Groupwise comparison for continuous variables

Load BONEDEN.DAT.txt as bone. Create a difference variable from fn1 and fn2

bone <- read.csv("BONEDEN.DAT.txt", quote = "'")
bone$fn.diff <- bone$fn1 - bone$fn2

One-sample t-test for the difference variable

t.test(bone$fn.diff, mu = 0)

    One Sample t-test

data:  bone$fn.diff 
t = 0.0503, df = 40, p-value = 0.9601
alternative hypothesis: true mean is not equal to 0 
95 percent confidence interval:
 -0.02866  0.03013 
sample estimates:
mean of x 
0.0007317 

Paired t-test for the variable pair is the same thing

t.test(bone$fn1, bone$fn2, paired = TRUE)

    Paired t-test

data:  bone$fn1 and bone$fn2 
t = 0.0503, df = 40, p-value = 0.9601
alternative hypothesis: true difference in means is not equal to 0 
95 percent confidence interval:
 -0.02866  0.03013 
sample estimates:
mean of the differences 
              0.0007317 

Independent two group comparison

t.test(age ~ zyg, data = bone, var.equal = TRUE)

    Two Sample t-test

data:  age by zyg 
t = 1.427, df = 39, p-value = 0.1615
alternative hypothesis: true difference in means is not equal to 0 
95 percent confidence interval:
 -2.163 12.525 
sample estimates:
mean in group 1 mean in group 2 
          51.38           46.20 

Independent two group comparison using summary data

library(BSDA)
tsum.test(mean.x = 51.38, s.x = 10.74, n.x = 21, mean.y = 46.20, s.y = 12.48, n.y = 20, var.equal = TRUE)

    Standard Two-Sample t-Test

data:  Summarized x and y 
t = 1.427, df = 39, p-value = 0.1616
alternative hypothesis: true difference in means is not equal to 0 
95 percent confidence interval:
 -2.164 12.524 
sample estimates:
mean of x mean of y 
    51.38     46.20 

To compare variance

var.test(age ~ zyg, data = bone)

    F test to compare two variances

data:  age by zyg 
F = 0.7406, num df = 20, denom df = 19, p-value = 0.5106
alternative hypothesis: true ratio of variances is not equal to 1 
95 percent confidence interval:
 0.2952 1.8382 
sample estimates:
ratio of variances 
            0.7406 


One-sample Wilcoxon signed rank test (no continuity correction). Symmetry required.

wilcox.test(bone$fn.diff, mu = 0, correct = FALSE)

    Wilcoxon signed rank test

data:  bone$fn.diff 
V = 403.5, p-value = 0.8505
alternative hypothesis: true location is not equal to 0 

One-sample Wilcoxon signed rank test (no continuity correction). No assumption of normality or symmetry.

library(BSDA)
SIGN.test(bone$fn.diff, md = 0)

    One-sample Sign-Test

data:  bone$fn.diff 
s = 22, p-value = 0.5224
alternative hypothesis: true median is not equal to 0 
95 percent confidence interval:
 -0.04  0.04 
sample estimates:
median of x 
       0.02 

                  Conf.Level L.E.pt U.E.pt
Lower Achieved CI     0.9404  -0.04   0.04
Interpolated CI       0.9500  -0.04   0.04
Upper Achieved CI     0.9725  -0.04   0.04

Wilcoxon signed rank test for paired data (no continuity correction)

wilcox.test(bone$fn1, bone$fn2, paired = TRUE, correct = FALSE)

    Wilcoxon signed rank test

data:  bone$fn1 and bone$fn2 
V = 403.5, p-value = 0.8505
alternative hypothesis: true location shift is not equal to 0 

Independent two group comparison (Wilcoxon rank sum test)

wilcox.test(age ~ zyg, data = bone, correct = FALSE)

    Wilcoxon rank sum test

data:  age by zyg 
W = 280, p-value = 0.06747
alternative hypothesis: true location shift is not equal to 0 


Load BETACAR.DAT.txt as vitA (4-types of beta-carotene supplements and plasma carotene concentration)

vitA <- read.csv("BETACAR.DAT.txt", quote = "'")

Independent three or more group comparison

anova(lm(Base1lvl ~ factor(Prepar), data = vitA))
Analysis of Variance Table

Response: Base1lvl
               Df Sum Sq Mean Sq F value Pr(>F)
factor(Prepar)  3   8379    2793    0.73   0.55
Residuals      19  73051    3845               

Independent three or more group comparison (distribution-free)

kruskal.test(Base1lvl ~ factor(Prepar), data = vitA)

    Kruskal-Wallis rank sum test

data:  Base1lvl by factor(Prepar) 
Kruskal-Wallis chi-squared = 2.268, df = 3, p-value = 0.5187