Health & Place 5 (1999) 259±270

www.elsevier.com/locate/healthplace

Social capital: a guide to its measurement

Kimberly Lochner a, Ichiro Kawachi a,*, Bruce P. Kennedy b
a
Department of Health and Social Behavior, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA
b
Division of Public Health practice, Harvard School of Public Health, Boston, MA, USA

Abstract

The primary aims of this paper are to review the concept of social capital and related constructs and to provide a
brief guide to their operationalization and measurement. We focus on four existing constructs: collective ecacy,
psychological sense of community, neighborhood cohesion and community competence. Each of these constructs
taps into slightly di€erent, yet overlapping, aspects of social capital. The existence of several instruments to measure
each of these constructs calls for further study into their use as measures of social capital. Despite di€erences in the
approach to measurement, there is general agreement that community characteristics, such as social capital, should
be distinguished from individual characteristics and measured at the community level. # 1999 Elsevier Science Ltd.
All rights reserved.

Keywords: Social capital; Cohesion; Community; Neighborhood

1. Introduction (Haan et al., 1987; LeClere et al., 1997). The emphasis

Researchers are increasingly interested in studying
the e€ects of the social environment on health
(Marmot, 1998). Recent examples of studies in which
researchers have attempted to establish aspects of the
social environment as determinants of health include:
the e€ects of income inequality on mortality (Kaplan
et al., 1996; Kennedy et al., 1996; Wilkinson, 1996);
the links between residential segregation and black
infant mortality (LaVeist, 1993; Polednak, 1997); and
the impact of neighborhood deprivation on coronary
risk factors (Diez-Roux et al., 1997), low birthweight
(Roberts, 1997), homicide (Shihadeh and Flynn, 1996),
morbidity (Robert, 1998) as well as all-cause mortality
* Corresponding author. Tel.: +1-617-432-0235; fax: +1-
617-432-3755.
E-mail address: ichiro.kawachi@channing.harvard.edu (I.
Kawachi)
on the characteristics of places in these studies is in
marked contrast to much of contemporary epidemiolo-
gical practice, which tends to focus almost exclusively
on the individuals who live in them and their risk
behaviors (Macintyre et al., 1993). However, even with
the growing interest in examining the relationship of
the social environment to health, there remains the
challenging task of identifying which ecologic charac-
teristics of communities and societies that matter for
population health status.
One characteristic that has generated considerable
attention is the concept of `social capital'. The concept
of social capital originated in the ®elds of sociology
and political science to explain how citizens within cer-
tain communities cooperate with each other to over-
come the dilemmas of collective action, but recently
public health researchers have turned to the notion of
social capital to explain heterogeneities in population
health status across geographic areas (Kawachi and
Kennedy, 1997; Kawachi et al., 1997a). Indeed the rel-

