American Journal of Community Psychology, Vol. 13, No.

2, 1985

Social Support Measurement

Charles H . Tardy ~
University of Southern Mississippi

Instruments designed to measure social support are evaluated. Five aspects

of social support are identified and instruments capable of assessing these
components are described. Reliabifity and validity evidence associated with
the measures are reviewed. The purpose o f this discussion is to facilitate the
study of social support by clarifying the decisions faced by researchers and
increasing awareness of the capabilities and limitations of currently avail-
able instruments.

Evidence of mankind's desire to affiliate abounds. Clubs, churches, bridge

parties, car pools, and thousands of other types of formal and informal groups
attest to this social fact. Only recently, however, have the positive physical
and mental outcomes of such affiliations received attention (Cassel, 1976;
Heller, 1979; Kaplan, Cassel, & Gore, 1977). The growth of research on the
topic of social support is indicated by the frequency of articles on that sub-
ject published in this journal.
From the abundance of recently published research there appear to be
no barriers to progress in the study of social support. However, the lack of
agreement concerning its conceptualization and measurement impedes the
production of valid generalizations about the development and functioning
of social support (Sandler & Barrera, 1984). As stated in a recent review of
the literature on psychological disorder and social support, "Most support
questionnaires are ad hoc measures with questionable reliability and unknown
validity" (Leavy, 1983, p. 16). Another writer phrased this problem less
diplomatically when he said "the concept of social support has been opera-

1All correspondence should be sent to Charles H. Tardy, University of Southern Mississippi,

Speech Communication, Southern Station, Box 5131, Hattiesburg, Mississippi 39406.

187

0091-0562/85/04000187504.50/0© I985 PlenumPublishingCorporation

188 Tardy

tionalized in a somewhat bewildering assortment of ways" (Wilcox, 1981,

p. 98). Most of the few rigorous studies of the measurement properties of
social support instruments do not attempt to build systematically on previ-
ous methodologies. Rather, authors frequently appear to start the develop-
ment process anew, a practice which is costly and inefficient.
This article attempts to resolve these problems by providing a review
of the currently available measures of social support. A description of issues
related to the conceptualization of social support precedes a discussion of
strategies utilized in previous noteworthy studies. Readers are informed of
methodological options available to researchers interested in the measure-
ment of social support.

CONCEPTUAL ISSUES

The varied operational definitions of social support appearing in the

literature are symptomatic of the multiple interpretations of the meaning of
social support. Unfortunately the differences too frequently go unnoticed,
contributing to misunderstandings and inaccurate generalizations. The so-
lution, however, is not to get people to reach consensus on a single defini-
tion. Rather, the solution is to recognize and discuss issues involved in defining
the concept at the theoretical and operational levels. This paper describes
five such issues in an attempt to both clarify the decisions facing researchers
and to organize the differences among the approaches taken by various
authors. Figure 1 illustrates these issues.
1. Direction. Social support is both given and received. The distinc-
tion between these two directions in which social support occurs is clear and
fundamental. Though most research focuses on the receipt of social support,
a few studies additionally examine the conveyance of support (Cohen &
Sokolovsky, 1979; Fischer, 1982; McFarlane, Neale, Norman, Roy, &
Steriner, 1981; Miller & Ingham, 1976; Tolsdorf, 1976). Researchers must
decide to investigate one or both.
2. Disposition. Some studies examine the availability of support (e.g.,
Barrera, 1981; Sarason, Levine, Basham, & Sarason, 1983) while others ex-
amine its enactment (e.g., Barrera, Sandier, & Ramsay, 1981; Burke, 1978;
Carveth & Gottlieb, 1979). Support availability refers to the quantity or qual-
ity of support to which people have access. The actual utilization of these
support resources is referred to as enacted support.
3. Description~Evaluation. Evaluation and description constitute two
distinct facets of social support. Illustrating evaluation are the studies of peo-
ple's satisfaction with their social support (e.g., Barrera, 1981; Cauce, Fel-
net, & Primavera, 1982). On the other hand, some studies propose only to
Social Support Measuremenl 189

SOCIAL SUPPORT

DIRECTION RECEIVED PROVI DED

DISPOSITION AVAILABLE ENACTED

DESCRIPTION/ DESCRIBED EVALUATEO

EVALUATION

CONTENT

zNN
EMOTIONAL INSTRUMENTAL INFORMATIONAL APPRAISAL

NETWORK FAMILY CLOSE NEIGHBORS CO-WORKERS COMMUNITY PROFESSIONALS

FRIENDS

Fig. 1. Aspects of social support.

