(rhetoric, logic) To select only evidence which supports an argument, and reject or ignore contradictory evidence.
Example: From all the available statistics, the politician cherry-picked only those that backed up his ideas.
1980, Television/radio age, Volume 28, page 80:
Those stations that will send CNN tapes of their local and regional news for CNN to cherry-pick from, in turn will be allowed to cherry-pick themselves.
2005, Kiplinger's Personal Finance, The PC Way to Pick Winners, v. 59, no. 4, page 44:
Screeners can help you cherry-pick a promising group of stocks that meet your exact specifications.
2008, Donna Andrews, The Penguin Who Knew Too Much, page 98:
"We're not offering you a chance to cherry-pick the Caerphilly Zoo to fill in the gaps in your own collection.
2012 April 29, Nathan Rabin, â€œTV: Review: THE SIMPSONS (CLASSIC): â€œTreehouse of Horror IIIâ€ (season 4, episode 5; originally aired 10/29/1992)â€:
In time The Simpsons would, indeed, resort to spoofing such decidedly non-spooktacular fare like E.T and Mr. And Mrs. Smith (both in â€œTreehouse Of Horror XVIIIâ€) but in 1992 the field was wide-open and the show could cherry-pick the most iconic and beloved fright fare of all time.
1999, Nicholas Alexander and Gary Akehurst, The Emergence of Modern Retailing, 1750-1950, p.38:
The reason for this concentration is to attempt to alleviate the problems of fragmentation which usually accompany the study of retailing, particularly at the end of the early period, and has so often happened in the past, to curtail the temptation to 'cherry-pick' the best evidence from a wide, but not necessarily related, area.
2001, Ian St. John in sci.environment
You are *asking* for a specific *subset* of the data to *prove* a specific theory. That is 'cherry picking'. It is no more valid to 'cherry pick' data to disprove global warming than it is to 'cherry pick' data to prove it. That is *science*. You *must* deal with the entire data set. What part of this is not clear?
2006, Scientific developments relating to the Abortion Act 1967: twelfth report of session 2006-07, Vol. 2: Oral and written evidence, p.32:
Yes, I do think in principle is is a good idea to have clearing houses where data is assessed and has as objective an account as possible given of the data because of the tendency of all organisations to cherry-pick those studies which may support a pre-existing position.
2008, Warren Davies, How to Study Psychology, p.80:
So it's possible to 'cherry pick' only the evidence that supports a position and ignore the rest. To the reader who does not know about the rest of the evidence, it may seem like a strong case. The whole body of evidence must be considered.