DA0-002 Sample Questions

DA0-002 Sample Questions & Answers

Choosing statistical methods and troubleshooting analysis issues carries the top weight, alongside acquiring and transforming data, visual-element and delivery-method choices for reports, basic data concepts and infrastructure, and privacy practices.

Launch the full DA0-002 simulator →

Showing 6 of 12 free samples.

  1. Question 1Intermediate

    Data Concepts and Environments · Identify infrastructure concepts

    To ensure consistency across development and production environments, a data engineering team wants to package their Python-based data processing scripts along with all necessary libraries, dependencies, and a lightweight operating system. This package must be able to run identically on a developer's local laptop, an on-premises server, or in AWS. Which infrastructure concept accomplishes this goal?

    Show answer & explanation

    Correct answer: B

    Containerization (using tools like Docker) packages code, libraries, and dependencies into a single isolated unit called a container. This ensures the application runs identically across different computing environments (local, on-premises, or cloud). Block storage is a storage architecture. A Data Lakehouse is a data repository. Object storage handles unstructured data storage at scale.

    graph TD subgraph Application Code[Python Scripts] Libs[Pandas/Libraries] Bins[OS Binaries] end Application --> Container[Container Engine / Docker] Container --> OS[Host Operating System] OS --> Hardware[Infrastructure: Laptop / On-Prem / AWS]
  2. Question 2Beginner

    Data Concepts and Environments · Identify common data analysis tools

    True or False: MongoDB Compass is a graphical user interface specifically designed for querying and managing relational SQL databases such as PostgreSQL and Microsoft SQL Server.

    Show answer & explanation

    Correct answer: B

    This statement is False. MongoDB Compass is the official graphical user interface (GUI) specifically designed for MongoDB, which is a NoSQL document database. Tools like DBeaver, SQL Server Management Studio (SSMS), or pgAdmin are typically used for relational SQL databases like PostgreSQL or SQL Server.

  3. Question 3Intermediate

    Data Acquisition and Preparation · Given a scenario, use data acquisition methods

    A retail enterprise recently migrated to a highly scalable cloud data warehouse. They need to ingest massive volumes of raw web clickstream data quickly. To maximize efficiency, the data engineering team decides to load the raw data directly into the warehouse first, and then utilize the cloud warehouse's massive compute power to process and clean the data. Which integration pattern does this describe?

    Show answer & explanation

    Correct answer: B

    This describes ELT (Extract, Load, Transform). In modern cloud data warehouses, it is often more efficient to Extract raw data, Load it directly into the target system, and then use the target system's robust compute capabilities to Transform the data natively. ETL performs the transformation in a separate staging server before loading. Delta loads and full loads refer to the volume of data being moved, not the architectural pattern of the transformation.

    flowchart LR Source[(Data Sources)] -->|Extract| Pipeline[Data Pipeline] Pipeline -->|Load Raw Data| Target[(Cloud Data Warehouse)] Target -->|Transform Natively| Target
  4. Question 4Advanced

    Data Acquisition and Preparation · Given a scenario, use data acquisition methods

    A financial institution runs a critical end-of-day reconciliation query that joins a 500-million row Transactions table with a 10-million row Customers table. Recently, the query has been timing out before completion.

    Upon reviewing the code, you note the query currently uses SELECT *, filters the transactions by a non-indexed TransactionDate column to isolate just today's records, and then performs an INNER JOIN on CustomerID.

    The database server is under heavy load, and you cannot alter the table schema to add permanent tables, but you need to optimize this query to return results immediately. What is the BEST combination of optimization techniques to resolve the timeout?

    Show answer & explanation

    Correct answer: B

    This is the optimal combination. Indexing the 'TransactionDate' speeds up the filtering of the 500-million row table. Replacing 'SELECT *' reduces the I/O and memory overhead by only retrieving needed columns. Using a temporary table to isolate today's transactions creates a much smaller subset of records before attempting the resource-intensive join operation with the Customers table. The other options either create massive Cartesian products (CROSS JOIN), fail to reduce data overhead, or do not address the missing index issue.

  5. Question 5Intermediate

    Data Acquisition and Preparation · Given a scenario, perform appropriate data transformation and cleansing techniques

    An analyst is preparing a real estate dataset for a machine learning model. The HomePrice column has several missing values. Upon profiling the data, the analyst notices the distribution is heavily right-skewed because of a few ultra-luxury properties in the area. Which imputation technique is MOST appropriate to fill the missing values without artificially inflating the typical home value?

    Show answer & explanation

    Correct answer: B

    Median imputation is the best choice for heavily skewed data. The mean is highly sensitive to outliers (like the ultra-luxury properties), which would artificially inflate the imputed values. The median represents the middle value and is robust against extreme outliers. Mode imputation is generally used for categorical data, and record deletion reduces the dataset size, which may discard otherwise useful data.

  6. Question 6Beginner

    Data Acquisition and Preparation · Given a scenario, use data acquisition methods

    A marketing analyst needs to identify all registered users who have NEVER made a purchase. The database has two tables: Users (u) and Purchases (p). To retrieve records from the Users table that have no matching record in the Purchases table, the analyst writes the following query:

    SELECT u.UserID
    FROM Users u
    _____ JOIN Purchases p
    ON u.UserID = p.UserID
    WHERE p.PurchaseID IS NULL;

    Which type of JOIN correctly fills in the blank?

    Show answer & explanation

    Correct answer: C

    A LEFT JOIN (or LEFT OUTER JOIN) is required here. It returns all records from the left table (Users) and the matched records from the right table (Purchases). By adding the WHERE p.PurchaseID IS NULL clause, it filters the result to show ONLY the users who do not have a corresponding record in the Purchases table. An INNER JOIN would only return users who have made purchases. A CROSS JOIN returns a Cartesian product.

Ready for the real thing?

The full DA0-002 simulator has every exam-style question, timed mode, and instant scoring.

Go to the DA0-002 simulator →