9 Cleaning and Wrangling Data

Learning Objectives

• Introduce dplyr and tidyr functions to clean and wrangle data for analysis
• Learn about the Split-Apply-Combine strategy and how it applies to data wrangling
• Describe the difference between wide vs. long table formats and how to convert between them

9.1 Introduction

The data we get to work with are rarely, if ever, in the format we need to do our analyses. It’s often the case that one package requires data in one format, while another package requires the data to be in another format. To be efficient analysts, we should have good tools for reformatting data for our needs so we can do further work like making plots and fitting models. The dplyr and tidyr R packages provide a fairly complete and extremely powerful set of functions for us to do this reformatting quickly. Learning these tools well will greatly increase your efficiency as an analyst.

Let’s look at two motivating examples.

Example 1

Suppose you have the following data.frame called length_data with data about salmon length and want to calculate the average length per year.

year	length_cm
1990	5.673318
1991	3.081224
1991	4.592696
1992	4.381523
1992	5.597777
1992	4.900052

The dplyr R library provides a fast and powerful way to do this calculation in a few lines of code:

length_data %>% 
  group_by(year) %>% 
  summarize(mean_length_cm = mean(length_cm))

Example 2

Another process we often need to do is to “reshape” our data. Consider the following table that is in what we call “wide” format:

site	1990	1991	…	1993
gold	100	118	…	112
lake	100	118	…	112
…	…	…	…	…
dredge	100	118	…	112

You are probably familiar with data in the above format, where values of the variable being observed are spread out across columns. In this example we have a different column per year. This wide format works well for data entry and sometimes works well for analysis but we quickly outgrow it when using R (and know it is not tidy data!). For example, how would you fit a model with year as a predictor variable? In an ideal world, we’d be able to just run lm(length ~ year). But this won’t work on our wide data because lm() needs length and year to be columns in our table.

The tidyr package allows us to quickly switch between wide format and long format using the pivot_longer() function:

site_data %>% 
  pivot_longer(-site, names_to = "year", values_to = "length")

site	year	length
gold	1990	101
lake	1990	104
dredge	1990	144
…	…	…
dredge	1993	145

This lesson will cover examples to learn about the functions you’ll most commonly use from the dplyr and tidyr packages:

Common dplyr functions

Function name	Description
mutate()	Creates modify and deletes columns
group_by()	Groups data by one or more variables
summarise()	Summaries each group down to one row
select()	Keep or drop columns using their names
filter()	Keeps rows that matches conditions
arrange()	order rows using columns variable
rename()	Rename a column

Common tidyr functions

Function name	Description
pivot_longer()	transforms data from a wide to a long format
pivot_wider()	transforms data from a long to a wide format
unite()	Unite multiple columns into one by pasting strings together
separate()	Separate a character column into multiple columns with a regular expression or numeric locations

9.2 Data cleaning basics

To demonstrate, we’ll be working with a tidied up version of a data set from Alaska Department of Fish & Game containing commercial catch data from 1878-1997. The data set and reference to the original source can be found at its public archive.

Setup

First, open a new Quarto document. Delete everything below the setup chunk, and add a library chunk that calls dplyr, tidyr, and readr

library(dplyr)
library(tidyr)
library(readr)

A note on loading packages

You may have noticed the following messages pop up when you ran your library chunk.

Attaching package: ‘dplyr’

The following objects are masked from ‘package:stats’:

    filter, lag

The following objects are masked from ‘package:base’:

    intersect, setdiff, setequal, union

These are important messages. They are letting you know that certain functions from the stats and base packages (which are loaded by default when you start R) are masked by different functions with the same name in the dplyr package. It turns out, the order that you load the packages in matters. Since we loaded dplyr after stats, R will assume that if you call filter(), you mean the dplyr version unless you specify otherwise.

Being specific about which version of filter(), for example, you call is easy. To explicitly call a function by its unambiguous name, we use the syntax package_name::function_name(...). So, if we wanted to call the stats version of filter() in this Rmarkdown document, I would use the syntax stats::filter(...).

