anova的步骤

差别

这里会显示出您选择的修订版和当前版本之间的差别。

到此差别页面的链接

--- anova的步骤 [2023/03/27 12:34] – 创建戴娉婷
+++ anova的步骤 [2024/06/09 08:55] (当前版本) – zzzz
@@ 行 1: / 行 1: @@
-==== ANOVA的步骤 ====
+==== ANOVA的步骤 (Steps of ANOVA) ====
-=== ===
+、陈述原假设H<sub>0</sub>和备择假设H<sub>1</sub>，确定进行检验的显著性水平α；
+ * State H<sub>0</sub> and H<sub>1</sub>.Determine the level of significance α.
+、确定检验的方向性，需要注意的是ANOVA检验总是**单尾检验**；
+ * Determine the directionality of the test. Notice: the ANOVA test is always a **single tail test**.
+、指出检验的自由度df，注意在ANOVA检验中有两个自由度，分别是组间变异的自由度（k-1）和组内变异的自由度（N-k），其中k为设置水平数，N为总的样本容量；
+ *  Indicate the degree of freedom of the test. Notice: there are two degrees of freedom in the ANOVA test, namely the degree of freedom of inter-group variation (k-1) and the degree of freedom of intra-group variation (N-k), where k is the number of setting levels and N is the total sample capacity.
+、根据组间自由度、组内自由度以及显著性水平α，查表找出临界F统计量的值；
+ * Find the critical F statistic by looking up the table according to inter-group degrees of freedom, intra-group degrees of freedom and significance level α.
+、对于样本计算F统计量的值，画出如下图所示的方差分析表
+{{ ::方差分析表.jpg?nolink&350 |}}
+ * Calculate the F value of your sample, and draw the ANOVA table as shown in the figure above.
+、比较计算得出的F值和临界F值，如果F<F<sub>crit</sub>，则接受原假设H<sub>0</sub>，如果F>F<sub>crit</sub>，则拒绝H<sub>0</sub>，接受备择假设H<sub>1</sub>；
+ * Compare the calculated F value with the critical F value.If F<sub>obs</sub> < F<sub>crit</sub>，then accept the null hypothesis; if not, then refuse it and accept the alternative hypothesis.
+、进行事后检验。
+ * Carry out the post hoc test.

anova的步骤.1679920495.txt.gz · 最后更改: 2023/03/27 12:34 由戴娉婷