DAC Cheat Sheet 2026

The 30 highest-yield DAC facts, distilled from real exam questions. Print it, save it as a PDF, or study it here — free, no sign-up.

1. A report double-counts revenue because the same order appears twice. What cleaning step fixes this? → Deduplication
2. A histogram is most useful for showing what about a single numeric variable? → Its distribution
3. Which of the following is an example of a data quality issue related to completeness? → A customer record has no email address
4. Which term describes the practice of using data collected for one purpose to make decisions about individuals in an unrelated context? → Function creep
5. What does feature importance from a tree-based model tell you? → How much each feature contributes to reducing impurity
6. Which security practice is essential for protecting cloud data? → Data Encryption
7. Which chart type is best for displaying the distribution of a single continuous variable? → Histogram
8. Which framework is commonly used to structure enterprise data governance programs in the US? → DAMA-DMBOK
9. Which data type would 'customer satisfaction rated 1 to 5' represent? → Ordinal
10. A heatmap is most effective for what purpose? → Showing intensity across a two-dimensional matrix
11. Which probability distribution is symmetric and bell-shaped, often used as a basis for statistical inference? → Normal distribution
12. Which data type best represents a 'yes/no' survey response? → Boolean
13. Which regulation primarily governs data privacy and protection for individuals in the United States healthcare sector? → HIPAA
14. What is a data steward's primary responsibility? → Overseeing data quality and enforcing data policies within a domain
15. A dataset column stores customer ages but a few entries read 'unknown'. What data quality issue does this represent? → Missing values
16. In Power BI, what does establishing a relationship between two tables enable? → Filtering and aggregating across related tables
17. Which of the following best defines data governance in the context of data analytics? → A framework of policies and processes ensuring data is accurate, available, and secure
18. Which type of database is best suited for storing and querying highly connected data such as social networks? → Graph database
19. What does a data catalog primarily provide in a data governance program? → A searchable inventory of data assets with metadata and lineage information
20. What is a Common Table Expression (CTE) in SQL? → A temporary named result set defined within a WITH clause for use in a single query
21. Which of the following is an unsupervised learning algorithm? → K-Means Clustering
22. Which visualization best shows how a value flows and splits between stages or categories? → Sankey diagram
23. A model shows high error on both training and test sets. What is the most likely issue? → Underfitting
24. Which tool is commonly used for data visualization? → Tableau
25. Which practice reduces misleading interpretation in bar charts? → Starting the value axis at zero
26. Which of the following is an example of a 'privacy by design' principle in data systems? → Building privacy safeguards into the system architecture from the start
27. Which machine learning algorithm is best suited for classifying emails as spam or not spam? → Naïve Bayes
28. What is 'data-ink ratio' as described by Edward Tufte? → The proportion of ink used to display actual data versus decoration
29. What is the purpose of a confidence interval in statistical analysis? → To provide a range of values likely to contain the true population parameter
30. In SQL for reporting, which clause groups rows so aggregate functions apply per category? → GROUP BY