knitr::opts_chunk$set(echo = TRUE)Data manipulation consists of modifying data to make it easier to read and more organized. We manipulate data for analysis and visualization.
There are different ways to manipulate data in R, and we will use a
couple of functions in the package dplyr. We are going to
look at filtering data, grouping data, and summarizing data.
In this lab, you will learn the following:
Data for NYC film permits used in this lab instructions was obtained from the NYC Open Data website. The Film_Permits.csv file for this lab is already conveniently in the lab’s folder in your project’s directory on Cocalc.com. But you can also find it here: Film_Permits.csv
NYC makes data about many things open and free for anyone to download and look at. This is the NYC Open Data website: https://opendata.cityofnewyork.us.
We also are going to use a dataset called us-cities-demographics.csv. This dataset contains information about the demographics of all US cities and census-designated places with a population greater or equal to 65,000. This data comes from the US Census Bureau’s 2015 American Community Survey. The data file is available here: https://public.opendatasoft.com/explore/dataset/us-cities-demographics/information/
There are numerous resources for learning about R; you will find some available on this webpage: resouces page.
It is not required for any of the MA336 Labs, but if you are interested in downloading and installing R and R Studio at home to your computer (it’s free), you can check how to do so here: general introduction to R and Rstudio.
To create a frequency table, graph data, and calculate the mean and standard deviation, we need to have some data first.
With R, that’s not entirely true. The default installation of R comes with several data sets.
Here are some of the pre-loaded data sets in R:
iris -> Edgar Anderson’s iris datawomen -> Average heights and weights for American
womenpressure -> Vapor pressure of mercury as a function
of temperaturemtcars -> Motor trend car road testscars -> Speed and stopping distances of carsLet’s look at the data set called women.
women## height weight
## 1 58 115
## 2 59 117
## 3 60 120
## 4 61 123
## 5 62 126
## 6 63 129
## 7 64 132
## 8 65 135
## 9 66 139
## 10 67 142
## 11 68 146
## 12 69 150
## 13 70 154
## 14 71 159
## 15 72 164Because R is a language built for statistics, it contains many functions that allow you generate random data.
rnorm(20, mean=8, sd=1.5)## [1] 8.991169 10.670567 9.612825 7.280626 9.947510 8.744634 10.557744
## [8] 5.670601 5.130413 8.570563 8.751443 9.159175 6.652833 9.058252
## [15] 5.912634 11.352714 7.455835 7.833809 9.370763 7.091717With the code above, you just made R to sample 20 numbers with a normal distribution with a mean of 8 and a standard deviation of 1.5 (later in the course, we will cover this type of distribution).
In future labs, we will look at how to use R to generate samples of numbers according to other types of distributions.
Let’s do something that might be more interesting for now. We will focus on grabbing data from a local file or from a URL.
Suppose we would like to know what kind of film productions are being shot in NYC.
To answer this question, let’s use the data set in the file Film_Permits.csv.
Searching through the data, we can find a data file that lists the locations of film permits for shooting movies throughout NYC’s five boroughs. There are multiple ways to load/import this data into R.
Use the following commands to load (import) the data.
library(data.table)
nyc_films <-fread("Film_Permits.csv")
nyc_films## Event ID EventType StartDateTime EndDateTime
## 1: 446040 Shooting Permit 10/19/18 14:00 10/20/18 4:00
## 2: 446168 Shooting Permit 10/19/18 14:00 10/20/18 2:00
## 3: 186438 Shooting Permit 10/30/14 7:00 10/31/14 2:00
## 4: 445255 Shooting Permit 10/20/18 7:00 10/20/18 18:00
## 5: 128794 Theater Load in and Load Outs 11/16/13 0:01 11/17/13 6:00
## ---
## 70985: 533990 Shooting Permit 2/26/20 6:00 2/26/20 20:00
## 70986: 465761 Shooting Permit 2/27/19 7:30 2/27/19 20:30
## 70987: 511630 Shooting Permit 10/11/19 10:00 10/11/19 23:00
## 70988: 584697 Shooting Permit 7/4/21 6:00 7/4/21 23:59
## 70989: 491040 Shooting Permit 6/12/19 7:00 6/12/19 20:00
## EnteredOn EventAgency
## 1: 10/16/18 11:57 Mayor's Office of Film, Theatre & Broadcasting
## 2: 10/16/18 19:03 Mayor's Office of Film, Theatre & Broadcasting
## 3: 10/27/14 12:14 Mayor's Office of Film, Theatre & Broadcasting
## 4: 10/9/18 21:34 Mayor's Office of Film, Theatre & Broadcasting
## 5: 11/7/13 15:48 Mayor's Office of Film, Theatre & Broadcasting
## ---
## 70985: 2/21/20 14:40 Mayor's Office of Film, Theatre & Broadcasting
## 70986: 2/22/19 14:57 Mayor's Office of Film, Theatre & Broadcasting
## 70987: 10/8/19 14:05 Mayor's Office of Film, Theatre & Broadcasting
## 70988: 6/11/21 0:39 Mayor's Office of Film, Theatre & Broadcasting
## 70989: 6/10/19 12:38 Mayor's Office of Film, Theatre & Broadcasting
## ParkingHeld
## 1: THOMPSON STREET between PRINCE STREET and SPRING STREET, SPRING STREET between WOOSTER STREET and 6TH AVENUE, SPRING STREET between THOMPSON STREET and 6TH AVENUE, 6TH AVENUE between VANDAM STREET and BROOME STREET, SULLIVAN STREET between WEST HOUSTON STREET and PRINCE STREET, PRINCE STREET between SULLIVAN STREET and 6 AVENUE
## 2: MARBLE HILL AVENUE between WEST 227 STREET and WEST 225 STREET, WEST 228 STREET between ADRIAN AVENUE and MARBLE HILL AVENUE
## 3: LAUREL HILL BLVD between REVIEW AVENUE and RUST ST, REVIEW AVE between VAN DAM STREET and LAUREL HILL BOULEVARD, 59 ROAD between 60 LANE and 61 STREET, 59 ROAD between 60 LANE and 61 STREET, 61 STREET between 59 ROAD and FRESH POND ROAD, FRESH POND ROAD between 59 AVENUE and 59 DRIVE, 59 DRIVE between FRESH POND ROAD and 63 STREET, 59 DRIVE between FRESH POND ROAD and 64 STREET
## 4: JORALEMON STREET between BOERUM PLACE and COURT STREET
## 5: WEST 31 STREET between 7 AVENUE and 8 AVENUE, 8 AVENUE between WEST 31 STREET and WEST 33 STREET
## ---
## 70985: CENTER BOULEVARD between NORTH BASIN ROAD and 46 AVENUE, 21 STREET between 43 AVENUE and 44 AVENUE, 46 AVENUE between 5 STREET and VERNON BOULEVARD, 46 AVENUE between CENTER BOULEVARD and 5 STREET, 22 STREET between 44 AVENUE and 43 AVENUE, CENTER BOULEVARD between 46 AVENUE and 47 AVENUE, 5 STREET between NORTH BASIN ROAD and 46 AVENUE, 5 STREET between 46 AVENUE and 46 ROAD, 43 AVENUE between 21 STREET and 22 STREET, 44 AVENUE between 21 STREET and 23 STREET, 21 STREET between 44 AVENUE and 44 ROAD, 21 STREET between 44 ROAD and 44 DRIVE, 44 ROAD between 11 STREET and 21 STREET
## 70986: GREENWICH STREET between BARROW STREET and WEST 10 STREET, GREENWICH STREET between BARROW STREET and MORTON STREET, HUDSON STREET between BARROW STREET and GROVE STREET, HUDSON STREET between GROVE STREET and CHRISTOPHER STREET, CHRISTOPHER STREET between GREENWICH STREET and WASHINGTON STREET, BARROW STREET between HUDSON STREET and WASHINGTON STREET, HUDSON STREET between BARROW STREET and GROVE STREET, HUDSON STREET between GROVE