Measuring Agile Effectiveness: Why This was Harder Than We Thought, and Why That is Important
Abstract
Large volumes of Transactional Data are different from conventional. Transactional Data has some characteristics of Big Data. Observational, uncontrolled, needs Context, Adapted, Merged (from Kaiser Fung and William of Occam). Data are observational rather than designed; definitions may vary. There are no controls or quasi-controls to test causality. We adapted data collected by different people with different purpose. Get the context; for example, identify how users use the tool. Merged data exacerbates the issue of lack of definition and misaligned objectives. We must come to terms with variability of data early on; significance does not mean real predictable effects.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2017
- Accession Number
- AD1087004
Entities
People
- William R. Nichols
Organizations
- Carnegie Mellon University