Combining and Reshaping Data

Welcome back! You've been doing an excellent job mastering data manipulation using the dplyr package. So far, we've covered selecting, filtering, summarizing, and mutating data. Now, it's time to step up your skills further by combining and reshaping datasets. This lesson will give you powerful tools to handle complex data structures more efficiently.

We will dive into two new essential techniques:

1. Combining Data: Merging multiple data frames into one cohesive dataset using the bind_rows function.
2. Reshaping Data: Changing the layout of your data for easier analysis using functions like gather from the tidyr package.

Let's look at these techniques step-by-step.

First, let's create example data frames that we'll use for combining and reshaping.

# Example data frames
df1 <- data.frame(ID = 1:2, Name = c("Alice", "Bob"))
df2 <- data.frame(ID = 3:4, Name = c("Charlie", "David"))

# Print the example data frames
print(df1)
print(df2)

# Output of df1:
#   ID  Name
# 1  1 Alice
# 2  2   Bob

# Output of df2:
#   ID    Name
# 1  3 Charlie
# 2  4   David

Here, we have two data frames df1 and df2 with a column for ID and Name.

We can combine these two data frames into one using the bind_rows function from dplyr.

# Example data frames
df1 <- data.frame(ID = 1:2, Name = c("Alice", "Bob"))
df2 <- data.frame(ID = 3:4, Name = c("Charlie", "David"))

# Combining data frames by rows
combined_data <- bind_rows(df1, df2)

# Print the combined data frame
print(combined_data)

# Output of combined_data:
#   ID    Name
# 1  1   Alice
# 2  2     Bob
# 3  3 Charlie
# 4  4   David

The bind_rows function stacks the rows of df1 and df2 to create a new data frame combined_data.

Now, let's reshape our combined data frame using the gather function from the tidyr package. The gather function converts wide-format data to long-format data. In our case, it will transform the data such that each row represents a single observation.

# Example data frames
df1 <- data.frame(ID = 1:2, Name = c("Alice", "Bob"))
df2 <- data.frame(ID = 3:4, Name = c("Charlie", "David"))

# Combining data frames by rows
combined_data <- bind_rows(df1, df2)

# Reshape data using gather
gathered_data <- gather(combined_data, key = "Variable", value = "Value", -ID)

# Display the reshaped data
print(gathered_data)

# Output of gathered_data:
#   ID Variable   Value
# 1  1     Name   Alice
# 2  2     Name     Bob
# 3  3     Name Charlie
# 4  4     Name   David

Here, gathered_data will be the reshaped version of combined_data, where the columns (except ID) are gathered into key-value pairs.

The gather function changes data from a wide format to a long format.

• Wide Format: Each variable is in a separate column. For example, in combined_data, we have the columns ID and Name, where each row represents a unique ID and its associated name.
• Long Format: Each observation is in a separate row. The gather function consolidates multiple columns into key-value pairs, making it easier to perform certain types of analysis.

In gather:

• key becomes the new column name that holds the original column names.
• value becomes the column name for the values.
• -column_name tells gather to exclude the column_name column from gathering.