Please add alt text (alternative text) to all of your posted graphics
for #TidyTuesday.
Twitter provides guidelines for how to add alt text to your images.
The DataViz Society/Nightingale by way of Amy Cesal has an article on writing good alt text for plots/graphs.
Here's a simple formula for writing alt text for data visualization: ### Chart type It's helpful for people with partial sight to know what chart type it is and gives context for understanding the rest of the visual. Example: Line graph ### Type of data What data is included in the chart? The x and y axis labels may help you figure this out. Example: number of bananas sold per day in the last year ### Reason for including the chart Think about why you're including this visual. What does it show that's meaningful. There should be a point to every visual and you should tell people what to look for. Example: the winter months have more banana sales ### Link to data or source Don't include this in your alt text, but it should be included somewhere in the surrounding text. People should be able to click on a link to view the source data or dig further into the visual. This provides transparency about your source and lets people explore the data. Example: Data from the USDA
Penn State has an article on writing alt text descriptions for charts and tables.
Charts, graphs and maps use visuals to convey complex images to users. But since they are images, these media provide serious accessibility issues to colorblind users and users of screen readers. See the examples on this page for details on how to make charts more accessible.
The {rtweet} package includes the ability to post
tweets with
alt text programatically.
Need a reminder? There are extensions that force you to remember to add Alt Text to Tweets with media.
The data this week comes from NOAA's National Weather Service Storm Prediction Center Severe Weather Maps, Graphics, and Data Page. Thank you to Evan Gower for the suggestion!
Evan investigated a version of this dataset on Kaggle.
# Get the Data
# Read in with tidytuesdayR package
# Install from CRAN via: install.packages("tidytuesdayR")
# This loads the readme and all the datasets for the week of interest
# Either ISO-8601 date or year/week works!
tuesdata <- tidytuesdayR::tt_load('2023-05-16')
tuesdata <- tidytuesdayR::tt_load(2023, week = 20)
tornados <- tornados
# Or read in the data manually
tornados <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2023/2023-05-16/tornados.csv')
| variable | class | description |
|---|---|---|
| om | integer | Tornado number. Effectively an ID for this tornado in this year. |
| yr | integer | Year, 1950-2022. |
| mo | integer | Month, 1-12. |
| dy | integer | Day of the month, 1-31. |
| date | date | Date. |
| time | time | Time. |
| tz | character | Canonical tz database timezone. |
| datetime_utc | datetime | Date and time normalized to UTC. |
| st | character | Two-letter postal abbreviation for the state (DC = Washington, DC; PR = Puerto Rico; VI = Virgin Islands). |
| stf | integer | State FIPS (Federal Information Processing Standards) number. |
| mag | integer | Magnitude on the F scale (EF beginning in 2007). Some of these values are estimated (see fc). |
| inj | integer | Number of injuries. When summing for state totals, use sn == 1 (see below). |
| fat | integer | Number of fatalities. When summing for state totals, use sn == 1 (see below). |
| loss | double | Estimated property loss information in dollars. Prior to 1996, values were grouped into ranges. The reported number for such years is the maximum of its range. |
| slat | double | Starting latitude in decimal degrees. |
| slon | double | Starting longitude in decimal degrees. |
| elat | double | Ending latitude in decimal degrees. |
| elon | double | Ending longitude in decimal degrees. |
| len | double | Length in miles. |
| wid | double | Width in yards. |
| ns | integer | Number of states affected by this tornado. 1, 2, or 3. |
| sn | integer | State number for this row. 1 means the row contains the entire track information for this state, 0 means there is at least one more entry for this state for this tornado (om + yr). |
| f1 | integer | FIPS code for the 1st county. |
| f2 | integer | FIPS code for the 2nd county. |
| f3 | integer | FIPS code for the 3rd county. |
| f4 | integer | FIPS code for the 4th county. |
| fc | logical | Was the mag column estimated? |
# All packages used in this script:
library(tidyverse)
library(here)
url <- "https://www.spc.noaa.gov/wcm/data/1950-2022_actual_tornadoes.csv"
# Some of the automatic column types are imperfect Get that spec and then
# update it.
tornados <- read_csv(url)
spec(tornados) # Copy/pasted into col_types below then edited.
