setwd("C:/Users/irina/Documents/DND/EOHI/eohi2")

options(scipen = 999)

df <- read.csv("eohi2.csv")
library(psych)
library(dplyr)
library(knitr)
library(irr)

# Your named variable sets (replace df with your actual dataframe name)
present_pref_vars <- c("present_pref_read", "present_pref_music", "present_pref_tv", "present_pref_nap", "present_pref_travel")
past_5_pref_vars  <- c("past_5_pref_read", "past_5_pref_music", "past_5_pref_TV", "past_5_pref_nap", "past_5_pref_travel")
past_10_pref_vars <- c("past_10_pref_read", "past_10_pref_music", "past_10_pref_TV", "past_10_pref_nap", "past_10_pref_travel")
fut_5_pref_vars   <- c("fut_5_pref_read", "fut_5_pref_music", "fut_5_pref_TV", "fut_5_pref_nap", "fut_5_pref_travel")
fut_10_pref_vars  <- c("fut_10_pref_read", "fut_10_pref_music", "fut_10_pref_TV", "fut_10_pref_nap", "fut_10_pref_travel")

present_pers_vars <- c("present_pers_extravert", "present_pers_critical", "present_pers_dependable", "present_pers_anxious", "present_pers_complex")
past_5_pers_vars  <- c("past_5_pers_extravert", "past_5_pers_critical", "past_5_pers_dependable", "past_5_pers_anxious", "past_5_pers_complex")
past_10_pers_vars <- c("past_10_pers_extravert", "past_10_pers_critical", "past_10_pers_dependable", "past_10_pers_anxious", "past_10_pers_complex")
fut_5_pers_vars   <- c("fut_5_pers_extravert", "fut_5_pers_critical", "fut_5_pers_dependable", "fut_5_pers_anxious", "fut_5_pers_complex")
fut_10_pers_vars  <- c("fut_10_pers_extravert", "fut_10_pers_critical", "fut_10_pers_dependable", "fut_10_pers_anxious", "fut_10_pers_complex")

present_val_vars <- c("present_val_obey", "present_val_trad", "present_val_opinion", "present_val_performance", "present_val_justice")
past_5_val_vars  <- c("past_5_val_obey", "past_5_val_trad", "past_5_val_opinion", "past_5_val_performance", "past_5_val_justice")
past_10_val_vars <- c("past_10_val_obey", "past_10_val_trad", "past_10_val_opinion", "past_10_val_performance", "past_10_val_justice")
fut_5_val_vars   <- c("fut_5_val_obey", "fut_5_val_trad", "fut_5_val_opinion", "fut_5_val_performance", "fut_5_val_justice")
fut_10_val_vars  <- c("fut_10_val_obey", "fut_10_val_trad", "fut_10_val_opinion", "fut_10_val_performance", "fut_10_val_justice")

all_scales <- list(
  "Present_Preferences" = present_pref_vars,
  "Past5_Preferences"   = past_5_pref_vars,
  "Past10_Preferences"  = past_10_pref_vars,
  "Fut5_Preferences"    = fut_5_pref_vars,
  "Fut10_Preferences"   = fut_10_pref_vars,
  "Present_Personality" = present_pers_vars,
  "Past5_Personality"   = past_5_pers_vars,
  "Past10_Personality"  = past_10_pers_vars,
  "Fut5_Personality"    = fut_5_pers_vars,
  "Fut10_Personality"   = fut_10_pers_vars,
  "Present_Values"      = present_val_vars,
  "Past5_Values"        = past_5_val_vars,
  "Past10_Values"       = past_10_val_vars,
  "Fut5_Values"         = fut_5_val_vars,
  "Fut10_Values"        = fut_10_val_vars
)

# Reliability analysis loop
alpha_results <- list()
summary_data <- data.frame()
item_data <- data.frame()
icc_data <- data.frame()

for(scale_name in names(all_scales)) {
  vars <- all_scales[[scale_name]]
  scale_data <- df %>% select(all_of(vars))
  
  # Only run if there is more than one column present
  if(ncol(scale_data) > 1) {
    alpha_out <- psych::alpha(scale_data, check.keys = TRUE)
    alpha_results[[scale_name]] <- alpha_out
    
    # Print full output for diagnostics
    cat("\n----------", scale_name, "----------\n")
    print(alpha_out$total)
    print(alpha_out$item.stats)
    print(alpha_out$alpha.drop)
    
    # Calculate ICC(2,1)
    tryCatch({
      icc_result <- icc(scale_data, model = "twoway", type = "consistency", unit = "single")
      cat("ICC(2,1) for", scale_name, ":", round(icc_result$value, 4), "\n")
      