Remove messages and warnings

Messages and warnings are important, but we might not want them in our final document. After you have read the packages in, adjust the chunk settings in your library chunk to suppress warnings and messages by adding #| message: false or #| warning: false. Both of these chunk options, when set to false, prevents messages or warnings from appearing in the rendered file.

Now that we have introduced some data wrangling libraries, let’s get the data that we are going to use for this lesson.

Setup

1. Go to KNB Data Package Alaska commercial salmon catches by management region (1886- 1997)

2. Find the data file byerlySalmonByRegion.csv. Right click the “Download” button and select “Copy Link Address”

3. Paste the copied URL into the read_csv() function

The code chunk you use to read in the data should look something like this:

catch_original <- read_csv("https://knb.ecoinformatics.org/knb/d1/mn/v2/object/df35b.302.1")

Note for Windows users: Keep in mind, if you want to replicate this workflow in your local computer you also need to use the url() function here with the argument method = "libcurl".

It would look like this:

catch_original <- read.csv(url("https://knb.ecoinformatics.org/knb/d1/mn/v2/object/df35b.302.1", method = "libcurl"))

This data set is relatively clean and easy to interpret as-is. While it may be clean, it’s in a shape that makes it hard to use for some types of analyses so we’ll want to fix that first.

Exercise

Before we get too much further, spend a minute or two outlining your Quarto document so that it includes the following sections and steps:

• Data Sources
  • Read in the data
  • Explore data
• Clean and Reshape data
  • Using select() function
  • Check column types
  • Replace values in a column
  • Reshape data
  • Rename columns rename()
  • Summary stats using group_by() and summarize()
  • Filtering rows using filter()
  • Sort data using arrange()
  • Split and combine values in columns

9.3 Data exploration

Similar to what we did in our Intro to Literate Analysis lesson, it is good practice to skim through the data you just read in.

Doing so is important to make sure the data is read as you were expecting and to familiarize yourself with the data.

Some of the basic ways to explore your data are:

## Prints the column names of my data frame
colnames(catch_original)

## First 6 lines of the data frame
head(catch_original)

## Summary of each column of data
summary(catch_original)

## Prints unique values in a column (in this case, the region)
unique(catch_original$Region)

## Opens data frame in its own tab to see each row and column of the data (do in console)
View(catch_original)

9.4 About the pipe (%>%) operator

Before we jump into learning tidyr and dplyr, we first need to explain the pipeline operator %>%.

Both the tidyr and the dplyr packages use the pipe operator (%>%), which may look unfamiliar. The pipe is a powerful way to efficiently chain together operations. The pipe will take the output of a previous statement, and use it as the input to the next statement.

Say you want to both filter() out rows of a data set, and select() certain columns.

Instead of writing:

df_filtered <- filter(df, ...)
df_selected <- select(df_filtered, ...)

You can write:

df_cleaned <- df %>% 
    filter(...) %>%
    select(...)

If you think of the assignment operator (<-) as reading like “gets”, then the pipe operator would read like “then”.

So you might think of the above chunk being translated as:

The cleaned data frame gets the original data, and then a filter (of the original data), and then a select (of the filtered data).

The benefits to using pipes are that you don’t have to keep track of (or overwrite) intermediate data frames. The drawbacks are that it can be more difficult to explain the reasoning behind each step, especially when many operations are chained together. It is good to strike a balance between writing efficient code (chaining operations), while ensuring that you are still clearly explaining, both to your future self and others, what you are doing and why you are doing it.

Quick Tip

RStudio has a keyboard shortcut for %>%

• Windows: Ctrl + Shift + M
• Mac: cmd + shift + M

9.5 Selecting or removing columns using select()

We’re ready to go back to our salmon dataset. The first issue is the extra columns All and notesRegCode. Let’s select only the columns we want, and assign this to a variable called catch_data.

catch_data <- catch_original %>%
    select(Region, Year, Chinook, Sockeye, Coho, Pink, Chum)

head(catch_data)

Much better!

