The critique of methodological nationalism is widely accepted: Many social phenomena, such as migration, professional fields, social movements, or elite education, span more than one location and cannot be well understood within the bounds of one or more nation-states. It remains, however, difficult to study these and other phenomena. Quantitative data sets claim to be representative of populations that are bounded by administrative units. It is difficult to overcome data limitations, but network analyses based on big data are an interesting and relatively recent new option. Qualitative research often resorts to ‘banal’ nationalism (Billig): concepts are implicitly informed by the particularity of a specific state. ‘Groupism’ (Brubaker) occurs when migration research studies mobile populations but homogenizes migrants and refugees along the lines of the country of origin and the country of arrival. Some mixed methods research designs such as the ethnosurvey invented by Massey offer an interesting combination of claims to represent a transnational population and in-depth case studies. Comparison in ethnography, global ethnography and multi-level designs that use cities or organizations rather than states also attempt to study contexts beyond the nation-state. The session invites papers on research designs that address and overcome the problem of methodological nationalism. (a) How exactly does methodological nationalism show in research design and data? (b) What kind of data might mitigate the problem? (c) How can concepts be operationalized in a manner that overcomes banal nationalism and groupism? (d) Which entities are suitable contexts for comparison beyond the nation-state? (e) How can comparative designs become glocal or transnational? The session gives preference to methodological papers that are exemplified by empirical studies. Other papers are also welcome.