Doing well by doing good. The relationship between formal volunteering and self-reported health and happiness

https://doi.org/10.1016/j.socscimed.2008.01.011Get rights and content

Abstract

In this paper, we examine whether engaging in voluntary work leads to greater well-being, as measured by self-reported health and happiness. Drawing on data from the USA, our estimates suggest that people who volunteer report better health and greater happiness than people who do not, a relationship that is not driven by socio-economic differences between volunteers and non-volunteers. We concentrate on voluntary labor for religious groups and organizations and using second stage least square regressions we find that religious volunteering has a positive, causal influence on self-reported happiness but not on self-reported health. We explore reasons that could account for the observed causal effect of volunteering on happiness. Findings indicate that low relative socio-economic status is associated with poor health both among those who volunteer and those who do not. Low status, however, is associated with unhappy states only among those who do not volunteer, while volunteers are equally likely to be happy whether they have high or low status. We propose that volunteering might contribute to happiness levels by increasing empathic emotions, shifting aspirations and by moving the salient reference group in subjective evaluations of relative positions from the relatively better-off to the relatively worse-off.

Introduction

By engaging in formal voluntary activities people provide unpaid work: they donate their time to groups and organization and do not ask for a monetary compensation in return. According to recent figures from the Independent Sector's survey of Giving and Volunteering in the United States, volunteering represents a sizable resource. Approximately 44% of the adult population in the United States engages in formal voluntary activities every year and on average volunteers donate 3.5 h each week. The literature suggests that people who volunteer enjoy good physical and mental health: they have lower rates of mortality and are more likely to have good health (Moen et al., 1993, Musick et al., 1999, Oman et al., 1999, Post, 2005). They are also more likely to report being happy and are less likely to suffer from depression (Musick and Wilson, 2003, Thoits and Hewitt, 2001, Wheeler et al., 1998, Whiteley, 2004). While the correlation between volunteering and well-being is well established, issues of omitted variable bias, self-selection and reverse causation remain mostly unresolved. The observed link could be spurious and hide the effect of other characteristics determining both a high propensity to volunteer and to feel well (omitted variable bias). The underlying characteristics that make individuals select themselves in the volunteering and non-volunteering groups might also be correlated with their well-being (self-selection). Finally, while volunteering may lead to higher well-being, the reverse is possible: people who feel well may be more likely to engage in volunteer labor (reverse causation).

Some studies have employed longitudinal datasets to mitigate omitted variable problems (Musick and Wilson, 2003, Thoits and Hewitt, 2001) and establish how changes in volunteering are correlated with changes in well-being, however, reverse causation remains a problem. Moreover, correlations could still be due to changing circumstances determining both the decision to start volunteering and higher well-being. When studies look at the relationship between volunteering and well-being measured in different periods to establish causation (Li & Ferraro, 2005), individual heterogeneity and self-selection remain unresolved. Helliwell and Putnam (2004) conclude their study on the relationship between social context and well-being calling for the use of instrumental variable or quasi-experimental settings to solve outstanding issues.

We use data on self-reported health and happiness to explore the link between volunteering and different aspects of well-being. The literature indicates that self-reported health and happiness are correlated (Rosenkranz et al., 2003, Ryff and Singer, 2001, Subramanian et al., 2005), but also that they do not overlap and capture different features of individual welfare. For example, in our sample over 80% of people suffering poor or fair health report being happy/very happy, possibly due to adaptation processes (Layard, 2005), and 5% of people in good/excellent health are not happy. Self-reported health is well suited at capturing the “life-ability of the person” while self-reported happiness is an indicator of “subjective appreciation of life” (Veenhoven, 2000).

Our results suggest that volunteer labor is positively correlated with both indicators of well-being. We concentrate on formal volunteering for religious groups and organizations and employ a second stage least squares regression framework to address reverse causation, self-selection and omitted variable bias. Results indicate that religious volunteering has a substantial, causal effect on happiness but not on health. We review the literature investigating the causes of well-being premiums among volunteers and hypothesize that volunteering might increase happiness by reducing people's concerns for status.

Section snippets

Data sources

The paper uses the Social Capital Community Benchmark Survey (SCCBS) dataset. The SCCBS collects data on voluntary work to a range of groups and organizations as well as on self-reported health and happiness. The SCCBS was designed by the Saguaro Seminar at the John F. Kennedy School of Government at Harvard University and was conducted in 2000 by telephone. The SCCBS contains both a sample of the U.S. population and specific samples from 41 communities across 29 states, for a total of nearly

Results

Models 1a and 1b in Table 1 constitute the base model where we regress the well-being indicators on basic individual socio-economic characteristics. Observations are clustered by state to take into account the correlation that exists among different respondents who live in the same state, robust standard errors are used to control for heteroskedasticity and estimates are weighted to take into account the design structure of the SCCBS which, for example, contains an oversample of minority groups

Discussion

We examined the association between volunteering and well-being using data on self-reported health and happiness. Our analysis indicates that volunteering is highly associated with greater health and happiness, while other forms of altruistic behavior, such as donations of money or donations of blood, are not. The main contribution of the study is that it examines whether the widely documented positive association between volunteering and well-being is causal. Our estimates suggest that the

Acknowledgements

The author would like to thank the British Academy for financial support; John Hills, Michael O'Hare, Richard Layard, Carmen Huerta, Franco Sassi and participants to seminars at the LSE Centre for the Analysis of Social Exclusion, the School of Psychology, University of Exeter, the Goldman School of Public Policy at the University of California at Berkeley for comments on an earlier version of this paper; Nick Warner, Palmer Phomika and Marilyn Milliken for help with the data. The views

References (54)

  • M. Bertrand et al.

    Do people mean what they say? Implications for subjective survey data

    American Economic Review

    (2001)
  • F. Borgonovi

    Divided we stand, united we fall: religious pluralism, giving and volunteering

    American Sociological Review

    (2008)
  • P. Bourdieu

    Distinction

    (1984)
  • M. Chaves et al.

    Religious pluralism and religious participation

    Annual Review of Sociology

    (2001)
  • R. Dawkins

    The selfish gene

    (1989)
  • E. Diener

    Subjective well-being

    Psychological Bulletin

    (1984)
  • E. Diener et al.

    The satisfaction with life scale

    Journal of Personality Assessment

    (1985)
  • M.M. Farmer et al.

    Distress and perceived health: mechanisms of health decline

    Journal of Health and Social Behavior

    (1997)
  • R.H. Frank

    Choosing the right pond: Human behavior and the quest for status

    (1985)
  • R.H. Frank

    Luxury fever

    (1999)
  • B.S. Frey et al.

    Happiness and economics

    (2001)
  • W.A. Fuller

    Some properties of a modification of the limited information estimator

    Econometrica

    (1977)
  • W. Greene

    Econometric analysis

    (2003)
  • J.F. Helliwell

    Well-being, social capital and public policy: what's new?

    The Economic Journal

    (2006)
  • J.F. Helliwell et al.

    The social context of well-being

    Philosophical Transactions of the Royal Society B: Biological Sciences

    (2004)
  • E.L. Idler et al.

    Self-rated health and mortality: a review of twenty-seven community studies

    Journal of Health and Social Behavior

    (1997)
  • R. Kammann et al.

    Affectometer 2. A scale to measure current level of happiness

    Australian Journal of Psychology

    (1983)
  • Cited by (328)

    View all citing articles on Scopus
    View full text