How to create columns in pandas

Raw tables often contain the quantities needed for analysis but not the derived values a report or export requires. pandas can calculate those values as new DataFrame columns without iterating over rows.

Bracket assignment writes one column at a time to the existing DataFrame. A scalar is broadcast to every row, while a Series or vectorized expression supplies one value per row; assigning a new label appends it to the column order.

Reusing an existing label replaces that column, and an assigned Series aligns by index label rather than row position. pandas 3 also rejects chained assignment for changes to a subset, so conditional updates must target the DataFrame in one statement.

1. Save the order data and DataFrame constructor in create_columns.py.

   create_columns.py

   import pandas as pd
    
   df = pd.DataFrame(
       {
           "item": ["notebook", "pencil", "eraser"],
           "qty": [3, 10, 5],
           "unit_price": [2.5, 0.4, 0.8],
       }
   )

2. Append the calculated line_total column beneath the DataFrame constructor.

   df["line_total"] = df["qty"] * df["unit_price"]

3. Append the scalar currency column beneath the calculated column.

   df["currency"] = "USD"

   pandas broadcasts the scalar string to every row.

4. Append the boolean bulk_order column beneath the scalar column.

   df["bulk_order"] = df["qty"].ge(10)

   Series.ge() returns one boolean per row, so this assignment avoids row-by-row loops and chained updates.
   Related: How to migrate pandas code for Copy-on-Write

5. Append the assertion and output section beneath the column assignments.

   expected_columns = [
       "item",
       "qty",
       "unit_price",
       "line_total",
       "currency",
       "bulk_order",
   ]
    
   assert df.columns.tolist() == expected_columns
   assert df["line_total"].tolist() == [7.5, 4.0, 4.0]
   assert df["currency"].eq("USD").all()
   assert df["bulk_order"].tolist() == [False, True, False]
   assert str(df["line_total"].dtype) == "float64"
   assert str(df["bulk_order"].dtype) == "bool"
    
   result_columns = ["item", "line_total", "currency", "bulk_order"]
    
   print(df[result_columns].to_string(index=False))
   print()
   print("Dtypes:")
   print(df[["line_total", "bulk_order"]].dtypes)

   A wrong column order, value, or dtype raises AssertionError before the table is printed.

6. Verify the calculated, scalar, and boolean columns with the completed create_columns.py script.

   $ python3 create_columns.py
       item  line_total currency  bulk_order
   notebook         7.5      USD       False
     pencil         4.0      USD        True
     eraser         4.0      USD       False
   
   Dtypes:
   line_total    float64
   bulk_order       bool
   dtype: object