The select() function also allows you to say which columns you don’t want, by passing unquoted column names preceded by minus (-) signs:

catch_data <- catch_original %>%
    select(-All,-notesRegCode)

9.6 Quality check

Now that we have the data we are interested in using, we should do a little quality check to see that everything seems as expected. One nice way of doing this is the glimpse() function.

dplyr::glimpse(catch_data)

Exercise

Examine the output of the glimpse() function call. Does anything seem amiss with this data set that might warrant fixing?

Answer:

The Chinook catch data are character class. Let’s fix it using the function mutate() before moving on.

9.7 Changing column content using mutate()

We can use the mutate() function to change a column, or to create a new column. First, let’s try to convert the Chinook catch values to numeric type using the as.numeric() function, and overwrite the old Chinook column.

catch_clean <- catch_data %>%
    mutate(Chinook = as.numeric(Chinook))

head(catch_clean)

We get a warning "NAs introduced by coercion" which is R telling us that it couldn’t convert every value to an integer and, for those values it couldn’t convert, it put an NA in its place. This is behavior we commonly experience when cleaning data sets and it’s important to have the skills to deal with it when it comes up.

To investigate, let’s isolate the issue. We can find out which values are NAs with a combination of is.na() and which(), and save that to a variable called i.

i <- which(is.na(catch_clean$Chinook))
i

It looks like there is only one problem row, lets have a look at it in the original data.

catch_data[i,]

Well that’s odd: The value in catch_thousands is the letter I. It turns out that this data set is from a PDF which was automatically converted into a csv and this value of I is actually a 1.

Let’s fix it by incorporating the if_else() function to our mutate() call, which will change the value of the Chinook column to 1 if the value is equal to I, then will use as.numeric() to turn the character representations of numbers into numeric typed values.

catch_clean <- catch_data %>%
    mutate(Chinook = if_else(condition = Chinook == "I", 
                             true = "1", 
                             false = Chinook),
           Chinook = as.numeric(Chinook))

##check
catch_clean[i, ]

9.8 Changing shape using pivot_longer() and pivot_wider()

The next issue is that the data are in a wide format and we want the data in a long format instead. The function pivot_longer() from the tidyr package helps us do this conversion. If you do not remember all the arguments that go into pivot_longer() you can always call the help page by typing ?pivot_longer in the console.

catch_long <- catch_clean %>% 
    #pivot longer all columns except Region and Year
    pivot_longer(
        cols = -c(Region, Year),
        names_to = "species",
        values_to = "catch"
    )

head(catch_long)

The syntax we used above for pivot_longer() might be a bit confusing so let’s walk though it.

• The first argument to pivot_longer is the columns over which we are pivoting. You can select these by listing either the names of the columns you do want to pivot, or in this case, the names of the columns you are not pivoting over.

• The names_to argument: this is the name of the column that you are creating from the column names of the columns you are pivoting over.

• The values_to argument: the name of the column that you are creating from the values in the columns you are pivoting over.

The opposite of pivot_longer() is the pivot_wider() function. It works in a similar declarative fashion:

catch_wide <- catch_long %>%
    pivot_wider(names_from = species,
                values_from = catch)

head(catch_wide)

Same than we did above we can pull up the documentation of the function to remind ourselves what goes in which argument. Type ?pivot_wider in the console.

9.9 Renaming columns with rename()

If you scan through the data, you may notice the values in the catch column are very small (these are supposed to be annual catches). If we look at the metadata we can see that the catch column is in thousands of fish, so let’s convert it before moving on.

Let’s first rename the catch column to be called catch_thousands:

catch_long <- catch_long %>%
    rename(catch_thousands = catch)

head(catch_long)

names() versus rename()

Many people use the base R function names() to rename columns, often in combination with column indexing that relies on columns being in a particular order. Column indexing is often also used to select columns instead of the select() function from dplyr. Although these methods work just fine, they do have one major drawback: in most implementations they rely on you knowing exactly the column order your data is in.