1353-8292/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S 1 3 5 3 - 8 2 9 2 ( 9 9 ) 0 0 0 1 6 - 7
260 K. Lochner et al. / Health & Place 5 (1999) 259±270
evance of `social capital' to public health has deep his-
torical roots. Just over a hundred years ago, Emile
Durkheim demonstrated that social integration was
inversely related to the suicide rate in societies
(Durkheim, 1897 (1951). Since Durkheim, there have
been a variety of attempts to conceptualize and
measure societal characteristics such as social inte-
gration, social cohesion and most recently, social capi-
tal.
The aims of the present paper are: to provide a brief
overview of the concept of social capital, to review the
theory and approaches to measuring constructs appar-
ently related to social capital; and to point out direc-
tions for further research.
We focus our review on four existing constructs: col-
lective ecacy, psychological sense of community,
neighborhood cohesion and community competence.
Each of these existing constructs seem to capture some
fundamental aspect of social capital. We discuss each
of these constructs in terms of their measurement and
their relationship to social capital.
2. Social capital: overview and de®nition
Social capital, as de®ned by its principal theorists
(Coleman, 1990; Putnam, 1993a,b), consists of those
features of social organization Ð such as networks of
secondary associations, high levels of interpersonal
trust and norms of mutual aid and reciprocity Ð
which act as resources for individuals and facilitate
collective action. For example, a community rich in
stocks of social capital is supposedly more likely to
possess e€ective civic institutions and, hence, to pros-
per (Putnam, 1993a) and more likely to be e€ective in
maintaining law and order (Hagan et al., 1995;
Sampson, 1996). According to Robert Putnam, a pol-
itical scientist whose seminal book Making Democracy
Work (1993a) stimulated much of the recent enthu-
siasm for this concept, the level of social capital in so-
ciety can be measured by indicators such as the density
of membership in voluntary associations of all kinds,
the extent of interpersonal trust between citizens, and
their perceptions of the availability of mutual aid
(Putnam, 1993b, 1995). Although several other de®-
nitions of social capital exist (Loury, 1992; Bordieu
and Wacquant, 1992), which di€er slightly from
Putnam's, there is enough consensus to draw some im-
portant generalizations about the nature of social capi-
tal. The most important of these is that social capital
is a collective dimension of society external to the indi-
vidual. Social capital is a feature of the social struc-
ture, not of the individual actors within the social
structure; it is an ecologic characteristic. In this way,
social capital can be distinguished from the concepts
of social networks and support, which are attributes of
individuals.
3. Social capital and public health
The relationship of social capital to population
health was examined recently in an ecologic analysis
based on state-level mortality rates within the United
States (Kawachi et al., 1997b). The indicators of social
capital (adapted from the work of Putnam) were
obtained from the General Social Surveys conducted
by the National Opinion Research Center. The indi-
cators of social capital were measured by responses to
several survey questions. The survey asked respondents
about their membership in a wide variety of voluntary
associations, including church groups, sports groups,
hobby groups, fraternal organizations, labor unions
and so on. Individual responses to these questions
were aggregated up to the state level to create an indi-
cator of per capita membership in voluntary groups.
Respondents in each state were also asked whether
they thought ``Most people can be trusted Ð or you
can't be too careful in dealing with people'' (an indi-
cator of interpersonal trust) as well as whether they
agreed with the question ``Most of the time people try
to be helpful Ð or are they mostly looking out for
themselves?'' (an indicator of perceived norms of reci-
procity). Thirty-nine states were covered by the
sampling frame of the survey. After adjusting for po-
tential sampling bias in the surveys, wide variations
were observed between states in their stocks of social
capital as measured by the above questions. For
example, less than 10% of those surveyed in North
Dakota believed that ``most people can't be trusted'',
compared to more than 40% in Louisiana.
The indicators of social capital were strongly corre-
lated with each other: civic mistrust with group mem-
bership (r=ÿ0.65); mistrust with perceived lack of
reciprocity (r = 0.81); and group membership to lack
of reciprocity (r=ÿ0.54) (all correlations, P < 0.05).
Each indicator of social capital was also strikingly cor-
related with state mortality rates, e.g., the level of mis-
trust in each state was strongly inversely correlated
with age-adjusted all-cause mortality rates (r = 0.79,
p < 0.0001) (Fig. 1).
In regression analyses adjusted for household pov-
erty rates, a 1% increment in the level of mistrust was
associated with an increase in total mortality rate of
4.5 deaths per 100,000 (P < 0.0001). Variations in the
level of trust explained 58% of the variance in total
mortality across states. Lower levels of trust were also
associated with higher rates of most major causes of
death, including coronary heart disease, malignant
neoplasms, stroke, unintentional injury and infant
mortality. If these associations are causal, then it was
 K. Lochner et al. / Health & Place 5 (1999) 259±270
Fig. 1. The relationship between age-adjusted mortality rates and lack of social trust.
estimated that a increase in level of trust by one stan-
dard deviation, or 10%, would be associated with
about a 9% lower level of overall mortality. The two
other indicators of social capital, group membership
and perceived lack of reciprocity, were similarly corre-
lated with age-adjusted mortality rates (r=ÿ0.49, and
0.71, respectively; p < 0.0001).
In addition to social capital's relationship to state
mortality rates, social capital is strongly inversely cor-
related with rates of violent crime (Wilkinson et al.,
1998, in press). The lower the level of social capital,
the higher the rate of violent crime, including homi-
cide, aggravated assault and burglary. Criminologists
have theorized for some time that low social capital is
associated with higher delinquency and violent crime.
According to social disorganization theory (Sampson,
1988; Sampson and Groves, 1989; Sampson, 1996), the
ability of a neighborhood to control crime depends
upon levels of informal social control, e.g., the moni-
toring of spontaneous play groups among children, or
a willingness to intervene to prevent acts such as tru-
ancy and vandalism. In turn, the willingness of local
residents to intervene on behalf of the common good
depends on conditions of mutual trust and solidarity
(social cohesion) among neighbors.
These studies linking social capital and community
