Loneliness and intersectionality: A progressive conditional approach

•Loneliness and intersectionality have been established as distinctive areas of research with little conscious and direct connection until very recently.•The progressive conditional approach keeps the study of loneliness strictly aligned with the spirit of intersectionality and systematically examines the empirical evidence for the intersectional effects on loneliness.•In 2014 to 2015, young people (under 30) of ethnic minority and born inside Great Britain suffered from the highest percentage of frequent loneliness, whilst their counterparts born outside the country enjoyed the lowest rate.•Among older people (60+) born outside Great Britain, regardless of ethnicity, the percentage of frequent loneliness was as high as 10%.•For the middle-aged, it was ethnicity, not country of birth, that determined how vulnerable they were to frequent loneliness. Both loneliness and intersectionality have become well established areas of academic research since the 1970s and 1980s. Nevertheless, only very recently some meaningful connections were made between the two, although researchers have paid attention to the interactive effects of two or more socio-demographic attributes on loneliness. For intersectionality, much of academic research is invested in establishing it as a theoretical approach in tackling social injustice, whilst how it should be studied empirically remains a major challenge. In contrast, research on loneliness has been predominantly empirical, and the small number of studies on loneliness from the intersectional perspective have adopted different research methodologies. This paper proposes and illustrates an approach loyal to the fundamental principles of intersectionality and simple to conduct in empirical investigations at the same time. First, it focuses exclusively on intersectional cross-classifications rather than both the main and the interactional effects; second, it demands a rationale of starting from one attribute and then moving on to include an additional attribute at a time; third, it examines the intersectional cross-classifications and their relationships with the interested outcome systematically without transforming the data in set memberships. The approach is illustrated with analyses of the data collected in Great Britain in the seventh round (2014/15) of European Social Survey. Young people (under 30) of ethnic minority and born inside Great Britain suffered from the highest percentage of frequent loneliness (15%), whilst their c