Abstract:

In rural areas of China, more than 50 million elderly people are left-behind, and they have a strong sense of loneliness. The relationship between husband and wife plays an important role in alleviating the loneliness feeling of the elderly, therefore, it is crucial to understand the association between marital attachment and loneliness of left-behind elderly. Previous study showed that Chinese elderly’s marital attachment was composed in three dimensions: security, avoidance and anxiety. However, the three-dimensional structure of Chinese elderly’s marital attachment has not been tested in the rural elderly population. In addition, the association between Chinese elderly’s loneliness and their three-dimensional marital attachment is also not clear. Therefore, further studies are needed. The current study, firstly, tested the three-dimensional structure of the Chinese elderly’s marital attachment, and secondly, investigated the relationship between marital attachment and loneliness of left-behind elderly.

The Older Adults’ Marital Attachment Scale, Scale of Perceived Social Support, Core Self-Evaluations Scale, and ULS-8 Loneliness Scale were administered to 510 left-behind elderly from rural areas in China. All cases were randomly assigned to two subsets: one set was for exploratory factor analysis, and the other was for confirmatory factor analysis. The SPSS 20.0 and AMOS 20.0 were used for this analysis, and the structural equation model and the bootstrapping analysis were employed to examine the hypothesized mediating model.

The key findings were as follows: (1) the 14-item revised OAMAS showed a three-dimensional construct of attachment, namely security, avoidance, and anxiety; (2) the marital attachment of left-behind elderly could be clustered into three types, including secure marital attachment (37.84%), refusal marital attachment (31.18%), and anxious marital attachment (30.98%). The proportion of secure marital attachment in left-behind elderly was significantly lower than that in community elderly (Z = -5.63, p < 0.001), and the proportion of refusal marital attachment in left-behind elderly was significantly higher than that in community elderly (Z = 7.01, p < 0.001) ; (3) perceived social support fully mediated the association among security attachment, avoidance attachment and loneliness. Perceived social support and core self-evaluations had multiple mediation effects on the association among security attachment, avoidance attachment and loneliness. While the association between anxiety attachment and loneliness was partially mediated by core self-evaluations.

The results indicated that three-dimensional structure of Older Adults’ Marital Attachment Scale was an effective measurement of marital attachment in left-behind elderly group. There was a close relationship between the marital attachment and loneliness of the left-behind elderly, safe marital attachment could effectively relieve the loneliness of left-behind elderly. The results may help to expand the study range of elderly’s marital attachment, and provide a reference for reducing the loneliness of the left-behind elderly. implications and limitations are discussed.

Key words: marital attachment, perceived social support, core self-evaluation, loneliness, left behind elderly