well-being (low crime and mortality rates), although
few in number, indicate the potential importance of
261
this construct for public health. On the other hand, the
loose de®nition of social capital leaves ample room for
further re®nement of both theory and di€erent
measurement approaches. In the remainder of this
review, we describe four alternative community-level
constructs that overlap to varying extents with the
notion of social capital. These constructs are collective
ecacy, psychological sense of community, neighbor-
hood cohesion and community competence.
4. Collective ecacy
Based on the work of Bandura (1986), social psy-
chologists have de®ned collective ecacy in several
ways with all de®nitions sharing the notion that group
members believe in the overall ability of the collective
to act e€ectively. Collective ecacy has been de®ned
as ``a sense of collective competence shared among in-
dividuals when allocating, coordinating and integrating
their resources in a successful concerted response to
speci®c situational demands'' (Zaccaro et al., 1995, p.
309). Like individual self-ecacy, collective ecacy is
situational, not universal.
Sampson et al. (1997) proposed that the di€erential
ability of neighborhoods to realize the common values
of residents and maintain public order is a determinant
of neighborhood variation in violence. They developed
262 K. Lochner et al. / Health & Place 5 (1999) 259±270
a measure of `collective ecacy', which is de®ned as
social cohesion among neighbors combined with their
willingness to intervene on behalf of the common
good. Their scale of `collective ecacy' represents a
combination of two separate subscales: `social cohe-
sion' and `informal social control'. `Social cohesion' is
comprised of ®ve items, asking respondents how
strongly they agreed (on a 5-point Likert scale) that:
``people in this neighborhood can be trusted''; ``this is
a close-knit neighborhood''; ``people around here are
willing to help their neighbors''; ``people in this neigh-
borhood generally don't get along with each other'';
and ``people in this neighborhood do not share the
same values'' (the last two statements were reverse
coded). In other words, the `social cohesion' subscale
tapped several elements (trust, norms of reciprocity)
that resonate with the notion of social capital.
The `informal social control' subscale, by contrast, is
much more rooted in criminology. Respondents were
asked about the likelihood (on a 5-point Likert scale)
that their neighbors could be counted on to intervene
in the following situations: children were skipping
school and hanging out on a street corner; children
spray-painting grati on a local building; children
showing disrespect to an adult; a ®ght breaking out in
front of their house; and the ®re station closest to their
home being threatened with budget cuts.
The scales were developed on a 1995 survey of 8783
residents residing in 343 neighborhoods of Chicago,
IL. Importantly, individual responses to the survey
were aggregated to the neighborhood level (equivalent
in area to 2±3 census tracts), and analyzed as an ecolo-
gic, not an individual, characteristic. The `social cohe-
sion' and `informal social control' subscales turned out
to be strongly correlated across neighborhoods
(r = 0.8, P < 0.001), hence they were combined to
form a single scale of `collective ecacy'. Evidence for
the internal consistency reliability of these subscales
was provided from an earlier pilot study (Barnes et al.,
1997). The alpha coecients for the pilot versions of
the `social cohesion' and `informal social control' sub-
scales were 0.89 and 0.82, respectively. Consistent with
the theory of social capital, the measure of `collective
ecacy' was signi®cantly correlated with voluntary
group participation (r = 0.45; p < 0.01), as well as to
neighborhood services (r = 0.21).
The predictive ability of the `collective ecacy' scale
was examined in a cross-sectional multilevel analysis of
crime in the 343 neighborhoods of Chicago enrolled in
the Project on Human Development in Chicago
Neighborhoods. After controlling for individual-level
sociodemographic characteristics, neighborhood collec-
tive ecacy showed a strong inverse association with
measures of perceived neighborhood violence, violent
victimization and homicide events. For example, a 2
standard deviation elevation in neighborhood collective
ecacy was associated with a 39.7% reduction in the
expected homicide rate (Sampson et al., 1997).
5. Psychological sense of community
Turning to the ®eld of community psychology, at
least 30 separate studies have been published since the
1970s on the concept of ``psychological sense of com-
munity'' (Hill, 1996). Originally formulated by Sarason
(1974), the construct was later re®ned by MacMillan
and Chavis (1986) to include four separate dimensions:
membership (the sense of feeling part of a group); in¯u-
ence (a bidirectional concept that refers to the sense
that the individual matters to the group, and that the
group can in¯uence its members, thereby creating
cohesiveness through community norms); integration
(the sense that members' needs will be met by the
resources received through their membership in the
group); and shared emotional connection (the sense of
shared history in the community). Selected examples of
existing instruments, along with evidence on their re-
liability and validity, are shown in Table 1.
From Table 1, it is clear that the existing measures
of ``psychological sense of community'' include several
individual items that tap into the same indicators of a
community's stock of social capital as de®ned by
Putnam (1993a,b). For example, questions regarding
involvement/participation of residents in churches or
local neighborhood associations; whether community
leaders can be trusted; whether people in the neighbor-
hood are sociable; and whether people can depend on
each other for help. All these items could be con-
sidered to tap aspects of a community's stock of social
capital.
Despite the considerable progress that community
psychologists have made toward measuring sense of
community, a recent review concluded that (Hill, 1996,
p. 431):
The development of a standardized, operational
de®nition of the construct has eluded researchers.
At least ®ve measures of the construct have been
developed, and there is still a lack of agreement as
to what speci®c dimensions make up psychological
sense of community.
Notwithstanding di€erences in the approach to
measurement across instruments, there is some consen-
sus that ``psychological sense of community'' refers to
a collective characteristic, not to individual relation-
ships and behaviors; and that, being an aggregate vari-
able, it is most usefully measured and studied at the
community level (Hill, 1996). For example Pretty et al.
(1994, 1996) hypothesized that although `sense of com-
munity' is related to individual-level social support and
Table 1
Summary of concepts used in the measurement of Sense of Communitya (NR=not reported)