describe social support (e.g., Barrera et al., 1981; Hammer, 1981). That these
two components are distinct parts rather than competing conceptualizations
of social support is indicated by the fact that several studies examine them
both (e.g., Monroe, 1983; Sarason et al., 1983).
4. Content. From situation to situation the content of support varies
greatly. Loaning someone money involves different commitments and
processes than does patting someone on the back, though both acts may be
supportive. Perhaps the most useful typology of support content eminates
from the discussion by House (1981). He distinguishes among four types:
emotional, instrumental, informational, and appraisal. The first type refers
to the provision of trust, empathy, and love. Emotional support involves
caring. Helping behaviors such as loaning money or giving one's time and
skill are examples of instrumental support. Informational support refers to
advice while appraisal support refers to evaluative feedback, e.g., "you're
doing a good job!" Although this category system accounts for most types
of support content, it may not be exhaustive. Alternatives are suggested by
190 Tardy

Fiore, Becker, and Coppel (1983), Gottlieb (1978), and Walker, MacBride,
and Vachon (1977). 2
5. Network. The social dimension of social support constitutes the
fourth issue. 3 The term "network" was selected because the more common
term "source" entails a specific direction of the support. The categories list-
ed in Fig. 1 are intended only to illustrate the possible members of the net-
work. Some studies are concerned with the mere existence of the network
while others consider the characteristics of the people in the support network/
Studies must consider the people providing a n d / o r receiving support.
Though not exhaustive, these five issues encompass the primary ele-
ments of social support. All research on this topic makes assumptions about
the direction, disposition, evaluation/description, content, and network of
support. Additionally, these issues are interdependent. Operational defini-
tions of social support make assumptions about each one. For example, one
author may be interested in a measure which describes the social support
received by a person from his/her family while another may want to describe
the social support available to the person from his/her family. Too frequently,
however, the author's position on these issues must be inferred from rather
limited information. These issues should be discussed explicitly so that con-
ceptualizations are clear. Appropriate selections among alternative measure-
ment strategies necessitate decisions on these conceptual issues.

MEASUREMENT DEVICES

The author conducted a systematic search for studies containing meas-

ures of social support. Though computerized searches of Psychological Ab-
stracts and Sociological Abstracts were undertaken, examination of journal
volumes and articles' references provided most citations. More than 60 rele-
vant studies were located. Literally dozens of different measures have been
published. In selecting instruments for discussion in this paper three criteria

2The content categories can be examined individually and in combination. For example, a study
might observe the range and variety of support content associated with particular support net-
work members.
aThis paper examines support networks. Several authors (Hammer, 1981; Hirsch, 1981; Wellman,
1981) argue that the entire social network should be examined rather than only the network
segment which provides support. These advocates say social network analysis allows the ex-
amination of the larger interpersonal context in which support is embedded and avoids the
positive bias inherent in the examination of only contacts which are supportive. However, the
topic of social networks exceeds the scope of this paper. See Mitchell and Trickett (1980) for
a review of the conceptual issues and operational definitions associated with this approach
to the study of social support.
4Additionally, some authors consider the neighborhood and other environmental features as
sources of support (e.g., Lin, Dean, & Ensel, 1981).
Social Support Measuremen! 191

were used. Instruments which clearly measure some aspect of social sup-
port were chosen over those that did not distinguish among the many aspects
discussed above. Instruments which measure supportive aspects of social re-
lations were chosen over those that focused on other aspects such as social
participation, integration, or contact. Instruments demonstrating reliability
and validity were chosen over those for which no data were available.
These criteria precluded the inclusion of ad hoc measures of social sup-
port, such as marital status, and conglomerate measures, which include items
about job satisfaction, participation in religious activities, etc. (House, 1981).
Though necessary and advantageous in some circumstances, use of such meas-
ures typically poses several significant problems. Such measures inherently
assess something other than or in addition to social support. This inclusion
problem may mask or attentuate the relationship between social support and
other variables, thus leading to the rejection of true hypotheses (Gore, 1981).
On the other hand, that which is included may have an effect independent
of social support, resulting in the acceptance of false hypotheses. Clearly,
absence of precise measurement practices impedes development of theory
about social support and hinders the application of research findings by psy-
chologists and community health practitioners: The purpose of this paper
is to encourage the development and use of precise, reliable, and valid meas-
ures of social support.

