"TEST LATER"

A SCIENTIFIC STUDY SHOWS THAT THERE IS NO BENEFIT TO TEST DRIVEN DEVELOPMENT OVER AND ABOVE A "TEST LATER" APPROACH.


 
This study, whilst interesting and thought-provoking, detracts from something that we at Piccadilly Group believe is far more crucial: intelligent application of an approach, regardless of which is followed, by skilled and experienced practitioners. The study is well thought through and the demographic pool has excellent programming experience, but their experience of unit testing is significantly less with 50% having no experience; and those with no experience of TDD is even greater at nearly 80%.
 
For testing to be a highly efficient and effective tool, whether that’s unit testing, TDD, BDD, system testing, UAT or NFT to name a few, the individuals and team involved should be appropriately skilled and experienced, allowing them to intelligently apply that knowledge within the context of the delivery. That said, the study and indeed any attempt to quantify the perceived benefits of a test delivery model should be welcomed – they don’t happen very often!
David Mansfield, Piccadilly Group

Test-driven development (TDD) is an agile practice claimed to improve the quality of a software product, as well as the productivity of its developers. A previous study (i.e., baseline experiment) at the University of Oulu (Finland) compared TDD to a test-last development (TLD) approach through a randomized controlled trial. The results failed to support the claims.  In this paper we reported a replication of the experiment and we confirmed the baseline results: TDD does not affect testing effort.

Sam Gormley