Causal conditions for loneliness : a set-theoretic analysis on an adult sample in the UK.

While age has been identified as a risk factor for loneliness, whether it is a necessary or sufficient condition for loneliness has never been examined. This is the first study that applies fuzzy-set QCA, a special type of set-theoretic method, to discover the necessary and sufficient causal conditions for loneliness, respectively, among adults in the UK, analysing the data collected from the UK sample of Round 6 of the European Social Survey (ESS, 2012, n = 2163). It firstly examines the configurations of five conditions: being female, old age, not living with spouse/partner, bad health, and not being frequently social with others. Gender was found neither a necessary nor a sufficient condition for loneliness, and old age was close to being a necessary condition and became necessary when united with any of the other conditions; the configuration of not living with spouse/partner and not healthy and not frequently social with others is a sufficient condition. Robustness of results was tested with two different conditions (a limiting illness and a confidante), and a separate analysis on the absence of loneliness was conducted. The effect of the unbalanced distribution of cases across different values of the outcome was highlighted as a source of uncertainty, and the results on the absence of loneliness are different from those on its presence