COSMIC Function Points

Methodologies of the COSMIC Application in Practice  ◾  225

© 2011 by Taylor & Francis Group, LLC

extraneous factors, such as nonhomogeneity of software types across the organiza-

tions where the measurements were derived.

ese analyses also provide an indication that convertibility can be fairly accurate

for the majority of the projects within a data set, but on the other hand, there are

some larger variations for a few projects. is means that convertibility of a full port-

folio could be reasonably accurate overall, but that some individual projects would

show some larger dispersion from the values predicted by the convertibility models.

Further research is required to investigate factors that could explain such larger indi-

vidual project v ...