A six-item scale for measuring loneliness: testing the validity of the scale and explaining differences in loneliness in several countries of the Gender and Generation Programme

Jenny Gierveld, Netherlands Interdisciplinary Demographic Institute (NIDI)

Internationally validated measuring instruments are a dire necessity for country comparative research. In this paper I will provide the test outcomes for loneliness, one of the crucial social well-being outcome variables in the Gender and Generation Surveys. Loneliness is an important indicator of social embeddedness and well-being of younger, but especially of older adults. In accordance with Weiss (1973) I differentiate between emotional loneliness, stemming from the absence of an intimate figure (partner, best friend), and social loneliness related to the absence of a wider network of people with common interests. The De Jong Gierveld loneliness scale (De Jong Gierveld & Van Tilburg, 1999) has proved to be a valid and reliable measuring instrument for emotional and social loneliness. In this paper a shorter version of the scale -- six items -- is empirically tested on data originating from several countries participating in the Gender and Generation Survey: Bulgaria, Russia, and the Netherlands. The results of analyses showed that this scale measures very well the emotional, social and overall loneliness of respondents aged 18 to 80. Moreover, the relationship between this scale and indicators of objective social isolation proved to be optimal. It is concluded that the six-item scale is a reliable and valid measuring instrument for loneliness, and suitable for large surveys in countries that might differ significantly according to social and economic conditions. Additionally, the outcomes of multivariate analyses will be presented that describe and explain social well-being and loneliness of adults aged fifty and over, taking into account several aspects of their life courses, such as partner status and partner history, household structure, material living conditions, and health status.

  See extended abstract

Presented in Session 59: Gender and generations