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 contributionposted 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.