Adding convert_cell_number to estimation and small changes to convert…

…_cell_number function
ONSdigital · Jan 28, 2025 · ada4855 · ada4855
1 parent 27793a5
commit ada4855
Show file tree

Hide file tree

Showing 4 changed files with 14 additions and 11 deletions.
diff --git a/mbs_results/estimation/pre_processing_estimation.py b/mbs_results/estimation/pre_processing_estimation.py
@@ -1,4 +1,5 @@
 from mbs_results.utilities.utils import read_colon_separated_file
+from mbs_results.staging.data_cleaning import convert_cell_number
 
 
 def get_estimation_data(
@@ -90,7 +91,7 @@ def derive_estimation_variables(
     reference: Str
         the name of the reference column
     **config: Dict
-       main pipeline configuration. Can be used to input the entire config dictionary
+        main pipeline configuration. Can be used to input the entire config dictionary
 
     Returns
     -------
@@ -99,6 +100,7 @@ def derive_estimation_variables(
 
     """
 
+    population_frame = convert_cell_number(population_frame, cell_number)
 
     population_frame = population_frame.merge(
         calibration_group_map, on=[cell_number], how="left"

diff --git a/mbs_results/staging/data_cleaning.py b/mbs_results/staging/data_cleaning.py
@@ -379,22 +379,23 @@ def convert_cell_number(
     cell_number: str
 ):
     """
-
+    Convert NI and GB cell numbers to UK by changing the first digit to 5 if this is 7.
+    
     Parameters
     ----------
     df : pd.DataFrame
-        DESCRIPTION.
-    cell_number : TYPE
-        DESCRIPTION.
-     : TYPE
-        DESCRIPTION.
+        Dataframe with cell_number column to convert
+    cell_number : str
+        Column name for cell_number in df
 
     Returns
     -------
-    None.
+    pd.DataFrame
+        Dataframe with converted cell_number column (and original cell_number
+        in separate column)                                                     
 
     """
-    df["ni_uk_cell_number"] = df[cell_number]
+    df["ni_gb_cell_number"] = df[cell_number]
     df[cell_number] = (
         df[cell_number]
         .astype(str)

diff --git a/tests/data/staging/data_cleaning/test_convert_cell_number.csv b/tests/data/staging/data_cleaning/test_convert_cell_number.csv
@@ -1,4 +1,4 @@
-cell_number,ni_uk_cell_number
+cell_number,ni_gb_cell_number
 1111,1111
 2222,2222
 3333,3333

diff --git a/tests/staging/test_data_cleaning.py b/tests/staging/test_data_cleaning.py
@@ -116,7 +116,7 @@ def test_convert_cell_number(filepath):
 
     expected_output = pd.read_csv(filepath / "test_convert_cell_number.csv")
 
-    df_in = expected_output.drop(columns=["cell_number"]).rename(columns = {"ni_uk_cell_number" : "cell_number"})
+    df_in = expected_output.drop(columns=["cell_number"]).rename(columns = {"ni_gb_cell_number" : "cell_number"})
 
     actual_output = convert_cell_number(df_in, "cell_number")