MacArthur SES & Health Network
MacArthur SES & Health Network

printable version

Support & Social Conflict: Section One - Social Support

Summary prepared by Teresa Seeman in collaboration with the Psychosocial Working Group. Last revised April 2008.

Chapter Contents

  1. Definition and Background
  2. Measurement
  3. Relation to SES
  4. Relationship to Health
  5. Limitations
  6. Network Usage
  7. Conclusions
  8. Selected Bibliography

Definition and Background

Social support refers to the various types of support (i.e., assistance/help) that people receive from others and is generally classified into two (sometimes three) major categories: emotional, instrumental (and sometimes informational) support. Emotional support refers to the things that people do that make us feel loved and cared for, that bolster our sense of self-worth (e.g., talking over a problem, providing encouragement/positive feedback); such support frequently takes the form of non-tangible types of assistance. By contrast, instrumental support refers to the various types of tangible help that others may provide (e.g., help with childcare/housekeeping, provision of transportation or money). Informational support represents a third type of social support (one that is sometimes included within the instrumental support category) and refers to the help that others may offer through the provision of information.


There are a variety of instruments currently used to assess social support; there is no single, "best" measure. This situation may be partially due to the fact that a wide range of different measurement strategies have yielded "scores" that have successfully been related to various health outcomes. Instruments range from single items used to assess whether or not major types of support (emotional, instrumental) are available ("yes/no")(Seeman & Berkman, 1988) to more extensive instruments which include multiple items asking about various types of emotional support and various types of instrumental/informational support (Seeman et al, 1994; Cohen et al, 1985; Schuster et al, 1990). These latter measures also vary with respect to whether or not the items are asked regarding specific social relationships (e.g., types of support available from children versus friends; Seeman et al, 1994; Schuster et al, 1990) versus items asking more generally about support available from "others" (Cohen et al, 1985). Psychometric data are available for the multiple item measures.

back to top

  1. Mid-Mac Study Measures - Coefficient alphas of .83 - .91 for emotional support from spouse, family and friends indicate good internal reliability (data documentation from Mid-Mac).
  2. MacArthur Successful Aging Study - Two-month test retest data indicate reasonable stability for measures of emotional support (.73) and for levels of demand/criticism (.80), and somewhat lower stability for instrumental support (.44 [.55 for rank correlation]) (Seeman et al, 1994).
  3. Interpersonal Support Evaluation List - Test-retest data reveal correlations between .77-.86 and internal alpha estimates of .88-.90.

Choice of the appropriate measure(s) for use in research projects will likely be driven by such factors as: a) available time (some measures require considerably more time than others) and, b) whether assessments for different types of relationships are hypothesized to be important.

Relationship to SES

Social support has been found to vary positively with socio-economic status in studies in the US (Brim et al, 2004; Matthews et al, 1989; Huang & Tausig, 1990; Campbell et al, 1986), England (Marmot et al, 1997) and Sweden (Ostergren, 1991). These patterns are seen for both emotional and instrumental support and for both men and women (though the differences appear to be somewhat greater for men; Marmot et al, 1997). Notably however, despite the evidence of statistically significant differences, the actual size of the observed variations is relatively small (e.g., r's <.20, Mid-Mac data; maximum difference of 10% in the relative prevalence of low support across occupational grades, Marmot et al, 1997). SES and social support may be most importantly linked to health in the context of lower SES. Evidence from the MIDUS study points to stronger links between more positive social engagement and support and better health (as well as stronger negative health impacts from poorer social relationship profiles) among those of lower SES (Ryff et al, 2004)

Research also suggests that social stressors that tend to be more prevalent in lower SES environments (e.g., residential crowding, fear of crime, financial strain) are associated with lower perceived support (Evans et al., 1989; Lepore et al, 1991a, b, c) and that these social stressors contribute to reductions in reported levels of social support because they tend to foster a distrust of others (Krause, 1992). Although studies have documented cases where high levels of support can be available within ethnic enclaves (e.g., Gans, 1962; Stack, 1977; MacLeod, 1995), more general population trends indicate that lower income adults in Western Europe and the U.S. tend to have smaller social networks and fewer organizational involvements (Cochran, Larner, Riley, Gunnarson & Henderson, 1990; House, Umberson & Landis, 1988; Whelan, 1993). They also experience less social support both from the community as from family members (Atkinson et al., 1986; Conger & Elder, 1994; Schoon & Parsons, 2002; Voydanoff & Donnelly, 1988; Whelan, 1993; Wright et al., 1998).

back to top

Social resources also vary by neighborhood SES. Residents of disadvantaged neighborhoods compared with more advantaged counterparts have weaker social ties, experience less interpersonal trust, and perceive lower levels of instrumental support and mutual aid (Kawachi, 1999; Leventhal & Brooks-Gunn, 2000; Sampson et al., 1997). Poor neighborhoods have fewer social resources and diminished capacity for informal social controls. Social capital is often in short supply in the very communities where it is most needed.

