Workshop: Dealing with Data in R

Getting Help with R

After this workshop

steffilazerte
@steffilazerte@fosstodon.org
@steffilazerte
steffilazerte.ca

Compiled: 2024-02-21

First things first

Save previous script
Consider taking notes during this section

Artwork by @allison_horst

Troubleshooting

Artwork by @allison_horst

Bit by bit

Line by line

R is sequential
If you skip lines, you’re not running that part (and R has no idea)

#a <- 1
b <- 2
a + b

Error in eval(expr, envir, enclos): object 'a' not found

Error? Start at the beginning and go line by line

a <- 1
b <- 2
a + b

[1] 3

Bit by bit

Line by line

library(tidyverse)

# Load Data
size <- read_csv("./data/grain_size2.csv")

# First modification
size <- mutate(size, 
               total_sand = coarse_sand + medium_sand + fine_sand,
               total_silt = coarse_silt + medium_silt + fine_silt)

# Second modification
size <- size |>
  group_by(plot) |>
  summarize(n = n(),
            total_sand = sum(total_sand),
            mean_sand = mean(total_sand),
            sd_sand = sd(total_sand),
            se_sand = sd_sand / sqrt(n))

Especially important if loading and modifying data

Can’t run 1st modification
after 2nd modification

Bit by bit

Section by section

library(tidyverse)

size <- read_csv("./data/grain_size2.csv") |>
  mutate(total_sand = coarse_sand + medium_sand + fine_sand,
         total_silt = coarse_silt + medium_silt + fine_silt) |>
  group_by(plot) |>
  summarize(n = n(),
            total_sand = sum(total_sand),
            mean_sand = mean(totall_sand),
            sd_sand = sd(total_sand),
            se_sand = sd_sand / sqrt(n))

Error in `summarize()`:
ℹ In argument: `mean_sand = mean(totall_sand)`.
ℹ In group 1: `plot = "CSP01"`.
Caused by error:
! object 'totall_sand' not found

Bit by bit

Section by section

library(tidyverse)

size <- read_csv("./data/grain_size2.csv")

No error

size <- read_csv("./data/grain_size2.csv") |>
  mutate(total_sand = coarse_sand + medium_sand + fine_sand,
         total_silt = coarse_silt + medium_silt + fine_silt)

No error

size <- read_csv("./data/grain_size2.csv") |>
  mutate(total_sand = coarse_sand + medium_sand + fine_sand,
         total_silt = coarse_silt + medium_silt + fine_silt) |>
  group_by(plot)

No error

Bit by bit

Section by section

size <- read_csv("./data/grain_size2.csv") |>
  mutate(total_sand = coarse_sand + medium_sand + fine_sand,
         total_silt = coarse_silt + medium_silt + fine_silt) |>
  group_by(plot) |>
  summarize(n = n(),
            total_sand = sum(total_sand),
            mean_sand = mean(totall_sand),
            sd_sand = sd(total_sand),
            se_sand = sd_sand / sqrt(n))

Error in `summarize()`:
ℹ In argument: `mean_sand = mean(totall_sand)`.
ℹ In group 1: `plot = "CSP01"`.
Caused by error:
! object 'totall_sand' not found

Ah ha!

Bit by bit

Applies to error messages too

First, don’t panic!
Look at the error bit by bit

Error: Problem with `summarise()` column `mean_sand`.
ℹ `mean_sand = mean(totall_sand)`.
x object 'totall_sand' not found
ℹ The error occurred in group 1: plot = "CSP01".

Bit by bit

Applies to error messages too

Error: Problem with 'summarise()' column 'mean_sand'`

Okay, we know the problem is in the summarize() part and then mean_sand part of that

ℹ 'mean_sand = mean(totall_sand)'
x object 'totall_sand' not found

Looks like this is the line with the problem.
And the problem is object 'totall_sand' not found.

Ooops! Typo!

ℹ The error occurred in group 1: plot = "CSP01".

Lastly, it’s telling us that the problem was when working with this group of data.

(This can be useful when troubleshooting, because you can filter() your data and take a look)

Artwork by @allison_horst

R is never wrong

Just sometimes unhelpful!

Getting Help

Cheat Sheets

Help
- Cheatsheets

Take a look yourself

Vignettes

Many packages come with vignettes (tutorials)

List Vignettes

vignette(package = "ggplot2")

Vignettes in package ‘ggplot2’:

ggplot2-specs                      Aesthetic specifications (source, html)
extending-ggplot2                  Extending ggplot2 (source, html)
profiling                          Profiling Performance (source, html)

Load Vignettes

vignette("ggplot2-specs", package = "ggplot2")

Try it!

Tutorials

Vignettes are also online

e.g., ggplot2
e.g., tidyverse

Organizations/Websites

A screenshot of the ggplot2 website

Books!

Free Online

R for Data Science (read it!)
R Graphics Cookbook (how to do X)
ggplot2 (next level)
Data Visualization: A practical introduction
Geocomputation with R (spatial, GIS, maps)
Statistical Inference via Data Science: A ModernDive into R and the tidyverse (stats)

Communities!

rOpenSci
#RStats Twitter
#RStats Mastodon
(e.g., Fosstodon.org or Hachyderm.io)
R4DS Online learning community on Slack
(ask any question, they’re really nice!)

Specific Groups

rOpenSci logo rLadies logo Minorities in R hex logo AsiaR hex logo AfricaR hex logo Two fuzzy monsters standing side-by-side outside of a door frame through which is a magical wonderland of different R communities, with a "mind blown" rainbow coming out of the one closest to the door. A welcome mat says "Welcome."

