Bridgstock, Martin. "A Sociological Approach to Fraud in Science," Austrialian and New
Zealand Journal of Sociology 18 (November, 1982), 364-383.
This is a functional piece, one which suggests that by studying different types of "science," one
can develop a model of cheating in "mainstream" science. Thus, Bridgstock suggests that one
examine fraud or cheating in pseudoscience, in cheating among "hired hands," and cheating in
industrial science. Mainstream science is very hard to study directly and it may be easier to get an
image of cheating in these less than mainstream areas.
To take just one example: industrial science is not straight science. Employers have different
expectations of what an employee-scientist is about: he is, at best, doing applied work and using
what skills he possesses for the good of the company for which he works. If he cuts corners, if he
"pushes" results, if he does less than sophisticated research, all that can be tolerated by the
employer. Applying this to mainstream science one can suggest that "employers" such as deans
and vice-presidents have different expectations of an employee than a co-professional may have.
One could develop, (though it is not developed here), a view of different "scientific expectations"
of co-professionals and administrators.
One bothersome thing about this article is that it lists references incompletely and leaves others
out in the bibliography.
In his discussion of why replication is not done in mainstream science, he provides nothing
beyond what might easily be found in Psycholog Today.