Paul K. Feyerabend (1976):
[...This result can be generalised...] Any hypothesis, however
implausible, can widen our horizon. It has not led to a better
understanding of science and it is even a hindrance to such
a better understanding because of its habit of beclouding
facts with semons and moralizing phrases (p. 136).
Paul K. Feyerabend (1978):
We may use hypothesis that contradict well confirmed
theories and/or well-established experimental results.
We may advance science by proceeding counterinductively
[FEYERABEND78, p. 5].
Paul F. Lazarsfeld (1956):
You never test hypotheses! First you run everything against
everything. Then find out what the relationships are to the
0.05 level. Then you write a hypothesis for a relationship
and say you tested it. And that's how Lazarsfeld said to do it
[GLASER98, p. 23].
Robert I. Sutton and Anat Rafaeli (1992):
When your hypotheses aren't confirmed, you don't admit it, you
change the hypotheses in your introductions with to fit the data.
[...] Of course, we always write our hypotheses after we get
our results. It is more efficient that way!
[FROSTSTABLEIN92, p. 122].
References
[FEYERABEND76]
Feyerabend, Paul K.
1976.
On The Critique of Scientific Reason.
In Essays in Memory of Imre Lakatos (Cohen, R. S., Feyerabend, P. K., and Wartofsky, M. W editor),
D. Reidel Pub., pp. 109-143.
[Q175 Bsps 39 CLMS, ISBN 9-0277-0654-9].
[FEYERABEND78]
Feyerabend, Paul K.
1978.
Against Method -- Outline of an Anarchistic Theory of Knowledge.
Verso, pp. 339.
[Q175 Fey CLMS, ISBN 0-8609-1700-2]
[FROSTSTABLEIN92]
Frost, Peter J. and Stablein, Ralph E.
1992.
Doing Exemplary Research.
Sage Pub., pp. 321.
[H62 Doi HLRBR, ISBN 0-8039-3908-6]
[GLASER98]
Glaser, Barney G.
1998.
Doing Grounded Theory: Issues and Discussions.
Sociology Press, pp. 254.
[-, -]