|
|
||||||||
Letters to the Editor |
Freeland, Oxford
Dear Sir,
It has been stated in the scientific press over several years that the frequency method of
statistical testing of experimental data by P-values and `confidence
intervals' (CI)
are subject to an unacceptable level of subjectivity giving inaccurate support to scientific
`discoveries' and medical `breakthroughs' (Matthews,
1998
).
This discussion has now moved into the public forum via the national press.
Are there any plans for the Editorial Committee to review the above findings to assess if they have any effect on the objective integrity of research material submitted for publication in the British Journal of Orthodontics?
References
Berger, J. and Selke, T. (1987) Testing a point null hypothesis: The irreconcilability of P-values and evidence, Journal of the American Statistical Association, 81, 112.
Matthews, R. (1998) Facts versus Factions: the use and abuse of subjectivity in scientific research, European Science and Environment Forum, a working paper.
This article has been cited by other articles:
![]() |
R. G. Newcombe Statistical Applications in Orthodontics. Part I. Confidence Intervals: an Introduction J. Orthod., September 1, 2000; 27(3): 270 - 272. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |