Goodson Patricia, Buhi Eric R, Dunsmore Sarah C
Department of Health & Kinesiology, Texas A&M University, College Station, Texas 77843-4243, USA.
J Adolesc Health. 2006 Mar;38(3):310-9. doi: 10.1016/j.jadohealth.2005.05.026.
This review systematically examined the relationship between self-esteem and adolescents' sexual behaviors, attitudes, and intentions, as empirically investigated by researchers over the last 20 years.
Based on the Matrix Method for systematic reviews, five electronic databases and reference lists of published studies were searched. Thirty-eight (n = 38) publications met specific inclusion and exclusion criteria and comprised the final sample. Each study was assigned a methodological quality score (MQS; maximum = 20 points). Each study's empirical test of the relationship between self-esteem and sexual behaviors/attitudes/intentions was counted as a single finding and was classified as exhibiting an inverse linear relationship, no statistically significant linear relationship, or a positive linear relationship.
A total of 189 findings (average = 4.97/publication, range 1-28) were analyzed. Most findings (n = 138, 73%) consisted of tests of self-esteem and sexual behaviors. Regarding the nature of the relationship between self-esteem and the various behavior/attitude/intention variables, 62% of behavior findings and 72% of the attitudinal findings exhibited no statistically significant association. All of the intention-related findings exhibited some type of relationship (either positive or inverse). Studies' average MQS was 11.71 (SD = 1.60). Higher quality studies yielded findings of positive or absence of relationship (Cramer's V = .329, p = .001).
If most findings exhibit no association between self-esteem and adolescent sexual behaviors/attitudes/intentions, questions are inevitably raised whether the emphasis placed on self-esteem by public health professionals is innocuous. Recommendations for methodological improvement of this body of research are also presented.
本综述系统地研究了过去20年研究人员通过实证调查得出的自尊与青少年性行为、态度及意图之间的关系。
基于系统综述的矩阵法,检索了五个电子数据库及已发表研究的参考文献列表。38篇出版物符合特定的纳入和排除标准,构成了最终样本。每项研究都被赋予一个方法学质量评分(MQS;满分 = 20分)。每项研究对自尊与性行为/态度/意图之间关系的实证检验被视为一项独立发现,并被归类为呈现负线性关系、无统计学显著线性关系或正线性关系。
共分析了189项发现(平均每项出版物4.97项,范围为1 - 28项)。大多数发现(n = 138,73%)涉及自尊与性行为的检验。关于自尊与各种行为/态度/意图变量之间关系的性质,62%的行为相关发现和72%的态度相关发现未呈现统计学显著关联。所有与意图相关的发现均呈现出某种类型的关系(正相关或负相关)。研究的平均MQS为11.71(标准差 = 1.60)。质量较高的研究得出了正相关或无关联的结果(克莱默V系数 = 0.329,p = 0.001)。
如果大多数发现表明自尊与青少年性行为/态度/意图之间无关联,那么公共卫生专业人员对自尊的强调是否无害就不可避免地会引发疑问。本文还针对这一研究领域的方法学改进提出了建议。