Information Technology Reference
In-Depth Information
Chapter 7
Reporting Your Results
“Cut out the appropriate part of the computer output and paste
it onto the draft of the paper.” George Dyke (tongue in cheek)
[1997].
T
HE FOCUS OF THIS CHAPTER IS ON WHAT
to report and how to report it.
Reportable elements include the experimental design and its objectives, its
analysis, and the sources and amounts of missing data. Guidelines for table
construction are provided. The bootstrap is proposed as an alternative to
the standard error as a measure of precision. The value and limitations of
p
values and confidence intervals are summarized. Practical significance is
distinguished from statistical significance and induction from deduction.
FUNDAMENTALS
Few experimenters fail to list number of subjects, doses administered, and
dose intervals in their reports. But many fail to provide the details of
power and sample size calculations. Feng et al. [2001] found that such
careless investigators also report a higher proportion of nonsignificant
intervention effects, indicating underpowered studies.
Too often inadequate attention is given to describing treatment alloca-
tion and the ones who got away. We consider both topics in what follows.
Treatment Allocation
1
Allocation details should be fully described in your reports including
dictated allocation versus allocation discretion, randomization, advance
preparation of the allocation sequence, allocation concealment, fixed versus
varying allocation proportions, restricted randomization, masking, simulta-
1
This material in this section relies heavily on a personal communication from Vance W.
Berger and Costas A. Christophi.
