PubMed · 6436613
The source did not provide an abstract. Follow the original record for more information.
Explore connections, maps & timelines
K E Barnard. MCN keys to research. Planning experiments.. https://pubmed.ncbi.nlm.nih.gov/6436613/
Cite the original work for its findings. Save a collection to share your selection of sources.
KEEP EXPLORING
Recent empirical evidence supports the importance of adequate randomization in controlled trials. Trials with inadequate allocation concealment have been associated with larger treatment effects compared with trials in which authors reported adequate allocation concealment. While that provides empirical evidence of bias being interjected into trials, trial investigators rarely document the sensitive details of subverting the intended purpose of randomization. This article relates anonymous accounts run the gamut from simple to intricate operations, from transillumination of envelopes to searching for code in the office files of the principal investigator. They indicate that deciphering is something more frequent than a rate occurrence. These accounts prompt some methodological recommendations to help prevent deciphering. Randomized controlled trials appear to annoy human nature--if properly conducted, indeed they should.
Because the goal of discriminant analysis (DA) is to optimise classification, it designedly exaggerates between-group differences. This bias complicates validation of DA. Jack-knifing has been used for validation but is inappropriate when stepwise selection (SWDA) is employed. A simple randomisation test is presented which is shown to give correct decisions for SWDA. The general superiority of randomisation tests over orthodox significance tests is discussed. Current work on non-parametric methods of estimating the error rates of prediction rules is briefly reviewed.