Partner stalking is often assumed to be a dangerous problem that is distinct from intimate partner violence (IPV) overall, or from other types of IPV (e.g., physical, sexual, and psychological violence). Such assumptions imply that partner stalking requires identification, assessment, and response processes distinct from other types of harm within both intact and separated intimate relationships. However, considerable confusion—about how best to (a) define partner stalking and (b) distinguish partner stalking from other harm types—seems to persist among researchers and practitioners alike. Accordingly, this thesis contains three empirical studies that investigate these assumptions (and the corresponding confusion) by focusing on partner stalking definitions and identification, as well as partner stalking-related risk prediction. The first study examined police-recorded partner stalking in episode reports across 1,150 IPV cases. Few police officers explicitly used the label ‘stalking’, but we found evidence of partner stalking within one in every seven IPV cases reported to New Zealand Police. The second study was a qualitative analysis of practice-based definitions of partner stalking, based on interviews with 14 specialist IPV practitioners. Rather than providing a summary definition of partner stalking, practitioners defined partner stalking as a list of behaviours. There was also considerable heterogeneity in whether—and, if so, how—practitioners distinguished partner stalking from other types of IPV harm. Finally, the third study examined whether partner stalking predicted subsequent IPV outcomes reported to police across a 12-month follow up in 1,126 cases. Partner stalking was a weak predictor of any IPV recurrence (in univariate analysis only) and did not predict physical IPV recurrence (in univariate or multivariate analyses). Taken together, and broadly mirroring the previous research literature, the findings in this thesis suggest that police and specialist IPV practitioners lack a shared understanding of partner stalking; in turn, hindering their ability to clearly define and identify this phenomenon, especially in relation to other types of IPV. And within this current practice context, partner stalking may not warrant special consideration for risk prediction, given that (with one exception) partner stalking did not predict subsequent IPV outcomes reported to police. Overall, these findings challenge assumptions that partner stalking—at least based on current research- and practice-based conceptualisations of the construct—is meaningfully understood as a distinct concern from other types of harm within abusive relationships. We discuss the theoretical and practical implications of this conclusion, while considering recommendations for ongoing research and theory development.