tornados <- read_csv(
url,
col_types = cols(
om = col_integer(),
yr = col_integer(),
mo = col_integer(),
dy = col_integer(),
date = col_date(format = ""),
time = col_time(format = ""),
tz = col_integer(),
st = col_factor(),
stf = col_integer(),
stn = col_integer(),
mag = col_integer(),
inj = col_integer(),
fat = col_integer(),
loss = col_double(),
closs = col_double(),
slat = col_double(),
slon = col_double(),
elat = col_double(),
elon = col_double(),
len = col_double(),
wid = col_integer(),
ns = col_integer(),
sn = col_integer(),
sg = col_integer(),
f1 = col_integer(),
f2 = col_integer(),
f3 = col_integer(),
f4 = col_integer(),
fc = col_integer()
)
)
glimpse(tornados)
# This table only contains one segment per tornado, so we can drop the sg
# column.
tornados$sg <- NULL
# The tz column is confusing in the provided dictionary
# (https://www.spc.noaa.gov/wcm/data/SPC_severe_database_description.pdf).
# Investigate it to make sense of the various values.
tornados |>
count(tz)
# The doc says 3 == CST, and 9 == GMT. 0 appears to be NA. What is 6?
tornados |>
filter(tz == 6) |>
count(st)
# All tornados with tz == 6 are in Mountain Time states, so we'll make that
# assumption. Update time encoding.
tornados <- tornados |>
# We can't really judge even what day the recording was on for unknown tz, so
# drop those values.
filter(tz != 0) |>
mutate(
# Make the remaining tz's more meaningful. We'll assume they meant Central
# (daylight or standard) for "CST", and likewise that they meant what we now
# call UTC for "GMT". "GMT" sometimes includes BST so we'll avoid using that
# name.
tz = case_match(
tz,
3 ~ "America/Chicago",
6 ~ "America/Denver",
9 ~ "UTC"
),
# Add a datetime_utc column to normalize the times. ymd_hms only wants a
# single timezone (not a vector of them), so break it up with a case_match.
datetime_utc = case_match(
tz,
"America/Chicago" ~ lubridate::ymd_hms(
paste(date, time),
tz = "America/Chicago"
),
"America/Denver" ~ lubridate::ymd_hms(
paste(date, time),
tz = "America/Denver"
),
"UTC" ~ lubridate::ymd_hms(
paste(date, time),
tz = "UTC"
)
) |>
lubridate::with_tz("UTC"),
.after = tz
) |>
# Drop stn because it was discontinued and was inconsistent before being
# discontinued. closs (crop loss) has an unexplained discontinuity in 2016 and
# it isn't entirely clear what changed.
select(-"stn", -"closs") |>
# Recode some more weird columns.
mutate(
# The mag column uses -9 for NA.
mag = na_if(mag, -9),
# The loss column is confusingly coded. Let's attempt to make it make sense.
# The documentation (last updated in 2010) explains that the coding changed in
# 1996. Observationally, it's clear that it changed again in 2016.
loss = case_when(
loss == 0 ~ NA,
yr < 1996 & loss == 1 ~ 50,
yr < 1996 & loss == 2 ~ 500,
yr < 1996 & loss == 3 ~ 5000,
yr < 1996 & loss == 4 ~ 50000,
yr < 1996 & loss == 5 ~ 500000,
yr < 1996 & loss == 6 ~ 5000000,
yr < 1996 & loss == 7 ~ 50000000,
yr < 1996 & loss == 8 ~ 500000000,
yr < 1996 & loss == 9 ~ 5000000000,
yr >= 1996 & yr < 2016 ~ loss * 1e6,
TRUE ~ loss
),
# The fc column is really a "was mag estimated" column
fc = as.logical(fc)
)
# Some of the remaining columns are confusing, but we'll explain them in the
# dictionary and see what people find!
write_csv(
tornados,
here(
"data",
"2023",
"2023-05-16",
"tornados.csv"
)
)