Differences

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

--- public:rem4:rem4-15:t-tests_and_linear_models [2015/10/08 15:33] – created thorisson
+++ public:rem4:rem4-15:t-tests_and_linear_models [2024/04/29 13:33] (current) – external edit 127.0.0.1
@@ Line 77: / Line 77: @@
 | Main limitation of linear models  | Assumes -- apriori! -- a linear relationship  |
+\\
+\\
+\\
+\\
+=== Beyond Linear Models & T-Tests ===
+| Pick the appropriate statistics  | Depending on your experimental design, or on the nature of your field test, some tests are more appropriate than others.  |
+|  **What you study**  |  **What you use**  |
+| Two factors varying along a continuum  |  Correlation/regression measures  |
+| Two factors, where independent variable has (or can have) a few discrete values  |  Kolmogorov-Smirnov two-sample test, t-test (if distribution is normal), t-test for unequal variances (if variances between underlying populations),Wilcoxon-Mann-Whitney (if distribution is not normal)  \\ cf. \\ http://beheco.oxfordjournals.org/content/17/4/688.full  \\ http://advan.physiology.org/content/34/3/128  \\ http://rsos.royalsocietypublishing.org/content/1/3/140216  \\  https://xkcd.com/882/   |
+| One dependent variable, multiple independent variables, each with two or more levels  |  ANOVA - Analysis of variance   |
+| Many dependent variables, many independent variables  |  MANOVA (multiple analysis of variance)  |
+\\
+\\
+\\
+\\
+=== p<0.05: A Word of Warning  ===
+| 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    |
+| 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.  |
 \\
 \\