![](/img/trans.png)
[英]How to structure dataset to run Binomial GLM of ratio of counts over time?
[英]How to structure dataset to run a PCA?
基本上,我的問題是我想運行PCA分析,但是我的數據結構不正確。 希望這張圖片能使您理解我的意思:
trial.one.two <- na.omit(trial.one.one)
head(trial.one.two)
v79 v81 v82 Q.One Q.Two Q.Three
2 Disagrees a little Agrees a little Disagrees a little 3 2 3
3 Agrees a lot Agrees a lot Disagrees a little 1 1 3
4 Agrees a little Disagrees a lot Disagrees a lot 2 4 4
5 Agrees a lot Agrees a lot Disagrees a lot 1 1 4
6 Agrees a little Agrees a lot Agrees a little 2 1 2
8 Agrees a lot Agrees a little Agrees a lot 1 2 1
我正在使用的數據是在5000多個個人中進行的一項調查,我想知道有多少人回答了例如“非常同意”:2253,“有點同意”:2005等。我需要這些數據來按以下方式分配:
1“非常同意” 2“有點同意” 3“有點不同意” 4“很多不同意”
其中1是組件1,2是組件2,依此類推,基本上,我想運行PCA。
誰能指導我該怎么做?
----------更新-------------
在實施之后:
convert.factor <- function(val){
if(val == "Agrees a lot"){
return(1)
} else if(val == "Agrees a little") {
return(2)
} else if(val == "Disagrees a little") {
return(3)
} else if(val == "Disagrees a lot") {
return(4)
}
}
trial.one.two$v79 <- sapply(trial.one.two$v79, convert.factor)
trial.one.two$v81 <- sapply(trial.one.two$v81, convert.factor)
trial.one.two$v82 <- sapply(trial.one.two$v82, convert.factor)
head(trial.one.two)
v79 v81 v82 Q.One Q.Two Q.Three
2 3 2 3 3 2 3
3 1 1 3 1 1 3
4 2 4 4 2 4 4
5 1 1 4 1 1 4
6 2 1 2 2 1 2
8 1 2 1 1 2 1
您可以按照以下方式進行操作
convert.factor <- function(val){
if(val == "Agrees a lot"){
return(1)
} else if(val == "Agrees a little") {
return(2)
} else if(val == "Disagrees a little") {
return(3)
} else if(val == "Disagrees a lot") {
return(4)
}
}
trial.one.two$v79 <- sapply(trial.one.two$v79, convert.factor)
trial.one.two$v81 <- sapply(trial.one.two$v81, convert.factor)
trial.one.two$v82 <- sapply(trial.one.two$v82, convert.factor)
另外,如果您只是在尋找人們回答每個類別的頻率,您可以執行以下操作:
table(trial.one.two$v79)
請注意,在這種情況下,沒有理由先轉換變量。
聲明:本站的技術帖子網頁,遵循CC BY-SA 4.0協議,如果您需要轉載,請注明本站網址或者原文地址。任何問題請咨詢:yoyou2525@163.com.