Data element

alpha <- 1 # this is data element

alpha  # If we call alpha, the computer will display the value assigned to alpha.

## [1] 1

Vector

To assign vector in R, we use c(), with comma , as separators. For example:

beta <- c(1, 2) # this is a vector

beta

## [1] 1 2

gamma <- c(alpha, beta)  # This is also a vector

gamma

## [1] 1 1 2

Remember vectors are ordered.

delta <- c(beta, alpha) # it is still a vector

delta

## [1] 1 2 1

When you call beta in R, it will retrieve data associated with beta. So, how do we assign characters (letters or categorical variable)? R recognizes character values with quotation marks, "". For instance:

delta_2 <- c("beta", "alpha") 

delta_2

## [1] "beta"  "alpha"

With quotation marks, the R will recognize inputs as a new character data, not the data we assigned previously.

Selecting variables (vectors) in data frame

A data frame has columns, where each column represents one variable (vector).

We use the dollar sign $ to select a specific variable from a data frame.

epsilon$gamma

## [1] 1 1 2

epsilon$delta

## [1] 1 2 1

epsilon$gamma returns the exact same vector as just gamma, which was used for constructing epsilon.

Now, how can we generate a larger dataset for analysis?

Data frame example

We can make data by adding multiple data elements to vectors, and we can bind those vectors to make one dataframe.

subject <- c("Joe", "Trump", "Obama", "George")  # Assigning multiple data elements as a vector
height <- c(183, 190, 187, 182)
IsTall <- c("short", "tall", "tall", "short")

example_dataframe <- data.frame(subject, height, IsTall)  # Combining 3 vectors into one data frame

example_dataframe  # Displaying the example_data frame

Loading Data

However, we won’t be typing all the data manually every time it’s needed. To save time, we will directly load data frames from our computer using the read.csv() function for .csv files. Excel .xlsx files need another function, such as readxl::read_excel().

Here is an example using systolic blood pressure (SBP) data from NHISS.

dataset_sbp <- read.csv(file = "dataset/sbp_dataset_korea_2013-2014.csv")


reactable::reactable(head(dataset_sbp, 50), sortable = TRUE)

Here, dataset/ is the folder inside our R project. Open BTE3207_Advanced_Biostatistics.Rproj first, then open this document. The path is relative to that project folder. The table above displays the first 50 rows.

When working with data in R, it’s often necessary to load data from your computer instead of typing everything manually. The most common file formats for data are .csv (Comma-Separated Values) and .xlsx (Excel files).

To load a .csv file, we use the read.csv() function. However, to load the file correctly, you need to know the file path, which tells R where to find the file on your computer.

Meanwhile, actually you can download the files at your current working directory with a valid URL for the file as well.

This is optional. The file is already in our dataset/ folder. Run this chunk manually only if you need another copy.

url_link <- "https://raw.githubusercontent.com/minsiksudo/BTE3207_Advanced_Biostatistics/refs/heads/main/dataset/sbp_dataset_korea_2013-2014.csv"

download.file(url = url_link, 
              destfile = "sbp_dataset_korea_2013-2014.csv") # the name that you want to store the linked file at your computer. 

list.files() # This shows all the files in your current working directory

dataset_sbp <- read.csv("sbp_dataset_korea_2013-2014.csv") # as the file is in your current folder, you don't need to specify the path.

#setwd("/Users/username/Documents")  # For Mac/Linux - this is example file path
#setwd("C:/Users/username/Documents")  # For Windows - this is example file path

getwd()

## [1] "/Volumes/macdrive/Dropbox/10_Inha/50_Lectures/10_BTE3207_Biostats/2026F/scripts/BTE3207_Advanced_Biostatistics"

# Example for Mac/Linux
#dataset_sbp <- read.csv(file = "/Users/username/Documents/sbp_dataset.csv")

# Example for Windows
#dataset_sbp <- read.csv(file = "C:/Users/username/Documents/sbp_dataset.csv")

#dataset_sbp <- read.csv(file = file.choose())

Mean Calculation

The function for calculating the mean is mean().

mean(dataset_sbp$SBP)

## [1] 121.8718

This is the mean value of all the SBP data (1,000,000 rows).

SD Calculation

The function for calculating the standard deviation is sd().

sd(dataset_sbp$SBP)

## [1] 14.56171

This is the sample standard deviation. R uses n - 1 in the sample variance calculation.

x <- c(110, 120, 130)
sqrt(sum((x - mean(x))^2) / (length(x) - 1))

## [1] 10

sd(x)

## [1] 10

Median

The function for calculating the median is median().

median(dataset_sbp$SBP)

## [1] 120

Function for Calculating Quantiles: quantile()

The quantile() function can be used to calculate multiple percentiles at once. Without specifying any percentile, it will return the 0%, 25%, 50%, 75%, and 100% percentiles.

quantile(dataset_sbp$SBP)

##   0%  25%  50%  75% 100% 
##   82  110  120  130  190

Since the output of quantile() contains multiple data elements, the result is a vector. Conveniently, this output also includes names. To check the structure of this output, we can use the str() function.

str(quantile(dataset_sbp$SBP))

##  Named num [1:5] 82 110 120 130 190
##  - attr(*, "names")= chr [1:5] "0%" "25%" "50%" "75%" ...

The str() function indicates that it is a named numeric variable.

Quantile() - Continued

But what if we want to find the value at a specific percentile? In that case, we need to provide more information to the function.

These pieces of information are called arguments.

Since functions in R can accept multiple inputs, we separate them with commas (,).

For example, to calculate the 10th percentile of the SBP dataset:

quantile(x = dataset_sbp$SBP, probs = 0.1)

## 10% 
## 103