Carnegie Mellon University

File(s) stored somewhere else

Please note: Linked content is NOT stored on Carnegie Mellon University and we can't guarantee its availability, quality, security or accept any liability.

Statistical Sampling for Process Assessments

journal contribution
posted on 2009-01-01, 00:00 authored by David M. Northcutt, Mark C. Paulk
Many framework-based process improvement programs, such as those based on the eSourcing Capability Model for Service Providers (eSCM-SP) and Capability Maturity Model Integration for Development (CMMIDEV), rely on a sample of engagements (or projects) in drawing conclusions about organizational capability. The intent is to determine the normal behavior in the organization based on a “representative” sample, but a rigorous definition of what characterizes a representative sample is usually not provided. The purpose of this article is to define in statistical terms, with explicitly stated assumptions, what a reasonable representative sample should be so a reliable characterization of organizational capability can be established. The analysis is illustrated in the context of the eSCM-SP but can be extended to other frameworks.