Working smarter with {dplyr} 1.2.0

A talk on the new functions introduced in the latest {dplyr} release.
Presented by

Isabella Velásquez

Presented on

June 24, 2026

  Event   Slides   Repo   Recording


Details

  • 👥 R-Ladies Abuja
  • 📆 24 June 2026 // 06:00 PM WAT
  • 💻️ Virtual

Description

Discover how the latest features in {dplyr} 1.2.0 can help you write cleaner, faster, and more maintainable data transformation workflows. This session explores practical techniques for filtering, joining, summarizing, and reshaping data more efficiently, with real-world examples that demonstrate how to streamline your R code and improve productivity.

Slides

Recording

Summary

This summary was generated by Claude Sonnet 5 and reviewed by me.

dplyr 1.2.0 shipped a set of functions that make two very common jobs — dropping rows and recoding values — say what they actually mean.

TipThe short version
Instead of… Reach for…
filter(!cond) filter_out(cond)
filter() with an OR operator filter(when_any(...))
& groups nested inside an OR when_all() inside when_any()
case_when(x == "a" ~ ...) recode_values(x, "a" ~ ...)
case_when(..., .default = x) to tweak a few values replace_values() / replace_when() 
library(dplyr)
library(palmerpenguins)

filter_out()

Drops rows where all conditions match — and keeps NA rows, because a row with a missing value was never shown to match.

penguins |>
  filter_out(island == "Torgersen" & body_mass_g > 4000)
# A tibble: 333 × 8
   species island    bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
   <fct>   <fct>              <dbl>         <dbl>             <int>       <int>
 1 Adelie  Torgersen           39.1          18.7               181        3750
 2 Adelie  Torgersen           39.5          17.4               186        3800
 3 Adelie  Torgersen           40.3          18                 195        3250
 4 Adelie  Torgersen           NA            NA                  NA          NA
 5 Adelie  Torgersen           36.7          19.3               193        3450
 6 Adelie  Torgersen           39.3          20.6               190        3650
 7 Adelie  Torgersen           38.9          17.8               181        3625
 8 Adelie  Torgersen           34.1          18.1               193        3475
 9 Adelie  Torgersen           37.8          17.1               186        3300
10 Adelie  Torgersen           37.8          17.3               180        3700
# ℹ 323 more rows
# ℹ 2 more variables: sex <fct>, year <int>

Animation: filter_out(island == "Torgersen" & body_mass_g > 4000) highlights the rows matching each condition, then drops only the row matching both, leaving the NA row in place.

Rule of thumb: if there’s a ! in your filter(), reach for filter_out().

when_any()

Expresses OR logic as a list of alternatives: keep the row if it matches at least one.

penguins |>
  filter(when_any(
    species == "Adelie",
    body_mass_g > 5000
  ))
# A tibble: 213 × 8
   species island    bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
   <fct>   <fct>              <dbl>         <dbl>             <int>       <int>
 1 Adelie  Torgersen           39.1          18.7               181        3750
 2 Adelie  Torgersen           39.5          17.4               186        3800
 3 Adelie  Torgersen           40.3          18                 195        3250
 4 Adelie  Torgersen           NA            NA                  NA          NA
 5 Adelie  Torgersen           36.7          19.3               193        3450
 6 Adelie  Torgersen           39.3          20.6               190        3650
 7 Adelie  Torgersen           38.9          17.8               181        3625
 8 Adelie  Torgersen           39.2          19.6               195        4675
 9 Adelie  Torgersen           34.1          18.1               193        3475
10 Adelie  Torgersen           42            20.2               190        4250
# ℹ 203 more rows
# ℹ 2 more variables: sex <fct>, year <int>

Animation: filter(when_any(species == "Adelie", body_mass_g > 5000)) highlights rows matching either condition and keeps their union.

Rule of thumb: if your filter() has complex OR expressions, reach for when_any().

when_all()

The AND counterpart: keep the row only if it matches every condition.

penguins |>
  filter(when_all(
    species == "Gentoo",
    body_mass_g > 5000
  ))
# A tibble: 61 × 8
   species island bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
   <fct>   <fct>           <dbl>         <dbl>             <int>       <int>
 1 Gentoo  Biscoe           50            16.3               230        5700
 2 Gentoo  Biscoe           50            15.2               218        5700
 3 Gentoo  Biscoe           47.6          14.5               215        5400
 4 Gentoo  Biscoe           46.7          15.3               219        5200
 5 Gentoo  Biscoe           46.8          15.4               215        5150
 6 Gentoo  Biscoe           49            16.1               216        5550
 7 Gentoo  Biscoe           48.4          14.6               213        5850
 8 Gentoo  Biscoe           49.3          15.7               217        5850
 9 Gentoo  Biscoe           49.2          15.2               221        6300
10 Gentoo  Biscoe           48.7          15.1               222        5350
# ℹ 51 more rows
# ℹ 2 more variables: sex <fct>, year <int>

Animation: filter(when_all(species == "Gentoo", body_mass_g > 5000)) keeps only the rows matching both conditions.