Specific help

Examples

In R

?geom_boxplot

Copy and paste the examples into your console

Examples

On the web

Nice to see expected output
Helps figure out if it’s your system or your code

Screenshot of some examples on the ggplot2 website

Web searches

Always include “R” in the search
Include the package name!
Use keywords
Some errors are very general

Web searches

Always include “R” in the search
Include the package name!
- Try “R boxplots” vs. “R boxplots ggplot2”
Use keywords
- Try “R boxplots ggplot2 notch”
Some errors are very general
- Try “R Error: object ‘m’ not found”

Stackoverflow etc.

“R how to remove duplicate rows”

Stackoverflow etc.

Things to consider

Date (i.e., R version, Package Version)
Packages used (tidyverse? R base? A mix?)
What are the example data?
- mtcars and iris are commonly used data sets built into R base
- msleep and diamonds are commonly used data sets built into ggplot2
What are the example columns?
What is actually required to answer your question?

Asking for Help

Not useful

“I got an error”
“It didn’t work”

Better!

“I got this error”
“It didn’t give me this”

Best!!

“I did this and I got this error”
“I expected it to do this, but in fact the output was this”

Best of the Best!!!

“I did this [small reproducible code, including data set] and I got this [exact error/output]”

Reproducible Examples

Minimal code and data required to reproduce the error
Often preparing this actually helps you solve the error!
Includes
- packages (library())
- data
- runnable code

Reproducible Examples

How do I change the order of `vore`?

Not reproducible

ggplot(data = m, aes(x = vore, y = awake, fill = `Body Size`)) +
  theme_bw() + 
  theme(axis.title.x = element_blank()) +
  geom_boxplot() +
  scale_fill_viridis_d() +
  labs(y = "Awake time (hrs)",
       title = "Awake time by Diet")

Error in eval(expr, envir, enclos): object 'm' not found

No indication of packages
No indication of what m is

Reproducible Examples

How do I change the order of `vore`?

Reproducible, but not minimal

library(ggplot2)

m <- msleep |>
  mutate(`Body Size` = if_else(bodywt > median(bodywt),
                               "Large", "Small"))

ggplot(m, aes(x = vore, y = awake, fill = `Body Size`)) +
  theme_bw() + 
  theme(axis.title.x = element_blank()) +
  geom_boxplot() +
  scale_fill_viridis_d() +
  labs(y = "Awake time (hrs)",
       title = "Awake time by Diet")

Reproducible Examples

How do I change the order of `vore`?

Reproducible AND Minimal

library(ggplot2)

ggplot(msleep, aes(x = vore, y = awake)) +
  geom_boxplot()

Paying it forward

Citing Software

In-line Text

Software name
Version
Programmers/authors OR Journal article releasing the software (if available)

Bibliography

Journal article releasing the program OR
Programmers/authors
Year of release
Program Name
URL

Citing R

Inline

“All statistical analyses were performed with R statistical software (v4.3.2, R Core Team 2023).”

Bibliography

R Core Team (2023). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.

Citing R

Version information

R.Version()$version.string

[1] "R version 4.3.2 (2023-10-31)"

Citation information

citation()

To cite R in publications use:

  R Core Team (2023). _R: A Language and Environment for Statistical
  Computing_. R Foundation for Statistical Computing, Vienna, Austria.
  <https://www.R-project.org/>.

Citing R Packages

Inline

“All statistical analyses were performed with R statistical software (v4.0.3, R Core Team 2020). We performed Type III ANOVAs using the ‘car’ package for R (v3.0.10, Fox and Weisberg 2019).”

Bibliography

John Fox and Sanford Weisberg (2019). An R Companion to Applied Regression, Third Edition. Thousand Oaks CA: Sage.

Citing R Packages

Version information

packageVersion("car")

[1] '3.1.2'

Citation information

citation("car")

To cite the car package in publications use:

  Fox J, Weisberg S (2019). _An R Companion to Applied Regression_,
  Third edition. Sage, Thousand Oaks CA.
  <https://socialsciences.mcmaster.ca/jfox/Books/Companion/>.

See more about citing packages in my rOpenSci blog post: How to Cite R and R packages

Artwork by @allison_horst

You made it!

Thank you!

(Feedback!)

Getting Help with R

After this workshop

First things first

Troubleshooting

Bit by bit

Line by line

Bit by bit

Line by line

Bit by bit

Section by section

Bit by bit

Section by section

Bit by bit

Section by section

Bit by bit

Applies to error messages too

Bit by bit

Applies to error messages too

R is never wrong

Getting Help

Cheat Sheets

RStudio Menu

Vignettes

List Vignettes

Load Vignettes

Tutorials

Vignettes are also online

Organizations/Websites

Books!

Free Online

Communities!

Specific Groups

Specific help

Examples

In R

Examples

On the web

Web searches

Web searches

Stackoverflow etc.

Stackoverflow etc.

Things to consider

Asking for Help

Not useful

Better!

Best!!

Best of the Best!!!

Reproducible Examples

Reproducible Examples

How do I change the order of vore?

Not reproducible

Reproducible Examples

How do I change the order of vore?

Reproducible, but not minimal

Reproducible Examples

How do I change the order of vore?

Reproducible AND Minimal

Paying it forward

Citing Software

In-line Text

Bibliography

Citing R

Inline

Bibliography

Citing R

Version information

Citation information

Citing R Packages

Inline

Bibliography

Citing R Packages

Version information

Citation information

You made it!

Thank you!

How do I change the order of `vore`?

How do I change the order of `vore`?

How do I change the order of `vore`?