Department of Sociology, McGill University, Montreal, Canada.
Int Perspect Sex Reprod Health. 2011 Dec;37(4):181-90. doi: 10.1363/3718111.
It is generally believed that men and women misreport their sexual behaviors, which undermines the ability of researchers, program designers and health care providers to assess whether these behaviors compromise individuals' sexual and reproductive health.
Data on 1,299 recent sexual partnerships were collected in a 2007 survey of 1,275 men and women aged 18-24 and living in Kisumu, Kenya. Chi-square and t tests were used to examine how sample selection bias and selective partnership reporting may result in gender differences in reported sexual behaviors. Correlation coefficients and kappa statistics were calculated in further analysis of a sample of 280 matched marital and nonmarital couples to assess agreement on reported behaviors.
Even after adjustment for sample selection bias, men reported twice as many partnerships as women (0.5 vs. 0.2), as well as more casual partnerships. However, when selective reporting was controlled for, aggregate gender differences in sexual behaviors almost entirely disappeared. In the matched-couples sample, men and women exhibited moderate to substantial levels of agreement for most relationship characteristics and behaviors, including type of relationship, frequency of sex and condom use. Finally, men and women tended to agree about whether men had other nonmarital partners, but disagreed about women's nonmarital partners.
Both sample selection bias and selective partnership reporting can influence the level of agreement between men's and women's reports of sexual behaviors. Although men report more casual partners than do women, accounts of sexual behavior within reported relationships are generally reliable.
人们普遍认为男性和女性在报告性行为时存在偏差，这削弱了研究人员、项目设计者和医疗保健提供者评估这些行为是否会损害个人的性健康和生殖健康的能力。
2007 年对基苏木市 1275 名年龄在 18-24 岁之间的男女进行了一项调查，共收集了 1299 名最近的性伴侣的数据。使用卡方检验和 t 检验来检查样本选择偏差和选择性伴侣报告可能如何导致报告的性行为存在性别差异。在对 280 对匹配的婚姻和非婚姻伴侣的样本进一步分析中，计算了相关系数和 Kappa 统计量，以评估对报告行为的一致性。
即使在调整了样本选择偏差后，男性报告的性伴侣数量仍是女性的两倍（0.5 对 0.2），并且更多的是偶然的伴侣。然而，当控制了选择性报告时，性行为方面的总体性别差异几乎完全消失。在匹配的夫妇样本中，男性和女性对大多数关系特征和行为表现出中度到高度的一致性，包括关系类型、性频率和避孕套使用情况。最后，男性和女性倾向于就男性是否有其他非婚姻伴侣达成一致，但在女性的非婚姻伴侣方面存在分歧。
样本选择偏差和选择性伴侣报告都可能影响男性和女性对性行为报告的一致性水平。尽管男性报告的偶然伴侣比女性多，但报告的关系内性行为通常是可靠的。
