DA0-002 Data Acquisition and Preparation Practice Question
A data analyst is preparing a dataset for predictive modeling and must handle missing values in several numeric and categorical columns. The team needs defensible, documented choices rather than ad hoc deletion. Which TWO actions are appropriate for handling missing data in this scenario? (Choose two.)
⚠ Common exam trap
The trap here is treating missing-data handling as a mechanical fill step and overlooking that the pattern of missingness itself can be informative and must be diagnosed first.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Create an explicit missing-value indicator column alongside an imputed value so the model can learn from the missingness pattern.
Defensible missing-data handling starts with diagnosing the extent and pattern of missingness, then applies a treatment matched to that pattern. Combining an indicator column with an imputed value preserves both usability and the signal contained in absence. Blind mean or mode substitution, blanket row deletion, and zero-filling all introduce bias without documentation.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Create an explicit missing-value indicator column alongside an imputed value so the model can learn from the missingness pattern.
Why this is correct
An indicator preserves the information that a value was absent, which is valuable when missingness correlates with the target. Pairing it with an imputed value keeps the record usable while letting the model distinguish imputed from observed cases. This is a documented, reproducible technique that satisfies the demand for defensible handling.
- ✗
Impute numeric missing values with the column mean and categorical missing values with the mode, without further review.
Why it's wrong here
Blind mean or mode imputation shrinks variance and can fabricate relationships, especially when missingness is systematic rather than random. Applying it without first examining the pattern violates the requirement for defensible, documented choices. It also performs poorly for skewed numeric distributions where the mean is not a representative central value.
- ✓
Document the missingness rate per column and investigate whether values are missing at random before choosing a treatment.
Why this is correct
Quantifying missingness and assessing its pattern determines whether a simple treatment is defensible or whether the missingness itself carries signal. If values are missing not at random, mean imputation can distort relationships the model relies on. This diagnostic step is what makes the eventual treatment auditable and reproducible, which the team explicitly requires.
- ✗
Replace all missing numeric values with zero so the column contains no nulls and requires no further processing.
Why it's wrong here
Zero is a real, meaningful quantity in most numeric features, so substituting it silently asserts facts that were never observed and can badly distort distributions and model coefficients. It is especially harmful for variables such as income or temperature where zero is implausible. This shortcut removes nulls cosmetically while introducing a systematic bias.
- ✗
Drop every row that contains any missing value across all columns to guarantee a complete dataset.
Why it's wrong here
Listwise deletion can discard a large fraction of records and, when missingness is not random, introduces selection bias that skews the model. It is rarely defensible as a blanket policy because the loss of rows is often disproportionate to the number of incomplete columns. The scenario explicitly asks for reasoned treatment rather than ad hoc deletion.
Go deeper
Related to this question
About these practice questions
This DA0-002 question is part of Courseiva's 1,004-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official CompTIA exam blueprint
This DA0-002 practice question is part of Courseiva's free CompTIA certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the DA0-002 exam.