
138 ◾ Software Metrics
investigating has already occurred, you must perform a survey or case
study. If the activity has yet to occur, you may choose between a case study
and a controlled experiment.
e central factor in this choice is the level of control needed for an
experiment. If you have a high level of control over the variables that can
aect the outcome, then you can consider an experiment. If you do not
have that control, an experiment is not possible; a case study is the pre-
ferred technique. But the level of control satises the technical concerns;
you must also address the practical concerns of research. It may be pos-
sible but ...