In this article, we investigated the publishing practices of Australian sociology PhD students during enrolment. It examines a five-year cohort of PhD completions 2013-17 from sociology departments and interdisciplinary schools of social sciences for all Australian universities. The key question considered is: do sociology PhD students publish in sociology journals? We used the Web of Science (WoS) to analyse disciplinary classification of journals where students' articles were published. By a ratio of 1:10, students' articles mostly appeared in non-sociology journals. We then compared these data with a recent study of new sociology faculty in the United States, who showed a similar diversity in publication patterns beyond recognised sociology journals. These empirical data contribute to debates about the adequacy of the Excellence in Research for Australia (ERA) framework: is practically technically correct if that can be done in measuring sociology's contribution when a major part of this exists outside conventional output metrics. Formal systems like Scopus and WoS provide extensive sets of data for comparing research outputs including by disciplinary classification; these journal indexes of research articles are commonly taken as a basic measure of research quality by universities and ERA. WoS journal disciplinary classification provided a useful basis of comparison with Warren's data on sociology in the US.