R new syntax for FCSN

Indicators/Food-consumption-score-nutrition/FCSNutrition.R

@@ -0,0 +1,142 @@
+# ------------------------------------------------------------
+# WFP Standardized Scripts - Food Consumption Score - Nutrition (FCS-N)
+# Last Updated: January 2026
+# ------------------------------------------------------------
+
+library(dplyr)
+
+# Required subgroup variables (FCS-N module)
+fcsn_vars <- c(
+  "FCSNPrMeatF","FCSNPrMeatO","FCSNPrFish","FCSNPrEggs",
+  "FCSNVegOrg","FCSNVegGre","FCSNFruiOrg"
+)
+
+# Required main-group variables from the standard FCS module (for flags + aggregates)
+# Protein main group: FCSPr; Vegetable main group: FCSVeg; Fruit main group: FCSFruit
+# Also used in aggregates: FCSDairy, FCSPulse
+main_needed <- c("FCSPr","FCSVeg","FCSFruit","FCSDairy","FCSPulse")
+
+# ------------------------------------------------------------
+# 0) (Optional) quick descriptive check of subgroup distributions
+# ------------------------------------------------------------
+fcsn_stats_before <- df %>%
+  summarise(across(all_of(fcsn_vars),
+                   list(min = ~min(as.numeric(.x), na.rm = TRUE),
+                        max = ~max(as.numeric(.x), na.rm = TRUE),
+                        mean = ~mean(as.numeric(.x), na.rm = TRUE))))
+fcsn_stats_before
+
+# ------------------------------------------------------------
+# 1) Recode subgroup missing to 0 (Survey Designer behavior)
+# ------------------------------------------------------------
+df <- df %>%
+  mutate(across(all_of(fcsn_vars), ~ as.numeric(.x))) %>%
+  mutate(across(all_of(fcsn_vars), ~ ifelse(is.na(.x), 0, .x)))
+
+# ------------------------------------------------------------
+# 2) Clean impossible subgroup values
+# ------------------------------------------------------------
+df <- df %>%
+  mutate(across(all_of(fcsn_vars), ~ ifelse(.x < 0 | .x >= 8, NA_real_, .x)))
+
+# ------------------------------------------------------------
+# 3) Flags: subgroup exceeds main group
+# ------------------------------------------------------------
+df <- df %>%
+  mutate(
+    FCSN_flag_protein = ifelse(
+      !is.na(FCSPr) & (
+        ( !is.na(FCSNPrMeatF) & FCSNPrMeatF > FCSPr ) |
+          ( !is.na(FCSNPrMeatO) & FCSNPrMeatO > FCSPr ) |
+          ( !is.na(FCSNPrFish)  & FCSNPrFish  > FCSPr ) |
+          ( !is.na(FCSNPrEggs)  & FCSNPrEggs  > FCSPr )
+      ),
+      1L, 0L
+    ),
+    FCSN_flag_veg = ifelse(
+      !is.na(FCSVeg) & (
+        ( !is.na(FCSNVegOrg) & FCSNVegOrg > FCSVeg ) |
+          ( !is.na(FCSNVegGre) & FCSNVegGre > FCSVeg )
+      ),
+      1L, 0L
+    ),
+    FCSN_flag_fruit = ifelse(
+      !is.na(FCSFruit) & !is.na(FCSNFruiOrg) & (FCSNFruiOrg > FCSFruit),
+      1L, 0L
+    )
+  )
+
+# Flag distributions
+table(df$FCSN_flag_protein, useNA = "ifany")
+table(df$FCSN_flag_veg,     useNA = "ifany")
+table(df$FCSN_flag_fruit,   useNA = "ifany")
+
+# ------------------------------------------------------------
+# 4) Compute nutrient aggregates (Vitamin A, Protein, Hem Iron)
+# ------------------------------------------------------------
+
+# Helper: row sum that returns NA when all inputs are NA
+rowSum_na <- function(mat) {
+  s <- rowSums(mat, na.rm = TRUE)
+  all_na <- rowSums(!is.na(mat)) == 0
+  s[all_na] <- NA_real_
+  s
+}
+
+df <- df %>%
+  mutate(
+    FGVitA = rowSum_na(cbind(FCSDairy, FCSNPrMeatO, FCSNPrEggs, FCSNVegOrg, FCSNVegGre, FCSNFruiOrg)),
+    FGProtein = rowSum_na(cbind(FCSPulse, FCSDairy, FCSNPrMeatF, FCSNPrMeatO, FCSNPrFish, FCSNPrEggs)),
+    FGHIron = rowSum_na(cbind(FCSNPrMeatF, FCSNPrMeatO, FCSNPrFish))
+  )
+
+# ------------------------------------------------------------
+# 5) Recode aggregates into 3 categories:
+#    0 -> 1 (never), 1-6 -> 2 (sometimes), 7-42 -> 3 (daily)
+# ------------------------------------------------------------
+df <- df %>%
+  mutate(
+    VitA_Cat = case_when(
+      is.na(FGVitA)        ~ NA_integer_,
+      FGVitA == 0          ~ 1L,
+      FGVitA >= 1 & FGVitA <= 6  ~ 2L,
+      FGVitA >= 7 & FGVitA <= 42 ~ 3L,
+      TRUE ~ NA_integer_   # outside expected range
+    ),
+    Protein_Cat = case_when(
+      is.na(FGProtein)        ~ NA_integer_,
+      FGProtein == 0          ~ 1L,
+      FGProtein >= 1 & FGProtein <= 6  ~ 2L,
+      FGProtein >= 7 & FGProtein <= 42 ~ 3L,
+      TRUE ~ NA_integer_
+    ),
+    Haem_iron_Cat = case_when(
+      is.na(FGHIron)        ~ NA_integer_,
+      FGHIron == 0          ~ 1L,
+      FGHIron >= 1 & FGHIron <= 6  ~ 2L,
+      FGHIron >= 7 & FGHIron <= 42 ~ 3L,
+      TRUE ~ NA_integer_
+    ),
+
+    # Apply labels as factors (mirrors SPSS VALUE LABELS)
+    VitA_Cat = factor(VitA_Cat, levels = c(1,2,3),
+                      labels = c("0 time (never consumed)",
+                                 "1-6 times (consumed sometimes)",
+                                 "7 times or more (consumed at least daily)")),
+    Protein_Cat = factor(Protein_Cat, levels = c(1,2,3),
+                         labels = c("0 time (never consumed)",
+                                    "1-6 times (consumed sometimes)",
+                                    "7 times or more (consumed at least daily)")),
+    Haem_iron_Cat = factor(Haem_iron_Cat, levels = c(1,2,3),
+                           labels = c("0 time (never consumed)",
+                                      "1-6 times (consumed sometimes)",
+                                      "7 times or more (consumed at least daily)"))
+  )
+
+# ------------------------------------------------------------
+# 6) Check results
+# ------------------------------------------------------------
+table(df$VitA_Cat,      useNA = "ifany")
+table(df$Protein_Cat,   useNA = "ifany")
+table(df$Haem_iron_Cat, useNA = "ifany")
+