CoChat
Sign Up
Cited by 528Psychological Well-being and Life SatisfactionNonprofit Sector and VolunteeringMindfulness and Compassion Interventions

Happy to help? A systematic review and meta-analysis of the effects of performing acts of kindness on the well-being of the actor

Open access
Curry, Oliver Scott, Rowland, Lee A., Van Lissa, Caspar J., Zlotowitz, Sally, McAlaney, John, Whitehouse, Harvey·2018·Journal of Experimental Social Psychology
DOI 10.1016/j.jesp.2018.02.014
Open this paper in CoChat — free accountOpen PDF

Abstract

Do acts of kindness improve the well-being of the actor? Recent advances in the behavioural sciences have provided a number of explanations of human social, cooperative and altruistic behaviour. These theories predict that people will be ‘happy to help’ family, friends, community members, spouses, and even strangers under some conditions. Here we conduct a systematic review and meta-analysis of the experimental evidence that kindness interventions (for example, performing ‘random acts of kindness’) boost subjective well-being. Our initial search of the literature identified 489 articles; of which 24 (27 studies) met the inclusion criteria (total N = 4045). These 27 studies, some of which included multiple control conditions and dependent measures, yielded 52 effect sizes. Multi-level modeling revealed that the overall effect of kindness on the well-being of the actor is small-to-medium (δ = 0.28). The effect was not moderated by sex, age, type of participant, intervention, control condition or outcome measure. There was no indication of publication bias. We discuss the limitations of the current literature, and recommend that future research test more specific theories of kindness: taking kindness-specific individual differences into account; distinguishing between the effects of kindness to specific categories of people; and considering a wider range of proximal and distal outcomes. Such research will advance our understanding of the causes and consequences of kindness, and help practitioners to maximise the effectiveness of kindness interventions to improve well-being.

AI snapshot

Written from the paper’s full text.

TL;DR

This systematic review and meta-analysis of experimental kindness interventions found that performing acts of kindness improves the actor’s subjective well-being, with a small-to-medium overall effect (δ = 0.28). Across 27 studies (N = 4045), effects were not moderated by tested demographic, intervention, control, or outcome variables, and there was no indication of publication bias.

Key findings

  • The overall effect of kindness on actor well-being was δ = 0.28, 95% CI [0.16, 0.41], Z = 4.36, p < .001, described as approximately equivalent to a 0.6 increase on a standard 0–10 happiness scale.
  • The review identified 24 articles containing 27 studies (total N = 4045) and 52 effect sizes; individual effect sizes ranged from −0.46 to 1.25 (M = 0.25, SD = 0.32), and sample sizes ranged from 26 to 474 (M = 158.57, SD = 132.05).
  • The effect was not moderated by sex, age, type of participant, intervention, control condition, or outcome measure; metaforest screening yielded a median out-of-bootstrap explained variance of R²_oob = −0.11.
  • There was no indication of publication bias: Begg’s test was Z = 1.07, p = .28, and file drawer analysis indicated 1919 unpublished null studies would be needed to bring the average unweighted effect to nonsignificance.

Methods

  • Systematic searches of Web of Science and PsychInfo plus other sources yielded 489 articles after duplicate removal; 24 articles (27 studies) met inclusion criteria requiring experiments that randomly allocated participants to kind-behaviour interventions versus controls and subsequently measured well-being.
  • Study characteristics were coded, including age, sex, location, participant type, intervention type, recipient type, control condition, dependent measures, group sizes, and Cohen’s d effect sizes; effect sizes were taken from papers or computed from reported statistics.
  • Main analysis used three-level meta-analysis in R with metafor to account for dependent effect sizes within studies; metaforest random forests with 10,000 bootstrap samples were used to screen potential moderators.

