Davis Don E, Choe Elise, Meyers Joel, Wade Nathaniel, Varjas Kristen, Gifford Allison, Quinn Amy, Hook Joshua N, Van Tongeren Daryl R, Griffin Brandon J, Worthington Everett L
Department of Counseling and Psychological Services, Georgia State University.
Department of Psychology, Iowa State University.
J Couns Psychol. 2016 Jan;63(1):20-31. doi: 10.1037/cou0000107. Epub 2015 Nov 16.
A recent qualitative review by Wood, Froh, and Geraghty (2010) cast doubt on the efficacy of gratitude interventions, suggesting the need to carefully attend to the quality of comparison groups. Accordingly, in a series of meta-analyses, we evaluate the efficacy of gratitude interventions (ks = 4-18; Ns = 395-1,755) relative to a measurement-only control or an alternative-activity condition across 3 outcomes (i.e., gratitude, anxiety, psychological well-being). Gratitude interventions outperformed a measurement-only control on measures of psychological well-being (d = .31, 95% confidence interval [CI = .04, .58]; k = 5) but not gratitude (d = .20; 95% CI [-.04, .44]; k = 4). Gratitude interventions outperformed an alternative-activity condition on measures of gratitude (d = .46, 95% CI [.27, .64]; k = 15) and psychological well-being (d = .17, 95% CI [.09, .24]; k = 20) but not anxiety (d = .11, 95% CI [-.08, .31]; k = 5). More-detailed subdivision was possible on studies with outcomes assessing psychological well-being. Among these, gratitude interventions outperformed an activity-matched comparison (d = .14; 95% CI [.01, .27]; k = 18). Gratitude interventions performed as well as, but not better than, a psychologically active comparison (d = -.03, 95% CI [-.13, .07]; k = 9). On the basis of these findings, we summarize the current state of the literature and make suggestions for future applied research on gratitude. (PsycINFO Database Record
伍德、弗罗和杰拉蒂（2010年）最近进行的一项定性综述对感恩干预的效果提出了质疑，表明需要仔细关注对照组的质量。因此，在一系列元分析中，我们评估了感恩干预（k值 = 4 - 18；样本量 = 395 - 1755）相对于仅进行测量的对照组或替代活动条件在3个结果（即感恩、焦虑、心理健康）方面的效果。感恩干预在心理健康测量方面优于仅进行测量的对照组（d = 0.31，95%置信区间[CI = 0.04, 0.58]；k = 5），但在感恩方面则不然（d = 0.20；95%置信区间[-0.04, 0.44]；k = 4）。感恩干预在感恩测量（d = 0.46，95%置信区间[0.27, 0.64]；k = 15）和心理健康测量（d = 0.17，95%置信区间[0.09, 0.24]；k = 20）方面优于替代活动条件，但在焦虑方面则不然（d = 0.11，95%置信区间[-0.08, 0.31]；k = 5）。对于结果评估心理健康的研究，可以进行更详细的细分。在这些研究中，感恩干预优于活动匹配的对照组（d = 0.14；95%置信区间[0.01, 0.27]；k = 18）。感恩干预的表现与心理活跃的对照组相当，但并不比其更好（d = -0.03，95%置信区间[-0.13, 0.07]；k = 9）。基于这些发现，我们总结了当前文献的状况，并对未来关于感恩的应用研究提出了建议。（《心理学文摘数据库记录》
