Glass T A, Mendes de Leon C F, Seeman T E, Berkman L F
Harvard School of Public Health, Department of health and Social Behavior, Boston, MA 02115, USA.
Soc Sci Med. 1997 May;44(10):1503-17. doi: 10.1016/s0277-9536(96)00270-5.
While the health promoting influences of social networks have been shown in a number of studies, little attention has been paid to measurement issues within the field of epidemiology. The purpose of this paper is to propose a new set of measures of social networks for use in epidemiological research on the elderly. We use confirmatory factor analysis to test a multidimensional model of social networks using data from a large epidemiologic study of community-dwelling adults age 65 and over (U.S.A.). Confirmatory factor analysis conducted using LISREL showed that our model provides a good fit to the data after several adjustments for correlated measurement error were introduced. Based on this analysis, we developed new measures of four dimensions and a summary index of social networks. Bivariate relationships between our new measures and several sociodemographic variables of interest are also presented.
虽然一些研究已经表明社交网络对健康有促进作用，但流行病学领域内对测量问题却很少关注。本文的目的是提出一套新的社交网络测量方法，用于对老年人的流行病学研究。我们使用验证性因子分析，利用来自一项针对65岁及以上美国社区居住成年人的大型流行病学研究的数据，来检验社交网络的多维模型。使用LISREL进行的验证性因子分析表明，在对相关测量误差进行几次调整后，我们的模型与数据拟合良好。基于此分析，我们开发了四个维度的新测量方法以及社交网络的综合指数。还呈现了我们的新测量方法与几个感兴趣的社会人口学变量之间的双变量关系。
