a School of Social Work, University of South Florida , Tampa , FL , USA.
b School of Social Work, The University of Texas at Austin , Austin , TX , USA.
Aging Ment Health. 2018 May;22(5):631-638. doi: 10.1080/13607863.2017.1286456. Epub 2017 Mar 14.
The objectives of this study were to (1) develop an empirical typology of social networks in older Koreans; and (2) examine its effect on physical and mental health.
A sample of 6900 community-dwelling older adults in South Korea was drawn from the 2014 Korean National Elderly Survey. Latent profile analysis (LPA) was conducted to derive social network types using eight common social network characteristics (marital status, living arrangement, the number and frequency of contact with close family/relatives, the number and frequency of contact with close friends, frequency of participation in social activities, and frequency of having visitors at home). The identified typologies were then regressed on self-rated health and depressive symptoms to explore the health risks posed by the group membership.
The LPA identified a model with five types of social network as being most optimal (BIC = 153,848.34, entropy = .90). The groups were named diverse/family (enriched networks with more engagement with family), diverse/friend (enriched networks with more engagement with friends), friend-focused (high engagement with friends), distant (structurally disengaged), and restricted (structurally engaged but disengaged in family/friends networks). A series of regression analyses showed that membership in the restricted type was associated with more health and mental health risks than all types of social networks except the distant type.
Findings demonstrate the importance of family and friends as a source of social network and call attention to not only structural but also non-structural aspects of social isolation. Findings and implications are discussed in cultural contexts.
本研究旨在(1)为韩国老年人构建一个社会网络实证类型学;(2)检验其对身心健康的影响。
从 2014 年韩国全国老年人调查中抽取了 6900 名居住在社区的韩国老年人作为样本。采用潜在剖面分析(LPA),使用 8 个常见的社会网络特征(婚姻状况、居住安排、与亲密家人/亲戚的接触次数和频率、与亲密朋友的接触次数和频率、参与社会活动的频率以及家中访客的频率)来推导社会网络类型。然后,将识别出的类型学回归到自评健康和抑郁症状上,以探讨群体成员身份带来的健康风险。
LPA 确定了一个具有 5 种社会网络类型的最优模型(BIC = 153848.34,熵 =.90)。这些群体分别被命名为多样/家庭(与家人有更多互动的丰富网络)、多样/朋友(与朋友有更多互动的丰富网络)、朋友聚焦(与朋友有高度互动)、疏远(结构上不参与)和受限(结构上参与但与家人/朋友网络脱节)。一系列回归分析表明,与除疏远型以外的所有类型的社会网络相比,受限型的成员与更多的健康和心理健康风险相关。
研究结果表明家庭和朋友作为社会网络来源的重要性,并提醒人们不仅要关注社会隔离的结构方面,还要关注非结构方面。在文化背景下讨论了研究结果和意义。