DA0-002 Data Acquisition and Preparation Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  df['sales'].apply(sum)

  Why it's wrong here

  apply(sum) sums each element of the sales Series individually, returning the same values rather than totals per product. It is tempting because apply is the general mechanism for running functions across a Series or DataFrame. Aggregating sales per product requires groupby('product')['sales'].sum().

• ✗

  df.pivot_table(values='sales', index='product', aggfunc='sum')

  Why it's wrong here

  Although `pivot_table` can aggregate data, it requires a unique index–column combination to reshape the table, whereas the question only needs a simple grouped sum without restructuring. It is tempting because `pivot_table` does accept `aggfunc='sum'`, making it appear suitable for aggregation; it would be correct if the goal were to reorganise the DataFrame into a cross-tabulation with multiple column categories, not merely to compute totals per product.

• ✓

  df.groupby('product')['sales'].sum()

  Why this is correct

  groupby('product') partitions rows by category, then selecting ['sales'] and calling sum() aggregates each group into total sales per product. This satisfies the per-product aggregation requirement; pivot_table or value_counts would reshape or count rather than sum the sales column.

• ✗

  df.merge(df, on='product')

  Why it's wrong here

  merge joins DataFrames on a key and returns matched rows, so merging df with itself duplicates rows rather than aggregating sales. It is tempting because merge is the tool for combining datasets, which is correct when relating separate tables. Grouping by product and summing sales requires groupby with sum.