Limitations

  • The paper states that most reviewed studies were underpowered (average sample size per condition N = 79; power 1−β = 0.42 to detect d = 0.28; at least 202 per group needed for 1−β = 0.80), mostly used non-clinical student samples from W.E.I.R.D. societies, did not manipulate the motive behind kind acts, did not test fine-grained predictions from specific altruism theories or systematically vary recipient type, and mostly examined immediate rather than longer-term or distal outcomes.
  • The text also makes evident that some effect sizes could not be calculated because statistics were not reported in the paper or available from the authors (Table 1 note), and that the small sample size limited moderator analysis, requiring metaforest to avoid overfitting.
  • The paper notes that control conditions varied, and some studies compared kindness with doing nothing, which may confound kindness effects with the effects of performing any novel fun activity.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1016/j.jesp.2018.02.014

Cites

143 of 143 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

0 of 15 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    The data from Lyubomirsky, Sheldon, & Schkade (2005) are no longer available.

    “Email correspondence with the author revealed that the data are no longer available.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Lyubomirsky, Sheldon, & Schkade, 2005 source

      Not checked in the free run.

  • Findingcitation

    File drawer analysis revealed that 1919 unpublished, filed, or unretrieved studies averaging null results would be required to bring the average unweighted effect size to nonsignificance.

    “File drawer analysis (Rosenthal, 1979) revealed that 1919 unpublished, filed, or unretrieved studies averaging null results would be required to bring the average unweighted effect size to nonsignificance.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Rosenthal, 1979 source

      Not checked in the free run.

  • Findingown data

    The overall effect of kindness on the well-being of the actor is small-to-medium (δ = 0.28, 95% CI [0.16, 0.41], Z = 4.36, p < .001).

    “Multi-level modeling revealed that the overall effect of kindness on the well-being of the actor is small-tomedium (δ = 0.28).”
    • Also stated in: § 5.3. Meta-analysis · plain
      “Results revealed that the overall effect size estimate was δ = 0.28, 95% CI [0.16, 0.41], Z = 4.36, p < .001(see Fig. 2).This is a small-to-medium effect, approximately equivalent to an increase of 0.6 on a standard 0-10 happiness scale (Helliwell, Layard, & Sachs, 2016).”
    • Also stated in: § 6. Discussion · hedged
      “The results of this systematic review and meta-analysis of the experimental kindness literature suggests that performing acts of kindness improves the well-being of the actor (δ = 0.28).”
    Quote in the paperSubstantial claim
  • Findingreasoning

    Existing research has not tested the more fine-grained predictions that arise from the more specific theories of helping (kin altruism, mutualism, reciprocal altruism and competitive altruism).

    “Fourth, although the finding is consistent with the general evolutionary account of altruism outlined above, existing research has not tested the more fine-grained predictions that arise from the more specific theories of helping (kin altruism, mutualism, reciprocal altruism and competitive altruism).”
    Quote in the paperNot a substantial claim
  • Findingcitation

    Most reviewed studies used non-clinical samples of students from Western, Educated, Industrial, Rich, Democratic (WEIRD) societies.

    “Second, most of the reviewed studies used non-clinical samples of students, from Western, Educated, Industrial, Rich, Democratic societies (W.E.I.R.D.; Henrich, Heine, & Norenzayan, 2010).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Henrich, Heine, & Norenzayan, 2010 source

      Not checked in the free run.

  • Classified as probably not a substantial claim. 18
  • May combine two claims. 18

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    The three-level model had df=3, AIC=22.14, BIC=28.16, and log-likelihood=-8.07.

    “Three-level model | 3 | 22.14 28.16 -8.07 | | |”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    Sample sizes ranged from 26 to 474 participants (M = 158.57, SD = 132.05).

    “Sample sizes ranged from 26 to 474 participants (M = 158.57,SD = 132.05).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Likelihood ratio tests indicated that constraining the within-studies variance to zero would not worsen model fit, whereas constraining either the between-studies variance or both variance components to zero did lead to significant deteriorations in model fit.

    “Likelihood ratio tests also indicated that constraining the within-studies variance to zero would not worsen model fit, whereas constraining either the between-studies variance or both variance components to zero did lead to significant deteriorations in model fit (see Table 2).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Findingnone given

    Existing research has tended to look at the immediate effects of kindness on well-being.

