This article introduces the concept of the carrying capacity of data (CCD), defined as an integrated, evaluative judgment of the credibility of specific data-based inferences, informed by quantitative and qualitative analyses, leavened by experience. The sequential process of evaluating the CCD is represented schematically by a framework that can guide data analysis and statistical inference, as well as pedagogy. Aspects of each phase are illustrated with examples. A key initial activity in empirical work is data scrutiny, comprising consideration of data provenance and characteristics, as well as data limitations in light of the context and purpose of the study.  Relevant auxiliary information can contribute to evaluating the CCD, as can sensitivity analyses conducted at the modeling stage. It is argued that early courses in statistical methods, and the textbooks they rely on, typically give little emphasis to, or omit entirely, discussion of the importance of data scrutiny in scientific research. This inattention and lack of guided, practical experience leaves students unprepared for the real world of empirical studies. Instructors should both cultivate in their students a true respect for data and engage them in authentic empirical research involving real data, rather than the context-free data to which they are usually exposed.