R语言cjoint包联合实验：按受访者特征分面绘图报错求助

Hey there! The error you're seeing happens because the plot() method for amce objects returns a base R plot, not a ggplot2 object. That means you can't use ggplot2's facet_grid() or facet_wrap() functions directly with it—those only work with ggplot objects.

Here are two solid ways to create faceted plots for your grouped AMCE results:

方法1：用cjoint原生工具生成多组图并组合 #

cjoint has a built-in multiplot() function that lets you combine multiple base R plots into one page. Here's how to use it:

1. Split your data by the grouping variable (e.g., econ_class)
2. Calculate AMCEs for each group separately
3. Generate a plot for each group
4. Combine them into a single page

# 1. Split your dataset by economic class
data_groups <- split(cdata_chosen, cdata_chosen$econ_class)

# 2. Calculate AMCEs for each group
amce_results <- lapply(data_groups, function(sub_df) {
  amce(
    chosen ~ gender + residence + party + partisanship + education + copartisan + profession_type + caste_type + pol_connection,
    data = sub_df,
    cluster = TRUE,
    respondent.id = "serial_id",
    design = "uniform"
  )
})

# 3. Create plots for each group
group_plots <- lapply(names(amce_results), function(group) {
  plot(
    amce_results[[group]],
    main = paste("Candidate Preference -", group),
    xlab = "Change in E[Y]",
    text.size = 8,
    point.size = 0.3,
    dodge.size = 0.9
  )
})

# 4. Combine plots into one page (adjust cols to control number of columns)
multiplot(plotlist = group_plots, cols = 2)

方法2：提取AMCE数据并用ggplot2手动绘图（更灵活） #

If you want full control over styling and faceting, extract the AMCE results into a data frame, then use ggplot2 to build your plot. This lets you use all ggplot2 features like facet_wrap() or facet_grid():

library(ggplot2)

# Define a helper function to extract AMCE results with group labels
get_group_amce <- function(sub_df, group_label) {
  amce_out <- amce(
    chosen ~ gender + residence + party + partisanship + education + copartisan + profession_type + caste_type + pol_connection,
    data = sub_df,
    cluster = TRUE,
    respondent.id = "serial_id",
    design = "uniform"
  )
  # Convert AMCE results to data frame and add group info
  amce_df <- as.data.frame(amce_out$amce)
  amce_df$econ_class <- group_label
  return(amce_df)
}

# Apply helper function to all groups and combine results
all_amce_data <- do.call(rbind, lapply(names(data_groups), function(g) {
  get_group_amce(data_groups[[g]], g)
}))

# Build faceted plot with ggplot2
ggplot(all_amce_data, aes(x = term, y = estimate, ymin = lower, ymax = upper)) +
  geom_pointrange(position = position_dodge(width = 0.9)) +
  coord_flip() +  # Match the vertical orientation of cjoint's default plot
  facet_wrap(~ econ_class) +
  labs(
    title = "Overall Candidate Preference by Economic Class",
    x = "Candidate Feature",
    y = "Change in E[Y]"
  ) +
  theme_bw() +
  theme(axis.text.y = element_text(size = 8))

小提示 #

• Method 1 keeps the exact styling of cjoint's default plots, which is great if you want consistency with your original figure.
• Method 2 gives you full control over colors, fonts, and layout—perfect if you need to customize the plot for publications or presentations.

内容的提问来源于stack exchange，提问作者Haseeb