By Christian Bird

Published in Perspectives on Data Science for Software Engineering

We have found that interviews can be wonderful tools for exploratory investigation and they often can drive the formation of theories and hypotheses. We can then use complementary quantitative methods to further investigate and support these hypotheses. Interviews allow rich engagement and follow up questions. You can collect historical data that is not recorded anywhere as well as elicit opinions and impressions in richer detail than people would provide through written communication. Information from interviews can be triangulated with other data sources. In addition, interviews can be used to clarify things that have already happened (especially following an observation).  In short, the interview can be an important tool that you should have in your research toolbox.

In this chapter, I’ll provide a description and best practices for the different phases of interview research based on experiences that we have had conducting interviews for software engineering research at Microsoft.


  author = {Christian Bird},
  title = {{Interviews}},
  booktitle = {{Perspectives on Data Science for Software Engineering}},
  publisher = {Morgan Kaufmann},
  year = {2016}