📊 births per delivery (#3686)

* 📊 births per delivery * add ratio children/delivery * add rate
owid · Dec 3, 2024 · bbb6735 · bbb6735
1 parent a7ee8cf
commit bbb6735
Show file tree

Hide file tree

Showing 2 changed files with 19 additions and 6 deletions.
diff --git a/etl/steps/data/garden/demography/2024-11-26/multiple_births.meta.yml b/etl/steps/data/garden/demography/2024-11-26/multiple_births.meta.yml
@@ -89,6 +89,11 @@ tables:
         description_short: |-
           The rate of deliveries that are multiple deliveries, per 1,000 deliveries.
       children_delivery_ratio:
+        title: "Children per delivery"
+        unit: "children per 1,000 deliveries"
+        description_short: |-
+          The average number of children born per delivery. This is estimated by dividing the number of children born by the total number deliveries, per 1,000 people.
+      children_multiple_delivery_ratio:
         title: "Children per multiple delivery"
         unit: "children per 1,000 multiple deliveries"
         description_short: |-

diff --git a/etl/steps/data/garden/demography/2024-11-26/multiple_births.py b/etl/steps/data/garden/demography/2024-11-26/multiple_births.py
@@ -101,17 +101,24 @@ def run(dest_dir: str) -> None:
     # Estimate singleton_rate
     tb["singleton_rate"] = (1_000 * tb["singletons"] / tb["total_deliveries"]).round(2)
 
-    # Estimate children_per_delivery
-    tb["children_delivery_ratio"] = (1_000 * tb["multiple_children"] / tb["multiple_deliveries"]).round(3)
+    # Estimate ratios
+    tb["children_delivery_ratio"] = (
+        1_000 * (tb["multiple_children"] + tb["singletons"]) / tb["total_deliveries"]
+    ).round(3)
+    tb["children_multiple_delivery_ratio"] = (1_000 * tb["multiple_children"] / tb["multiple_deliveries"]).round(3)
     tb["multiple_to_singleton_ratio"] = (1_000 * tb["multiple_deliveries"] / tb["singletons"]).round(3)
 
     # Remove outliers
     flag = (tb["country"] == "England and Wales") & (tb["year"] == 1938)
-    assert (tb.loc[flag, "children_delivery_ratio"] >= 4000).all(), "Unexpected outlier for England and Wales in 1938"
-    tb.loc[flag, ["multiple_children", "children_delivery_ratio"]] = pd.NA
+    assert (
+        tb.loc[flag, "children_multiple_delivery_ratio"] >= 4000
+    ).all(), "Unexpected outlier for England and Wales in 1938"
+    tb.loc[flag, ["multiple_children", "children_multiple_delivery_ratio", "children_delivery_ratio"]] = pd.NA
     flag = (tb["country"] == "England and Wales") & (tb["year"] == 1939)
-    assert (tb.loc[flag, "children_delivery_ratio"] <= 1500).all(), "Unexpected outlier for England and Wales in 1938"
-    tb.loc[flag, ["multiple_children", "children_delivery_ratio"]] = pd.NA
+    assert (
+        tb.loc[flag, "children_multiple_delivery_ratio"] <= 1500
+    ).all(), "Unexpected outlier for England and Wales in 1938"
+    tb.loc[flag, ["multiple_children", "children_multiple_delivery_ratio", "children_delivery_ratio"]] = pd.NA
 
     # Keep relevant columns
     tb = tb[
@@ -130,6 +137,7 @@ def run(dest_dir: str) -> None:
             "multiple_rate",
             # Ratios
             "children_delivery_ratio",
+            "children_multiple_delivery_ratio",
             "multiple_to_singleton_ratio",
             # Births
             "multiple_children",