Doolittle and Glynn, Bachrach and Chavis et al., Davidson and Davidson Perkins et al., Julian et al.,
MacDonald, 1981 Zautra, 1985 1986 Cotter, 1986; and Cotter, 1990 1997
1978 1991 1993

Membership
Participation in churches or local organizations [ [ [ [
(e.g. PTA, youth groups, business/civic groups)
Have a sense of civic duty [
Desire for organizational participation [
Being a member of this community is like [ [
being a member of a group of friends
Feel that you belong/at home in community [ [ [ [ [
In¯uence
Involvement in political clubs/organizations or [
issue- or action- oriented groups
Political ecacy, e.g. I feel I can contribute to [ [
local/city politics if I want to
Neighbors ask your opinions; you ask your [
neighbors opinions
Feel that I am an important part of this [ [
community
Care about what my neighbors think of my [
actions
Interested in knowing what goes on in community [ [
Can trust community leaders [ [
Sharing of values/ful®llment of needs
Perception of block's ability to solve problems (e.g. [ [ [ [
if there is a problem the community can solve it)
K. Lochner et al. / Health & Place 5 (1999) 259±270

Have in¯uence on improving block/what this [ [ [

community is like
Perception of block/community, e.g.: [
Good place for children to play; good place to [ [
raise teenagers
Feel safe here [ [
Police in community are friendly; e€ective [ [
Pretty [
Important that community does well and care [ [ [
what community is like
Satisfaction with block [ [ [ [ [
Similarity with others, e.g.:
(continued on next page)
263
Table 1 (continued )
264

Doolittle and Glynn, Bachrach and Chavis et al., Davidson and Davidson Perkins et al., Julian et al.,
MacDonald, 1981 Zautra, 1985 1986 Cotter, 1986; and Cotter, 1990 1997
1978 1991 1993

People in community share the same values; [ [ [ [

degree to which you agree with values and beliefs
of neighbors
Feel similar to most people in my community [ [
Feel good when someone does something good [ [
for my community
Believe that what is good for my community is [ [ [
good for me; my neighbors and I want the same
things from this community
Shared Emotional Connection
Own or rent home [
Planned length of residency [ [ [
Length of residency [ [ [
Neighboring patterns and relationships with
people in neighborhood, e.g.:
Number of neighbors one can identify by ®rst [ [ [
name or recognize
Visited by neighbors frequently; visit neighbors [ [
frequently
Number of people known well enough to visit; [ [
know most of the adult residents in the
neighborhood
Number of people who run a business or store [
in community one can identify by ®rst name
Friends in my community are part of my [ [
everyday activities
K. Lochner et al. / Health & Place 5 (1999) 259±270

People in neighborhood know/like each other; [ [ [

people on this block generally don't get along
Do/do not like my neighbors [ [
Dicult to make good friends here [ [
Very few of my neighbors know me [ [
Attachment to the community, e.g.:
My neighborhood is a good place for me to live; [ [
it is important for me to live here
Like my neighborhood/house [ [
Proud to tell others where I live [ [
Would be sorry if I had to move [ [ [
Supportive relationships, e.g.:
People can depend on each other/willing to help [ [
 K. Lochner et al. / Health & Place 5 (1999) 259±270 265

Julian et al.,
social networks, it should be viewed as a distinct, eco-
logical attribute that is not dependent on actual experi-

a=0.90
ence of social support. To illustrate this point, Pretty
1997

NR
15
et al. (1994) quoted an adolescent in their study who

[
stated he would feel much more comfortable asking
Perkins et al.,

for help from a stranger who lived on his block than

he would approaching a stranger on any other block.

a=0.80
This was true even though the adolescent had not met
1990

NR
12
most of his neighbors.
The concept of social capital, like `sense of commu-

scales=0.80±
nity', applies to communities in both the geographic

Correlation
and Cotter,
Davidson

and relational sense, and should be considered distinct

between
a=0.84
from individual characteristics. For example, a particu-
1993

0.87
lar individual may be mistrustful of others (scoring
5

high on a personality scale such as the Cook±Medley

Factor analysis;

Hostility inventory), but his/her experience of commu-

Davidson and
Cotter, 1986;

nity is also likely to depend on the extent of trust

among others around them. In other words, whether
emerged
1 factor
a=0.75

or not the community to which the individual belongs

1991

17

is socially cohesive depends to a large extent on the

[

behaviors of others. Identical individuals might there-

Chavis et al.,

fore experience quite divergent health outcomes

explained

depending on the characteristics of the social environ-

criterion
variance
a=0.97

96% of

ment in which they grow up and live. Proof of the

1986

existence of such `contextual e€ects' on health out-

23

comes is the ultimate challenge confronting those who

Bachrach and

investigate the social environment.

Face validity
Zautra, 1985

The four domains of ``sense of community'' are based on MacMillan and Chavis (1986).
a=0.76

6. Neighboring and neighborhood cohesion

7

Reported

A related concept to `sense of community' is that of

validity
tests of
Glynn,

KR-20

several
=0.97

`neighboring' and `neighborhood cohesion'. There is

1981

some debate about the importance of neighbors in a

60
[

time when the de®nition of community is less

Factor analysis;

explained 54%

depended on geographical boundaries and when many

Doolittle and
MacDonald,

of the social relationships people engage in occur out-

of variance
6 factors

side their residential neighborhood (Unger and

Wandersman, 1985). Be that as it may, the measure-
1978

NR
26
[

ment of neighboring is relevant for capturing locally

available resources, in the form of a€ective and instru-
I can always ®nd someone to talk to; there are

mental support. Neighboring involves social inter-

people in community other than my friends who

Most friends in community are here to stay

actions, by which residents establish social connections

that are either personal or at the neighborhood level.
Neighborhood social networks frequently come about
as a result of the formation of neighborhood organiz-
ations or block groups, which in turn serve several
Can be alone when need to

functions besides encouraging sociability. In fact, block

groups are an example of what Coleman (1988)
referred to as `appropriable social organizations',
really care about me
Table 1 (continued )

which are themselves a form of social capital. For

Number of items

example, Coleman described a resident's organization

in an urban housing project, which initially formed for
Reliability

the purpose of pressuring builders to ®x various pro-

Validity

blems (leaks, crumbling sidewalks, etc). After the pro-

a

blems were solved, the organization remained as

266 K. Lochner et al. / Health & Place 5 (1999) 259±270

Table 2
Summary of concepts used in the measurement of neighboringa and neighborhood cohesion. (NR=not reported)

