
225
Chapter 6
Analyzing Software
Measurement Data
*
I
C , general techniques for designing empir-
ical studies. en, in Chapter 5, we explained the importance of good
data-gathering to support rigorous investigation. In this chapter, we turn
to the analysis of the data.
Data analysis involves several activities and assumptions:
• We have a number of measurements of one or more attributes from
a number of soware entities (products, processes, or resources). We
call the set of measurements a dataset or a batch.
• We expect the soware items to be comparable in some way. For exam-
ple, we may compare modules from the same so