Differences

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

--- r:twoway_anova [2018/05/23 09:04] – [E.g. 1] hkimscil
+++ r:twoway_anova [2018/10/19 08:39] (current) – [e.g. 5] hkimscil
@@ Line 218: / Line 218: @@
 [{{:r:interaction effect eg04.png|Interaction effect example 4}}]
-===== 5 =====
+===== e.g., 5 =====
 download {{:r:dataset_anova_twoWay_comparisons.csv|dataset_anova_twoWay_interactions.csv}}
-<code>data <- read.csv("http://commres.net/wiki/_media/r/dataset_anova_twoway_comparisons.csv")
+<code>stressdata <- read.csv("http://commres.net/wiki/_media/r/dataset_anova_twoway_comparisons.csv")
 > #display the data
-> data
+> stressdata
    Treatment   Age StressReduction
      mental young              10
@@ Line 253: / Line 253: @@
    medical   old               0</code>
-<code>> aov(d$StressReduction~d$Treatment*d$Age, d)
+<code>
-Call:
+> stressdata <- read.csv("http://commres.net/wiki/_media/r/dataset_anova_twoway_comparisons.csv")
-   aov(formula = d$StressReduction ~ d$Treatment * d$Age, data = d)
+>
+> a.mod <- aov(StressReduction~Treatment*Age, data=stressdata)
-Terms:
-                d$Treatment d$Age d$Treatment:d$Age Residuals
-Sum of Squares           18   162                 0        18
-Deg. of Freedom           2     2                 4        18
-Residual standard error: 1
-Estimated effects may be unbalanced
-> a.mod <- aov(d$StressReduction~d$Treatment*d$Age, d)
 > summary(a.mod)
-                  Df Sum Sq Mean Sq F value  Pr(>F)
+              Df Sum Sq Mean Sq F value  Pr(>F)
-d$Treatment        2     18       9       9 0.00195 **
+Treatment      2     18       9       9 0.00195 **
-d$Age              2    162      81      81   1e-09 ***
+Age            2    162      81      81   1e-09 ***
-d$Treatment:d$Age  4      0       0       0 1.00000
+Treatment:Age  4      0       0       0 1.00000
-Residuals         18     18       1
+Residuals     18     18       1
 ---
 Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
-></code>
+>
+</code>
-<code>> pairwise.t.test(d$StressReduction, d$Treatment, p.adj = "none")
+<code>> TukeyHSD(a.mod, which="Treatment")
+  Tukey multiple comparisons of means
+% family-wise confidence level
-	Pairwise comparisons using t tests with pooled SD
+Fit: aov(formula = StressReduction ~ Treatment * Age, data = stressdata)
-data:  d$StressReduction and d$Treatment
+$Treatment
+                 diff        lwr       upr     p adj
+mental-medical      2  0.7968988 3.2031012 0.0013531
+physical-medical    1 -0.2031012 2.2031012 0.1135025
+physical-mental    -1 -2.2031012 0.2031012 0.1135025
-         medical mental
+>
-mental   0.13    -
+</code>
-physical 0.45    0.45
+<code>
+> TukeyHSD(a.mod, which="Age")
+  Tukey multiple comparisons of means
+% family-wise confidence level
-P value adjustment method: none</code>
+Fit: aov(formula = StressReduction ~ Treatment * Age, data = stressdata)
-<code>> pairwise.t.test(d$StressReduction, d$Age, p.adj = "none")
-	Pairwise comparisons using t tests with pooled SD
+$Age
+          diff       lwr       upr    p adj
+old-mid     -3 -4.203101 -1.796899 1.54e-05
+young-mid    3  1.796899  4.203101 1.54e-05
+young-old    6  4.796899  7.203101 0.00e+00
+>
+</code>
-data:  d$StressReduction and d$Age
-      mid     old
-old   2.5e-05 -
-young 2.5e-05 2.3e-10
-P value adjustment method: none
->
-</code>
 <WRAP info>위는 아래의 linear model 을 이용하여도 가능. 사실 모든 ANOVA 테스트는 linear model이기도 함(lm)
@@ Line 316: / Line 314: @@
 >
-</code>
-<code>> pairwise.t.test(d$StressReduction, d$Treatment, p.adj = "none")
-	Pairwise comparisons using t tests with pooled SD
-data:  d$StressReduction and d$Treatment
-         medical mental
-mental   0.13    -
-physical 0.45    0.45
-P value adjustment method: none</code>
-<code>> pairwise.t.test(d$StressReduction, d$Age, p.adj = "none")
-	Pairwise comparisons using t tests with pooled SD
-data:  d$StressReduction and d$Age
-      mid     old
-old   2.5e-05 -
-young 2.5e-05 2.3e-10
-P value adjustment method: none
->
 </code>