensils, and must forage for scrap wood or cardboard to provide fuel
for a cooking fire. Those female pavement dwellers who are employed
usually occupy positions related to domestic service (e.g., cooking
and cleaning), while males are frequently rickshaw pullers or market
employees. Pavement dwellers are vulnerable to police harassment
and inclement weather. The current sample (N = 67) includes 25
men and 42 women who range in age from 15 to 70.
2. Homeless men and women in Fresno, California. The homeless
population of Fresno, California is a relatively heterogeneous group.
The men and women who obtain meals at Poverello House are
married, divorced, and widowed; young and old; Spanish and English
speaking; and of many ethnicities. Further, some of them receive
public welfare benefits, while others find work as day laborers, and a
few find steady employment driving taxis, engaging in prostitution,
selling drugs, or working at traveling carnivals. Likewise, the Fresno
homeless have a variety of sleeping arrangements, ranging from
homeless shelters and temporary rented rooms to tent camps and
abandoned cars and buildings. The relatively mild weather in Cali-
fornia allows the homeless to sleep outside more comfortably than
their peers in our other two samples. The Fresno homeless have
certain elements in common as well: the vast majority reported a
history of sexual or physical abuse, a history of drug use, a history of
arrests and/or incarceration, and complain frequently of police
harassment. The incidence of violent crime is relatively high in
Fresno, and the homeless are at particular risk. The research par-
ticipants frequently complained of a lack of trust of their peers and
the need for constant vigilance. The homeless people in Fresno have
relatively few possessions – usually only clothes, a coat, and a bedroll
– but they have relatively good access to services including free dental
care, hair cuts, religious ministry, food, and blankets. The current
sample (N = 78) includes 62 men and 16 women who range in age
from 21 to 62. Care was taken to ensure that women and men, and
SWB OF THE HAPPINESS
191

people of all ethnicities and ages had the opportunity to participate in
the study.
3. Dignity Village, Portland, Oregon. Dignity Village (DV) is a
self-governing cooperative tent community in Portland, Oregon. It
has legal non-profit status and was founded by the homeless as a
transitional living arrangement for homeless adults. DV is unusual in
the United States in that it is was founded by homeless people, for
homeless people, and is populated and governed by homeless people.
Despite general prohibitions against long-term camping in public
spaces the City of Portland has allowed DV to locate in a parking lot
near the airport and adjacent to a state prison. According to the
Dignity Village charter, residents are obligated to donate time to DV
community activities including security, fund raising, sanitation, and
outreach. Residents of DV live in tents or wooden structures made
from discarded scrap wood. These structures are elevated off the
ground by wooden palettes to protect them the frequent flooding that
accompanies the heavy winter rains. Most residents of Dignity Vil-
lage own bedding, clothing, books, candles, simple cookware, and a
few own pets, musical instruments, and bicycles. There are also
community-wide resources available to residents of Dignity Village: a
basketball hoop, an organic garden, a common area with wood-
burning stove, and shower and bathroom facilities. In addition, DV
has internet access (run from locally generated wind-power), and the
residents have made all community facilities wheelchair accessible to
accommodate disabled residents. Most DV residents are not em-
ployed full time, and report histories of drug abuse and failed rela-
tionships. The charter of Dignity Village allows for only 60 full time
residents. The current sample (N = 41) includes 29 men and 12
women who range in age from 22 to 59. The sample, according to
unpublished statistics kept by the DV governing council, is approx-
imately representative of Dignity Village in terms of age, sex, and
race.
Measures
By using multiple measures we were able to capture both cognitive
(satisfaction judgments) and affective (emotions and moods) as-
pects of subjective well-being. Members of all three samples an-
swered questions related to demographics, life satisfaction, positive
ROBERT BISWAS-DIENER AND ED DIENER
192