Hunter, Riger and Unger and Buckner, Perkins et al.,

1975 Lavrakas, Wandersman, 1988 1990
1981 1983

Social Interaction
Casual interaction, e.g.:
Number of people you can recognize on your [ [
block; number of people who live on this
block that you know by name
Frequency that you chat with neighbors [ [
Frequency that you visit informally in [ [ [
neighbors home; number of people whom
have a neighborly relationship with
Social support, e.g.:
socio-emotional support [ [ [
instrumental support [ [ [ [
informational support [ [ [ [
Social networks, e.g.:
personal networks (both within and outside [ [
neighborhood)
neighborhood social networks (e.g., [
neighborhood groups)
A€ective Bonds
Sense of mutual aid (e.g. feeling that you can [
rely on your neighbors; believing that potential
help is available)
Sense of community (e.g. membership, [ [ [
in¯uence, sharing of values and emotional
connection)
Attachment to place (e.g. plan to remain a [ [
resident of this neighborhood for a number of
years; would be sad if I had to move from this
neighborhood)
Number of items 13 6 10 18 10

Reliability NR a=0.56±0.59 a=0.88 a=0.95 a=0.76

NR Factor analysis; NR Criterion validity, NR
Validity
2 factors emerged p < 0.001

a
The dimensions of neighboring are based on Unger and Wandersman (1985). The cognitive component of neighboring
described in Unger and Wandersman (1985) is not shown because no studies measured this component.

available social capital to improve the quality of life
for residents. The general point is that an organization,
once brought into existence for one set of purposes
can also be appropriated for other uses, thus constitut-
ing a form of social capital.
Selected instruments for measuring neighboring and
neighborhood cohesion are shown on Table 2. In ad-
dition to tapping neighborhood networks and casual
interaction with neighbors, existing instruments also
attempt to directly measure the availability of instru-
mental and emotional social support within the neigh-
borhood. This approach captures the public goods
aspect of social capital, i.e., the notion that resources
are potentially available to all members within the
social structure, not just those who `invest' in main-
taining social relationships.
The important aspect about the measurement of
neighborhood social support is that it is an aggregate,
community-level characteristic, not an individual-level
measure. In other words, individual responses to an
instrument such as Buckner's Neighborhood Cohesion
Index (1988) are aggregated up to the level of the com-
munity in order to derive an overall neighborhood
cohesion score. As Buckner (1988) (p. 775) notes:
. . . it is important to distinguish between an individ-
ual variable which is used to measure an individual-
level attribute and an aggregate individual variable
 K. Lochner et al. / Health & Place 5 (1999) 259±270
(a mean) which is used to infer a neighborhood-
level attribute . . . if these individual-level scores are
aggregated and averaged the resultant mean score
(an aggregate individual-level variable) is said to
form a measure of the cohesiveness of that collec-
tive of neighborhood residents.
This method for measuring `cohesion' follows tra-
ditional methods employed by social psychologists,
which often use a mean score from aggregated individ-
ual-level data to quantify some collective attribute.
Buckner's neighborhood-level analysis (n = 3) found
that using the mean value scores did distinguish
between neighborhood cohesion levels in the predicted
order (Buckner, 1988).
7. Community competence
Turning ®nally to the ®eld of health promotion, the
concept of `community competence' has potential over-
lap with social capital. Community competence can be
thought of as the problem solving ability of a commu-
nity that arises through collective e€ort. According to
its principal theorist, Cottrell, it is a distinctly group
phenomenon. Cotrell (1976) proposed eight essential
preconditions of a competent community (Table 3).
Among the conditions that are particularly relevant to
social capital are: commitment to the community (e.g.,
whether residents volunteer for community activities);
participation (whether residents belong to civic clubs,
fraternal organizations and church); and community
social support, a dimension added by Eng and Parker
(1994). The remaining dimensions tap predominantly
into the stock of political skills (or political ecacy)
that are available within the community. These dimen-
sions include: articulateness; management of relations
with larger society; and the machinery for facilitating
participant interaction and decision making. The latter
dimensions re¯ect the practical interest of health pro-
motion and health education experts in assisting com-
munities to mobilize resources to implement social
change.
Interestingly, theorists of social capital would view
political ecacy as a consequence of social capital. For
example, Verba et al. (1995) have put forward the
`civic voluntarism model' to explain patterns of politi-
cal participation. According to this model, political
skills within the community are developed as a by-pro-
duct of civic engagement in secondary associations.
Thus (Verba et al., 1995, p. 40):
Political activity is embedded in the nonpolitical in-
stitutions of civil society . . . Undertaking activities
that themselves have nothing to do with politics Ð
for example, running the PTA fund drive or mana-
267
ging the soup kitchen Ð can develop organizational
and communications skills that are transferable to
politics. In addition, these nonpolitical institutions
can act as the locus of attempts at political recruit-
ment.
In other words, the measurement of `community
competence' seems to combine the outcome (political
ecacy) with its predictor (social capital). As with the
concept of `sense of community', a review of existing
measures of community competence concluded that:
`the lack of a clear operational de®nition of commu-
nity competence has resulted in an inconsistent array
of indices . . . The concept itself is still too abstract,
requiring more grounded theory building through dis-
covery in the ®eld' (Eng and Parker, 1994, p. 204).
8. Discussion
Having reviewed the concept of social capital and
potentially related measures, what concluding com-
ments can be made? First, we have at present no single
de®nition of social capital, just as there are no single
de®nitions of the other community-level concepts
reviewed. Instead, the di€erent instruments tap slightly
di€erent Ð but also substantially overlapping Ð
characteristics of social relations at the community
level, including civic participation in voluntary associ-
ations, norms of mutual aid and reciprocity, as well as
levels of interpersonal trust. The availability of several
instruments calls for a study in which all of them are
simultaneously administered, and the extent of shared
variance examined.
Despite di€erences in the approach to measurement,
a strong and consistent theme emerging from this
review is the almost universal agreement that commu-
nity characteristics ought to be distinguished from indi-
vidual characteristics, and measured at the community
level. On the other hand, much work remains to be
carried out to determine the most valid ways to go
about measuring collective attributes. To date, investi-
gators have used the social survey approach (with indi-
vidual responses aggregated up to the community
level). However, alternative approaches Ð such as sys-
tematic social observation, or using intrinsic measures
of community characteristics Ð may yield additional
insights into the processes of social cohesion. By
`intrinsic' measures, we mean the measurement of com-
munity characteristics that do not rely on aggregating
the responses of individuals. Creative examples of such
intrinsic measures might include directly observable
features of a community, such as the extent to which
neighborhood sidewalks are cleared after a snow storm
(an indicator of reciprocity), or whether local gas
268 K. Lochner et al. / Health & Place 5 (1999) 259±270

Table 3
Summary of concepts used in the measurement of community competencea

Goeppinger Eng and Parker,

and Baglioni, 1994
1985

Commitment
Length of residence [
Care about look of community [ [
Would be sad if had to move [
Casual interaction with other [ [
Use services in town [
People feel they have an active role in making community function [
People volunteer for community activities [
People engage in favors, e.g. lending of goods [
People feel that what happens in community can a€ect them [
Participation
People belong to local service or civic clubs, fraternal organizations, church [ [
People stay in the community for fun [
When things are done in the community, it is often the same few people doing [
all the work
Number of registered voters in community [
Self±other awareness and clarity of situational de®nition
Residents have strong opinions about way things are done [
Local government tries to in¯uence what goes on in community [
Compared with other towns, this town provides services, e.g. parks, libraries, [
health services, to take care of most people's needs
Social supportb
Instrumental, emotional and informational support [
Articulateness/E€ective communication
People will express an unpopular or dissenting opinion [
People are willing to stand before an outside group and state community needs [
Con¯ict containment and accommodation
People speak out about di€erences and work together/®nd ways to solve [ [
di€erences
Management of relations with the larger society
People in this community try to in¯uence what goes on in the larger [ [
surrounding areas
Local elected ocials represent my views [
Number of people in community who own their home [
Number of people who know about services/resources o€ered in larger society [
Machinery for facilitating participant interaction and decision making
People in town try to in¯uence what happens in this community [ [
Town council is e€ective [
People in community have most (some, few, or none) of friends living in [
community
People are willing to help work during a political campaign [
People are willing to contact, e.g. phone, write a letter, to local ocials [
22 41
Number of items
NR Reported separately
Reliability
for each dimension
Validity Discriminant and Face validity
criterion validity
assessed