Rule of thumb: for a simple AND you don’t need it — commas in filter() already mean AND. Its real job is nesting.

when_all() inside when_any()

Nest them and complex logic reads as an outline instead of an expression.

penguins |>
  filter(when_any(
    when_all(species == "Adelie", island == "Torgersen", body_mass_g > 3700),
    when_all(species == "Gentoo", body_mass_g > 5000)
  ))
# A tibble: 83 × 8
   species island    bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
   <fct>   <fct>              <dbl>         <dbl>             <int>       <int>
 1 Adelie  Torgersen           39.1          18.7               181        3750
 2 Adelie  Torgersen           39.5          17.4               186        3800
 3 Adelie  Torgersen           39.2          19.6               195        4675
 4 Adelie  Torgersen           42            20.2               190        4250
 5 Adelie  Torgersen           38.6          21.2               191        3800
 6 Adelie  Torgersen           34.6          21.1               198        4400
 7 Adelie  Torgersen           42.5          20.7               197        4500
 8 Adelie  Torgersen           46            21.5               194        4200
 9 Adelie  Torgersen           41.8          19.4               198        4450
10 Adelie  Torgersen           39.7          18.4               190        3900
# ℹ 73 more rows
# ℹ 2 more variables: sex <fct>, year <int>

Animation: filter(when_any(when_all(...), when_all(...))) showing two nested condition groups and the rows each contributes to the result.

Rule of thumb: reach for the nested form when each alternative needs more than one condition.

recode_values()

Matches values directly instead of writing column == over and over. Name the column once, then list the mappings.

penguins |>
  mutate(island_location = island |> recode_values(
    "Biscoe"    ~ "Southwest",
    "Dream"     ~ "Northwest",
    "Torgersen" ~ "Northwest"
  ))
# A tibble: 344 × 9
   species island    bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
   <fct>   <fct>              <dbl>         <dbl>             <int>       <int>
 1 Adelie  Torgersen           39.1          18.7               181        3750
 2 Adelie  Torgersen           39.5          17.4               186        3800
 3 Adelie  Torgersen           40.3          18                 195        3250
 4 Adelie  Torgersen           NA            NA                  NA          NA
 5 Adelie  Torgersen           36.7          19.3               193        3450
 6 Adelie  Torgersen           39.3          20.6               190        3650
 7 Adelie  Torgersen           38.9          17.8               181        3625
 8 Adelie  Torgersen           39.2          19.6               195        4675
 9 Adelie  Torgersen           34.1          18.1               193        3475
10 Adelie  Torgersen           42            20.2               190        4250
# ℹ 334 more rows
# ℹ 3 more variables: sex <fct>, year <int>, island_location <chr>

Animation: mutate(island_location = island |> recode_values("Biscoe" ~ "Southwest", "Dream" ~ "Northwest", "Torgersen" ~ "Northwest")) filling the new column.

Add unmatched = "error" and a stray category raises an error instead of silently becoming NA.

Rule of thumb: if you’re writing == inside case_when(), reach for recode_values().

recode_values() with a lookup table

Because the mappings are values rather than code, they can live in a data frame — which means they can live in a spreadsheet, a database, or someone else’s repo.

lookup <- tibble(
  from = c("Biscoe", "Dream", "Torgersen"),
  to   = c("Southwest", "Northwest", "Northwest")
)

penguins |>
  mutate(
    island_location = recode_values(island, from = lookup$from, to = lookup$to)
  )
# A tibble: 344 × 9
   species island    bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
   <fct>   <fct>              <dbl>         <dbl>             <int>       <int>
 1 Adelie  Torgersen           39.1          18.7               181        3750
 2 Adelie  Torgersen           39.5          17.4               186        3800
 3 Adelie  Torgersen           40.3          18                 195        3250
 4 Adelie  Torgersen           NA            NA                  NA          NA
 5 Adelie  Torgersen           36.7          19.3               193        3450
 6 Adelie  Torgersen           39.3          20.6               190        3650
 7 Adelie  Torgersen           38.9          17.8               181        3625
 8 Adelie  Torgersen           39.2          19.6               195        4675
 9 Adelie  Torgersen           34.1          18.1               193        3475
10 Adelie  Torgersen           42            20.2               190        4250
# ℹ 334 more rows
# ℹ 3 more variables: sex <fct>, year <int>, island_location <chr>

Animation: recode_values(island, from = lookup$from, to = lookup$to) with a two-column lookup table mapping each island to a location.

Rule of thumb: once you have more than a handful of mappings, put them in a lookup table.

replace_values() and replace_when()

case_when() and recode_values() create a column, so every row needs a value and anything uncovered needs a .default. The replace_*() family modifies a column instead — unmatched values simply stay as they are.

# swap specific values
penguins_located |>
  mutate(
    island_location = island_location |>
      replace_values(
        "Northwest" ~ "Northern Palmer Archipelago"
      )
  )

# swap values matching a condition
penguins |>
  mutate(body_mass_g = replace_when(body_mass_g, body_mass_g > 5000 ~ 5000))

Rule of thumb: if you want to change some values and keep the rest, reach for replace_*() rather than case_when() with a .default.

Summary

⬢ Use filter() to keep rows

⬢ Use filter_out() to drop rows — especially if you’re reaching for !

⬢ Use when_any() for OR logic, when the expressions get complex

⬢ Use when_all() for AND logic, especially nested inside when_any()

⬢ Use recode_values() for exact matches, instead of == in case_when()

⬢ Use replace_values() / replace_when() to modify an existing column