    “Fifth and finally, existing research has tended to look at the immediate effects of kindness well-being.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingreasoning

    The reviewed studies varied whether actions benefitting others were performed, but did not vary whether the benefits were intended; they did not manipulate the motive behind the action.

    “The studies reviewed here varied whether actions benefitting others were performed, they did not vary whether the benefits were intended or notin other words, they did not manipulate the motive behind the action.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingreasoning

    Most reviewed studies were under-powered: the average sample size per condition was N = 79, giving power of only 1-β = 0.42 to detect a typical effect size of d = 0.28.

    “First, most of the reviewed studies were under-powered.The average sample size per condition was N = 79; this gives power of only 1-β = 0.42 to detect a typical effect size of d = 0.28.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The within-studies variance component was negligible (0.00, 95% CI [<0.01, 0.02]), while the between-studies variance component differed significantly from zero (0.08, 95% CI [0.04, 0.18]).

    “The within-studies variance component σ w 2 was negligible, 0.00, 95% CI [ < 0.01, 0.02].The between-studies variance σ b 2 , on the other hand, differed significantly from zero, 0.08, 95% CI [0.04, 0.18].”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The median estimated explained variance in out-of-bootstrap cases was negative (R oob 2 = -0.11), with a large standard deviation across replications (SD = 0.19).

    “The median estimated explained variance in out-of-bootstrap cases was negative (R oob 2 = -0.11),with a large standard deviation across replications (SD = 0.19).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    Constraining between-studies variance to zero resulted in df=2, AIC=33, BIC=36.76, log-likelihood=-14.37, LRT=12.61, and p=0.

    “Between-studies variance constrained | 2 | 33 | 36.76 -14.37 12.61 0 | |”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingnone given

    Lyubomirsky, Sheldon, & Schkade (2005) does not report the sample size, the dependent measure, or any inferential statistics.

    “However, the article does not report the size of the sample, the dependent measure, or any inferential statistics (for example, effect size or significance).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Lyubomirsky, Sheldon, & Schkade, 2005 source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    The effect of kindness on well-being was not moderated by sex, age, type of participant, intervention, control condition or outcome measure.

    “The effect was not moderated by sex, age, type of participant, intervention, control condition or outcome measure.”
    • Also stated in: § 5.4. Moderation · plain
      “we did not find evidence for associations between the moderators and effect size.”
    • Also stated in: § 6. Discussion · plain
      “The effect was not moderated by sex, age, type of participant, intervention, control condition or outcome measure.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The systematic review identified 24 articles containing 27 studies that met the inclusion criteria, with a total N of 4045 participants.

    “Our initial search of the literature identified 489 articles; of which 24 (27 studies) met the inclusion criteria (total N = 4045).”
    • Also stated in: § 5.1. Study characteristics · plain
      “These 27 studies included a total of 4045 participants (mean proportion male = 35%, mean age = 25.04,SD = 11.05).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingcitation

    Lyubomirsky, Sheldon, & Schkade (2005) purports to provide evidence that kind acts boost the well-being of the actor.

    “The most highly cited paper in the kindness literature (with citations at the time of the last search) purports to provide evidence that kind acts boosts the well-being of the actor(Lyubomirsky, Sheldon, & Schkade, 2005).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Lyubomirsky, Sheldon, & Schkade, 2005 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    The 27 included studies yielded 52 effect sizes.

    “These 27 studies, some of which included multiple control conditions and dependent measures, yielded 52 effect sizes.”
    • Also stated in: § 5.1. Study characteristics · plain
      “Some studies had multiple control conditions, and/or multiple outcome measures, and hence provided more than one effect size; there were 52 in total.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    Constraining within-studies variance to zero resulted in df=2, AIC=20.14, BIC=24.15, log-likelihood=-8.07, LRT=0, and p=1.

    “Within-studies variance constrained | 2 | 20.14 24.15 -8.07 | 0 | 1 |”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    Constraining both variance components to zero resulted in df=1, AIC=78, BIC=80.25, log-likelihood=-38.12, LRT=60.1, and p=0.

