Testing model outputs in dbt - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the main purpose of testing model outputs in dbt?

To ensure that the data produced by your models is accurate, consistent, and meets expected quality standards before using it for analysis or reporting.

beginner

Name two common types of tests used in dbt to validate model outputs.

1. Unique tests - to check if a column has unique values.<br>2. Not null tests - to ensure no missing values in a column.

intermediate

How does a dbt 'schema test' help in testing model outputs?

Schema tests check the structure and data quality of your model outputs by validating constraints like uniqueness, non-null values, and referential integrity.

intermediate

What is a 'custom test' in dbt and when would you use it?

A custom test is a user-defined SQL query that checks specific conditions in your data. Use it when built-in tests don't cover your unique data quality rules.

beginner

Why is it important to run tests regularly on model outputs in dbt?

Regular testing catches data issues early, prevents bad data from spreading, and maintains trust in your data pipelines and reports.

Which dbt test checks if a column contains no missing values?

A Referential integrity test

B Unique test

C Not null test

D Custom test

The not null test ensures that a column has no missing (null) values.

What does a unique test in dbt verify?

A That data types are correct

B That no values are missing

C That values match another table

D That all values in a column are different

A unique test checks that all values in a column are distinct with no duplicates.

When should you create a custom test in dbt?

A When you want to speed up model runs

B When built-in tests do not cover your specific data rules

C When you want to change data types

D When you want to create new models

Custom tests are used to check data conditions that built-in tests do not support.

What is the benefit of running tests on model outputs before analysis?

A To ensure data accuracy and reliability

B To reduce storage space

C To speed up SQL queries

D To create visualizations

Testing ensures the data is accurate and reliable for analysis.

Which of the following is NOT a typical dbt test?

A Data type test

B Unique test

C Not null test

D Referential integrity test

dbt does not have a built-in data type test; it focuses on uniqueness, nulls, and referential integrity.

Explain how you would use dbt tests to ensure your model outputs are trustworthy.

Think about built-in tests and when to create your own.

Describe the difference between a schema test and a custom test in dbt.

Consider what each test type validates.