Key points are not available for this paper at this time.
Although recent data creation efforts in international relations have begun to focus on issues of reliability and validity more explicitly than previously, current efforts still contain significant problems. This essay focuses on three recent data generation projects that study international relations (the ICOW, ATOP, and River Treaty datasets) and shows the successes and failures of each in assessing reliability when generating data from qualitative evidence. All three datasets attempt to generate reliable data, document the procedures used, and present indications of data reliability. However, their efforts face problems when assessing the reliability of their case selection variables, in the development of reliability indicators, and in the presentation of reliability statistics. In addition to evaluating these recent efforts to generate large-N databases, this essay clarifies the difference between generating data from qualitative and quantitative evidence, explains the importance of reliability when coding qualitative evidence, and provides ways to improve the assessment of the quality of one's data.
Steven B. Rothman (Sat,) studied this question.