Table 2

---

This lab journal replicates table 2.

---

1 Custom functions

• fpackage.check: Check if packages are installed (and install if not) in R (source).

rm(list = ls())

fpackage.check <- function(packages) {
    lapply(packages, FUN = function(x) {
        if (!require(x, character.only = TRUE)) {
            install.packages(x, dependencies = TRUE)
            library(x, character.only = TRUE)
        }
    })
}

---

2 Packages

• tidyverse: tidy data manipulations
• kableExtra: creating tables in consistent formatting

packages = c("tidyverse", "kableExtra")

fpackage.check(packages)

---

3 Input

We use one processed dataset.

• df_stopping.rda: dataset of PhDs with all relevant variables: gender + ethnicity + field + university + PhD year
  • For construction of this dataset see Dependent Variables : Starting and Stopping to Publish
    
  • name of dataset: df_ppf3

load(file = "./data/processed/df_stopping.rda")

4 Table 2

columns <- c("Min", "Max", "Mean", "SD")

rows <- c("Gender", "Man", "Woman", "Missing", "Ethnicity", "Ethnic majority", "Ethnic minority", "Other",
    "University", "Erasmus University", "Leiden University", "Radboud University", "University of Groningen",
    "Delft University of Technology", "Eindhoven University of Technology", "Tilburg University", "Maastricht University",
    "University of Twente", "Utrecht University", "University of Amsterdam", "Vrije Universiteit Amsterdam",
    "Wageningen University and Research Centre", "Field", "Biological and health sciences", "Physical and mathematical sciences",
    "Social and behavioural sciences", "Engineering", "Agricultural sciences", "Humanities", "Missing ",
    "<strong>PhD Cohort</strong></p>", "<strong>Previous publications</strong></p> <sup>a</sup></var>",
    "<strong>Stopping to publish</strong></p>", "Source: Dutch Doctoral Recipients (D-cubed) data 1990-2018")


t2 <- data.frame(matrix(nrow = length(rows), ncol = length(columns)))
colnames(t2) <- columns
rownames(t2) <- rows

df_ppf3[!duplicated(df_ppf3$id), ] -> sample

t2[c(2:4), 3] <- prop.table(table(sample$gender))[1:3]
t2[c(6:8), 3] <- prop.table(table(sample$ethnicity2))[1:3]
t2[c(10:22), 3] <- prop.table(table(sample$uni))[1:13]
t2[c(24:30), 3] <- prop.table(table(sample$field2))[1:7]
t2[31, c(1:3)] <- summary(sample$phd_cohort)[c(1, 6, 4)]
t2[31, 4] <- sd(sample$phd_cohort)
t2[32, c(1:3)] <- summary(df_ppf3$npubs_prev_s)[c(1, 6, 4)]
t2[32, 4] <- sd(df_ppf3$npubs_prev_s)
t2[33, 3] <- prop.table(table(df_ppf3$inactive))[2]


t2[c(2:4, 6:8, 10:22, 24:30, 33), c(1, 2)] <- c(rep(0, times = 27), rep(1, times = 27))  # binary variables min/max

t2[c(2:4, 6:8, 10:22, 24:33), 3] <- format(round(t2[c(2:4, 6:8, 10:22, 24:33), 3], 2), nsmall = 2)  # rounding to 2 decimals

t2[c(31, 32), c(1, 2, 4)] <- format(round(t2[c(31, 32), c(1, 2, 4)], 2), nsmall = 2)  # rounding

t2[is.na(t2)] <- ""


# N
nrow(sample)

#> [1] 15021

t2 %>%
    kable(format = "html", caption = "<b>Table 2.</b> Descriptive statistics for analyses `stopping to publish` (N=15,021)",
        escape = FALSE) %>%
    row_spec(row = 34, italic = T) %>%
    row_spec(row = c(1, 5, 9, 23), bold = T) %>%
    kable_classic(full_width = F, html_font = "Cambria") %>%
    footnote(alphabet = "Summary statistics for 'previous publications' are calculated at the level of person-periods, instead of at the level of individuals") %>%
    kable_styling(font_size = 14) -> table2

table2

Table 2. Descriptive statistics for analyses stopping to publish (N=15,021)

	Min	Max	Mean	SD
Gender
Man	0	1	0.44
Woman	0	1	0.34
Missing	0	1	0.22
Ethnicity
Ethnic majority	0	1	0.67
Ethnic minority	0	1	0.02
Other	0	1	0.32
University
Erasmus University	0	1	0.04
Leiden University	0	1	0.03
Radboud University	0	1	0.19
University of Groningen	0	1	0.02
Delft University of Technology	0	1	0.01
Eindhoven University of Technology	0	1	0.01
Tilburg University	0	1	0.01
Maastricht University	0	1	0.09
University of Twente	0	1	0.12
Utrecht University	0	1	0.13
University of Amsterdam	0	1	0.10
Vrije Universiteit Amsterdam	0	1	0.09
Wageningen University and Research Centre	0	1	0.16
Field
Biological and health sciences	0	1	0.36
Physical and mathematical sciences	0	1	0.15
Social and behavioural sciences	0	1	0.13
Engineering	0	1	0.10
Agricultural sciences	0	1	0.09
Humanities	0	1	0.03
Missing	0	1	0.14
PhD Cohort	0.00	28.00	18.46	7.04
Previous publications a	0.00	2.05	0.48	0.30
Stopping to publish	0	1	0.08
Source: Dutch Doctoral Recipients (D-cubed) data 1990-2018
a Summary statistics for ‘previous publications’ are calculated at the level of person-periods, instead of at the level of individuals

---

---

Copyright © 2023