1.1 Prepare and Process Data ASSOCIATE-DATA-PRACTITIONER Practice Questions (Page 5)

Questions 21–24

1. 21foundation · easy

   A data analyst is reviewing a dataset and notices that some customer records have the same email address but different customer IDs. Which data quality issue does this represent?

   • AOutliers
   • BMissing values
   • CInconsistencies
   • DDuplicates
   Reveal Solution Discussion Select an answer first
2. 22application · medium

   A data analyst is reviewing a table of sales transactions and finds that the 'quantity' column contains negative values, which are invalid. They also notice that some rows have duplicate transaction IDs. Which SQL approach should they use to clean the data?

   • ASELECT DISTINCT * FROM sales WHERE quantity > 0;
   • BCREATE OR REPLACE TABLE sales AS SELECT * EXCEPT(row_num) FROM (SELECT *, ROW_NUMBER() OVER (PARTITION BY transaction_id ORDER BY timestamp) AS row_num FROM sales WHERE quantity >= 0) WHERE row_num = 1;
   • CDELETE FROM sales WHERE quantity < 0; DELETE FROM sales WHERE transaction_id IN (SELECT transaction_id FROM sales GROUP BY transaction_id HAVING COUNT(*) > 1);
   • DUPDATE sales SET quantity = ABS(quantity) WHERE quantity < 0; DELETE FROM sales WHERE transaction_id IN (SELECT transaction_id FROM sales GROUP BY transaction_id HAVING COUNT(*) > 1);
   Reveal Solution Discussion Select an answer first
4. 23application · medium

   A streaming pipeline in Dataflow receives clickstream events. Some events have a missing user_id, and the team wants to drop those events but log them to a dead-letter topic for later analysis. What is the best approach?

   • AUse a Combine.PerKey to aggregate events and then filter nulls.
   • BUse a GroupByKey to combine events with the same user_id and then discard nulls.
   • CUse a ParDo with a DoFn that checks for null user_id and outputs to a side output for invalid events.
   • DUse a Filter transform to remove null user_id events and ignore them.
   Reveal Solution Discussion Select an answer first
5. 24expert · hard

   A data team is using Cloud Data Fusion to build a data quality pipeline. They need to detect outliers in a numeric column, but the definition of an outlier depends on the category of the record. For example, a sales amount of $10,000 might be normal for one product category but an outlier for another. What is the best way to implement this?

   • AUse a Wrangler transform to manually inspect and flag outliers.
   • BUse a Joiner to combine the data with a reference table of thresholds.
   • CUse a Conditional plugin to split data by category, then apply a Validator with category-specific thresholds to each branch.
   • DUse a single Validator plugin with a fixed threshold for all records.
   Reveal Solution Discussion Select an answer first