Exploring R Consortium ISC Grants

A contribution to the 2024-02-20 #tidytuesday social data project

Photo by Markus Winkler

library(tidyverse)
library(plotly)
library(skimr)
library(tidytext)
library(here)
library(scales)

Background

I’ve never really contributed to tidytuesday. Recently, I’ve been trying to spark some inspiration, so I thought contributing to this social data project would be a good start. I used this post as an opportunity to get more comfortble using plotly and Tableau for creating data visualizations.

data_isc_grants <- 
  read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2024/2024-02-20/isc_grants.csv')

Data description

The data represents information about past projects funded by the R Consortium Infrastructure Committee (ISC) Grant Program. The purpose of these grants is to support projects contributing to the R community. Learn more about the most recent round of funding by checking out their blog post announcing this round of grants.

The data includes columns like: year, group (i.e., funding cycle), title, funded (i.e., funding amount), and summary. Before creating some data visualizations, let’s do some quick exploratory analysis.

glimpse(data_isc_grants)

Rows: 85
Columns: 7
$ year        <dbl> 2023, 2023, 2023, 2023, 2023, 2022, 2022, 2022, 2022, 2022, 2022, 2022, 2022, …
$ group       <dbl> 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 1, 1, 1, 1, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, …
$ title       <chr> "The future of DBI (extension 1)", "Secure TLS Communications for R", "volcalc…
$ funded      <dbl> 10000, 10000, 12265, 3000, 15750, 8000, 8000, 22000, 6000, 25000, 15000, 20000…
$ proposed_by <chr> "Kirill Müller", "Charlie Gao", "Kristina Riemer", "Mark Padgham", "Jon Harmon…
$ summary     <chr> "This proposal mostly focuses on the maintenance and support for {DBI}, the {D…
$ website     <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA…

skim(data_isc_grants)

Data summary

Name	data_isc_grants
Number of rows	85
Number of columns	7
_______________________
Column type frequency:
character	4
numeric	3
________________________
Group variables	None

Variable type: character

skim_variable	n_missing	complete_rate	min	max	empty	n_unique	whitespace
title	0	1.00	4	120	0	85	0
proposed_by	0	1.00	8	63	0	66	0
summary	0	1.00	31	2210	0	85	0
website	33	0.61	21	224	0	48	0

Variable type: numeric

skim_variable	n_missing	complete_rate	mean	sd	p0	p25	p50	p75	p100	hist
year	0	1	2019.14	2.08	2016	2017	2019	2021	2023	▇▆▇▅▅
group	0	1	1.40	0.49	1	1	1	2	2	▇▁▁▁▅
funded	0	1	13781.14	11325.80	0	6000	10000	16000	62400	▇▂▁▁▁

What’s the trend for grant funding?

Let’s take a look at the funding trend by funding cycle (i.e., fall and spring).

data_by_year_grp <- data_isc_grants |>
  mutate(group = case_when(
    group == 1 ~ "Spring", 
    group == 2 ~ "Fall")
  ) |>
  group_by(year, group) |>
  summarise(funded = sum(funded), .groups = "drop") |>
  arrange(group, year) |> 
  pivot_wider(names_from = group, values_from = funded)

plot_ly(
  data_by_year_grp, 
  x = ~year, 
  y = ~Fall, 
  name = "Fall", 
  type = 'scatter', 
  mode = 'lines',
  line = list(width = 5),
  text = ~paste(
    "Funding awarded: $", comma(Fall),
    "<br>Year: ", year
  ),
  hoverinfo = "text"
) |>
add_trace(
  y = ~Spring,
  name = "Spring",
  text = ~paste(
    "Funding awarded: $", comma(Spring),
    "<br>Year: ", year
  ),
  hoverinfo = "text"
) |>
layout(
  title = list(
    text = "<b>Funding trend for R Consortium ISC grants by funding round</b>",
    xanchor = "center",
    yanchor = "top",
    font = list(family = "arial", size = 24)
  ),
  xaxis = list(title = ""),
  yaxis = list(title = "Funding amount ($US)")
)

What words are used most often within descriptions of funded projects?

Now, let’s explore the words used within descriptions most often in awarded grant applications.

data_word_fund_trend <- data_isc_grants |>
  mutate(
    summary = str_remove_all(str_to_lower(summary), "[[:punct:]]"),
    summary = str_remove_all(summary, "[0-9]"),
  ) |>
  unnest_tokens(word, summary) |>
  anti_join(get_stopwords()) |>
  group_by(year) |>
  count(word) |>
  arrange(word, year) |>
  group_by(word) |>
  mutate(
    n_cume = cumsum(n)
  )

top_words <- data_word_fund_trend |>
  ungroup() |>
  summarise(top = quantile(n_cume, .99)) |>
  pull(top)

data_top_words <- data_word_fund_trend |>
  filter(n_cume >= top_words) |>
  distinct(word)

plot_ly(
  data = data_word_fund_trend, 
  x = ~year,
  y = ~n_cume,
  mode = "lines",
  line = list(color = "#d3d3d3", width = 3),
  type = "scatter",
  mode = "lines",
  name = "",
  text = ~paste(
    "Word: ", word,
    "<br>Cumulative mentions: ", n_cume,
    "<br>Year: ", year
  ),
  hoverinfo = "text"
) |>
add_lines(
  data = data_word_fund_trend |> semi_join(data_top_words),
  x = ~year,
  y = ~n_cume,
  line = list(color = "#0C2D48", width = 3),
  type = "scatter",
  mode = "lines",
  name = ""
) |>
layout(
  title = list(
    text = "<b>Aiming for RConsortium grant funding? Consider using these words</b>",
    xanchor = "center",
    yanchor = "top",
    font = list(family = "arial", size = 24)
  ),
  xaxis = list(title = ""),
  yaxis = list(title = "Cumulative mentions"),
  showlegend = FALSE
)

An attempt using Tableau

To learn more about using Tableau, I took this week’s data as an opportunity to learn more. Here’s what I came up with.