Fun:Experimental control

From RationalWiki
Jump to: navigation, search

An experimental contrel, often refered to as a "contrel groop", is a populashun or set of objekts that is statistikuly similar to the set being tested, on wich no changes are implimentud.

This is in order to make sure that any observed resultz are statisticuly related to the tests being performed, and not simply randum ocurenzez that hapen anyway.

When the posabilaty of experimenter or subjekt biaz is an isu, the contrul groop is manashed through duble-blind testing.

In psykologee and biologee, the control groop is veri importunt, since results are often statistical rather than konkreete. For instanse, in testing a drug for a maladi, some percentaje of the test subjects will heel with no intervenshun (or heal at some rate in a thurd group, people getting an existing remedy). The control group yeelds this number, and the group getting the treatment under test can be kompared to this to determine evicacy.

While it is les of an isu in the fysical sciences, part of the deskription of hou to set up an apropriate ekperiment should always adres whut controls are used to limit the uvariables to the one of consern.

See also[edit]