Differences

This shows you the differences between two versions of the page.

--- public:rem4:rem4-15:t-tests_and_linear_models [2015/10/12 10:43] – thorisson2
+++ public:rem4:rem4-15:t-tests_and_linear_models [2024/04/29 13:33] (current) – external edit 127.0.0.1
@@ Line 96: / Line 96: @@
 \\
-=== A Word of Warning: What Does p<0.005 Mean? ===
+=== p<0.05: A Word of Warning  ===
-| XXX say  |  If there were actually no effect (if the true difference between means were zero) then the probability of observing a value for the difference equal to, or greater than, that actually observed would be p=0.05. In other words there is a 5% chance of seeing a difference at least as big as we have done, by chance alone.  |
+| What Does p<0.005 Mean?  | David Colquhoun says: If there were actually no effect (if the true difference between means were zero) then the probability of observing a value for the difference equal to, or greater than, that actually observed would be p=0.05. In other words there is a 5% chance of seeing a difference at least as big as we have done, by chance alone.   \\  http://beheco.oxfordjournals.org/content/17/4/688.full    |
-| In more detail | Of course the number will be right only if all the assumptions made by the test were true. Note that the assumptions include the proviso that subjects were assigned randomly to one or the other of the two groups that are being compared. This assumption alone means that significance tests are invalid in a large proportion of cases in which they are used. Here, however, we shall deal only with the perfect case of properly randomized, bias-free tests. |
+| The number will be right only if all the assumptions made by the test were true  | One of the assumptions is that the measurements are truly randomized -- that there is no relationship between the measurements of the dependent variable on the dimensions of the independent variable being tested. This assumption is however frequently broken.  |
 \\
 \\