a
The dimensions for `community competency' are based on Cottrell (1976). (NR=not reported).
b
Not one of Cottrell's eight dimensions included in Eng and Parker (1994).
 K. Lochner et al. / Health & Place 5 (1999) 259±270
stations demand prepayment before motorists can ®ll
their tanks (an indicator of trust).
The issue of measurement raises an additional issue:
what is the relevant unit of aggregation? As with any
ecological characteristic, there are several possible
choices for aggregating information on social capital.
It makes sense to measure social capital within neigh-
borhoods, or at the level of states, or even whole
countries. There is every reason to believe that social
capital at each of these levels can exert an in¯uence on
population health. But the processes determining the
causes and consequences of social capital may be quite
di€erent at various levels of aggregation. For example,
at the level of the local community, social capital
depends much more on the day-to-day interactions
between neighbors than on distal social policies. In
contrast, the level of social capital at the state or
country level is more likely to re¯ect the in¯uence of
culture, social and economic policies and other macro-
social forces. More theoretical and empirical work is
needed to clarify these di€erences.
In summary, despite the promise suggested by the
title of this paper, more questions have been raised
than answered. For example, is there added value in
talking about `social capital', over and above the exist-
ing notions of social relations and community cohe-
sion? Does social capital merely represent `old wine in
new bottles'? At the present time, insucient theoreti-
cal and empirical work has been carried out to o€er a
de®nitive guide to the measurement of social capital.
Still, a variety of evidence from di€erent disciplines
has linked elements of social capital to collective out-
comes ranging from crime prevention, civic insti-
tutional performance, to population longevity. The
potential sweep of this concept in explaining such
diverse outcomes is sucient by itself to justify further
explorations in this area.
Acknowledgements
I.K. and B.P.K. are recipients of the Robert Wood
Johnson Foundation Investigator Awards in Health
Policy Research.
References
Bachrach, K.M., Zautra, A.J., 1985. Coping with a commu-
nity stressor: the threat of a hazardous waste facility.
Journal of Health and Social Behavior 26, 127±141.
Bandura, A., 1986. Social Foundations of Thought and
Action: a Social Cognitive Theory. Prentice-Hall,
Englewood Cli€s, NJ.
Barnes, J., Sampson, R.J., Kindlon, D. and Reiss, A.J. Jr
1997. A community survey approach to ecological assess-
269
ment: results from a pilot survey. In: Earls, F., Buka, S.L.
(Eds.), Project on Human Development in Chicago
Neighborhoods. A Technical Report to the National
Institute of Justice, pp. 34±45.
Bordieu, P., Wacquant, L., 1992. Invitation to Re¯exive
Sociology. Chicago University Press, Chicago.
Buckner, J.C., 1988. The development of an instrument to
measure neighborhood cohesion. American Journal of
Community Psychology 16, 771±791.
Chavis, D.M., Hogge, J.H., McMillan, D.W., 1986. Sense of
community through Brunswick's Lens: a ®rst look.
Journal of Community Psychology 14, 24±40.
Coleman, J.S., 1988. Social capital in the creation of human
capital. American Journal of Sociology 94, S95±S120.
Coleman, J.S., 1990. Foundations of Social Theory. Harvard
University Press, Cambridge, MA.
Cottrel, L.S., 1976. The competent community. In: Kaplan,
B.H., Wilson, R.N., Leighton, A.H. (Eds.), Further
Explorations in Social Psychiatry. Basic Books, New
York, pp. 195±209.
Davidson, W., Cotter, P., 1986. Measurement of sense of
community within the sphere of the city. Journal of
Applied Social Psychology 16, 608±619.
Davidson, W., Cotter, P., 1991. The relationship between
sense of community and subjective well-being: a ®rst look.
Journal of Community Psychology 19, 246±253.
Davidson, W., Cotter, P., 1993. Psychological sense of com-
munity and support for public school taxes. American
Journal of Community Psychology 21, 59±66.
Diez-Roux, A.V., Nieto, J., Muntaner, C., Tyroler, H.A.,
Comstock, G.W., Shahar, E., Cooper, L.S., Watson, R.L.,
Szklo, M., 1997. Neighborhood environments and coron-
ary heart disease: a multilevel analysis. American Journal
of Epidemiology 146, 48±63.
Doolittle, R., MacDonald, D., 1978. Communication and a
sense of community in a metropolitan neighborhood: a
factor analytic explanation. Communication Quarterly 26,
2±7.
Durkheim, E., 1897. In: Simpson, G. (Ed.), Suicide. The Free
Press, New York, p. 1951 (Spaulding, J.A. and Simpson,
G. (transl)).
Eng, E., Parker, E., 1994. Measuring community competence
in the Mississippi Delta: the interface between program
evaluation and empowerment. Health Education Quarterly
21, 199±220.
Glynn, T., 1981. Psychological sense of community: measure-
ment and application. Human Relations 34, 789±818.
Goeppinger, J., Baglioni, A.J., 1985. Community competence:
a positive approach to needs assessment. American
Journal of Community Psychology 13, 507±523.
Haan, M., Kaplan, G.A., Camacho, T., 1987. Poverty and
health: prospective evidence from the Alameda County
Study. American Journal of Epidemiology 125, 989±998.
Hagan, J., Merkens, H., Boehnke, K., 1995. Delinquency and
disdain: social capital and the control of right-wing extre-
