What software complexity do we need for software effort estimation? How ISBSG data analysis may contribute to deconstruct cognitive biases
R. Meli – DPO
An empirical study, using ISBSG data, challenges the long-held belief that Unadjusted Function Points (UFP) provide more accurate effort estimates for complex projects than simplified measures like Simple Function Points (SFP).
The research indicates that for median productivity, both UFP and SFP methods yielded estimates of comparable accuracy. While UFP showed a slight edge with mean productivity, the transactional component of SFPs alone achieved a similar accuracy to UFPs. The study concludes that the traditional definition of complexity, based on Data Element Types (DETs) and File Type Referenced (FTR), is not a significant factor in improving the accuracy of software effort estimation. This suggests that a new definition of complexity may be needed.
See PDF of Roberto’s presentation
About Roberto Meli
Over the past 30 years, Roberto Meli has developed focused competences in project management and software measurement areas. He has written papers for technical magazines and international conferences. He is a consultant and lecturer in training courses on project management and software measurement for many major Italian companies and public organizations. He has also had key roles within GUFPI-ISMA (Italian Software Measurement Association), COSMIC and the Simple Function Point Association (SiFPA – www.SiFPA.org).
