SEED (the Software Engineering Evidence Database), is at http://www.evidencebasedse.com
. I found it via Janzen and Ryoo's paper "Engaging the net generation with evidence-based software engineering through a community-driven web database", Journal of Systems and Software
, 82 (2009), pp. 563-70, DOI:10.1016/j.jss.2008.12.047. The site contains paragraph-sized summaries of over 200 empirical software engineering studies, indexed by them. There's no rating/ranking system, but lots of useful detail---a typical entry
is included below to give you a flavor. According to the paper, contributions are actively sought...
An Exploratory Longitudinal Case Study of Agile Methods in a Small Software
This thesis introduced TDD, SCRUM, and pair programming to a developer team in PetroSleuth. One main threat to the external validity of the study was that the team was relatively small with many developers that just joined the team from just coming out from a university. They found that implementing a TDD practice had its difficulties. One was that the build would break due to the complexity of the build rather than the source code, another was that some developers had trouble translating tasks into tests, and they had difficulties learning how to write tests and use the test framework. At the beginning developers were not using TDD, and pair programming as much but towards the end of the case study, they were using these practices much more often. The case study is considered a success even though more needs to be done in order to refine the study and gather more accurate data.
C. Mann, An Exploratory Longitudinal Case Study of Agile Methods in a Small Software Company, master's thesis, Dept. Computer Science, Univ. of Calgary, 2004.
Type of Study: Case Study
Duration: 8 months
Type of project: Windows-based oil and gas project management with
statistical modeling elements
# of developers: 4-7
project life: legacy
Team Language Experience: Experienced