To illustrate why your knowledge of column order isn’t reliable enough for these operations, considering the following scenario:

Your colleague emails you letting you know that she has an updated version of the conductivity-temperature-depth data from this year’s research cruise, and sends it along. Excited, you re-run your scripts that use this data for your phytoplankton research. You run the script and suddenly all of your numbers seem off. You spend hours trying to figure out what is going on.

Unbeknownst to you, your colleagues bought a new sensor this year that measures dissolved oxygen. Because of the new variables in the data set, the column order is different. Your script which previously renamed the fourth column, SAL_PSU to salinity now renames the fourth column, O2_MGpL to salinity. No wonder your results looked so weird, good thing you caught it!

If you had written your code so that it doesn’t rely on column order, but instead renames columns using the rename() function, the code would have run just fine (assuming the name of the original salinity column didn’t change, in which case the code would have thrown an error in an obvious way). This is an example of a defensive coding strategy, where you try to anticipate issues before they arise, and write your code in such a way as to keep the issues from happening.

9.10 Adding columns using mutate()

Now let’s use mutate() again to create a new column called catch with units of fish (instead of thousands of fish).

catch_long <- catch_long %>%
    mutate(catch = catch_thousands * 1000)

head(catch_long)

Let’s remove the catch_thousands column for now since we don’t need it. Note that here we have added to the expression we wrote above by adding another function call (mutate) to our expression. This takes advantage of the pipe operator by grouping together a similar set of statements, which all aim to clean up the catch_clean data frame.

catch_long <- catch_long %>%
    mutate(catch = catch_thousands * 1000) %>%
    select(-catch_thousands)

head(catch_long)

We’re now ready to start analyzing the data.

9.11 Summary statistics using group_by() and summarize()

Suppose we are now interested in getting the average catch per region. In our initial data exploration we saw there are 18 regions, we can easily see their names again:

unique(catch_original$Region)

Think about how we would calculate the average catch per region “by hand”. It would be something like this:

0. We start with our table and notice there are multiple regions in the “Regions” column.

1. We split our original table to group all observations from the same region together.

2. We calculate the average catch for each of the groups we form.

3. Then we combine the values for average catch per region into a single table.

Analyses like this conform to what is known as the Split-Apply-Combine strategy. This strategy follows the three steps we explained above:

1. Split: Split the data into logical groups (e.g., region, species, etc.)
2. Apply: Calculate some summary statistic on each group (e.g. mean catch by year, number of individuals per species)
3. Combine: Combine the statistic calculated on each group back together into a single table

The dplyr library lets us easily employ the Split-Apply-Combine strategy by using the group_by() and summarize() functions:

mean_region <- catch_long %>%
    group_by(Region) %>%
    summarize(mean_catch = mean(catch))

head(mean_region)

Let’s see how the previous code implements the Split-Apply-Combine strategy:

1. group_by(Region): this is telling R to split the dataframe and create a group for each different value in the column Region. R just keeps track of the groups, it doesn’t return separate dataframes per region.

2. mean(catch): here mean is the function we want to apply to the column catch in each group.

3. summarize(catch = mean(catch)) the function summarize() is used to combine the results of mean(catch) in each group into a single table. The argument mean_catch = mean(catch) indicates that the column having the results of mean(catch) will be named mean_catch.

Another common use of group_by() followed by summarize() is to count the number of rows in each group. We have to use a special function from dplyr, n().

n_region <- catch_long %>%
    group_by(Region) %>%
    summarize(n = n())

head(n_region)

Try using count()

If you are finding that you are reaching for this combination of group_by(), summarize() and n() a lot, there is a helpful dplyr function count() that accomplishes this in one function!

Exercise

• Find another grouping and statistic to calculate for each group.
• Find out if you can group by multiple variables.

## for example:
catch_year_sp <- catch_long %>%
    group_by(Year, species) %>%
    summarize(total_year = sum(catch, na.rm = T))

9.12 Filtering rows using filter()

We use the filter() function to filter our data.frame to rows matching some condition. It’s similar to subset() from base R.

