From Pixels to Patterns: Strengthening Sampling in Video-Based Criminology

• In Event: Video Data Analysis for Policing Research: Innovations and Next Steps

Abstract

As large-N inference-based studies gain prominence in video-based criminology, questions of sampling, representativeness, and generalizability rise in significance, as well. Despite this trend, numerous methodological publications do not discuss this subject, and empirical video-based research often include at most cursory discussions of sampling issues. To address this gap, this presentation applies insights from sampling methodology to video-based social science research. We give an overview of how sampling issues have been discussed in video-based criminology, reflect on sampling issues in the context of video-based social science research, and develop a decision-tree flow chart that can help practitioners identify fruitful sampling strategies and potential pitfalls. We illustrate how the tree can be used in three common video-based sampling scenarios. We thereby contribute to establishing clear guidelines and standards for sampling in video-based criminology, which will be helpful to current and future practitioners, as well as reviewers and readers of video-based studies.

Authors

• Nicolas M. Legewie, University of Münster

• Anne Nassauer, University of Erfurt

• Simon Kühne, Bielefeld University