New metric for measuring programmer productivity

Measuring programmer productivity is a challenging exercise because it is not well understood since it is without bounded rules and ill-defined meanings of complicated set of parameters. This is made worse by new software development paradigms such as agile and mashup methods and the ensuing advances in software engineering practices and ways of designing and developing software. The existing metrics for measuring performance of solo and pairs of programmers does not appear to be suitable to meet these new forms of software development. This paper presents a new metrics which can be used to measure performance productivity of programming effort by solo and pair programmers in a much more comprehensive manner. All of the key parameters are looked at objectively and defined for and from software engineering lifecycle points of view to make productivity performance calculations for solo and pair programmers. The methodology used for formulating and composing the formulae and defining the parameters as coefficients is explained. These parameters are based on a weighting system to derive the effective performance results. It is a more objective and quick way for the performance analysts to define a table of weights for the parameters based on several criteria within the different aspects of the software development lifecycle. The composition of the formulae and the metric for measuring programmer productivity performance is presented and examples are used to illustrate the method and the outcomes.