STREET and CHRISTOPHER STREET
## 70987: 44 ROAD between 24 STREET and HUNTER STREET, 24 STREET between 44 ROAD and 43 AVENUE, JACKSON AVENUE between PEARSON STREET and COURT SQUARE, COURT SQUARE between JACKSON AVENUE and DEAD END, 43 AVENUE between CRESCENT STREET and 24 STREET, COURT SQUARE between JACKSON AVENUE and DEAD END, COURT SQUARE between JACKSON AVENUE and DEAD END, THOMPSON AVENUE between COURT SQUARE and 44 DRIVE, 44 DRIVE between JACKSON AVENUE and THOMPSON AVENUE, COURT SQUARE WEST between JACKSON AVENUE and DEAD END
## 70988: CENTER BOULEVARD between 54 AVENUE and 55 AVENUE, 55 AVENUE between CENTER BOULEVARD and 2 STREET, 2 STREET between 54 AVENUE and 55 AVENUE
## 70989: SPOFFORD AVENUE between COSTER STREET and FAILE STREET, HUNTS POINT AVENUE between SPOFFORD AVENUE and BRYANT AVENUE, HUNTS POINT AVENUE between BRYANT AVENUE and RANDALL AVENUE, WALNUT AVENUE between EAST 139 STREET and EAST 140 STREET, BRYANT AVENUE between HUNTS POINT AVENUE and RANDALL AVENUE, RANDALL AVENUE between BRYANT AVENUE and LONGFELLOW AVENUE, EAST 140 STREET between LOCUST AVENUE and WALNUT AVENUE, EAST 139 STREET between WALNUT AVENUE and LOCUST AVENUE
## Borough CommunityBoard(s) PolicePrecinct(s) Category
## 1: Manhattan 2 1 Television
## 2: Manhattan 12, 8 34, 50 Film
## 3: Queens 2, 5 104, 108 Television
## 4: Brooklyn 2 84 Still Photography
## 5: Manhattan 4, 5 14 Theater
## ---
## 70985: Queens 2 108 Television
## 70986: Manhattan 2 6 Television
## 70987: Queens 2 108 Television
## 70988: Queens 2 108 Television
## 70989: Bronx 1, 2, 7 40, 41, 72 Television
## SubCategoryName Country ZipCode(s)
## 1: Cable-episodic United States of America 10012
## 2: Feature United States of America 10034, 10463
## 3: Episodic series United States of America 11378
## 4: Not Applicable United States of America 11201
## 5: Theater United States of America 10001, 10121
## ---
## 70985: Episodic series United States of America 11101, 11109
## 70986: Episodic series United States of America 10014
## 70987: Episodic series United States of America 11101
## 70988: Special/Awards Show United States of America 11101
## 70989: Episodic series United States of America 10454, 10474, 11220If you have Internet access enabled in your project, you can also try loading the data from the source:
library(data.table)
nyc_films <- fread("https://data.cityofnewyork.us/api/views/tg4x-b46p/rows.csv?accessType=DOWNLOAD")Note: The code above will not be evaluated in R
because the option eval is set to FALSE. To
run it, you must change the value of eval to
TRUE.
The data above is stored in a data frame. A data frame
is a rectangular collection of variables (in the columns) and
observations (in the rows). It looks like an Excel spreadsheet if you
are familiar with Excel.
The variable nyc_films stores the data frame imported
from the file Film_Permits.csv.
It is possible to build a data frame from scratch. But most of the time, we will use some actual data in the labs to obtain our data frames from reading files.
Let’s look at how the data is structured. For example, to the size of
this data set (data frame), we can use the function
dim.
data_size <- dim(nyc_films)
data_size## [1] 70989 14It’s helpful to know about it if you need to look at the data frame itself. But this data frame is massive; it has 70989 rows and 14 columns of data. That’s a lot to look at.
We examine what is in a data frame in R using different functions.
Here are some examples: head(), tail(),
str(), nrow, ncol,
colnames(), and summary.
