Rebecca
Research Question:
How do high school students’ subgroup makeup (i.e., Race/ethnicity, Male vs. Female, economically disadvantaged, Limited English, Migrant status, Disability status, and Homelessness) differ among states/regions?
- Design choices:
- Color-blind friendly gradient palette
- My original idea was to show differences in test scores by gender (e.g. male and female). However, I decided to show all groups in one plot.
- Expanded height and width so that the axes are legible.
- Findings: There are no clear regional differences. Some eastern states have highest proficiency (Virginia and Maryland) and some have lowest (Rhode Island and New Hampshire).
- Prior versions: I initially wanted to show the ridge lines by state and thought that the gradient palette was very striking and visually representative of the mean proficient scores.
Plot
Prior version 1
Prior version 2
Code
df_pivot %>%
ggplot(aes(x = mean_pct,
y = fct_reorder(STNAM.y, mean_pct),
fill = stat(x))) +
geom_density_ridges_gradient(scale = 2, size = 0.3, rel_min_height = 0.001) +
scale_fill_viridis_c(name = "%", option = "C") +
coord_cartesian(clip = "off") +
theme_ridges(grid = FALSE) +
labs(title = 'Mean percentage of students that scored at or above proficient',
x = 'Mean percent proficient',
y = 'State')