Unit Testing, Agile Development, Leadership & .NET - By Roy Osherove
Here's an interesting analysis of that same study with a bit more scrutiny:
scruffylookingcatherder.com/.../tdd-proven-effective-or-is-it.aspx
Note that I am a big proponent of TDD and actually do TDD at work and in private projects.
What the study seems to miss is that TDD IMHO tends to improve the design and make the resulting code easier to refactor.
Roy Osherove wrote "To read a good summary of that paper go to Phil's post and read it to the end."
To read a good summary of that paper go to the end of the paper and read the "CONCLUSIONS AND FUTURE WORK" section ;-)
"Test-First programmers did not achieve better quality on average, ... Writing more tests improved the minimum quality achieved and decreased the variation, but this effect does not appear to be specific to Test-First."
Pingback from TDD really works! :: Fat Penguin
Thanks for updating with my analysis. It's an interesting report with some support for TDD, but also some fascinating data points that I think are troubling as well (that I think should have made it to the abstract).