[Moller] Blind analysis in Physics experiments: Is this trip necessary?

[Jay Benesch]

Given the systematic error requirement, perhaps the collaboration should think about doing the first run unblinded to make studying them easier.  First ten percent? 


https://arxiv.org/abs/2311.13542

Robert Golub

    Based on the work of Klein and Roodman we present an alternate conclusion as to the charm of blind analysis in physics experiments. 

Discussion

Based on the paper by Klein and Roodman [2] written in support of the use
of blind analysis in physics experiments we have attempted to show that the
method, introduced into science by medical researchers to solve a real problem,
has little place in physics research, since, as the authors admit there is surprisingly 
little evidence for experimenter bias in physics research even taking into
account the cases cited by Feynman [1] and Galison [3]. and the introduction of
the technique has real costs in both the ability to study systematic errors that
are often unknown at the planning stage of the experiment and come to light
during the measurement and analysis and the ability to follow up unexpected
results which may only show up after opening the box.

The experience in medical research has shown that even double blind analysis 
is not sufficient to avoid experimental bias when the experimenters are
really determined. Physicists on the other hand still have the ability to follow
the advice of my mentor, Prof Gerrold R. Zacharias who often repeated the
statement that experimental physics was really all about Character.

[2] Klein, J. and Roodman, A. ’Blind analysis in nuclear and particle physics’
Ann. Rev. Nucl. Part. Sci. 55, 141-63, (2005)