Empirically Guided Software Effort Guesstimation

Software project effort estimation is frequently seen as complex and expensive for individual software engineers. In Project LEAP, we are investigating tools and methods to support low-cost, empirically-based software developer improvement. In a recent case study, we investigated effort estimation and the relative accuracy of a dozen different analytic estimation procedures. Student programmers could estimate the effort required using any of the analytic methods, or else provide their own “guesstimate”. Our study provides evidence that “guesstimates”, when informed by low-cost analytical methods, may be the most accurate of all.