[SystemSafety] Analysis of some Work Breakdown Structure projects

Derek M Jones derek at knosof.co.uk
Thu Jun 10 03:56:19 CEST 2021


Rod,

> That's the data from the original version of the PSP training course. The course involves writing 10 small-ish programs, 
> hence "over 8000" programs for 810 developers.

Thanks.  I did not know that.

> I asked Watts if they'd mine and publish the data-set, but SEI were never inclined to do so... one of the things about 
> PSP is that individuals' personal data stays that way. I have data for about 12 people (I was the Instructor after all) 
> taking the course at Altran UK in about 2010/11.

Everybody has to start at zero, and 12 is progress.
When you get to around 100 it would be worth trying
to fit a regression model.  Here is one I did earlier:
https://shape-of-code.coding-guidelines.com/2018/08/19/impact-of-team-size-on-planning-when-sitting-around-a-table/

I think the interesting output is what is not there, and the
variability between people.  So many models blithely assume
that human performance is more or less the same.

-- 
Derek M. Jones           Evidence-based software engineering
tel: +44 (0)1252 520667  blog:shape-of-code.coding-guidelines.com


More information about the systemsafety mailing list