      # Collect ICC data for HTML
      icc_row <- data.frame(
        Scale = scale_name,
        ICC_2_1 = icc_result$value,
        ICC_CI_Lower = icc_result$lbound,
        ICC_CI_Upper = icc_result$ubound,
        F_Value = icc_result$Fvalue,
        p_Value = icc_result$p.value,
        stringsAsFactors = FALSE
      )
      icc_data <- rbind(icc_data, icc_row)
    }, error = function(e) {
      cat("ICC calculation failed for", scale_name, ":", e$message, "\n")
    })
    
    # Collect summary data for HTML
    total_row <- alpha_out$total
    total_row$Scale <- scale_name
    summary_data <- rbind(summary_data, total_row)
    
    # Debug: print what we're collecting
    cat("Collecting data for", scale_name, "- rows in summary_data:", nrow(summary_data), "\n")
    
    # Collect item data for HTML
    item_row <- alpha_out$item.stats
    alpha_drop_row <- alpha_out$alpha.drop
    item_row$Scale <- scale_name
    item_row$AlphaIfDropped <- alpha_drop_row$raw_alpha
    item_data <- rbind(item_data, item_row)
  }
}

# Debug: check summary_data
cat("Final summary_data has", nrow(summary_data), "rows\n")
if(nrow(summary_data) > 0) {
  cat("Column names:", paste(colnames(summary_data), collapse = ", "), "\n")
}

# Create simple HTML report
html_content <- paste0("<!DOCTYPE html>
<html>
<head>
    <title>Reliability Analysis Results</title>
    <style>
        body { font-family: Arial, sans-serif; margin: 20px; }
        h1 { color: #2c3e50; }
        h2 { color: #34495e; border-bottom: 2px solid #3498db; padding-bottom: 5px; }
        table { border-collapse: collapse; width: 100%; margin: 20px 0; }
        th, td { border: 1px solid #ddd; padding: 8px; text-align: left; }
        th { background-color: #3498db; color: white; }
        tr:nth-child(even) { background-color: #f2f2f2; }
        .summary-stats { background-color: #ecf0f1; padding: 15px; border-radius: 5px; }
    </style>
</head>
<body>
    <h1>Reliability Analysis Results</h1>
    <p>Generated on: ", Sys.time(), "</p>
    <h2>Overall Scale Statistics</h2>
    <div class='summary-stats'>
    <table>
    <tr><th>Scale</th><th>Raw Alpha</th><th>Std Alpha</th><th>G6(SMC)</th><th>Average r</th><th>S/N</th><th>ASE</th><th>Mean</th><th>SD</th><th>Median r</th></tr>")

for(i in 1:nrow(summary_data)) {
  row <- summary_data[i,]
  html_content <- paste0(html_content, sprintf("<tr><td>%s</td><td>%.4f</td><td>%.4f</td><td>%.4f</td><td>%.4f</td><td>%.4f</td><td>%.4f</td><td>%.4f</td><td>%.4f</td><td>%.4f</td></tr>",
    row$Scale, row$raw_alpha, row$std.alpha, row$G6.smc, row$average_r, row$S.N, row$ase, row$mean, row$sd, row$median_r))
}

html_content <- paste0(html_content, "</table></div>")

# Add item statistics with alpha if dropped
html_content <- paste0(html_content, "<h2>Item Statistics</h2>
    <table>
    <tr><th>Scale</th><th>Item</th><th>n</th><th>Raw r</th><th>Std r</th><th>r.cor</th><th>r.drop</th><th>Mean</th><th>SD</th><th>Alpha if Dropped</th></tr>")

for(i in 1:nrow(item_data)) {
  row <- item_data[i,]
  html_content <- paste0(html_content, sprintf("<tr><td>%s</td><td>%s</td><td>%.0f</td><td>%.4f</td><td>%.4f</td><td>%.4f</td><td>%.4f</td><td>%.4f</td><td>%.4f</td><td>%.4f</td></tr>",
    row$Scale, rownames(row), row$n, row$raw.r, row$std.r, row$r.cor, row$r.drop, row$mean, row$sd, row$AlphaIfDropped))
}

html_content <- paste0(html_content, "</table>
</body>
</html>")

# Write HTML file
writeLines(html_content, "reliability_analysis_report.html")


# Check for reversed items
cat("\n=== REVERSED ITEMS CHECK ===\n")
for(scale_name in names(all_scales)) {
  vars <- all_scales[[scale_name]]
  scale_data <- df %>% select(all_of(vars))
  
  if(ncol(scale_data) > 1) {
    alpha_out <- alpha_results[[scale_name]]
    
    # Check if any items were reversed by looking at the keys
    if(!is.null(alpha_out$keys)) {
      keys_df <- as.data.frame(alpha_out$keys)
      reversed_items <- rownames(keys_df)[keys_df[,1] < 0]
      if(length(reversed_items) > 0) {
        cat(scale_name, "reversed items:", paste(reversed_items, collapse = ", "), "\n")
      } else {
        cat(scale_name, ": No items reversed\n")
      }
    } else {
      cat(scale_name, ": No keys available\n")
    }
  }
}