Relationship to Health

The strongest associations between social support (particularly emotional support) and a health outcome are seen in relation to psychological well-being. A large literature documents lower risk for depression and for psychological distress more generally for those who enjoy greater social support (for review see George, 1989 and most recently, e.g., Stansfeld et al, 1997). Relationships to physical health outcomes are less well documented. This may partly reflect the longer history of epidemiologic research using measures of social integration (i.e., network size) rather than social support (House et al, 1988; Seeman, 1996). Some evidence links social support (as opposed to social integration) to incidence of major disease such as MI, stroke or cancer (Seeman 1996) as well as to less extensive development of coronary atherosclerosis (Seeman & Syme, 1987, Blumenthal et al, 1987). The strongest evidence linking support to disease is research demonstrating better survival post-myocardial infarction for those with more emotional support (Berkman et al, 1992, Williams et al, 1992). Similar data indicate beneficial effects on prognosis post-stroke (Glass & Maddox, 1992). More generally, evidence suggests that emotional support is protective with respect to preservation of better physical (Seeman et al, 1995) and cognitive (Seeman et al, 2001; Fratiglioni et al, 2004) functioning at older ages. The effects of instrumental support, however, appear considerably more mixed with higher levels of such support leading to greater disability in some cases (Seeman et al, 1996; for review, see also Burg & Seeman, 1994).

Data on health outcomes among children indicate that emotional support in particular impacts both psychological and physical health outcomes. Children exposed to deficit nurturing are at increased risk for depression (Kaslow et al, 1994, Leweinsohn et al, 1994) and suicidal ideation (Adams et al, 1994). Children born to mothers who lacked family support are at increased risk for low birth weight (Collins et al, 1993) and childhood exposure to less responsive parenting has been related to increased risk for childhood illness (Gottman & Katz, 1989) and substance abuse among adolescents (Shedler & Block, 1990; Baumrind, 1991) as well as poorer socioemotional development (Bradley & Corwyn, 2003; Grant et al, 2003; McLoyd, 1998; Repetti, Taylor & Seeman, 2002).

In addition to research on support and disease outcomes, there is a growing body of evidence linking social support to physiological regulatory processes. Among children, presence of a supportive caregiver has been shown to lower HPA to maternal separation reactivity (as indexed by salivary cortisol levels) (Gunnar et al, 1992). For adults, social support has likewise been found to predict lower levels of HPA and SNS activity in laboratory-based challenge paradigms as well as community settings (Seeman & McEwen, 1996; Uchino, 2006). Evidence also links social support to lower risk of decline in CD4 counts among HIV infected men (Theorell et al, 1995). These data suggest that support may indeed play a role in the development of (or protection against) various disease outcomes. The generally weaker evidence linking support to health outcomes (as compared with associations seen for measures of social integration, social ties) may reflect the greater variability of support over time making it more difficult to predict the disease outcomes which are themselves the result of developmental processes.

back to top


Social support measures suffer from two major limitations. One is the lack of established, 'gold standard' measures. The variety of different measures currently in use makes it difficult to draw conclusions based on comparisons of results across studies. The second limitation relates to the variability of support over time and our inability (to date) to assess these variations and their impact on relationships between social support and health outcomes. As a result, we currently have little evidence linking social support to the occurrence of major physical health outcomes, despite strong theoretical reasons for believing that such effects exist.

Network Usage

Measures of social support have been used extensively in research supported by various MacArthur Networks, including the current SES network (e.g., S. Cohen, T. Seeman), the Mid-Life Network (collaborative project Ryff & Seeman; R. Kessler), and the Successful Aging Network (Seeman et al, 1995; Seeman et al, 1994).


Existing data indicate that higher levels of social support, particularly emotional support are both associated with higher SES and appear to be protective with respect to a number of health outcomes. This pattern of relationships suggests that social support may function as one of the mediators of SES effects on health and should be a focus of on-going research into relationships between SES and health. Further support for the potentially important role of social support comes from evidence linking such support to differences in physiological reactivity (Seeman & McEwen, 1996). There are, however, a number of outstanding issues. For example, to date, research has largely examined social support as a main effect and as a mediator of SES (and other) effects on health. However, recent evidence suggests important moderating effects of social support with respect to SES links to health (Ryff et al, 2004) as well as in relation to risk factors for declines in physical functioning (Unger et al, 1999). Such evidence points to the importance of greater consideration of a moderating role for social support. Also, as indicated above, there are a number of different measures in use and there is no consensus regarding which, if any, is the best and little or no direct comparisons of the ability of different measures to predict outcomes in a given study. Happily, the construct appears to be quite sturdy in the face of such diversity of measures: consistent findings have generally been seen across different measures of social support. Nonetheless, development of a more commonly used set of measures would be advantageous, particularly for future comparative research. Overall, this construct appears likely to be useful in research on SES and health.

back to top

Selected Bibliography

(also see bibliography for section two - Social Conflict)

Atkinson, T., Liem, R., & Liem, J. (1986) Social costs of unemployment: Implications for social support. Journal of Health and Social Behavior, 27, 317-331.

