在R

Question

我正在处理数据集，该数据集包含大量关于使用公共交通工具的人员调查的变量。

附加数据集和csv文件。 Data Variables

数据链接：https://drive.google.com/open?id=1MvfnwR4IkUyUzSnCuAOL8fxAYiDjBoBi

读取数据集的代码。

df = read.csv("PublicTransportSurvey.csv",sep=";", header = T, stringsAsFactors=TRUE)
# Display the dataset and obtain overall summary of the dataset 
df <- subset(df, select = -Row_Num)
View(df)

变量也可归纳如下：

项目是李克特量表（1-5），可能有回应：非常不同意（1），不同意（2），中立（3），同意（3）和非常同意（5）。

Perceived Usefulness and Ease of Use
PU1: PT information is easily accessible
PU2: PT infrastructure is easily accessible 
PU3: The maps on PT infrastructure are helpful and clear
PU4: PT tickets are easy to purchase 
PU5: PT connections in Adelaide are well integrated
PU6: Waiting times for PT services are reasonable


Perceived Enjoyment 
ENJ1: The views from PT in Adelaide are scenic
ENJ2: Fellow passengers on PT in Adelaide are friendly 

Quality 
QU1: PT in Adelaide is reliable
QU2: PT in Adelaide supports disabled travellers
QU3: PT in Adelaide offers free wi-fi
QU4: PT in Adelaide has a low carbon footprint
QU5: PT in Adelaide is clean


Safety and Security
SS1: PT is safe in Adelaide
SS2: Adelaide PT drivers handle unruly passengers
SS3: PT shelters in Adelaide are well-lit at night-time 

Use Behaviour 
USE1: I use PT in the mornings only
USE2: I use PT during off-peak times
USE3: I use PT only during the evening
USE4: I use PT during the week
USE5: I use PT at the weekend


PT Incentives 
INC1: I use PT to save money
INC2: I use PT to protect the environment
INC3: I use PT to exercise more
INC4: I use PT to experience the city firsthand

Information Access
INF1: I access PT timetables and information using a mobile device 
INF2: I access PT timetables and information from a hotel concierge
INF3: I access PT timetables and information on the platform
INF4: I access PT timetables and information from a newsagency
INF5: I access PT timetables and information from other commuters

但是，如果我们看到我附加的数据集变量图片，它还包含一些超过1-5的值。

我在过去4个小时内陷入了这个问题并尝试搜索。

我的最终目标是从上述变量中​​删除异常值（大于5），然后绘制一个相似的图。请有人建议我，如何解决这个问题。

先谢谢。

Answer 1

我的解决方案：

library(likert)

df <- read.csv2("PublicTransportSurvey.csv")

df <- df[,12:54]
df[sapply(df, is.factor)] <- lapply(df[sapply(df, is.factor)], function(x) as.numeric(as.character(x)))
df[sapply(df, is.character)] <- lapply(df[sapply(df, is.character)], function(x) as.numeric(as.character(x)))

df <- data.frame(apply(df, 2, function(x) ifelse(x > 5, NA, x)))
df <- data.frame(lapply(df, function(x) as.factor(x)))

likert_df <- likert(df)
plot(likert_df)

首先，我删除了非Likert变量的列。然后我将因子和字符列转换为数字列，并用NA替换了大于5的所有值，因为据我所知，likert包会忽略这些值。

然后我将所有列转换回因子，因为likert函数需要这些因素。代码生成此图像：