Arizona Social Support Interview Schedule

Manuel Barrera (1980, 1981) reports the development of an instrument

to measure several aspects of social support. An interviewer first asks the
subject to give the names or initials of the people who provide support, e.g.,
"Who would you go to if a situation came up when you needed advice?"
(1981, p. 91). The response to this item describes social support available
to the subject. Following each of six such questions, three further questions
phrased to be consistent with each topic are asked: (a) Did the person actu-
ally give support in the past month? (b) Would you have liked more sup-
port? (c) Did you get the support you needed? The answer to the first question
describes the enactment of support while the answers to the second and third
evaluate the enacted support. The author includes a question about conflicts
or unpleasant interactions and a series of questions requesting descriptions
of the individuals named by the respondent. From these data the author
reported calculating the following measures: (a) total network size; (b) con-
flicted network size; (c) unconflicted network size; (d) support satisfaction;
and (e) support need.
With a sample of 45 college students, the total size measure yielded a
test-retest correlation coefficient of .88 over a period of 2 or more days.
192 Tardy

Seventy percent of the individuals named appeared on both administrations

of the interview. The size of the conflicted network resulted in a lower
test-retest correlation coefficient of .54 but was still statistically significant.
The internal reliabilities of the size measures are not reported. The coefficient
alphas for the satisfaction and need measures were low, .33 and .52, respec-
tively. The two measures' test-retest correlations were .61 and .80. Together
these analyses suggest that the need and satisfaction measures vary consider-
ably internally but remain relatively stable over time at the aggregate level.
In a study of pregnant adolescents, the measures of conflicted network size,
need, and satisfaction correlated significantly with several measures of
maladjustment-depression, anxiety, and somatization while unconflicted net-
work size, need, and satisfaction correlated significantly with a measure of
negative life events? In a more recent study of college students (Sandier &
Barrera, 1984), only support satisfaction and conflicted network size cor-
related with the three types of psychological maladjustment.
This instrument has the advantage of measuring several aspects of so-
cial support. In addition to the ones described above, the user could calcu-
late support content and source measures from these data. The negative
interaction question also may prove useful for some investigators (Sandler
& Barrera, 1984).
Similar instruments developed by Fischer (1982) and McFarlane et al.
(1981) are described below. When selecting from among these, the questions
should be examined carefully. The content areas assessed by the instruments
differ, though slightly, and consequently users must judge the appropriate-
ness of questions included by each one.

Inventory of Socially Supportive Behaviors (ISSB)

Another measure devised by Manuel Barrera focuses strictly on the

receipt of behaviors providing social support (1981, Barrera et al., 1981).
Subjects report the frequency of occurrence during the past month of such
behaviors as being given a ride to see a doctor or being loaned $25. The self-
report measure consists of 40 such statements describing the enactment of
various content types of support. The scale requests no information about
the support network.
As a univariate measure, the scale has shown a high degree of internal
consistency as indicated by a coefficient alpha of .93 with college students

Sin this and m a n y other studies discussed subsequently, tentative evidence of validity for the
instrument is adduced from the correlation of the social support measure with some other vari-
able. Kerlinger (1973) describes the use of this type of data in assessing construct validity. That
a relationship should be expected between social support and psychological disorder is strong-
ly supported by the literature on this subject (Beels, 1981; Leavy, 1983).
Social Support Measurement 193

and .92 with pregnant adolescents. A test-retest correlation of .88 was ob-
tained over at least 2 days. The ISSB correlated significantly with a measure
of negative life events in two studies but not in a third (Barrera, 1981; San-
dler & Barrera, 1984). Stokes and Wilson (1984) report that social network
characteristics of men but not women predicted ISSB scores.
Two studies factor analyzed the ISSB, both reporting similar but not
identical factor solutions (Barrera & Ainlay, 1983; Stokes & Wilson, 1984).
Reliability coefficients of .85, .71, .83, and .77 were obtained for the emo-
tional support, tangible assistance, cognitive information, and directive
guidance dimensions, respectively, in one study (Stokes & Wilson, 1984). Ad-
ditionally, a measure composed of emotional support items significantly
differentiated men from women, providing tentative evidence of the validity
of content measures derived from the ISSB. The other factor analytic study
did not report internal reliability coefficients or form scales from the four
factors.
The primary purpose of the instrument is to describe the enactment of sup-
portive behaviors. The goal is achieved by identifying such acts which oc-
cured in the past month. This strategy allows the identification of recent events
though neglecting prior ones. If desired, the time frame could be altered.
This instrument specifies no source for the support. However, one could
alter the instructional set to specify support from a specific source. In pretest-
ing the instrument, Barrera observed that the instrument with a family in-
structional set correlated significantly with the Moos family environment
questionnaire (Barrera et al., 1981). Additionally, the instructional set might
be altered to assess the provision instead of the receipt of support. Natural-
ly, the previously reported reliability and validity evidence would not apply
to the instrument if such changes were made. Nonetheless, a researcher would
be better off altering this instrument than designing a completely new one
for this purpose.

