Source: http://www.r-bloggers.com/two-sample-students-t-test-2/
两个独立群组平均的比较,这里的两个群组从两个方差未知且样本方差不同质的总体抽取。
我们想比较两个个体群组的身高,单位inches。这里是测量数据:
A: 175, 168, 168, 190, 156, 181, 182, 175, 174, 179
B: 120, 180, 125, 188, 130, 190, 110, 185, 112, 188
如前面的《两个样本的Student's t-test (1)》,我们必须首先用Fisher's F-test来检验方差是否同质:
a = c(175, 168, 168, 190, 156, 181, 182, 175, 174, 179)
b = c(120, 180, 125, 188, 130, 190, 110, 185, 112, 188)
var.test(b,a)
F test to compare two variances
data: b and a
F = 14.6431, num df = 9, denom df = 9, p-value = 0.0004636
alternative hypothesis: true ratio of variances is not equal to 1
95 percent confidence interval:
3.637133 58.952936
sample estimates:
ratio of variances
14.64308
我们获得的p-value小于0.05,于是两个方差不同质。实际上我们能比较计算的F值和查表的F值(alpha=0.05,分子自由度=9,分母自由度=9),由函数qf(p, df.num, df.den)计算:
qf(0.95, 9, 9)
[1] 3.178893
计算的F大于查表的F,因此我们能拒绝方差同质的null hypothesis H0。
为比较以上两个群组,我们使用不同质方差的t.test函数(var.equal = FALSE,因为是默认值,也可以省略),并且样本独立(paired = FALSE,也因为默认值而可以省略),过程如下:
t.test(a,b, var.equal=FALSE, paired=FALSE)
Welch Two Sample t-test
data: a and b
t = 1.8827, df = 10.224, p-value = 0.08848
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
-3.95955 47.95955
sample estimates:
mean of x mean of y
174.8 152.8
正如我们在标题中看到,你进行了两个样本的t-test,自由度的计算使用Welch-Satterthwaite的公式(公式结果为df = 10.224),其用于方差不同质的情况。Welch-Satterthwaite方程也称作Dixon-Massey公式。我们获得的p-value大于0.05,于是我们能论断两个群组的均值显著相似(尽管p-value非常接近阈值0.05)。实际上t值小于自由度为10.224的查表t值,R中计算为:
qt(0.975, 10.224)
[1] 2.221539
我们能接受均值相等的null hypothesis H0。
Welch-Satterthwaite formula:
$$df=\frac{\sum deviance(X)}{\sum df(X)}=\frac{\sum_{i=1}^{k} (\sum_{j=1}^{n} (X_i_j - \bar{X_i})^2}{\sum_{i=1}^{k}(n_i-1)}$$
Dixon-Massey formula:
$$df=\frac{\left(\frac{\displaystyle S_1^2}{\displaystyle n_1}+\frac{\displaystyle S_2^2}{\displaystyle n_2}\right)^2}{\frac{\displaystyle\left(\frac{\displaystyle S_1^2}{\displaystyle n_1}\right)^2}{\displaystyle n_1-1}+\frac{\displaystyle\left(\frac{\displaystyle S_2^2}{\displaystyle n_2}\right)^2}{\displaystyle n_2-1}}$$