Like most creationists Kyle Shockley makes much of geologists discarding discordant results. It is important to explain what this means. If a series of measurements are taken and (say) 9 give closely matching results and one is wildy out, it is quite reasonable to believe that the one discordant result is wrong. By analogy, if you were doing research on George Washington, and found his birth year given as 1732 in 9 texts and 1740 in one, this is not evidence for the non-existance of George Washington, but evidence of an error in one text. One bad result does not prove anything other than that errors occur.
On the subject of radio-dating tuf, this is extradorinarily difficult, and, for that reaon rarely undertaken. This is due to the nature of the material. Tuf is a mixture of molten and solid material exected from a volcano. Radio dating the molten part will return the date of the eruption, radio-dating the rest will return the date of the base rock. Thus the samples must be very carefully separated in order to achieve an accurate result.
On comapring with faunal evidence, radio dating in other areas has established a "range" in which certain species exist. This can be used to check the dating of the sampl.
------------------
For Whigs admit no force but argument.