    “Both variance components constrained 1 | 78 | 80.25 -38.12 60.1 | 0 | |”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationcitation

    To detect an effect size of d = 0.28 with power 1-β = 0.80, a sample size of at least 202 per group is needed.

    “In order to detect such an effect with power 1-β = 0.80, future researchers should use a sample size of at least 202 per group (Faul, Erdfelder, Lang, & Buchner, 2007).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Faul, Erdfelder, Lang, & Buchner, 2007 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Premisenone given

    Theories predict that people will be 'happy to help' family, friends, community members, spouses, and even strangers under some conditions.

    “These theories predict that people will be 'happy to help' family, friends, community members, spouses, and even strangers under some conditions.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The analysis was replicated 100 times to ensure the reliability of findings.

    “We replicated the analysis 100 times to ensure the reliability of findings.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    It remains unclear whether the current findings would generalise to clinical samples of participants diagnosed with specific mental health problems, in non-WEIRD societies.

    “Thus it remains unclear whether the current findings would generalise to clinical samples of participants diagnosed with specific mental health problems, in non-WEIRD societies.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The data and analysis script for this study are available on its Open Science Framework page.

    “The data and analysis script for this study are available on its Open Science Framework page (https://osf.io/sey6x/).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationown data

    Together, these results suggest that policy-makers and practitioners are correct to see kindness interventions as effective ways of improving wellbeing.

    “Together, these results suggest that policy-makers and practitioners are correct to see kindness interventions as effective ways of improving wellbeing.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The age and sex ratio averages in Table 1 are approximate because the age and sex ratio of the samples were not available for some studies.

    “These averages are approximate (~), because the age and sex ratio of the samples were not available for some studies.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Helping others makes you happy, but the effect is relatively modest.

    “Helping others makes you happy, but the effect is relatively modest.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    Variation in effect sizes was primarily accounted for by differences between studies, whereas differences between effect sizes within the same studies were negligible.

    “The fact that the between-studies component was larger than the within-studies component indicates that the variation in effect sizes was primarily accounted for by differences between studies, whereas differences between effect sizes within the same studies were negligible.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Previous research has found an association between motive and outcome; one longitudinal study found that volunteers motivated by a desire to help others lived longer than non-volunteers, but volunteers motivated by a desire to help themselves did not.

    “Previous research has found an association between motive and outcome; one longitudinal study found that volunteers motivated by a desire to help others lived longer than non-volunteers, but that volunteers motivated by a desire to help themselves did not (Konrath, Fuhrel-Forbis, Lou, & Brown, 2012).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Konrath, Fuhrel-Forbis, Lou, & Brown, 2012

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationnone given

    Biology and culture have equipped humans to help automatically, intuitively, and innocently.

    “Note that the argument here is that biology and culture have equipped us to help automatically, intuitively, innocentlythere is no suggestion that people are necessarily aware of the causes of their benevolent behaviour, or are acting from any ulterior motive.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationcitation

    The effect of kindness is small-to-medium and comparable to other positive psychology interventions (such as 'mindfulness', 'positive thinking' and 'counting your blessings'; d = 0.34, Bolier et al., 2013; d = 0.31, Sin & Lyubomirsky, 2009; d = 0.44, Weiss, Westerhof, & Bohlmeijer, 2016).

    “The effect of kindness is small-to-mediumcomparable to other positive psychology interventions (such as 'mindfulness', 'positive thinking' and 'counting your blessings'; d = 0.34, Bolier et al., 2013;d = 0.31, Sin & Lyubomirsky, 2009;d = 0.44, Weiss, Westerhof, & Bohlmeijer, 2016).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Bolier et al., 2013 source

      Not checked in the free run.

    • cites Sin & Lyubomirsky, 2009 source

      Not checked in the free run.

    • cites Weiss, Westerhof, & Bohlmeijer, 2016 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing). May combine two claims.

    Interpretationown data

    These results support the general claim that, as social animals, humans possess a range of psychological mechanisms that motivate them to help others, and that they derive satisfaction from doing so.

