Don’t embarrass yourself

Beware of bias in your data

C. Bird    Microsoft Research, Redmond, WA, United States


Bias may exist in data that you collect for quantitative analysis in software engineering research. Some forms of bias can negatively affect the validity of analysis, results, and conclusions. In this chapter, I discuss forms of bias, how to test for them, and in some cases how to address them.


Bias; Features of defects; Population distribution; Sample distribution; Histogram

Dewey Defeats Truman

The 1948 U.S. presidential election proved to be one of the ...

Get Perspectives on Data Science for Software Engineering now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.