Let’s go back to our original data.frame and do some filter()ing:

sse_catch <- catch_long %>%
    filter(Region == "SSE")

head(sse_catch)

Exercise

• Filter to just catches of over one million fish
• Filter to just Chinook from the SSE region

## Catches over a million fish
catch_million <- catch_long %>%
    filter(catch > 1000000)

## Chinook from SSE data
chinook_see <- catch_long %>%
    filter(Region == "SSE",
           species == "Chinook")

## OR
chinook_see <- catch_long %>%
    filter(Region == "SSE" & species == "Chinook")

9.13 Sorting your data using arrange()

The arrange() function is used to sort the rows of a data.frame. Two common cases to use arrange() are:

• To calculate a cumulative sum (with cumsum()) so row order matters
• To display a table (like in an .qmd document) in sorted order

Let’s re-calculate mean catch by region, and then arrange() the output by mean catch:

mean_region <- catch_long %>%
    group_by(Region) %>%
    summarize(mean_catch = mean(catch)) %>%
    arrange(mean_catch)

head(mean_region)

The default sorting order of arrange() is to sort in ascending order. To reverse the sort order, wrap the column name inside the desc() function:

mean_region <- catch_long %>%
    group_by(Region) %>%
    summarize(mean_catch = mean(catch)) %>%
    arrange(desc(mean_catch))

head(mean_region)

9.14 Splitting a column using separate() and unite()

The separate() function allow us to easily split a single column into numerous. Its complement, the unite() function, allows us to combine multiple columns into a single one.

This can come in really handy when we need to split a column into two pieces by a consistent separator (like a dash).

Let’s make a new data.frame with fake data to illustrate this. Here we have a set of site identification codes with information about the island where the site is (the first 3 letters) and a site number (the 3 numbers). If we want to group and summarize by island, we need a column with just the island information.

sites_df <- data.frame(site = c("HAW-101",
                                "HAW-103",
                                "OAH-320",
                                "OAH-219",
                                "MAU-039"))

sites_df %>%
    separate(site, c("island", "site_number"), "-")

Exercise

Split the city column in the data frame cities_df into city and state_code columns

## create `cities_df`
cities_df <- data.frame(city = c("Juneau AK",
                                 "Sitka AK",
                                 "Anchorage AK"))

colnames(cities_df)

cities_clean <- cities_df %>%
    separate(city, c("city", "state_code"), " ")

The unite() function does just the reverse of separate(). If we have a data.frame that contains columns for year, month, and day, we might want to unite these into a single date column.

dates_df <- data.frame(
    year = c("1930",
             "1930",
             "1930"),
    month = c("12",
              "12",
              "12"),
    day = c("14",
            "15",
            "16")
)

dates_df %>%
    unite(date, year, month, day, sep = "-")

9.15 Now, all together!

We just ran through the various things we can do with dplyr and tidyr but if you’re wondering how this might look in a real analysis. Let’s look at that now:

catch_original <- read_csv(url("https://knb.ecoinformatics.org/knb/d1/mn/v2/object/df35b.302.1", 
                               method = "libcurl"))

mean_region <- catch_original %>%
  select(-All, -notesRegCode) %>% 
  mutate(Chinook = ifelse(Chinook == "I", 1, Chinook)) %>% 
  mutate(Chinook = as.numeric(Chinook)) %>% 
  pivot_longer(-c(Region, Year), 
               names_to = "species", 
               values_to = "catch") %>%
  mutate(catch = catch*1000) %>% 
  group_by(Region) %>% 
  summarize(mean_catch = mean(catch)) %>% 
  arrange(desc(mean_catch))

head(mean_region)

We have completed our lesson on Cleaning and Wrangling data. Before we break, let’s practice our Git workflow.

Steps

1. Save the .qmd you have been working on for this lesson.

2. Render the Quarto file. This is a way to test everything in your code is working.

3. Stage (Add) > Commit > Pull > Push