Scribd
Upload
28 views4 pages

R Basics: Data Import & Management

This document provides an overview of basic R concepts including importing and working with data frames, descriptive statistics, and splitting data frames by factors. It discusses importing data from CSV files, the main data types in R, using assignment operators to create variables, and functions for summarizing and exploring data like str(), summary(), describeBy(), and by(). Exercises demonstrate how to extract subsets of data, check data types, produce summaries grouped by factors, and load data from other packages.

Uploaded by

Viem Anh
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
28 views4 pages

R Basics: Data Import & Management

This document provides an overview of basic R concepts including importing and working with data frames, descriptive statistics, and splitting data frames by factors. It discusses importing data from CSV files, the main data types in R, using assignment operators to create variables, and functions for summarizing and exploring data like str(), summary(), describeBy(), and by(). Exercises demonstrate how to extract subsets of data, check data types, produce summaries grouped by factors, and load data from other packages.

Uploaded by

Viem Anh
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

STA2 - R LAB 1

Review on Basics of R
1. Objectives
 Importing data
 Dataframe
 Basic descriptive functions
 The „by‟ function
This review lab is not intended to be comprehensive. Students thus are strongly advised to review all
labs in the course of Probability and Statistics (PAS).

2. Installing R & Rstudio


Detailed instruction on how to install R and Rstudio will be posted on your Google class.

3. How to create a variable


If we want to create a variable x that holds the result of the expression: 27*log10(3) + sqrt(65), we
can try:
 x <- 27*log10(3) + sqrt(65)
The “<-” is called the assignment operator, which assigns the value of the expression on the right
to whatever object we specify on the left. Note that R is a case-sensitive language, so x and X are
different. We are not required to declare the type of the variable. It is possible for a variable to hold
any object such as a function or a plot.
Exercise 1: Let‟s create a variable y that takes the value equaling the square root of ten times of x.
Then check the data type of y using the class function.
 class(y)

4. Main data types


In R, the basic data types include numeric, character and logical. Numeric data is the most common
data type. Numeric data comprise integers and decimals. For example, type in the following code:
 a <- 5
 class(a)
The next type of data is character. As its name suggests, character data are represented by strings of
characters and/or numbers.
 b <- "blue"
 class(b)
 b <- "blue5"
 class(b)
 b <- "5"
 class(b)

1|P a ge
STA2 - R LAB 1

Logical data can be either TRUE or FALSE. Let‟s type in:


 x <- FALSE
 y <- TRUE
 x==y
Notice that in the last statement, we use a comparison operator ==. Other comparison operators
are: <, >, !=, >=, <=. For example,
 5 >=7
 3<6
 TRUE !=FALSE
 9 <= 10

5. Managing working directory


To set our working directory, we can use setwd(“mydirectory”). Next, type getwd() at the
command prompt. Our current working directory should be listed. R will read files from and save
results to this directory. It is necessary that you set up your working directory in every lab
before you run any codes.
Exercise 2: Create a folder in your D directory and name it Lab1. Go back to R and change the
working directory to “D:/Lab1”. Now try to display our current working directory.

6. Importing data from a delimited text file


You can create a CSV file by entering data into Excel, and then save the file as a CSV (comma
delimited) file. Now, open the file [Link], and try saving it as a CSV file with the same name
mtcars. Copy and paste the file to be used for the lab into your working directory. We can use the
[Link]() or [Link]() function to import csv data into R.
The code for reading the .csv file is shown below:
 data1 <- [Link]("[Link]",header=TRUE, sep=",",
quote="\"", stringsAsFactors=FALSE)

Exercise 3:
a. What does the following code do?
 extracted1 <- data1[1:10,]
 extracted2 <- data1[,1:3]
b. Check the structure of data1 using str() function. How many variables and observations are
there? Extract data for mpg, cyl and am variables and assign it to a data frame called
extracted3.
7. Data frames
A data frame in R is analogous to a dataset in SPSS, SAS or Stata. It is more general than a matrix in
that it permits different data types (numeric, character…) for different columns. Usually, a column
represents data for a variable (column name = variable name), and each row represents data for one
observation (or case in the language of SPSS). Data for one column must be of the same type.

2|P a ge
STA2 - R LAB 1

Note down the variable names, the number of variables, and the number of observations in data1.
What does the first column represent?
Let‟s try the following
mean(data1$mpg)
sd(data1$mpg)
summary(data1$mpg)
plot(data1$mpg, data1$wt)

In the above code, we access the mpg variable of the data1 data frame via the $ notation. This
notation signifies that we are dealing with a specific variable from that data frame.

8. Descriptive statistics
The following code produces summary statistics for 3 variables simultaneously.
 summary(data1[c("mpg", "drat", "wt")])

If we want more descriptive measures, the function [Link]() in the pastecs package can do a
good job. Let‟s install the pastecs package. Then follow the code listing below:
 library(pastecs)
 myvars <- c("mpg", "drat", "wt")
 [Link](data1[myvars], basic=TRUE, desc=TRUE, norm=FALSE,
p=0.95)

Note that by default, basic=TRUE, desc=TRUE, norm=FALSE, p=0.95.