colnames(nyc_films)## [1] "Event ID" "EventType" "StartDateTime"
## [4] "EndDateTime" "EnteredOn" "EventAgency"
## [7] "ParkingHeld" "Borough" "CommunityBoard(s)"
## [10] "PolicePrecinct(s)" "Category" "SubCategoryName"
## [13] "Country" "ZipCode(s)"head(nyc_films)## Event ID EventType StartDateTime EndDateTime
## 1: 446040 Shooting Permit 10/19/18 14:00 10/20/18 4:00
## 2: 446168 Shooting Permit 10/19/18 14:00 10/20/18 2:00
## 3: 186438 Shooting Permit 10/30/14 7:00 10/31/14 2:00
## 4: 445255 Shooting Permit 10/20/18 7:00 10/20/18 18:00
## 5: 128794 Theater Load in and Load Outs 11/16/13 0:01 11/17/13 6:00
## 6: 43547 Shooting Permit 1/10/12 7:00 1/10/12 19:00
## EnteredOn EventAgency
## 1: 10/16/18 11:57 Mayor's Office of Film, Theatre & Broadcasting
## 2: 10/16/18 19:03 Mayor's Office of Film, Theatre & Broadcasting
## 3: 10/27/14 12:14 Mayor's Office of Film, Theatre & Broadcasting
## 4: 10/9/18 21:34 Mayor's Office of Film, Theatre & Broadcasting
## 5: 11/7/13 15:48 Mayor's Office of Film, Theatre & Broadcasting
## 6: 1/4/12 12:25 Mayor's Office of Film, Theatre & Broadcasting
## ParkingHeld
## 1: THOMPSON STREET between PRINCE STREET and SPRING STREET, SPRING STREET between WOOSTER STREET and 6TH AVENUE, SPRING STREET between THOMPSON STREET and 6TH AVENUE, 6TH AVENUE between VANDAM STREET and BROOME STREET, SULLIVAN STREET between WEST HOUSTON STREET and PRINCE STREET, PRINCE STREET between SULLIVAN STREET and 6 AVENUE
## 2: MARBLE HILL AVENUE between WEST 227 STREET and WEST 225 STREET, WEST 228 STREET between ADRIAN AVENUE and MARBLE HILL AVENUE
## 3: LAUREL HILL BLVD between REVIEW AVENUE and RUST ST, REVIEW AVE between VAN DAM STREET and LAUREL HILL BOULEVARD, 59 ROAD between 60 LANE and 61 STREET, 59 ROAD between 60 LANE and 61 STREET, 61 STREET between 59 ROAD and FRESH POND ROAD, FRESH POND ROAD between 59 AVENUE and 59 DRIVE, 59 DRIVE between FRESH POND ROAD and 63 STREET, 59 DRIVE between FRESH POND ROAD and 64 STREET
## 4: JORALEMON STREET between BOERUM PLACE and COURT STREET
## 5: WEST 31 STREET between 7 AVENUE and 8 AVENUE, 8 AVENUE between WEST 31 STREET and WEST 33 STREET
## 6: EAGLE STREET between FRANKLIN STREET and WEST STREET, WEST STREET between EAGLE STREET and FREEMAN STREET, FREEMAN STREET between WEST STREET and FRANKLIN STREET
## Borough CommunityBoard(s) PolicePrecinct(s) Category
## 1: Manhattan 2 1 Television
## 2: Manhattan 12, 8 34, 50 Film
## 3: Queens 2, 5 104, 108 Television
## 4: Brooklyn 2 84 Still Photography
## 5: Manhattan 4, 5 14 Theater
## 6: Brooklyn 1, 2 108, 94 Television
## SubCategoryName Country ZipCode(s)
## 1: Cable-episodic United States of America 10012
## 2: Feature United States of America 10034, 10463
## 3: Episodic series United States of America 11378
## 4: Not Applicable United States of America 11201
## 5: Theater United States of America 10001, 10121
## 6: Episodic series United States of America 11101, 11222str(nyc_films)## Classes 'data.table' and 'data.frame': 70989 obs. of 14 variables:
## $ Event ID : int 446040 446168 186438 445255 128794 43547 66846 104342 244863 446379 ...
## $ EventType : chr "Shooting Permit" "Shooting Permit" "Shooting Permit" "Shooting Permit" ...
## $ StartDateTime : chr "10/19/18 14:00" "10/19/18 14:00" "10/30/14 7:00" "10/20/18 7:00" ...
## $ EndDateTime : chr "10/20/18 4:00" "10/20/18 2:00" "10/31/14 2:00" "10/20/18 18:00" ...
## $ EnteredOn : chr "10/16/18 11:57" "10/16/18 19:03" "10/27/14 12:14" "10/9/18 21:34" ...
## $ EventAgency : chr "Mayor's Office of Film, Theatre & Broadcasting" "Mayor's Office of Film, Theatre & Broadcasting" "Mayor's Office of Film, Theatre & Broadcasting" "Mayor's Office of Film, Theatre & Broadcasting" ...
## $ ParkingHeld : chr "THOMPSON STREET between PRINCE STREET and SPRING STREET, SPRING STREET between WOOSTER STREET and 6TH AVENUE, "| __truncated__ "MARBLE HILL AVENUE between WEST 227 STREET and WEST 225 STREET, WEST 228 STREET between ADRIAN AVENUE and M"| __truncated__ "LAUREL HILL BLVD between REVIEW AVENUE and RUST ST, REVIEW AVE between VAN DAM STREET and LAUREL HILL BOULEVAR"| __truncated__ "JORALEMON STREET between BOERUM PLACE and COURT STREET" ...
## $ Borough : chr "Manhattan" "Manhattan" "Queens" "Brooklyn" ...
## $ CommunityBoard(s): chr "2" "12, 8" "2, 5" "2" ...
## $ PolicePrecinct(s): chr "1" "34, 50" "104, 108" "84" ...
## $ Category : chr "Television" "Film" "Television" "Still Photography" ...
## $ SubCategoryName : chr "Cable-episodic" "Feature" "Episodic series" "Not Applicable" ...
## $ Country : chr "United States of America" "United States of America" "United States of America" "United States of America" ...
## $ ZipCode(s) : chr "10012" "10034, 10463" "11378" "11201" ...
## - attr(*, ".internal.selfref")=<externalptr>Use the file us-cities-demographics.csv to code the following:
Import the file into R using the name demo for
storing the data.
Find out the size of the file and the columns’ names and read the top lines of the file.
Enter your code below:
If you are new to R and mainly from an Excel background, think more about column names than you usually might. Excel is very flexible when it comes to naming columns.
To ease this process, it is best to keep column names simple, without spaces, and without special characters (e.g. !, @, &, $, etc.).
But sometimes, we import a data set with spaces or other symbols in the name of the variables (columns).
Option 1: One way to deal with this is to invoke the
columns with spaces or special characters by surrounding their names
between single backticks: column_name.
head(nyc_films$`Event ID`)## [1] 446040 446168 186438 445255 128794 43547Note: the above code asks to display the top values
in the column Event ID in the data frame code
nyc_films. Note the use of the dollar sign $.
Option 2: Another way to do that is by renaming the columns without spaces or special characters.
colnames(nyc_films)[1]<-"Event_ID"
colnames(nyc_films)## [1] "Event_ID" "EventType" "StartDateTime"
## [4] "EndDateTime" "EnteredOn" "EventAgency"
## [7] "ParkingHeld" "Borough" "CommunityBoard(s)"
## [10] "PolicePrecinct(s)" "Category" "SubCategoryName"
## [13] "Country" "ZipCode(s)"Note: column 1 was renamed Event_ID in
the above code.
For the data frame corresponding to the file us-cities-demographics.csv, remove spaces from the variables Male Population, Female Population, Total Population, and State Code. Male Population, Female Population, Total Population, and State Code*.
Enter your code below:
The tidyverse package is designed for data science and
contains other packages that all work harmoniously.
The packages in tidyverse that we will cover in this and
future labs are dplyr and ggplot2. You can use
the function library() to load the whole tidyverse package
or the individual packages.
The package dplyr provides tools for the most common
data manipulation tasks. It is built to work directly with data
frames.
We are going to learn a couple of key dplyr functions
that allow you to deal with data manipulation challenges:
filter(): Pick observations (rows) by their
values.summarize(): Collapse many values into a single
summary.These functions can be used in conjunction with
group_by(), which changes each of the functions above from
operating on the entire dataset to working on it group by group.
Let’s dive in and see how these functions work.
The group_by() function splits the data into groups upon
which some operations can be run.
group_by() is often used together with
summarize(), which collapses each group into a single-row
summary of that group.
We might have questions about this data in the file Film_Permits.csv that can help us understand how to use these functions.
For which of the five boroughs of NYC most of the film permits are being requested?
We can find out by creating a frequency table for the data. We need to count how many film permits are made in each borough.
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:data.table':
##
## between, first, last## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unioncounts <- nyc_films %>%
group_by(Borough) %>%
summarize(count_of_permits = length(Borough))
counts## # A tibble: 5 × 2
## Borough count_of_permits
## <chr> <int>
## 1 Bronx 2280
## 2 Brooklyn 22087
## 3 Manhattan 34200
## 4 Queens 11484
## 5 Staten Island 938Pipes (%>%) let you take the output of one
function and send it directly to the next.