Perceived Social Support from Family and Friends

Procidano and Heller (1983) report a series of studies assessing measures

of perceived social support. The 20-item self-report measures require a simple
"yes," "no," or "don't know" response. Some statements may be word-
ed to refer to either friends or family, e.g., "My friends give me the moral
support I need" (p. 20), or "My family gives me the moral support I need"
(p. 21). The 20 items for friends yielded an alpha coefficient of .88, while
the family items yielded .90. Factor analysis of each instrument resulted in
a single-factor solution, further indication of the internal consistency of the
items. A pretest indicated a high test-retest reliability (r < .80). In a series
of studies both the friend and family scales correlated with measures of psy-
194 Tardy

chopathology and distress. In a study of conversational behavior, students

who were higher in perceived social support from friends talked more and
possessed lower trait anxiety than subjects scoring lower on this measure.
Subjects low in perceived social support from family evidenced marked ver-
bal inhibition when talking with a sibling than did subjects with high scores.
This study also observed a correlation between negative m o o d states and per-
ceived social support from friends, but not with perceived social support from
family. Neither social support measures correlated with a positive mood state.
This finding indicates a potentially significant problem: Individuals' reports
of social support may be a function of how individuals feel when they com-
plete the instrument. This possibility must be considered by all researchers
using self-report measures of social support. The behavioral evidence reported
by Procidano and Heller, however, is quite impressive and leads to the con-
clusion that self-reports can predict behaviors associated with social support.
The scale includes items related to both social support provision, e.g.,
"My friends come to me for emotional support" (p. 20) as well as receipt,
e.g., "My friends give me the moral support I need" (p. 20). The items refer
to both enactment and availability. Most, though not all, items refer to emo-
tional support. Consequently, this instrument is a general measure of social
support which does not differentiate among the dimensions of direction, dis-
position, and content. However since the receipt items overwhelmingly out-
number the provision items, the scale should be interpreted primarily as a
measure of support receipt. Likewise, since almost all the items assess emo-
tional support, the scale should be interpreted as measuring primarily this
type of social support. The secondary items unlikely contribute significantly
to the total support score. Though these imprecisions potentially reduce the
utility of the instrument, Procidano and Heller's measure certainly warrants
examination.

Social Relationship Scale

McFarlane et al. (1981) report an instrument designed to describe and

evaluate the enactment of social s u p p o r t ? This self-administered question-
naire first requests respondents to list the initials or names of people with
whom they discuss a specified topic, e.g., work and finances. Subjects next
describe their relationship to each listed person. Then the helpfulness of the
discussion is rated along a four-point continuum from "makes things a lot
worse" to "makes things a lot better." Finally, subjects answer the question,
"Would this person come to you to discuss topic X?" The last question ena-
8McFarlane et al. (1981) use the term "availability" to describe their measure. These authors
assume that support provided in the past is available in the future. This paper, as discussed
earlier, uses the term more precisely.
Social Support Measurement 195

bles an index of reciprocity to be computed in addition to the measures of sup-

port receipt.
The questionnaire covers six topics ranging from society to personal
health. These responses can be scored in multiple fashions. The simplest meas-
ures are the total number of different people listed and average helpfulness
for each topic. The test-retest correlation with a student sample over the peri-
od of 1 week for the six items ranged from .62 to .99 with a median of .91
for the number of individuals listed and from .54 to .94 with a median of
.78 for average helpfulness. In addition to detecting demographic group
differences with a general population sample, the measures of spouse help-
fulness differentiated significantly between a sample of parent therapists (n
= 18) and parents soliciting psychiatric counseling (n = 15). In sum, psy-
chometric evidence warrants continued use of this measure.
The questionnaire provides more information than most measures of
social support. It can be used to describe sources of support, types of sup-
port content, as well as availability of and satisfaction with support. Addi-
tionally, it assesses the giving as well as the receiving Of social support.
Another advantage of this instrument over others is the provision for
negative outcomes. The questionnaire is not limited to only interactions which
are positive. This flexibility allows the researcher to assess both positive and
negative influences as well as various combinations of the two.
However, the instrument may not be limited to the assessment of so-
cial support. Each question refers to "discussion" of topics. The items may
elicit people who converse without giving support. Additionally, the num-
ber of individuals listed may be limited to those who provide verbal sup-
port. These same individuals may or may not provide support with actions
and behaviors, e.g., instrumental support. Each user must closely inspect
the wording of the questions in order to make a proper decision concerning
the appropriateness of the instrument for his or her purposes.

