Towards standard for experiments in program comprehension

Program comprehension can make a unique contribution to the field of software engineering because it is feasible to validate its claims with inexpensive experiments. To fully realize this unique position, program comprehension researchers need to develop standards that will guide them in designing experiments and allow them to judge the strength of an experiment in supporting a claim. To begin the discussion leading to such standards, we propose that program comprehension experiments always measure and interpret the following dependent variables: accuracy, accurate response time, and inaccurate response time.