Question

你能告诉我我做错了什么吗？我正试图对5支NBA球队的工资进行ANOVA测试。我试图将薪水改为一个因素，但我怀疑我错过了一步。我非常感谢你的回应。请参阅下文。

> read.csv("~/DATA630/ANOVANBATeams.csv")
    Pistons    Wizards  Cavaliers   Clippers    Thunder
1 7,000,000 16,957,900 18,000,000 21,468,695 17,832,627
2 3,650,000  5,497,625 21,108,899  6,938,750  4,663,129
3   2772440  4,440,000  5,982,375 10,188,750  2,757,500
4 5,479,933  2,123,733  4,525,000  1,448,490  2,184,960
5 1,208,325  1,448,490    903,456    613,478    966,800

> > read.csv("~/DATA630/ANOVANBATeams.csv", header = TRUE)
Error: unexpected '>' in ">"
> salary <- read.csv("~/DATA630/ANOVANBATeams.csv", header = TRUE)
> salary$Team = as.factor(salary$Team)
Error in `$<-.data.frame`(`*tmp*`, "Team", value = integer(0)) : 
  replacement has 0 rows, data has 5

Answer 1

听起来您希望您的数据采用长格式，并以团队名称为因素。试试这个：

salary <- read.csv("~/DATA630/ANOVANBATeams.csv", header=T)

Teams <- colnames(salary)

salary2 <- reshape(salary, direction="long", varying=list(Teams), ids=Teams)
colnames(salary2)[2:3] <- c("Salary", "Team")

工资中的逗号会阻止R将其正确转换为数字，因此我们可以使用gsub()删除它们。

salary2$Salary <- as.numeric(gsub(",", "", salary2$Salary)))
salary2$Team <- as.factor(salary2$Team)

进行方差分析。

aov(Salary ~ Team, data=salary2)

R薪资的方差分析

1 个答案: