TEMPLATE

Modern Plain Text Social Science: Week FIXME

Kieran Healy
kieran.healy@duke.edu

August 27, 2023

This is an H1-level header

Hello there! I am just some regular text. You can see an H1-level slide is meant for section separation and big callouts and the like. The background color of H1 slides, and the color of H1 text, is set via a lua filter.

This is an H2 header

I am more your regular topical slide.

  • It will appear in the overview hierarchy correctly
  • i.e. below the earlier H1

This is also an H2-level header

  • Hello there! I am a bullet point
  • Everyone hates bullet points
  • And yet here we are

This is another H2-level header

I create a new slide when invoked.

This is an H3-level header

I do not create a new slide when invoked.

And this is H4

Me neither.

And this is H5

You get the idea.

And this is H6

The humblest header level

Another First Level Header

This time the foreground is light blue, but this text is off-white

Followed by an H2

With some content

Hello I’m green()?

Code should set up as green on this sort of H1 slide.

Does subhead structure still work?

  • Yes it does: things appear in the correct hierarchy when you press O

Testing classed styling of words

Classed styling in H2

Is green() or not?

There’s a class for huge text that’s not a header

We can specify its color, too

Three dashes makes a slide break too

Use two to give a blank slide.

---

---

Markdown

Standard Markdown Stuff

Unordered lists:

  • Bullet points again
    • You cannot escape me
    • Next slide please

Or ordered lists

  1. This is the first item
    1. Nesting lists is not a good idea on the whole
    2. You’re giving a talk, not reading a contract
  2. Just don’t do it.

Standard Markdown Stuff

Format text in bold, italic, or include URLs.

Maths

Inline mode: \(c^2 = a^2 + b^2\)

Or in Display mode:

\[p(x) \sim \frac{1}{\sqrt{2\pi\sigma^2}} \exp{ \bigg[-\frac{1}{2}\bigg( \frac{x-\mu}{\sigma}\bigg)^2 \bigg] }\]

Static Images

Image as background

Code

Code

The dot syntax for codeblocks we don’t want to run is nice.

gss_svy |> 
  group_by(year) |> 
  group_map_dfr(possibly(~ tidy(svyglm(fefam_n ~ age + sex + race, 
                       design = .x, 
                       family = quasibinomial(),
                       na.action = na.omit),
                       conf.int = TRUE), 
                    otherwise = NULL))

Executable Code

And we can run it in the usual way:

letters 
 [1] "a" "b" "c" "d" "e" "f" "g" "h" "i" "j" "k" "l" "m" "n" "o" "p" "q" "r" "s"
[20] "t" "u" "v" "w" "x" "y" "z"

Executable Code

library(tidyverse)
library(socviz)

Executable Code

gss_sm |> 
  group_by(bigregion, religion) |> 
  tally() |> 
  mutate(pct = round((n/sum(n))*100, 1)) |> 
  drop_na() |> 
  ggplot(mapping = aes(x = pct, y = reorder(religion, -pct), fill = religion)) +
  geom_col() +
    labs(x = "Percent", y = NULL) +
    guides(fill = "none") + 
    facet_wrap(~ bigregion, nrow = 1)

GSS Plot

Executable Code, but smaller

gss_sm |> 
  group_by(bigregion, religion) |> 
  tally() |> 
  mutate(pct = round((n/sum(n))*100, 1)) |> 
  drop_na() |> 
  ggplot(mapping = aes(x = pct, y = reorder(religion, -pct), fill = religion)) +
  geom_col() +
    labs(x = "Percent", y = NULL) +
    guides(fill = "none") + 
    facet_wrap(~ bigregion, nrow = 1)

GSS Plot