and negative affect, and domain satisfaction. In addition, research
participants in Calcutta and at Dignity Village answered questions
relating to history of homelessness, drug use, medical history,
family history, education, and current living conditions. All
questions were presented in English in the United States and in
Bengali or Hindi in India. The translator was an experienced
psychological interviewer who had been trained specifically for this
interview.
The respondents were approached where they lived or received
meals and asked to participate in the study. They were told that their
participation was voluntary, and that, in some cases they would re-
ceive minor compensation for their participation. All respondents
were informed about confidentiality and instructed not to use a
family name during the interview. At Dignity Village, where literacy
was higher than either of the other two groups, informed consent
forms were used. Verbal consent was obtained in Fresno and Cal-
cutta. Participants were told they were free not to answer any given
question and that they could ask questions about the research fol-
lowing the interview.
The interview began with the collection of basic demographic
information, a brief biography, and questions related to current living
conditions, arrest history, medical history, and drug history. Next,
the respondents were presented with the Satisfaction With Life Scale
(SWLS), a 5-item measure of general life satisfaction found to have
good psychometric properties (Pavot and Diener, 1993). A 1–7 Likert
scale was used for each of the items, but segmented rectangles were
used for illiterate respondents. The rectangles were tested for com-
prehension prior to the interview and have been used in other re-
search with non-literate populations (Biswas-Diener et al., 2004,
Submitted).
Following the SWLS, the participants were asked to respond to
questions about the frequency of their experience of positive and
negative emotions (affection, joy, sadness, anger, guilt, worry, and
pride) during the previous month. They answered using a 1 (never) to
7 (always) Likert scale. Past research has shown cultural variation in
which emotions are considered core pleasant or unpleasant feelings
(e.g., Kim-Prieto et al., 2004, submitted). Due to these differences in
the interpretation of the emotion ‘‘pride,’’ in particular, we chose not
to include it with either positive or negative emotions in our analyses.
SWB OF THE HAPPINESS
193

The participants were then asked to rate their satisfaction with 14
life domains (material resources, friendship, morality, intelligence,
food, romantic relationship, family, physical appearance, self, in-
come, housing, social life, privacy, and health). The domains were
presented in the same order to all respondents, who were asked to
rate their satisfaction with each on a 1 (totally dissatisfied) to 7 (to-
tally satisfied) scale. Identical measures of domain satisfaction have
been used in past research with homeless groups (e.g., Biswas-Diener
& Diener, 2001).
RESULTS
Demographic Variables
Demographic information for each group is presented in Table I. As
can be seen, men outnumbered women in the American samples, a
fact true of homeless people in the United States in general. It is
possible that men are underrepresented in the Indian sample because
the interviewer was a woman and same sex contacts were more
comfortable for both the interviewer and interviewees. Data com-
piled from the Fresno Rescue Mission, a shelter agency in Fresno,
shows that our sample approximates their service population for
gender and age, but that Hispanic people are under-represented in
our sample (Hispanic people constitute approximately 41% of all
TABLE I
Demographic variables
Calcutta Dignity village Fresno Total
N 67 41 78 186
Men 25 29 62 116
Women 42 12 16 70
Age 37.4
(12.0)
40.3
(10.2)
44
(9.0)
40.7
(10.8)
% Caucasian 83% 31%
% Married 72% 41%
Average Monthly
Income
24 dollars 270 dollars* 358 dollars*
*Includes government benefits such as ‘‘food stamps.’’
ROBERT BISWAS-DIENER AND ED DIENER
194

shelter stays; Fresno Rescue Mission web-site, 2001). Reasons for
Hispanic under-representation could include English language
comprehension difficulties, or temporary work that interfered with
participation in the research. The Dignity Village sample was rep-
resentative of Dignity Village as a whole on all key demographic
variables.
Life Satisfaction
The means and standard deviations for key satisfaction variables are
presented in Table II. The means appear in the left hand column and
the standard deviations appear in parentheses to the right. We
conducted an analysis of the SWLS’ internal reliability with each of
TABLE II
Satisfaction of the homeless
Group: Calcutta Dignity village Fresno
Life Satisfaction
(5–35; 20 is neutral)
22.20 (8.68)
A
17.27 (7.36)
B
14.12 (7.61)
B
Domain Satisfaction
(score range1–7)
Material Domains
1. Material Resources 3.46 3.84 3.42
2. Food 5.37 4.41 4.99
3. Income 3.75
A
2.41
B
2.68
B
4. Housing 2.38 3.28 3.17
5. Health 3.33
A
4.63
B
4.62
B
Social Domains
6. Social 5.08
A
4.34
B
4.04
B
7. Romantic Relations 4.97 4.19 4.04
8. Friends 4.22 5.00 4.07
9. Privacy 3.00 3.09 3.22
10. Family 5.93
A
4.31
B
4.36
B
Self-related Domains
11. Self 5.77
A
4.66
B
4.73
B
12. Morality 4.97 5.41 4.57
13. Physical Appearance 4.40 4.34 4.36
14. Intelligence 5.85 5.00 5.38
Note: Scores in the same row with different superscript letters differ from each other
by p < 0.05 or less.
SWB OF THE HAPPINESS
195

the three groups. For the Calcutta sample the Cronbach’s alpha was
0.78. For Dignity Village the Cronbach’s alpha was 0.78. The Sat-
isfaction with Life Scale showed the best reliability with the Fresno
homeless population with an overall alpha of 0.84. The substantial
alphas in all three samples suggested that the items formed a
coherent whole.
An ANOVA was performed to determine whether or not the three
groups differed from one another on general Life Satisfaction, and
the significant overall F(2, 179) = 18.16, p < 0.001 was followed up
by a Bonferroni post hoc multiple comparison to determine which
groups differed from one another. The Calcutta pavement dwellers
scored the highest on life satisfaction (M = 22.2), Dignity Villagers
in the middle (M = 17.27), and the Fresno homeless lowest
(M = 14.12). Because the neutral point of the SWLS is 20.0, only the
pavement dwellers in Calcutta scored in the positive range, and both
the Fresno homeless and the Dignity Villagers scored significantly
lower than the Calcutta group (p < 0.001 and p < 0.01, respec-
tively).
The mean scores for Domain Satisfaction are also shown in Ta-
ble II. All three samples scored in the positive range on the majority
of domains except, as expected, on material satisfaction, income,
housing, privacy, and health (Calcutta only). The only domain re-
lated to material satisfaction in which the three samples scored pos-
itively was food. A MANOVA was performed across the domains to
ascertain whether the groups differed significantly from one another.
The overall significant MANOVA (Pillai’s Trace Criterion,
F(28, 292) = 3.02, p < 0.001) was followed up by a Bonferroni post
hoc comparison of the three means for each domain, and the signif-
icant differences between groups are shown in Table II. The Calcutta
sample was found to differ significantly from the other two groups on
satisfaction with income, social life, family, health and self. Of these,
it was only on health satisfaction that the Calcutta respondents
scored significantly lower.
The means for the frequency of experience of positive and negative
affect are shown in Table III. The overall MANOVA [F(14,
288) = 1.55, p < 0.10] was followed by post hoc comparisons of the
groups for each emotion; although the overall MANOVA was only
of marginal significance. As can be seen in Table III, only a few
significant differences were found between the three sample groups. It
ROBERT BISWAS-DIENER AND ED DIENER
196

is noteworthy that all three samples reported unusually high levels of
negative affect. In fact, the levels of NA were higher than PA in
Calcutta, and almost equal to PA in the U.S. samples. In all three
samples in the current study the frequency of worry exceeded that of
joy. These results differ markedly from most samples, where the
difference between pleasant and unpleasant emotion scores is strongly
positive (Diener and Diener, 1996).
Variables Associated with Basic Needs
Because basic physical needs such as adequate shelter have long been
assumed to be a prerequisite for psychological health, special
attention to variables associated with basic needs is warranted. As
can be seen in Table II the three samples were generally dissatisfied
with their material domains of life. The high level of reported food
satisfaction is a notable exception. In Calcutta a single basic needs
variable in Calcutta – crowding – predicted general life satisfaction,
but not in the expected direction. The total number of people
sharing pavement space with the research participant correlated
significantly and positively with general life satisfaction (r = 0.37,
p < 0.05), with the more people sharing space reporting higher life
satisfaction! It is possible that this finding is caused by the social
benefits of living with friends and family members. Housing
appeared to be an important part of overall subjective well-being in
the American samples: Not only was satisfaction with ones’ own
TABLE III
Positive and negative affect of the homeless
Calcutta DV Fresno
1. Affection 3.65
A
4.62
A
4.68
B
2. Joy 4.08 3.95 4.22
3. Anger 4.18 3.72 3.94
4. Sadness 4.48
A
3.54
B
4.06
A
5. Guilt 3.44 2.92 3.44
6. Worry 4.32 4.33 4.48
7. Pride 3.95 3.90 4.56
Scores range from 1 (never) to 7 (always).Scores in the same row with a different
superscript letter differ by p < 0.05 or less.
SWB OF THE HAPPINESS
197

