There is a long-standing interest in gender differences in satisfaction in intimate relationships. Whereas prior research has focused on gender differences in central tendency (i.e., means), we conducted two studies – a secondary analysis of data from a probability sample of Australian married couples and a meta-analysis – to examine gender differences in variability (i.e., variances). We hypothesized that compared to males, females would demonstrate greater variability in intimate relationship satisfaction (i.e., greater female variability hypothesis), particularly at lower levels of relationship satisfaction. Results from a secondary analysis of data from 2,711 married couples in the Household, Income and Labour Dynamics in Australia (HILDA) survey and from a meta-analysis of 20 years of research (k = 171, N = 84,976), including independent samples from 33 countries, indicated that relative to males, females reported greater variability in relationship satisfaction. Obtained effect sizes (female-to-male variance ratios [VRs] of 1.42 for the HILDA sample and 1.19 for the meta-analysis) were larger than proposed cutoffs for meaningful group differences in variability. Analysis of tail ratios (ratios of the relative proportion of females divided by the relative proportion of males in the distributional tail regions) in the HILDA sample indicated that gender differences in variability were greater at lower (versus higher) levels of satisfaction. Findings support the greater female variability hypothesis and suggest that by focusing only on gender differences in means, the existing literature has underestimated gender differences in intimate relationship satisfaction.

长期以来，人们一直对亲密关系满意度的性别差异感兴趣。虽然之前的研究侧重于集中趋势（即平均值）的性别差异，但我们进行了两项研究——对澳大利亚已婚夫妇概率样本数据的二次分析和一项荟萃分析——以检查可变性（即方差）的性别差异。我们假设，与男性相比，女性在亲密关系满意度方面表现出更大的可变性（即更大的女性可变性假说），尤其是在关系满意度较低的情况下。对澳大利亚家庭、收入和劳动力动态 （HILDA） 调查中 2,711 对已婚夫妇的数据进行二次分析，以及对 20 年研究的荟萃分析（k = 171，N = 84,976），包括来自 33 个国家的独立样本，结果表明，相对于男性，女性报告的关系满意度差异更大。获得的效应量 （HILDA 样本的女性与男性方差比 [VR] 为 1.42，荟萃分析为 1.19）大于建议的变异性有意义组差异的临界值。HILDA 样本中尾部比率（女性的相对比例除以分布尾部区域中男性的相对比例的比率）分析表明，在较低（相对于较高）满意度水平下，可变性的性别差异更大。研究结果支持更大的女性可变性假说，并表明现有文献仅关注均值的性别差异，低估了亲密关系满意度的性别差异。