3
edits
Changes
Jump to navigation
Jump to search
Line 133:
Line 133: + +
Conservapedia talk:Lenski dialog (view source)
Revision as of 11:29, July 8, 2008
, 11:29, July 8, 2008→Reply to comments in archive
:::::Aschlafly, the purpose of a Peer Review is not to make sure the entire experiment wasn't a massive fraud - it is to ensure that the work adhered to scientific principles and guidelines. In this case, the work was short and did adhere to principal. Lenski has, as required, laid out how the experiment can be repeated. Are you saying that he risked his entire reputation on results that can be checked and refuted by another lab? {{unsigned|Falsehood}}
:::::Aschlafly, the purpose of a Peer Review is not to make sure the entire experiment wasn't a massive fraud - it is to ensure that the work adhered to scientific principles and guidelines. In this case, the work was short and did adhere to principal. Lenski has, as required, laid out how the experiment can be repeated. Are you saying that he risked his entire reputation on results that can be checked and refuted by another lab? {{unsigned|Falsehood}}
::::Aschlafly, peer reviews are not suitable to detect all experiment flaws or outright fraud. Reviewers can't always know whether a lab assistent made a mistake or even if the experiment was done at all. The primary goal is to make sure that the experiment can be repeated so completely independent researchers can verify the results. This is the only good verification. A mistake in the experiment means that the data cannot be trusted. That is true for fraudulent data as well. There are several reasons to repeat an experiment. First of all, a groundbreaking result will nearly always be repeated. Scientists want to be sure they can trust the result and not waste years on related experiments. The second reason is to test yourself. A scientist may want to repeat an important experiment before starting their own related experiments. It may also be done to test a new lab or test setup. Thirdly, indications of fraud are an obvious reason. Fourthly, highly complicated experiments that can easily go wrong may be replicated. Fifthly, new scientists are less trusted and more likely to be checked. Now, this experiment is not particularly likely to be repeated. It is not particularly groundbreaking (evolutionary changes have been observed in many other experiments), complicated and there are no indications of fraud. Of course, you/Conservapedia may want to appeal to colleages of mr. Lenski for a repeat of (parts of) the experiment, but I doubt that they will be swayed by your arguments that seem to boil down to: 'I don't understand the science, but I don't like the result, so there must be a mistake'. It would be wiser to point to an actual lack of evidence or methodology flaw in the paper (if it exists), so an experiment can be drafted to test the significance of that mistake. --[[User:Aapje|Aapje]] 07:29, 8 July 2008 (EDT)
== More replies to archived comments ==
== More replies to archived comments ==