housing a strong predictor of general life satisfaction, but with sat-
isfaction with many other life domains as well. At Dignity Village
housing satisfaction correlated with general LS (r=0.40,
p < 0.05), as well as with self satisfaction (r = 0.45, p < 0.05). In
Fresno, housing satisfaction correlated significantly (at the
p < 0.001 level) with general life satisfaction (r = 0.57), material
satisfaction ( r = 0.41), food satisfaction (r = 0.46), and income
satisfaction (r = 0.54).
In Calcutta income satisfaction appeared to be extremely impor-
tant to overall subjective well-being. Income satisfaction correlated
significantly (at the p < 0.001 level) with food satisfaction
(r = 0.50), health satisfaction (r = 0.47), and social satisfaction
(r = 0.45).
DISCUSSION
Common sense and common images of homelessness lead us to be-
lieve that ‘‘life on the street’’ is extremely undesirable, with both
physical and psychological downsides. The results of the current
study largely support this conclusion. In both our American samples
respondents reported below neutral levels of life satisfaction while all
three homeless samples reported high negative affect. Across the three
groups the respondents’ reported low levels of satisfaction with
material resources such as income and housing, where housing re-
ferred to sleeping accommodations ranging from tents to makeshift
shelters to community shelters. We also found strong correlations
between housing satisfaction and general life satisfaction. It is
interesting that the three samples in the current study, chosen for
their differences, reported similar trends in material satisfaction. This
suggests that there are certain experiences common to extreme pov-
erty – such as health problems – that transcend differences in culture.
The relatively low satisfaction and high negative affect scores in the
current study are consistent with a larger research literature showing
a relation between income and subjective well-being (e.g., Diener &
Suh, 1999; Argyle, 2001). Although expected, these results provide
additional support for the pressing need to address the problem of
homelessness.
ROBERT BISWAS-DIENER AND ED DIENER
198

Aside from these intuitive findings there are also some unexpected,
and encouraging, results from the current research. For example, all
three samples reported surprisingly high satisfaction with domains
related to the self, including morality, intelligence, and physical
appearance. In fact, scores for all self-related domains were sub-
stantially above the neutral point for all three samples. This suggests
that despite material and emotional hardships, homeless people tend
to like themselves. This is consistent with the literature on the positive
aspects of cognition, such as the finding that people tend to believe
they are better than average on most dimensions (e.g., Heady and
Wearing, 1988). If, in fact, homeless people view themselves in a
positive light, as our research suggests, this finding points to a po-
tential area of strength that may be useful in designing policies and
interventions targeting homelessness. This finding also indicates
directions for future research, suggesting that self-appraisal may be
an aspect of psychology largely invulnerable to the negative effects of
poverty.
Another interesting finding from the current study is the high level
of reported satisfaction with food by members of all three of samples
in the current study. Although the three groups were, on average,
dissatisfied with all domains related to material resources, including
income and housing, all three reported above-neutral satisfaction
with their food, despite large differences in quality, quantity, and
food security. This is consistent with past research, in which groups
of homeless people have reported high satisfaction with food (e.g.,
Marshal et al., 1996). Why would food satisfaction be one of the
most satisfying domains for the homeless people in our study? One
possibility is that the expectations of these groups are so low that any
amount or quality of food is appreciated. Another possible expla-
nation is that these groups are actually hungry, so that they truly
enjoy the food they eat. In the case of the U.S. samples, social service
agencies and government issued food coupons might provide ample
food that is beyond the level of resources obtained in other areas.
Further research needs to be conducted to better understand the
unusually high food satisfaction scores of the homeless.
Perhaps the most counter-intuitive finding is the relatively high
subjective well-being of our sample in India. Despite poorer access to
food, clean water, medical care, opportunities for employment, and
adequate shelter than their counterparts in the United States, the
SWB OF THE HAPPINESS
199

pavement dwellers in Calcutta reported higher levels of life satisfac-
tion. Not only was the general life satisfaction among the Calcutta
sample higher than that of the two American samples, but it was in
the positive range! This is consistent with our past research, in which
impoverished groups in Calcutta reported surprisingly high life sat-
isfaction, given their environmental conditions (Biswas-Diener and
Diener, 2001). What could account for this unexpected positive
quality of life? The answer may lie in specific life domains.
Despite similar trends among the 3 samples in self satisfaction, the
Calcutta sample reported significantly greater satisfaction in this
area. And, although we found a common pattern of dissatisfaction
with material domains, the Calcutta sample reported significantly
higher satisfaction with income. Finally, while people in both
American samples reported satisfaction with social domains that was
near or below the neutral point, the Calcutta sample expressed sig-
nificantly greater satisfaction with both their overall social lives and
their families, well above the neutral point in both instances. In
addition, sharing living quarters with many other people in Calcutta
was correlated with increased life satisfaction. This suggests that the
relations among pavement dwellers in Calcutta are of a higher quality
than among those in the United States, where respondents commonly
complained of mistrusting others. Further support for this can be
found in the fact that the Calcutta respondents reported higher levels
of affection than did members of the other two homeless groups. This
is consistent with our past results from research with homeless people
in Calcutta in which we suggested good social relations may help
protect against the harmful psychological effects of poverty (Biswas-
Diener and Diener, 2001).
There are many possible explanations for the relatively high social
satisfaction of the Calcutta sample, and we discuss 2 of them here: (1)
The state of West Bengal, in which Calcutta is located, has a com-
munist government that may be associated with widespread sympa-
thy for the poor, and therefore fewer feelings of persecution and
marginalization by the poor. In addition, because of the high overall
rate of poverty in India in general, and Calcutta in particular, there
might be less social stigma attached to homelessness, whereas similar
social standing in America is often seen as the result of personal
failure. This is supported by the finding that the Calcutta sample
reported significantly higher levels of satisfaction with ‘‘self’’ than the
ROBERT BISWAS-DIENER AND ED DIENER
200

other two groups. (2) Homelessness in Calcutta is often the result of
macroeconomic conditions, such as high competition for jobs, rather
than with personal pathology such as alcoholism, mental illness, or
physical disability. Because of this, homeless people in Calcutta tend
to stay together as family units, whereas their counterparts in the
Untied States frequently report being estranged from spouses, chil-
dren, and friends. Support for this explanation of higher life satis-
faction can be found in the fact that the Calcutta sample reported
significantly higher satisfaction with ‘‘family’’ than the U.S. samples.
More research needs to be conducted before we can determine the
causes of the unusually high social satisfaction and self satisfaction of
the Calcutta homeless sample. Regardless of the exact cause, though,
it seems likely that social satisfaction is an important component to
overall subjective well-being, and may be especially so for homeless
individuals. This is consistent with the work of Sheldon et al. (2001)
in which they suggest that ‘‘relatedness’’ rivals physical needs as a
fundamental human need, as well as that of Diener and Seligman
(2002), in which they found that good relationships are necessary for
happiness. The results of the current research suggest that social
needs merit renewed attention, and suggest that they may be as
important as physical needs.
Our findings run counter to Maslow’s (1954) classic basic needs
theory, in which he suggested that physical and material needs must
be met before social and psychological health can occur. Although
Maslow’s theory is widely accepted, little is understood about the
specific thresholds for ‘‘meeting needs.’’ For instance, do people need
to eat one meal a day before they are able to develop healthy social
relationships? What if they only eat every-other day? Our research
findings suggest that physical, social, and psychological needs might
interact in a more ‘‘horizontal’’ fashion to create overall psycholog-
ical well-being, and that a success in one can counterbalance a failure
in another. Indeed, we do not argue here that material needs are
unimportant, or less important than social needs, but that the two
may deserve an equal place in the hierarchy of psychological health.
While some physical needs such as a healthy body and adequate
housing are important for subjective well-being, the results of the
current study suggest the possibility that we are less psychologically
vulnerable to material deprivation than previously thought, at least in
comparison to social needs.
SWB OF THE HAPPINESS
201

It is also possible that adaptation plays a role in subjective well-
being as it relates to basic needs. Results from previous research
suggests that people have an extraordinary ability to adapt to certain
physical events such as a handicap (Silver, 1982), but that they adapt
less well to social ills such as the death of a spouse (Stroebe et al.,
1996). It could be that a lack of basic material needs takes a psy-
chological toll, but that people adapt rather well to it as compared to
a lack of good social relationships. It is noteworthy that with ex-
tremely low income the Calcutta sample shows higher income satis-
faction than the other two samples, indicating the importance of
one’s standards to satisfaction judgments. Further research needs to
be conducted to examine the role of adaptation in meeting basic
needs as well research focusing on the relation of physical, social, and
existential needs to one another.
In the end, our research reveals certain common characteristics of
homeless people in diverse settings: they are dissatisfied with their
material quality of life, especially their housing, income, and health.
Similarly, they are highly satisfied with themselves: their morality,
physical appearance, and intelligence. Unfortunately, liking oneself
does not seem sufficient to counterbalance the psychological harm
caused by material deprivation. Good social relationships, on the
other hand, are not only important to subjective well-being (Diener
and Seligman, 2002) but may to some extent avert the psychological
costs of material deprivation.
ACKNOWLEDGEMENTS
The authors would like to thank in Fresno: Robert Levine, Mike
McGarvyn and Poverello House, The Holy Cross Center for Women,
and Ari. At Dignity Village: Jack Tafari, Gaye, Tim, Ibrahim, and all
the research participants. In Calcutta: Avirupa Bhaduri
REFERENCES
Aptekar, L.: 1994, ‘Street children in the developing world: A review of their con-
dition’, Cross-Cultural Research: The Journal of Comparative Social Sciences 28,
pp. 195–224.
ROBERT BISWAS-DIENER AND ED DIENER
202