Note that all numerical descriptive measures introduced in the labs (e.g. sd, var, etc.) are
sample measures. If you wish to compute measures for a population, in some cases you
have to adjust the obtained values by an appropriate formula.
The describeBy() function in the psych package provides more summary statistics by group.
Let‟s try:
 describeBy(data1["mpg"], data1$cyl)

Exercise 4: Use describeBy() function to produce summary statistics for mpg variable grouped by
am and cyl. Use mat=TRUE to display results in the form of a matrix. You should use list to
include more than 1 grouping variable.

9. Apply a Function to a Data Frame Split by Factors


The by( ) function applies a function to each level of a factor or factors of a data frame.
The basic format of this function is as follows:
by(data, a factor or a list of factors, a function to be applied, ...,
simplify = TRUE)

3|P a ge
STA2 - R LAB 1

For instance, if we want to obtain a summary on mpg variable separately for each number of
cylinders in data1 data frame, we can use the following code:
 by(data1$mpg,data1$cyl,summary)
 by(data1$mpg,list(Cylinder=data1$cyl),summary)

If we want to produce a summary for mpg variable for each level of both am and cyl variables, we
can type in R the following code:
 by(data1$mpg,list(data1$am, data1$cyl),summary)
 by(data1$mpg,list(AM=data1$am, Cyl=data1$cyl),summary)

Exercise 5. Load the textbooks data from the openintro package. Examine a summary of the price
differences between new textbook sold at the UCLA bookstore and on Amazon for two subsets:
courses where one book is required and where more than one book is required.
 [Link]("openintro”)
 library(openintro)
 data(textbooks)
10. Recap questions
a. What does the assignment operator do?
b. What are the main data types?
c. What function is used to check for data type of an object?
d. Print the data1 dataset. Find the sample mean and standard deviation of the weight
variable. Display the five-number summary for the weight variable. Provide a summary
for hp variable grouped by cyl variable using the functions describeBy() and by().

4|P a ge

Common questions

Powered by AI

Summary statistics for specific variables in an R data frame can be retrieved using functions like summary(), which provides a basic overview including mean and quartiles, and stat.desc() from the pastecs package for more detailed statistics . The describeBy() function allows grouped summary stats, particularly useful for comparisons across categories . Compared to summary(), describeBy() offers more granularity by allowing grouping variables to produce statistics by levels, which is not directly available in summary().

The 'by()' function in R applies a function, like summary(), to each level of factors within a data frame, enabling separate summaries for each group, such as mpg by cylinders or transmission types . It allows for multi-factor grouping and gives detailed summaries. In contrast, 'describeBy()' from the psych package provides more comprehensive statistics grouped by factors, offering options for matrix form output which is particularly handy for broader statistical insights across multiple grouping variables with less manual coding. Both are robust for group-wise analysis, but 'describeBy()' offers more detailed statistical insights .

The 'by' function in R is used to apply a function to each level of a factor or list of factors within a data frame, allowing users to perform separate analyses for each group. This is particularly useful for obtaining summaries or statistics on subsets of data grouped by one or more factors, such as cylinders or transmission in a car dataset, and provides a way to compare data metrics across different groups .

Data frames in R enhance the flexibility of data analysis over matrices by allowing columns to have different data types, such as numeric, character, and logical, whereas matrices require consistent data types across all entries . Typical statistical operations on data frames include computing means and standard deviations (e.g., mean(data1$mpg) and sd(data1$mpg)), accessing descriptive statistics (e.g., summary(data1$mpg)), and plotting variables (e.g., plot(data1$mpg, data1$wt)). This flexibility facilitates more complex analyses similar to dataset handling in tools like SPSS or SAS .

To generate a summary for multiple variables simultaneously in R, you can use the summary() function on a subset of the data frame containing the desired variables, such as summary(data1[c("mpg", "drat", "wt"]). This approach benefits data analysis by providing a quick overview of key statistics across several variables at once, allowing for efficient preliminary analyses and comparisons across different measurements, which aids in identifying patterns or anomalies .

R's case sensitivity means that variable names like 'x' and 'X' are considered different. This requires programmers to be consistent with variable naming to avoid logical errors, as a reference to a variable that differs by case will not point to the intended object . In data frames, careful attention must be paid to consistency in referring to column names or variables to ensure operations are performed on the correct data, which is essential for accurate data type consistency checks and data manipulation .

The assignment operator "<-" in R is used to assign values to variables, meaning it stores the result of an expression on the right side into the object specified on the left side . The implications for case sensitivity in R mean that variable names are case-sensitive, thus 'x' and 'X' would be considered different variables . This requires careful naming of variables to avoid conflicts and errors.

To import data from a CSV file into R, one must save the Excel data as a CSV file and then use the read.table() or read.csv() function to import it into R, ensuring the header is TRUE to recognize column names, and setting the appropriate separator and quote characters . Once the data is imported, checking the structure with str() is crucial to verify the correct format, number of variables, and observations .

Managing the working directory in R involves setting the location from which files are read and to which results are saved using setwd() and verifying it with getwd(). It is crucial to set the working directory before running code because it ensures that file paths are correct and relevant input/output operations are conducted smoothly. This prevents errors related to file not found and ensures that data and results are accessed and saved in the intended location, streamlining the workflow .

The describeBy() function in R can be used to generate grouped summary statistics by using it with a list of grouping variables, such as 'am' and 'cyl', to see how a variable like 'mpg' varies across combinations of these factors . Using mat=TRUE displays results in matrix form, which provides a clear and organized view of statistics for each group combination, making comparisons more straightforward and visually digestible .

You might also like