Bradley, R. H., & Corwyn, R. (2003) Socioeconomic status and child development. Annual Review of Psychology, 53, 371-399.

Brim OG, C.D. Ryff, and R.C. Kessler, Editors. (2004) How Healthy are We?: A National Study of Well-Being at Midlife, University of Chicago Press: Chicago.

Cochran, M., Larner, M., Riley, D., Gunnarson, L., & Henderson, C. (1990) Extending families: The social networks of parents and their children. New York: Cambridge University Press.

Conger, R. D., & Elder, G. H., Jr. (1994) Families in troubled times. NY: Aldine de Gruyter.

Fratiglioni, L., S. Paillard-Borg, and B. Winblad, (2004) An active and socially integrated lifestyle in late life might protect against dementia. Lancet Neurol 3(6): p. 343-53.

Grant, K. E., Compas, B. E., Stuhlmacher, A. F., Thurm, A. E., McMahon, S. D., & Halpert, J. A. (2003) Stressors and child and adolescent psychopathology: Moving from markers to mechanisms of risk. Psychological Bulletin, 129, 447-466.

House, J. S., Umberson, D., & Landis, K. (1988) Structures and processes of social support. Annual Review of Sociology, 14, 293-318.

Kawachi, I. (1999) Social capital and community effects on population and individual health. In N.E. Adler, M. Marmot, B.S. McEwen, & J. Stewart (Eds.), Annals of the New York Academy of Sciences, Vol. 896. Socioeconomic status and health in industrial nations (pp. 120-130). New York: New York Academy of Sciences.

Leventhal, T., & Brooks-Gunn, J. (2000) The neighborhoods they live in: The effects of neighborhood residence on child and adolescent outcomes. Psychological Bulletin, 126, 309-337.

McLoyd, V. C. (1998) Socioeconomic disadvantage and child development. American Psychologist, 53, 185-204.

Repetti, R. L., Taylor, S. E., & Seeman, T. E. (2002). Risky families: Family social environments and the mental and physical health of offspring. Psychological Bulletin, 128, 330-366.

back to top

Ryff, C.D., B.H. Singer, and K.A. Palmersheim. (2004) Social Inequalities in Health and Well-Being: The Role of Relational and Religious Protective Factors, in How Healthy are We?: A National Study of Well-Being at Midlife, O.G. Brim, C.D. Ryff, and R.C. Kessler, Editors. University of Chicago Press: Chicago, IL. p. 90-123.

Sampson, R., Raudenbush, S., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918-924.

Schoon, I., & Parsons, S. (2002) Competence in the face of adversity: The influence of early family environment and long-term consequences. Children and Society, 16, 260-272.

Seeman, T.E., (1996) Social ties and health: the benefits of social integration. Ann Epidemiol, 6(5): p. 442-51.

Seeman TE, Lusignolo T, Berkman L, Albert M. (2001) Social Environment Characteristics and Patterns of Cognitive Aging: MacArthur Studies of Successful Aging. Health Psychology 20:243-255.

Uchino, B.N., (2006) Social support and health: a review of physiological processes potentially underlying links to disease outcomes. J Behav Med 29(4): p. 377-87.

Unger, J.B., et al., (1999) Variation in the impact of social network characteristics on physical functioning in elderly persons: MacArthur Studies of Successful Aging. J Gerontol B Psychol Sci Soc Sci, 54(5): p. S245-51.

Voydanoff, P., & Donnelly, B. W. (1988) Economic distress, family coping, and quality of family life. In P. Voydanoff, & Majka, L.C. (Ed.), Families and economic distress (pp. 97-116). Los Angeles: Sage.

Whelan, C. T. (1993) The role of social support in mediating the psychological consequences of economic stress. Sociology of Health and Illness, 15, 86-101.

Wright, L.B., Treiber, F., Davis, H., Bunch, C., & Strong, W.B. (1998) The role of maternal hostility and family environment upon cardiovascular functioning among youth two years later. Ethnicity and Disease, 8, 367-376.

back to top


UCSF Home About UCSF Search UCSF UCSF Medical Center