# Card (1993) Table 1 – sample characteristics (cols 2–3 reproducible).
library(tidyverse)
library(here)
library(gt)
source(here("04-topics/rep-card1993/Rcode/card1993-data-prep.R"))
source(here("04-topics/rep-card1993/Rcode/card1993-gt-quarto.R"))

data <- load_card93_data()
stopifnot(nrow(data) == 3613L)

s2 <- data
s3 <- card93_wage_sample(data)
stopifnot(nrow(s3) == 3010L)

pct <- function(x) 100 * mean(x, na.rm = TRUE)
mean_num <- function(x) mean(x, na.rm = TRUE)

region_ne <- function(d) pct(d$reg661 + d$reg662)
region_midwest <- function(d) pct(d$reg663 + d$reg664)
region_west <- function(d) pct(d$reg668 + d$reg669)
region_south <- function(d) pct(d$south66)

age_pct <- function(d, lo, hi) {
  pct(d$age66 >= lo & d$age66 <= hi)
}

section_row <- function(label) {
  tibble(
    stub = label,
    col1 = " ",
    col2 = " ",
    col3 = " ",
    is_header = TRUE
  )
}

stat_row <- function(label, fn, digits = 1L, col1 = "–") {
  tibble(
    stub = label,
    col1 = col1,
    col2 = card93_fmt_num(fn(s2), digits),
    col3 = card93_fmt_num(fn(s3), digits),
    is_header = FALSE
  )
}

# Section headers in stub column preserve @tbl-card-1 row order (no gt row_group reordering).
rows <- bind_rows(
  section_row("1. Age Distribution in 1966:"),
  stat_row("Age 14-15 (%)", function(d) age_pct(d, 14, 15)),
  stat_row("Age 16-17", function(d) age_pct(d, 16, 17)),
  stat_row("Age 18-20", function(d) age_pct(d, 18, 20)),
  stat_row("Age 21-24", function(d) age_pct(d, 21, 24)),
  section_row("2. Regional Distribution in 1966:"),
  stat_row("Northeast (%)", region_ne),
  stat_row("Midwest", region_midwest),
  stat_row("South", region_south),
  stat_row("West", region_west),
  stat_row("3. Lived in SMSA 1966 (%)", function(d) pct(d$smsa66r)),
  stat_row("4. Lived Near 4-year College in 1966 (%)", function(d) pct(d$nearc4)),
  section_row("5. Family structure at Age 14:"),
  stat_row("Mother & Father (%)", function(d) pct(d$momdad14)),
  stat_row("Mother Only (%)", function(d) pct(d$sinmom14)),
  section_row("6. Average Parental Education"),
  stat_row("Mother's Education (yrs)", function(d) mean_num(d$momed), digits = 1L),
  stat_row("Father's Education (yrs)", function(d) mean_num(d$daded), digits = 1L),
  stat_row("7. Percent Black", function(d) pct(d$black)),
  stat_row("8. Average score on KWW Test", function(d) mean_num(d$kww), digits = 1L),
  stat_row("9. Interviewed in 1976 (%)", function(d) 100),
  stat_row("10. Mean Education in 1976", function(d) mean_num(d$ed76), digits = 1L),
  stat_row("11. Live in south in 1976 (%)", function(d) pct(d$reg76r)),
  tibble(
    stub = "12. Sample size",
    col1 = "5225",
    col2 = as.character(nrow(s2)),
    col3 = as.character(nrow(s3)),
    is_header = FALSE
  )
)

header_rows <- which(rows$is_header)

paper_targets <- tribble(
  ~row, ~col2, ~col3,
  "Age 14-15", 25.3, 25.5,
  "Age 16-17", 23.8, 24.1,
  "Age 18-20", 26.1, 24.6,
  "Age 21-24", 26.7, 25.8,
  "Northeast", 20.0, 20.7,
  "Midwest", 26.3, 26.0,
  "South", 41.3, 41.4,
  "West", 12.5, 11.9,
  "SMSA 1966", 64.3, 65.0,
  "Near college", 67.8, 68.2,
  "Mom & Dad", 79.2, 78.9,
  "Mom only", 10.0, 10.1,
  "Mom ed", 10.4, 10.3,
  "Dad ed", 10.0, 10.0,
  "Black", 23.0, 23.0,
  "KWW", 33.5, 33.5,
  "Mean ed76", 13.2, 13.3,
  "South 1976", 40.0, 40.3,
  "N col2", 3613, NA_real_,
  "N col3", NA_real_, 3010
)

replication <- tibble(
  col2_n = nrow(s2),
  col3_n = nrow(s3),
  age_14_15_c2 = age_pct(s2, 14, 15),
  age_14_15_c3 = age_pct(s3, 14, 15),
  nearc4_c2 = pct(s2$nearc4),
  nearc4_c3 = pct(s3$nearc4),
  black_c2 = pct(s2$black),
  mean_ed_c2 = mean_num(s2$ed76),
  mean_ed_c3 = mean_num(s3$ed76)
)

message("Table I replication (selected stats):")
print(replication)

table_data <- ak91_quarto_blank_df(rows |> select(stub, col1, col2, col3, is_header))

gt_tbl <- table_data |>
  gt() |>
  cols_hide(columns = is_header) |>
  cols_label(
    stub = "",
    col1 = html("(1)<br>Overall<br>NLS-YM"),
    col2 = html("(2)<br>1976 interview;<br>valid education"),
    col3 = html("(3)<br>Valid wage<br>&amp; education")
  )

gt_tbl <- card93_gt_standard(
  gt_tbl,
  stub_col = "stub",
  data_cols = c("col1", "col2", "col3"),
  header_rows = header_rows
)

gt_tbl <- gt_tbl |>
  card93_gt_source_notes(
    c(
      paste(
        "**Notes:** Means are based on all available valid observations",
        "in each subsample (Card, Table 1 notes)."
      )
    )
  ) |>
  card93_gt_col_footnote(
    "col1",
    paste(
      "Column (1) shows published values for the original NLS-YM cohort (N = 5225).",
      "The proximity extract (`nls.dat`) contains only the 1976 cross-section;",
      "column (1) cannot be fully replicated from the distributed data file."
    )
  ) |>
  card93_gt_stub_footnote(
    table_data,
    "2. Regional Distribution in 1966:",
    paste(
      "Northeast = `reg661` + `reg662`; Midwest = `reg663` + `reg664`;",
      "West = `reg668` + `reg669`; South = `south66`."
    )
  )

gt_tbl <- card93_gt_finalize(gt_tbl)

save(table_data, gt_tbl, replication, paper_targets, file = here("04-topics/rep-card1993/Rcode/Table_I.RData"))