Social Support Network Interview

Claude Fischer and his colleagues (Jones & Fischer, 1978; Philips &
Fischer, 1981) developed a measure of individuals' exchange and support net-
work to be used in the Northern California Community Survey (Fischer,
1982). The original purpose of the instrument was to describe the individu-
al's network of significant relationships. Jones and Fischer (1978) argue that
the best way to do this is by asking individuals to name the people who pro-
vide specific supports, e.g., "Whose opinion do you consider in making im-
portant decisions?" Ten questions were developed, eight of which relate to
three types of social support content: counseling, companionship, and prac-
tical help. These might be translated into appraisal, emotional, and instrumen-
196 Tardy

tal support. Some of the questions refer to support availability while others
refer to enacted support. For each of the questions, interviewees supply the
names or initials of people providing support. Two series of questions are
then asked which elicit descriptions of the network members. These data pro-
vide the opportunity to assess not only the availability of support but also
the characteristics of supporters, e.g., number of kin, number of social com-
panions. Additionally, Fischer (1982) describes a measure of support ade-
quacy. A minimum number of supporters for each of the three content
categories are established for inadequate, minimum, and adequate support.
Unfortunately, the author does not report reliability estimates in his discus-
sion of these scales. Other studies using the name-elicitation strategy present
acceptable estimates of reliability. Fischer (1982) notes that the aggregate
measures, e.g., total network size, number of kin, are more stable than
the occurrence of individual names, i.e., the inclusion of a particular friend's
name on the list. Several network measures were observed to correlate with
a "happiness" index (Jones & Fischer, 1978) while the three content meas-
ures described above detected differences among demographic groups (Fisch-
er, 1982).
The authors provide no estimate of the time required for completion
of the social support interview. A considerable period probably would be
required. However, questions need not be asked about all the content areas
of support. The questions about the characteristics of the individuals provid-
ing support might also be reduced to only those of relevance to the project
being undertaken. However, these two actions would reduce the reliability
of the total measure and reduce the flexibility of scoring the instrument.
Perhaps the strength of this instrument lies in the specificity and breadth
of questions eliciting the names of individuals providing support. This charac-
teristic of the instrument enables researchers to use it for a wide range of
problems.

Social Support Questionnaire

In a series of studies, Sarason et al. (1983) report the testing of a self-

report questionnaire to measure availability and satisfaction of social sup-
port. Respondents supply the names or initials of supporters for each of 27
problems or subjects, e.g., "Whom can you really count on to listen to you
when you need to talk?" and then rate the level of satisfaction with the total
support for that issue along a six-point continuum from very satisfied to very
dissatisfied. The authors calculate a measure which describes availability by
dividing the number of people providing support by the number of items
and the satisfaction with available support score by dividing the summed score
by the number of items. The availability description score yielded a test-retest
Social Support Measurement 197

correlation of .90 over a period of 4 weeks with a sample of 107 college stu-
dents. The satisfaction score was slightly less reliable with a correlations of
.83. Both measures correlated significantly and negatively with depression
scores in another study of students. A third study, also with college students,
observed a positive relationship between support availability and the occur-
rence of positive life events, the perceived effect of those events, the extent
to which the events were expected, and the degree to which the subjects felt
they had control over the occurrence of those events. The satisfaction meas-
ure was associated negatively with negative life events and control over those
events. An additional experimental study found subjects classified as exter-
nals on a locus of control scale persisted longer and evidenced less cognitive
interference during a stressful task when the subjects had more as opposed
to less than the average number of supporters. This accumulation of evi-
dence suggests that this is a viable measure of social support.
In addition to the authors' suggested procedure for scoring the instru-
ment, one could calculate the number of supporters for each role, e.g., friend,
family, or the number of roles for each category of need. Sarason et al. (1983)
suggest that such measures are highly intercorrelated. Nonetheless, the data
gathered under this format are amenable to multiple scoring procedures.
Studies which focus on the source of support could use this measure.
Before using this questionnaire, the content of the items should be care-
fully examined. Most of the questions deal with emotional support, e.g., "Who
accepts you totally, including both your worst and best points?" Although
items representing other types of support are included, these items are in the
minority and are often ambiguous. For example, the following item, "Whom
do you feel would help if a good friend of yours had been in a car accident
and was hospitalized in serious condition?" could assess emotional support,
e.g., X would comfort me, or instrumental support, e.g., X would watch my
children while I went to the hospital. Thus this instrument might best be used
only to assess emotional support.

Social Support Vignettes

Turner (1981) reports the use of an instrument designed by Kaplan

(1977). The innovativeness of this measure warrants our attention. Subjects
are presented a series of vignettes which describe people who vary systemati-
cally in social support. Subjects compare themselves to the persons described
in the vignette. For example:
Debbie: People are devoted to Debbie and love her, listen to her. Theycare about her.
Leslie: People are usually fond of Leslie. They can be sympathetic, but do not al-
ways listen to her or support her.
Robin: People are not devoted to Robin. They do not support her or sympathize
with her. They do not care about her or love her. (p. 359)
198 Tardy

Subjects are then asked to check one: (a) I'm like Debbie; (b) I'm half-
way between Debbie and Leslie; (c) I'm like Leslie; (d) I'm halfway between
Leslie and Robin; or (e) I'm like Robin.
Turner (1981) reports four studies utilizing nine such vignettes to meas-
ure social support. With samples of normal parents, maladaptive parents,
mentally ill adults, and hearing-impaired adults, internal reliability of the
scale ranged from .79 to .83 using the coefficient alpha. In factor analyses
with three other sets of variables, the social support items loaded together.
And perhaps most important, the social support measure consistently relat-
ed to measures of psychological well-being.
The scale does not distinguish among sources of social support. Con-
sequently the use of the vignettes is limited to studies in which the network
of supporters is of no concern. Neither does the instrument distinguish among
the types of support content. Without a closer examination of the instru-
ment than is possible by reading Turner's (1981) report, the specific types
of support being measured cannot be identified. However, vignettes could
be composed which embody these other characteristics of social support. The
reliability and validity data from Turner's study indicate the potential utility
of this strategy.

DISCUSSION

Tables I and II summarize the description of social support measures

reviewed in this article. Clear patterns of strengths and weaknesses exist. The
subsequent discussion examines the characteristics of social support assessed
by these instruments and suggests priorities for the development of new
measures.
The first characteristic of social support, direction, consists of recep-
tion and provision. Of the eight instruments, five focus solely on the receipt
of support while none focus on its provision. One of the instruments includes
a few provision items among mostly receipt items. One scale, the Social Rela-
tionship Scale (McFarlane et al., 1981), provides for the assessment of
reciprocity of support, the degree to which a person gives support to the peo-
ple from whom she or he receives support. Though all of the instruments
reviewed adequately measure support receipt, none adequately assess the pro-
vision of support.
The second and third characteristics of social support are disposition
and description/evaluation. Both of the support dispositions are equally
represented by these measures. Two of the instruments yield indices of enact-
ment, one of availability, and one of both. Two other instruments contain
items or questions about availability and enactment without distinguishing
Social Support Measurement 199

Table I. Summary of Social Support Measures: Part 1

Characteristics
Description/
Scales Direction Disposition evaluation
Arizona Social Received Enacted and Described and
Support Interview available evaluated
Schedule (Barrera,
1981)
Inventory of Received Enacted Described
Socially
Supportive
Behaviors
(Barrera, 1981)
Perceived Social Mostly Part Described
Support from received enacted
Friends and and part
Family Scale available
(Procidano &
Heller, 1983)
Social Relationship Received Enacted Described and
Scale (McFarland evaluated
et al., 1981)
Social Support Received Part Described
Network Interview enacted
(Fischer, 1982) and part
available
Social Support Received Available Described and
Questionnaire evaluated
(Sarason et al.,
1983)
Social Support Received Uncertain Described
Vignettes
(Turner, 1981)

b e t w e e n t h e m . A l l seven i n s t r u m e n t s p r o v i d e d e s c r i p t i o n s o f social s u p p o r t .
T h r e e a d d i t i o n a l l y include m e a s u r e s o f e v a l u a t i o n . N o n e are strictly evalua-
tive. These o p t i o n s p r o v i d e flexibility a n d increased p r e c i s i o n in the selec-
tion o f an a p p r o p r i a t e i n s t r u m e n t .
M o s t o f the i n s t r u m e n t s evidence awareness a n d c o n c e r n for the multi-
ple c o n t e n t types o f social s u p p o r t , the f o u r t h characteristic. F o u r o f the
measures include a b r o a d range o f items or questions in o r d e r to be represen-
tative o f all types o f social s u p p o r t . O n l y two measures reflect p a r t i c u l a r con-
tent categories, P r o c i d a n o and Heller's (1983) Perceived Social S u p p o r t f r o m
F r i e n d s a n d F a m i l y Scale assesses e m o t i o n a l s u p p o r t while Fischer's (1982)
Social S u p p o r t N e t w o r k Interview contains item to m e a s u r e counseling, corn-
200 Tardy

Table II. Summary of Social Support Measures: Part 2

Characteristics
Additional
Scales Content Network features
Arizona Social Multiple Supporters Also measures
Support Interview listed by for support
Schedule (Barrera, respondents Contains
1981) separate
negative items
Inventory of Multiple Not Can be labeled
Socially measured to measure
Supportive network
Behaviors
(Barrera, 1981)
Perceived S o c i a l Emotional Friends None
Support from and
Friends and family
Family Scale
(Procidano &
Heller, 1983)
Social Relationship Multiple Supporters Contains a
Scale (McFarlane listed by separate ne-
et al., 1981) respondent gative support
item
Social Support Counseling, Supporters A measure of
Network Interview compan- listedby adequacy can
(Fischer, 1982) ionship respondent be computed
and helping
Social Support Supporters None
Questionnaire Mostly listed by
(Sarason et al., emotional respondent
1983)
Social Support
Vignettes Uncertain Not None
(Turner, 1981) measured

p a n i o n s h i p , a n d practical help. T h a t more i n s t r u m e n t s do not focus o n emo-

t i o n a l s u p p o r t is surprising since so m a n y writers associate social s u p p o r t
with only this type of content. Understanding the types of support con-
tent depends o n the d e v e l o p m e n t of i n s t r u m e n t s which delineate a m o n g the
types a n d adequately measure their properties.
The measures of social s u p p o r t use two different strategies for assess-
ing the network, the fifth characteristic. One procedure asks the respondents
to supply the names and descriptions of people in their support network. F r o m
this i n f o r m a t i o n , indices of network characteristics are constructed. The other
Social Support Measurement 201

procedure involves requesting r e s p o n d e n t s to describe the s u p p o r t provided

to or received f r o m individuals specified by the i n s t r u m e n t . Both strategies
yield acceptable results. Researchers m a y choose between the two procedures
or c o m b i n e them as did H e n d e r s o n (1981).
T h o u g h substantial progress has been made in the past few years, several
p r o b l e m s still exist. The most pressing c o n c e r n is the d e v e l o p m e n t of instru-
m e n t s which clearly delineate a m o n g the types o f social s u p p o r t c o n t e n t a n d
which assess the p r o v i s i o n o f support. W h e n there is no adequate m e a s u r e
for one of these characteristics, the p r o b l e m is multiplied geometrically
because all possible c o m b i n a t i o n s of characteristics are affected. F o r exam-
ple, since n o n e o f the measures a d e q u a t e l y assess s u p p o r t provision, new
measures which describe as well as evaluate each of the different types of sup-
port c o n t e n t a n d n e t w o r k characteristics m u s t be developed before this issue
can be investigated.
The need for further development is clear. New measures will u n d o u b t -
edly be constructed a n d the present ones revised. Awareness o f the limita-
tions as well as the capabilities of current instruments should promote efficient
a n d r a t i o n a l progress in the study of social support.

REFERENCES

Barrera, M. (1980). A method for the assessment of social support networks in community sur-
vey research. Connections, 3(3), 8-13.
Barrera, M. (1981). Social support in the adjustment of pregnant adolescents: Assessment is-
sues. In B. H. Gottlieb (Ed.), Social networks and social support (pp. 69-96). Beverly
Hills: Sage.
Barrera, M., & Ainlay, S. L. (1983). The structure of social support: A conceptual and empiri-
cal analysis. Journal of Community Psychology, 11, 133-143.
Barrera, M., Sandler, I. N., & Ramsay, T. B. (1981). Preliminary development of a scale of
social support: Studieson collegestudents. Journal of Community Psychology, 9, 435-447.
Beels, C. C. (1981). Social support and schizophrenia. Schizophrenia Bulletin, 7(1), 58-72.
Burke, R. J. (1978). Sex differences in adolescent life stress, social support, and well-being.
Journal of Psychology, 98, 277-288.
Carveth, W. B., & Gottlieb, B. H. (1979). The measurement of social support and its relation
to stress. Canadian Journal of Behavioral Science, 11, 111-116.
Cassel, J. (1976). The contribution of the social environmentto host resistance.American Journal
of Epidemiology, 102(2), 107-123.
Cauce, A. M., Felner, R. D., & Primavera, J. (1982). Social support in high-risk adolescents.
American Journal of Community Psychology, 10, 417-428.
Cohen, C. I., & Sokolovsky, J. (1979). Health-seekingbehavior and social networks of the aged
living in single-room occupancy hotels. Journal of the American Geriatrics Society, 27,
270-278.
Fiore, J., Becket, J., & Coppel, D. B. (1983). Social network interactions: A buffer or a stress.
American Journal of Community Psychology, 11, 423-439.
Fischer, C. S. (1982). To dwell among friends: Personal networks in town and city. Chicago:
University of Chicago Press.
Gore, S. (1981). Stress buffering functions of support. In B. S. Dohrenwend & B. P. Dohren-
wend (Eds.), Stressful life events and their contexts. New York: Prodist.
202 Tardy

Gottlieb, B. H. (1978). The development and application of a classification scheme of informal

helping behaviours. Canadian Journal o f Behavioral Science, 10, 105-115.
Hammer, M. (1981). Social supports, social networks and schizophrenia. Schizophrenia Bulle-
tin, 7, 45-56.
Heller, K. (1979). The effects of social support: Prevention and treatment implications. In A.
P. Goldstein & F. H. Kanfer (Eds.), Maximizing treatment gains (pp. 353-382). New
York: Academic Press.
Henderson, S. (1980). Social relationships, adversity, and neurosis: An analysis of prospective
observations. British Journal o f Psychiatry, 136, 574-183.
Henderson, S. (1981). Neurosis and the social environment. Sydney: Academic Press.
Hirsch, B. J. (1981). Social networks and the coping process: Creating personal communities.
In B. H. Gottlieb (Ed.), Social networks and social support (pp. 149-170). Beverly Hills:
Sage.
House, J. S. (1981). Work stress and social support. Reading, MA: Addison-Wesley.
Jones, L., & Fischer, C. S. (1978). A procedure for surveying personal networks. Sociological
Methods and Research, 7, 131-148.
Kaplan, A. (1977). Social support: The construct and its measurement. Unpublished bachelor's
thesis, Brown University, Providence, RI.
Kaplan, B. H., Cassel, J. C., & Gore, S. (1977). Social support and health. Medical Care,
25(Suppl.), 47-58.
Kerlinger, F. N. (1973). Foundations o f behavioral research. New York: Holt, Rinehart &
Winston.
Leavy, R. L. (1983). Social support and psychological disorder: A review. Journal o f Commu-
nity Psychology, 11, 3-21.
Lin, N., Dean, A., & Ensel, W. M. (1981). Social support scales. Schizophrenia Bulletin, 73-89.
McFarlane, A. H., Neale, K. A., Norman, G. R., Roy, R. G., & Streiner, D. L. (1981). Methodo-
logical issues in developing a scale to measure social support. Schizophrenia Bulletin,
7, 90-100.
Miller, P., & Ingham, J. (1976). Friends, confidants, and symptoms. SocialPsychiatry, 11, 51-58.
Mitchell, R. E., & Trickett, E. J. (1980). Task force report: Social networks as moderators of
social support. Community Mental Health Journal, 16, 27-44.
Monroe, S. M. (1983). Social support and disorder. American Journal o f Community Psychol-
ogy, 11, 81-97.
Philips, S. L., & Fischer, C. S. (1981). Measuring social support networks in general popula-
tions. In B. S. Dohrenwend & B. P. Dohrenwend (Eds.), Stressful life events and their
contexts (pp. 223-233). New York: Prodist.
Procidano, M. E., & Heller, K. (1983). Measures of perceived social support from friends and
from family: Three validation studies. American Journal o f Community Psychology 11,
1-24.
Sandier, 1. N., & Barrera, M. (1984). Towards a multimethod approach to assessing the effects
of social support. American Journal o f Community Psychology, 12, 37-52.
Sarason, I. G., Levine, H. M., Basham, R. B., & Sarason, B. R. (1983). Assessing social sup-
port: The social support questionnaire. Journal o f Personality and Social Psychology,
44, 127-139.
Stokes, J. P., & Wilson, D. G. (1984). The inventory of socially supportive behaviors: Dimen-
sionality, prediction, and gender differences. American Journal o f Community Psychol-
ogy, 12, 53-69.
Tolsdorf, C. C. (1976). Social networks, support and coping. Family Process, 15, 407-418.
Turner, R. J. (1981). Social support as a contingency in psychological well-being. Journal o f
Health and Social Behavior, 22, 357-367.
Walker, K. N., MacBride, A., & Vaehon, M. L. S. (1977). Social support networks and the
crisis of bereavement. Social Science and Medicine, 11, 35-41.
Wellman, B. (1981). Applying network analysis to the study of support. In B. H. Gottlieb (Ed.)
Social networks and social support (pp. 171-200). Beverly Hills: Sage.
Wilcox, B. L. (1981). Social support in adjusting to marital disruption. In B. J. Gottlieb (Ed.),
Social networks and social support (pp. 97-115). Beverly Hills: Sage.