Free Google Cloud Associate Data Practitioner exam Practice Questions

Sample Practice Questions

Try these 10 sample questions from the Google Cloud Associate Data Practitioner exam practice bank. Each includes the answer and explanation. Continue with the interactive practice above to review the full question set.

1. 1What distinguishes ETL from ELT when preparing data for a destination system?

   1. A. ETL transforms data before loading; ELT transforms it after loading.
   2. B. ETL loads raw data first; ELT transforms it before loading.
   3. C. ETL extracts data after transformation; ELT extracts it before transformation.
   4. D. ETL and ELT both transform data before loading it into the destination.

   Answer: A

   Explanation: Google states, "ETL processes data by transforming it prior to loading it into the destination system." It adds, "In ELT, data is loaded into the target system in its raw format and then transformed." The position of transformation relative to loading distinguishes these methods.

   Reference: What is ETL? — ETL vs. ELT

2. 2An analyst needs only sales rows whose status equals 'completed' before computing totals. Which GoogleSQL clause applies that row filter?

   1. A. ORDER BY status
   2. B. GROUP BY status
   3. C. WHERE status = 'completed'
   4. D. LIMIT 100

   Answer: C

   Explanation: Google states, "The WHERE clause filters the results of the FROM clause." Only rows whose predicate evaluates to TRUE are retained. Grouping and sorting do not select only the requested status, and LIMIT restricts quantity rather than status.

   Reference: Query syntax — WHERE clause

3. 3A team must move existing Apache Spark and Hadoop transformation jobs to managed Google Cloud clusters without rewriting them for a different processing framework. Which service matches this requirement?

   1. A. Dataform
   2. B. Cloud Composer
   3. C. Dataflow
   4. D. Dataproc (Managed Service for Apache Spark)

   Answer: D

   Explanation: Google states, "You don't need to learn new tools or APIs to use Managed Service for Apache Spark on clusters, which lets you move existing projects into Managed Service for Apache Spark on clusters without redevelopment." The service runs Spark and Hadoop workloads. Dataflow uses Beam, Dataform manages BigQuery SQL, and Composer orchestrates tasks rather than replacing the Spark runtime.

   Reference: Managed Service for Apache Spark on clusters overview — Advantages > Simple and familiar

4. 4A production data service account needs a small set of permissions already covered by a service-specific predefined role. Which approach best follows least privilege?

   1. A. Grant Owner to cover future service requirements.
   2. B. Grant Editor because the account runs automated jobs.
   3. C. Grant the limited predefined role that meets the task requirements.
   4. D. Grant Viewer plus Editor to separate reading from writing.

   Answer: C

   Explanation: Google advises, "Instead, grant the most limited predefined roles or custom roles that meet your needs." The available service-specific role meets the actual task. Owner and Editor include unnecessarily broad permissions, and adding Viewer to Editor does not narrow Editor.

   Reference: Role types — Basic roles > Caution

5. 5A team needs a managed transfer of 20 TiB of objects from Amazon S3 to Cloud Storage. Which service directly supports this source and destination?

   1. A. BigQuery Data Transfer Service
   2. B. Database Migration Service
   3. C. Storage Transfer Service
   4. D. Datastream

   Answer: C

   Explanation: Storage Transfer Service lists "Amazon S3, Azure Blob Storage, or Cloud Storage to Cloud Storage" among its supported movements. This is an object-storage transfer, rather than a warehouse ingestion, database migration, or database change-streaming task.

   Reference: Storage Transfer Service — Storage Transfer Service, supported storage systems

6. 6A query groups sales by customer_id and calculates SUM(amount). Which clause keeps only customer groups whose total exceeds 1000?

   1. A. HAVING SUM(amount) > 1000
   2. B. WHERE amount > 1000
   3. C. ORDER BY SUM(amount) DESC
   4. D. LIMIT 1000

   Answer: A

   Explanation: Google states, "The HAVING clause filters the results produced by GROUP BY or aggregation." HAVING applies the condition to each aggregated customer total. WHERE amount > 1000 filters individual sales, sorting keeps lower-total groups, and LIMIT does not impose a total threshold.

   Reference: Query syntax — HAVING clause

7. 7A team has Apache Beam transformation code and wants a managed execution service supporting both streaming and batch pipelines. Which service should execute this code?

   1. A. Dataflow
   2. B. Dataform
   3. C. Cloud Composer
   4. D. Storage Transfer Service

   Answer: A

   Explanation: Google states, "Dataflow executes Apache Beam pipelines." It also provides "unified stream and batch data processing at scale." Dataflow directly executes the Beam transformations; the alternatives manage SQL workflows, orchestrate tasks, or transfer stored data.

   Reference: Dataflow overview — Dataflow overview; Portable with Apache Beam

8. 8Which statement correctly describes Google Cloud IAM predefined roles?

   1. A. They provide service-specific granular access and are maintained by Google.
   2. B. They are customer-maintained permission sets created separately in each project.
   3. C. They provide the same broad cross-service access as Owner and Editor.
   4. D. They are individual permissions that cannot contain multiple operations.

   Answer: A

   Explanation: Google describes predefined roles as roles "which provide granular access for a specific service and are managed by Google Cloud." They differ from customer-managed custom roles, broad basic roles, and individual permissions.

   Reference: Role types — Role types

9. 9A customer dataset has blank values in a required delivery-address field. Which data quality dimension most directly measures this problem?

   1. A. Volume
   2. B. Completeness
   3. C. Freshness
   4. D. Consistency

   Answer: B

   Explanation: The documentation states, "Completeness assesses whether the data contains all of the information that's required for its intended purpose." The required delivery field is missing information. Volume measures expected data quantity, freshness measures update age, and consistency compares values across instances.

   Reference: Auto data quality overview — Dimensions > Completeness

10. 10A report must include every customer, including customers with no orders, and any matching order details. Customers is the left input and Orders is the right input. Which join should be used on customer_id?

    1. A. INNER JOIN
    2. B. RIGHT JOIN
    3. C. CROSS JOIN
    4. D. LEFT JOIN

    Answer: D

    Explanation: Google states, "All rows from the left from_item are retained". Unmatched right-side columns receive NULL values. A left join therefore retains customers without orders. An inner join drops them, a right join preserves orders instead, and a cross join does not match on customer_id.

    Reference: Query syntax — LEFT [OUTER] JOIN