Argyle M.: 2001, The Psychology of Happiness, 2nd edition (New York, Routledge).
Biswas-Diener, R. and E. Diener: 2000 [A survey of attitudes about homelessness].
Unpublished raw data.
Biswas-Diener, R. and E. Diener: 2001, ‘Making the best of a bad situation: Satis-
faction in the slums of Calcutta’, Social Indicators Research 55, pp. 329–352.
Brown, I.: 1996, The quality of life of marginal citizens, in R. Renwick,
I. Brown and M. Nagler (eds.), Quality of Life in Health Promotion and Reha-
bilitation (Sage Publications, Thousand Oaks), pp. 204–216.
Buhrich, N., T. Hodder and M. Teesson: 2000, ‘Lifetime prevalence of trauma
among homeless people in Sydney’, Australian & New Zealand Journal of Psy-
chiatry 34, pp. 963–966.
Christensen, R.C. and G.D. Grace: 1999, ‘The prevalence of low literacy in an
indigent psychiatric population’, Psychiatric Services 50, pp. 262–263.
DiBiase, R. and S. Waddell: 1995, ‘Some effects of homelessness on the psychological
functioning of preschoolers’, Journal of Abnormal Child Psychology 23, pp. 783–
792.
Diener, E. and R. Biswas-Diener: 2002, ‘Will money increase subjective well-being?:
A review and guide to needed literature’, Social Indicators Research 57, pp. 119–
169.
Diener, E. and C. Diener: 1995, ‘The wealth of nations, revisited: Income and quality
of life’, Social Indicators Research 36, pp. 275–286.
Diener, E. and R.E. Lucas: 2000, ‘Explaining differences in societal levels of hap-
piness: Relative standards, need fulfillment, culture, and evaluation theory’,
Journal of Happiness Studies 1, pp. 41–78.
Diener, E., J. Horowitz and R. Emmons: 1985, ‘Happiness of the very wealthy’,
Journal of Personality Assessment 49, pp. 71–75.
Diener, E. and E.M. Suh: 1999, National differences in subjective well-being, in D.
Kahneman, E. Diener and N. Schwarz (eds.), Well-Being: Foundations of He-
donic Psychology (Russell Sage Foundation, New York), pp. 434–450.
Diener, E., E.M. Suh, R.E. Lucas and H.E. Smith: 1999, ‘Subjective well-being:
Three decades of progress’, Psychological Bulletin 125, pp. 276–302.
Diener, E. and S. Oishi: 2000, Money and Happiness: Income and subjective well-
being across nations, in E. Diener and E.M. Suh (eds.), Culture and Subjective
Well-Being (MIT press, Cambridge, MA), pp. 185–218.
Diener, E. and M.E.P. Seligman: 2002, ‘Very happy people’, Psychological Science
13, pp. 80–83.
Edleman, M.W. and L. Mihaly: 1989, ‘Homeless families and the housing crisis in the
Untied States’, Children & Youth Services Review 11, pp. 91–108.
Frederick, S. and G. Loewenstein: 1999, Hedonic adaptation: From the bright side
to the dark side, in D. Kahneman, E. Diener and N. Schwarz (eds.), Well-Being:
Foundations of Hedonic Psychology (Russell Sage Foundation, New York), pp.
302–329.
Fresno Rescue Mission: 2001, Fresno Rescue Mission Statistics 1999 [On-line].
Available: http://www.fresnorescuemission.org/.
Hagen, J.L.: 1987, ‘The heterogeneity of homelessness’, Social Casework 68, pp. 451–
457.
Heady, B. and A. Wearing: 1988, ‘The sense of relative superiority - Central to well-
being’, Social Indicators Reserach 20, pp. 497–516.
SWB OF THE HAPPINESS
203