mism among East and West Berlin youth. American
Journal of Sociology 100, 1028±1052.
Hill, J.L., 1996. Psychological sense of community: sugges-
tions for future research. Journal of Community
Psychology 24, 431±438.
Hunter, A., 1975. The loss of community: an empirical test
270 K. Lochner et al. / Health & Place 5 (1999) 259±270
through replication. American Sociological Review 40,
537±552.
Julian, D.A., Reisch, T.M., Carrick, R.V., Katrencich, C.,
1997. Citizen participation-lessons from a local United
Way planning process. Journal of the American Planning
Association 63, 345±355.
Kaplan, G.A., Pamuk, E., Lynch, J.W., Cohen, R.D.,
Balfour, J.L., 1996. Income inequality and mortality in the
United States. British Medical Journal 312, 999±1003.
Kawachi, I., Kennedy, B.P., 1997. Health and social cohesion:
why care about income inequality? British Medical Journal
314, 1037±1040.
Kawachi, I., Kennedy, B.P., Lochner, K. 1997. Long live
community: social capital as public health, The American
Prospect, Nov./Dec., pp. 56±59.
Kawachi, I., Kennedy, B.P., Lochner, K., Prothrow-Stith, D.,
1997b. Social capital, income inequality and mortality.
American Journal of Public Health 87, 1491±1498.
Kennedy, B.P., Kawachi, I., Prothrow-Stith, D., 1996. Income
distribution and mortality: test of the Robin Hood Index
in the United States. British Medical Journal 312, 1004±
1007.
LaVeist, T.A., 1993. Segregation, poverty and empowerment:
health consequences for African Americans. The Milbank
Quarterly 71, 41±64.
LeClere, F.B., Rogers, R.G., Peters, K.D., 1997. Ethnicity
and mortality in the United States: individual and commu-
nity correlates. Social Forces 76, 169±198.
Loury, G. 1992. The economics of discrimination: getting to
the core of the problem. Harvard Journal of African
American Public Policy.
Macintyre, S., MacIver, S., Sooman, A., 1993. Area, class
and health: should we be focusing on places or people?
Journal of Social Policy 22, 213±234.
MacMillan, D.W., Chavis, D.M., 1986. Sense of community:
a de®nition and theory. Journal of Community Psychology
14, 6±23.
Marmot, M., 1998. Improvement of social environment to
improve health. Lancet 351, 57±60.
Perkins, D.D., Florin, P., Rich, R.C., Wandersman, A.,
Chavis, D.M., 1990. Participation and the social and
physical environment of neighborhood blocks: crime and
community context. American Journal of Community
Psychology 18, 83±115.
Polednak, A.P., 1997. Segregation, Poverty and Mortality in
urban African Americans. Oxford University Press, New
York.
Pretty, G., Andrews, L., Colett, C., 1994. Exploring adoles-
cents sense of community and its relationship to lonliness.
Journal of Community Psychology 22, 346±358.
Pretty, G., Conroy, C., Dugay, J., Fowler, K., 1996. Sense of
community and its relevance to adolescents of all ages.
Journal of Community Psychology 24, 365±379.
Putnam, R.D., 1993. Making Democracy Work. Civic
Traditions in Modern Italy. Princeton University Press,
Princeton, NJ.
Putnam, R.D., 1993b. The prosperous community. Social
capital and public life. The American Prospect 13, 35±42.
Putnam, R.D., 1995. Bowling alone: America's declining
social capital. Journal of Democracy 6, 65±78.
Riger, S., Lavrakas, P.J., 1981. Community ties: patterns of
attachment and social interaction in urban neighborhoods.
American Journal of Community Psychology 9, 55±66.
Robert, S.A., 1998. Community-level socioeconomic status
e€ects on adult health. Journal of Health and Social
Behavior 39, 18±37.
Roberts, E.M., 1997. Neighborhood social environments and
the distribution of low birthweight in Chicago. American
Journal of Public Health 87, 597±603.
Sampson, R.J., 1988. Local friendship ties and community
attachment in mass society: a multilevel systems model.
American Sociological Review 53, 766±779.
Sampson, R.J., Groves, W.B., 1989. Community structure
and crime: testing social disorganization theory. American
Journal of Sociology 94, 774±802.
Sampson, R.J., 1996. The community. In: Wilson, J.Q.,
Petersili, J. (Eds.), Crime. Institute for Contemporary
Studies, San Francisco, pp. 193±216.
Sampson, R.J., Raudenbush, S.W., Earls, F., 1997.
Neighborhoods and violent crime: a multilevel study of
collective ecacy. Science 227, 918±924.
Sarason, S., 1974. The Psychological Sense of Community.
Prospects for a Community Psychology. Josey-Bass, San
Francisco, CA.
Shihadeh, E.S., Flynn, N., 1996. Segregation and crime: the
e€ect of black social isolation on the rates of black urban
violence. Social Forces 74, 1325±1352.
Unger, D., Wandersman, A., 1983. Neighboring and its role
in block organization: an exploratory report. American
Journal of Community Psychology 11, 291±300.
Unger, D., Wandersmann, A., 1985. The importance of neigh-
bors: the social cognitive, and a€ective components of
neighboring. American Journal of Community Psychology
13, 139±169.
Verba, S., Lehman Schlozman, K., Brady, H.E., 1995. Voice
and Equality. Civic Voluntarism in American Politics.
Harvard University Press, Cambridge, MA.
Wilkinson, R.G., 1996. Unhealthy Societies. The A‚ictions of
Inequality. Routledge, London.
Wilkinson, R.G., Kawachi, I., Kennedy, B.P. 1998. Mortality,
the social environment, crime and violence. Sociology of
Health and Illness (in press).
Zaccaro, S.J., Blair, V., Peterson, C., Zazanis, M., 1995.
Collective ecacy. In: Maddux, J.E. (Ed.), Self-Ecacy,
Adaptation and Adjustment: Theory, Research and
Application. Plenum Press, New York, NY.