Skip to content

Instantly share code, notes, and snippets.

@cbgoodman
Created December 2, 2019 19:48
Show Gist options
  • Select an option

  • Save cbgoodman/e0f399f3e8d691a9907285a406d050ec to your computer and use it in GitHub Desktop.

Select an option

Save cbgoodman/e0f399f3e8d691a9907285a406d050ec to your computer and use it in GitHub Desktop.
Example standardized coefplot (by hand)
library(tidyverse)
library(readxl)
plot.data <- read_excel("popgrow-coeff.xlsx", sheet = "popgrow", col_types = c("guess", "numeric", "numeric", "numeric", "guess"))
multiplier <- qnorm(1 - 0.05 / 2)
plot.data <- plot.data %>%
mutate(std.est = estimate * sd,
std.se = se * sd,
ymin = std.est - (multiplier * std.se),
ymax = std.est + (multiplier * std.se),
term = factor(term, levels = rev(c("Central city population share",
"Suburban municipal fragmentation",
"Suburban unincorporated population",
"School district decentralization",
"Special district overlap",
"ln population",
"Real per capita money income",
"Growth in log population, 1950-1960",
"Unemployment rate",
"Manufacturing share",
"% population with 16+ year education")), ordered = TRUE),
group = factor(group, levels = c("1960 to 1970",
"1960 to 1980",
"1960 to 1990",
"1960 to 2000"), ordered = TRUE))
p <- ggplot(plot.data, aes(x = term, y = std.est)) +
geom_hline(yintercept = 0, colour = "#000000", size = 0.75) + # Line at 0
geom_pointrange(aes(ymin = ymin, ymax = ymax, group = rev(group), color = group), position = position_dodge(width = 0.75), size = 1, fatten = 1.25,
show.legend = T) + # Ranges for each coefficient
labs(x="Coefficient", y="Standardized Estimate", title="Population Growth") + # Labels
coord_flip() + # Rotate the plot
theme_bw() +
theme(legend.position="bottom",
legend.title = element_blank())
@E-Mendez
Copy link

Hi Chris,
I would like to reproduce your example but I cannot find the popgrow-coeff.xlsx excel file. Is there a chance you would upload it?

Cheers!
Eduardo.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment