Perform Data Exploration DA0-001 Practice Questions (Page 2)

Questions 6–10

1. 6application · medium

   A data engineer is optimizing a large dataset that stores website clickstream data. The dataset has a 'session_id' and a 'user_id' column. The engineer notices that 'user_id' is repeated for every click in a session. The goal is to reduce redundancy to improve query performance. What should the engineer do?

   • AKeep the data as is because redundancy is acceptable in clickstream data.
   • BMove 'user_id' to a separate session table and link it via 'session_id'.
   • CReplace 'user_id' with a hash of 'session_id'.
   • DRemove the 'user_id' column from the clickstream table.
   Reveal Solution Discussion Select an answer first
2. 7expert · hard

   A data analyst is exploring a dataset of customer transaction amounts. The dataset has 10,000 records. A box plot reveals several points beyond the upper whisker. The analyst needs to determine whether these points are true outliers or valid high-value transactions before deciding how to handle them. Which approach best supports this decision?

   • AInvestigate the source and business context of the flagged transactions before deciding on exclusion or transformation.
   • BApply a log transformation to the entire dataset to make the outliers less visible.
   • CImpute the flagged values with the median transaction amount to reduce their influence.
   • DRemove all points beyond the upper whisker to prevent them from skewing the mean.
   Reveal Solution Discussion Select an answer first
4. 8foundation · easy

   Which visualization is most effective for spotting outliers in a numeric variable?

   • ALine chart
   • BBar chart
   • CBox plot
   • DPie chart
   Reveal Solution Discussion Select an answer first
5. 9expert · hard

   A retail company's customer database contains duplicate records. A data analyst discovers that some duplicates have identical customer IDs but different email addresses, while others have the same email but different IDs. The analyst must consolidate the data without losing legitimate information. What is the best strategy?

   • AReplace all conflicting fields with the value from the most frequently occurring record.
   • BKeep the record with the most recent timestamp and delete all others.
   • CDelete all duplicates based on customer ID, keeping the first record in the dataset.
   • DMerge duplicate records by combining non-conflicting fields and flagging conflicting fields for review.
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   A data analyst is reviewing a dataset of employee records. The dataset includes 'employee_id', 'department_id', and 'department_name'. The analyst notices that 'department_name' is repeated for every employee in the same department. What is the most appropriate way to eliminate this redundancy?

   • AReplace 'department_name' with the department_id in the employee table.
   • BCreate a separate department table with unique department_id and department_name, and link it via department_id.
   • CDelete the 'department_name' column from the employee table.
   • DKeep the data as is because department_name is needed for analysis.
   Reveal Solution Discussion Select an answer first