    “And they support the general claim that, as social animals, humans possess a range of psychological mechanisms that motivate them to help others, and that they derive satisfaction from doing so.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    For some studies, statistics needed to calculate effect size were not reported in the paper, nor available from the authors.

    “*= Statistics neededto calculate effect size were not reported in the paper, nor available from the authors.”
    Quote in the paperNot a substantial claim
30 of 30

  • The paper relies on this without support.

    The experimental studies included in the meta-analysis validly manipulated kindness and measured well-being, so their effect sizes can be interpreted as causal.

    “experiments that randomly allocated participants to (b) interventions involving kind behaviour and controls and (c) subsequently measured and compared participant well-being.”
1 of 1

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Altruistic and prosocial social behavior appears to promote well-being for the helpers.

    “Altruistic and prosocial social behavior appears to promote well-being for the helpers [18, 39].”

    From Psychiatric symptoms, risk, and protective factors among university students in quarantine during the COVID-19 pandemic in China (2021) · cites this paper as [18]

  • Prosocial behavior has been associated with positive mental health outcomes, including increased positive affect, family and community impact, social network, overall mental health, and satisfaction with life.

    “Prosocial behavior has been associated with a plethora of positive mental health outcomes, including increased positive affect, family and community impact, social network, overall mental health, and satisfaction with life, as reviewed by Curry et al. ( 2018 ).”

    From A latent profile analysis of aggression and prosocial behavior in relation to adolescent wellbeing (2025) · cites this paper as Curry et al., 2018

  • Practising acts of kindness has a small-to-moderate effect on increasing well-being of the kindness-actors.

    “a meta-analysis of 27 experimental studies shows that practising acts of kindness has a small-to-moderate effect on increasing well-being of the kindness-actors 5 .”

    From Kindness as a public health action (2025) · cites this paper as [5]

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Consistentz = 4.36reported p < .001 · recomputed p < .001

    “Z = 4.36, p < .001”

  • Consistentz = 1.07reported p = .28 · recomputed p = .285

    “Z = 1.07, p = .28”

2 of 2

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Self-check results appear here once an independent check has confirmed them.

Cites

Flagged first, then by citations

Research works printed but not linked: 1

  1. Gardner, 2014, Special issue: Inclusive fitness; 50 years on, Philosophical Transactions of the Royal Society, B: Biological Sciences, 369

Cited by

Sorted by citations

Related

Authors
Curry, O. S., Rowland, L. A., Van Lissa, C. J., Zlotowitz, S., McAlaney, J., & Whitehouse, H.
Affiliations
  • University of Oxford (GB)
  • University College London (GB)
  • Bournemouth University (GB)
Corresponding author
Curry, Oliver Scott
ORCID
Venue
Journal of Experimental Social Psychology
Publisher
Elsevier BV
Published
2018-05
Volume
76
Pages
320-329
ISSN
0022-1031 (ISSN-L)
DOI
10.1016/j.jesp.2018.02.014
OpenAlex
W2776060271
Cited by
528
References
143
Citations by year
2026: 66 · 2025: 80 · 2024: 92 · 2023: 72 · 2022: 93 · 2021: 75 · 2020: 29 · 2019: 15 · 2018: 5
FWCI
4.81
Type
journal article
Language
en
Open access
Licence
Locations
  • Journal of Experimental Social Psychology (publishedVersion, cc-by-nc-nd)
  • Bournemouth University Research Online (Bournemouth University) (acceptedVersion)
  • Utrecht University Repository (Utrecht University) (submittedVersion, other-oa)
Funders
kindness.org (R46536/CN001)
Topics
  • Social Sciences › Psychology › Social Psychology › Psychological Well-being and Life Satisfaction
  • Social Sciences › Sociology and Political Science › Nonprofit Sector and Volunteering
  • Social Sciences › Psychology › Clinical Psychology › Mindfulness and Compassion Interventions
Keywords
acts of kindness, random acts of kindness, subjective well-being, altruism, prosocial behavior, happiness, individual differences

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-06.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-06.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review