DocumentCode
1871817
Title
Performance norms: An approach to rework reduction in software development
Author
Conroy, Patrick ; Kruchten, Philippe
Author_Institution
Electr. & Comput. Eng., Univ. of British Columbia, Vancouver, BC, Canada
fYear
2012
fDate
April 29 2012-May 2 2012
Firstpage
1
Lastpage
6
Abstract
Rework consumes large portions of software engineering budgets. Human factors, and Cognitive Bias in particular, have been shown in other disciplines to be implicated in the kinds of reasoning errors that lead to rework. Research of these phenomena in software engineering lags similar efforts in other disciplines. This study identifies the Performance Norms, standards by which Cognitive Biases are determined to have occurred, in a single but critically important software engineering task: Estimating. Analysis of data from professional practitioners regarding real-life situations indicates that several Performance Norms for Estimating are often `in play´, the least important being that assumed in previous, lab-based experiments. Most of these Norms require skills very different from those in which most technical personnel are trained. We conclude that rework reduction efforts will continue to falter until Performance Norms are recognized as key determinants in software engineering practice.
Keywords
cognitive systems; human factors; software cost estimation; cognitive bias; estimating task; human factors; performance norms; rework reduction; software engineering budgets; software engineering practice; Cognition; Humans; Interviews; Psychology; Software; Software engineering; Standards; cognitive bias; grounded theory; human factors; performance norms; rework;
fLanguage
English
Publisher
ieee
Conference_Titel
Electrical & Computer Engineering (CCECE), 2012 25th IEEE Canadian Conference on
Conference_Location
Montreal, QC
ISSN
0840-7789
Print_ISBN
978-1-4673-1431-2
Electronic_ISBN
0840-7789
Type
conf
DOI
10.1109/CCECE.2012.6335063
Filename
6335063
Link To Document