13.5 Plotly: Visualizing statistical results

13.5.1 Dotplots & error bars

fit <- lm(Fertility ~ Catholic + Education  + Examination, data = swiss)
# arrange by estimate, then make term a factor to order categories in the plot
d <- broom::tidy(fit) %>% 
  arrange(desc(estimate))
d <- d[-1,]

plot_ly(d, x = ~estimate, y = ~term, mode = "markers", 
        marker = list(color = "black")) %>%
  add_markers(error_x = ~list(type = "data",
                              array = d$std.error*2, 
                              color = "black", 
                              width = 3,
                              thickness = 1),
              text = paste(d$std.error)) %>%
  layout(margin = list(l = 200),
         yaxis = list(title=""),
         xaxis = list(title="Estimate"))

13.5.2 Exercise: Dotplots & error bars

1. Use the code below to generate a categorical variable that has three values depending on the share of catholics.
2. Build three subsets (new datasets) of the dataset according to this variable.
3. For all 3 subsets generate coefficient plots with Fertility as outcome variable and Agriculture, Examination, Education as explanatory variables.
4. Visualize those coefficient plots as small multiples (make the graph publicaton ready!).

swiss$Catholic.cat <- cut(swiss$Catholic, breaks = c(-Inf, 30, 60, Inf), c("Low","Middle","High"))