The purpose of this study is to synthesize the overall effect size of tourism on poverty alleviation and to unveil underlying factors explaining the heterogeneity of this effect size across estimates. Using a rigorous meta-analysis based on 298 estimates extracted from 33 studies, we calibrate a combined effect size of −0.14 with a 95% confidence interval of [-0.23, −0.05], indicating that tourism moderately reduces poverty. A meta-regression demonstrates that the effect size of the tourism–poverty nexus is susceptible to several factors, such as poverty and tourism measures, focal countries’ development level, and endogeneity treatment. In particular, the results show that use of the Gini coefficient, a popular proxy for the poverty gap, can unexpectedly underestimate the negative effect size. Moreover, funnel plot and Galbraith plot demonstrate that researchers are apt to report a positive tourism–poverty nexus in the literature. Lastly, research and policy implications are provided.

本研究的目的是综合旅游对扶贫的总体影响大小，并揭示解释该影响大小在估计之间的异质性的潜在因素。使用基于从 33 项研究中提取的 298 个估计值进行的严格荟萃分析，我们校准了 -0.14 的组合效应大小和 [-0.23, -0.05] 的 95% 置信区间，表明旅游业适度减少了贫困。元回归表明，旅游-贫困关系的影响大小易受多种因素的影响，例如贫困和旅游措施、重点国家的发展水平和内生性处理。特别是，结果表明，使用基尼系数（一种流行的贫困差距替代指标）可能会意外地低估负面影响的大小。而且，漏斗图和加尔布雷斯图表明，研究人员倾向于在文献中报告积极的旅游-贫困关系。最后，提供了研究和政策影响。