Hien, D. and C. Bukszpan: 1999, ‘Interpersonal violence in ‘‘normal’’ low-income
groups’, Woman & Health 29, pp. 1–16.
Inglehart, R. and H-D Klingemann: 2000, Genes, culture, democracy, and. Happi-
ness, in E. Diener and E. M. Suh (eds.), Culture and Subjective Well-Being (MIT
Press, Cambridge, Massechusettes), pp. 165–183.
Lankenau, S.E.: 1999, ‘Stronger than dirt: Public humiliation and status enhance-
ment among panhandlers’, Journal of Contemporary Ethnography 28, pp. 288–
318.
Link, B.G., S. Schwartz, R. Moore and J. Phelan: 1995, ‘Public knowledge, attitudes,
and beliefs about homeless people: Evidence for compassion fatigue’, American
Journal of Community Psychology 23, pp. 533–555.
Marshall, G.N., M.A. Burnam, P. Koegel, G. Sullivan and B. Benjamin: 1996,
‘Objective life circumstances and life satisfaction: Results from the Course of
Homelessness Study’, Journal of health & Behavior 37, pp. 44–58.
Maslow, A.H.: 1954, Motivation and Personality (Harper & Row, New York).
Morgan, M.M., H.W. Goddard and S.N. Givens: 1997, ‘Factors that influence
willingness to help the homeless’, Journal of Social Distress & the Homeless 6, pp.
45–56.
Nyamathi, A., S. Wenzel, C. Keenan, B. Leake and L. Gelberg: 1999, ‘Associations
between homeless women’s intimate relationships and their health and well-being’,
Research in Nursing & Health 22, pp. 486–495.
Pavot, W. and E. Diener: 1993, ‘Review of the Satisfaction with Life Scale’, Per-
sonality Assessment 5, pp. 164–172.
Salovey, P., A.J. Rothman, J.B. Detweiler and W.T. Steward: 2000, ‘Emotional
states and physical health’, American Psychologist 55, pp. 110–121.
Sheldon, K.M., A.J. Elliot, Y. Kim and T. Kasser: 2001, ‘What is so satisfying about
satisfying events? Testing 10 candidate psychological needs’, Journal of Person-
ality and Social Psychology 80, pp. 325–339.
Shlay, A.: 1994, ‘Running on empty: Monitoring the lives and circumstances of
formerly homeless families with children’, Journal of Social Distress & the
Homeless 3, pp. 135–162.
Silver, R.L.: 1982, Coping with an Undesirable Life Event: A Study Of Early
Reactions to Physical Disability (Doctoral Dissertation- Northwestern University,
Evanston, Illinois).
Stroebe, W., M. Stroebe, G. Abakoumkin and H. Schut: 1996, ‘The role of loneliness
and social support in adjustment to loss: A test of attachment versus stress theory’,
Journal of Personality and Social Psychology 70, pp. 1241–1249.
Thomas, F.C.: 1999, Calcutta: The Human Face of Poverty (Penguin Books, Cal-
cutta, India).
United Nations (United Nations Development Program, 2002). Human Develop-
ment Report: Deepening Democracy in a Fragmented World. New York: United
Nations.
Vostanis, P., S. Cumella, J. Briscoe and F. Oyebode: 1996, ‘A survey of psychosocial
characteristics if homeless families’, European Journal of Psychiatry 10, pp. 108–
117.
Wilkenson, R.G.: 1996, Unhealthy Societies: The Affliction of Inequality (Routl-
edge, London).
Wolf, J., A. Burnam, P. Koegel, G. Sullivan and S. Morton: 2001, ‘Changes in
subjective quality of life among homeless adults who obtain housing: A prospec-
ROBERT BISWAS-DIENER AND ED DIENER
204

tive examination’, Social Psychiatry and Psychiatric Epidemiology 36, pp. 391–
398.
Robert Biswas-DienerUniversity of Oregon
E-mail: jayajedi@comcast.net
Ed DienerDepartment of Psychology
University of Illinois and The Gallup Organization
603 E. Daniel Street
Champaign, IL 61820
USA
E-mail: ediener@cyrus.psych.uiuc.edu
SWB OF THE HAPPINESS
205