The above code grouped the data by each of the five boroughs and then
counted the number of times each Borough occurred (using the
length function). The result is a new variable called
counts.
Enter an R code below that shows how many cities each state had a population of more than 65K in 2015.
Enter your code below:
You might be skeptical of what you are doing here, copying and pasting things. Soon you’ll see how fast you can do something by copying, pasting, and making a few changes.
Let’s quickly ask another question about what kinds of films are
being made. The column Category gives us some information
about that. Let’s copy-paste the code we already made above and see what
categories the films fall into. See if you can tell what was changed in
the code to make this work:
counts <- nyc_films %>%
group_by(Category) %>%
summarize(count_of_permits = length(Category))
counts## # A tibble: 10 × 2
## Category count_of_permits
## <chr> <int>
## 1 Commercial 5569
## 2 Documentary 296
## 3 Film 11061
## 4 Music Video 236
## 5 Red Carpet/Premiere 1
## 6 Still Photography 4165
## 7 Student 433
## 8 Television 39840
## 9 Theater 6802
## 10 WEB 2586Let’s notice the changes.
Borough was changed to Category. That
was the main thing.
Note that none of the library() commands are used
again, and I didn’t re-run the very early code to get the data. R
already has those things in its memory, so we don’t need to do that
again. If you ever clear the memory of R, you will need to reload those
things.
Now let’s use another R function, sum, to add the female
population per state.
Enter your code below:
We can use the function group_by with more than one
parameter, i.e., group by more than one variable.
counts <- nyc_films %>%
group_by(Borough,Category) %>%
summarize(count_of_permits = length(Category))## `summarise()` has grouped output by 'Borough'. You can override using the
## `.groups` argument.head(counts)## # A tibble: 6 × 3
## # Groups: Borough [1]
## Borough Category count_of_permits
## <chr> <chr> <int>
## 1 Bronx Commercial 152
## 2 Bronx Documentary 11
## 3 Bronx Film 466
## 4 Bronx Music Video 6
## 5 Bronx Still Photography 64
## 6 Bronx Student 12tail(counts)## # A tibble: 6 × 3
## # Groups: Borough [1]
## Borough Category count_of_permits
## <chr> <chr> <int>
## 1 Staten Island Music Video 4
## 2 Staten Island Still Photography 2
## 3 Staten Island Student 12
## 4 Staten Island Television 589
## 5 Staten Island Theater 1
## 6 Staten Island WEB 4We did two critical things. First, we added Borough and
Category into the group_by() function. This
automatically gives separate counts for each film category for each
Borough.
Enter a chunk of code to group the demographic data by State Code and Race, which adds population (Count) for each race for each state.
Enter your code below:
filter() allows you to subset observations (rows) based
on their values. The first argument is the name of the data frame. The
second and subsequent arguments are the expressions that filter the data
frame.
The library dplyr allows us to filter data by a single
Borough. Let’s use the function filter() for the NYC
permits data to find the number of permits only for Manhattan.
counts <- nyc_films %>%
group_by(Category) %>%
filter(Borough=='Manhattan')%>%
summarize(count_of_permits = length(Category))
counts## # A tibble: 10 × 2
## Category count_of_permits
## <chr> <int>
## 1 Commercial 3290
## 2 Documentary 193
## 3 Film 5106
## 4 Music Video 108
## 5 Red Carpet/Premiere 1
## 6 Still Photography 2611
## 7 Student 249
## 8 Television 15583
## 9 Theater 5628
## 10 WEB 1431Now let’s do something similar with the demographic data.
Enter a chunk of code to group the demographic data by State Code and add the population of Hispanic or Latino (Count) for each state.
Enter your code below:
For all functions in dplyr, we have the following:
The first argument is a data frame.
The subsequent arguments describe what to do with the data frame using the variable names (without quotes).
The result is a new data frame.
This document contains excerpts and adaptions from:
Matthew J. C. Crump, Anjali Krishnan, Stephen Volz, and Alla Chavarga (2018) “Answering questions with data: Lab Manual.” Last compiled on 2019-04-06. https://www.crumplab.com/statisticsLab/
Data Carpentry. “Aggregating and analyzing data with dplyr.” https://datacarpentry.org/dc_zurich/R-ecology/04-dplyr.html
Hadley Wickham and Garrett Grolemund. “R for Data Science.”https://r4ds.had.co.nz/index.html