On Data (Unsourced)

"An extremely odd demand is often set forth but never met, even by those who make it; i.e., that empirical data should be presented without any theoretical context, leaving the reader, the student, to his own devices in judging it. This demand seems odd because it is useless simply to look at something. Every act of looking turns into observation, every act of observation into reflection, every act of reflection into the making of associations; thus it is evident that we theorize every time we look carefully at the world." (Johann Wolfgang von Goethe)

"Errors using inadequate data are much less than those using no data at all." (Charles Babbage)

"If you can’t have an experiment, do the best you can with whatever data you can gather, but do be very skeptical of historical data and subject them to all the logical tests you can think of." (Robert Hooke)

"In general, it is necessary to have some data on which to calculate probabilities.[...] Statisticians do not evolve probabilities out of their inner consciousness, they merely calculate them." (Leonard C Tippett)

"In these days of rapid scientific progress there is a tendency to accept the facts of nature, as at present known, without glancing back at the slow and difficult stages by which the knowledge of these facts has been arrived at. Yet such a retrospect is by no means unprofitable, since it warns us that hasty generalizations upon insufficient data retard rather than advance the progress of knowledge, and that the theories of the day must not be accepted as necessarily expressing absolute truths." (Archibald Garrod) 

"[…] numerous samples collected without a clear idea of what is to be done with the data are commonly less useful than a moderate number of samples collected in accordance with a specific design." (William C Krumbein)

"Progress in science depends not only upon new data but also upon the careful elaboration of new approaches to old data as well as new." (David Rindos)

"Too little attention is given to the need for statistical control, or to put it more pertinently, since statistical control (randomness) is so rarely found, too little attention is given to the interpretation of data that arise from conditions not in statistical control." (William E Deming)