Which Approach Best Handles Missing Values While Minimizing Bias for AI Models?
A company is preparing customer data for a predictive model. They notice that many records have missing values for the 'annual income' field. Which approach is best to handle this issue while minimizing bias?
Quick Answer
The answer is model-based imputation, which uses relationships between other features like education and job role to predict missing annual income values. This approach is best for minimizing bias because it preserves the natural data distribution and avoids the distortion caused by simple mean or zero imputation, while also retaining sample size better than deletion. On the Salesforce AI Associate AI Associate exam, this question tests your understanding of bias mitigation in data preparation—a core concept in ethical AI. A common trap is choosing mean imputation, which can artificially compress variance and introduce systematic bias, especially for skewed fields like income. Remember the memory tip: “Model maps the missing, mean masks the bias.”
⚠ Common exam trap
Salesforce often tests the misconception that mean imputation is a safe default, but the trap here is that it ignores feature dependencies and can artificially shrink variance, leading to overconfident model predictions and biased coefficients.
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
✓
Use model-based imputation considering other features.
Model-based imputation (Option B) is best because it uses relationships between features (e.g., education, job role) to predict missing 'annual income' values, preserving data distribution and minimizing bias. This approach avoids the distortion caused by simple mean/zero imputation and retains sample size better than deletion.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Remove all records with missing values.
Why it's wrong here
Deleting records reduces sample size and may introduce bias if missingness is not random.
- ✓
Use model-based imputation considering other features.
Why this is correct
Model-based imputation leverages other features to predict missing values, preserving relationships and minimizing bias.
- ✗
Replace missing values with the mean.
Why it's wrong here
Mean imputation distorts the distribution and can weaken correlations.
- ✗
Set missing values to zero.
Why it's wrong here
Setting to zero introduces a large bias and is not appropriate for income data.
Go deeper
Related to this question
About these practice questions
Courseiva writes every AI Associate question from scratch — 753 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
Same concept, more angles
1 more way this is tested on AI Associate
These questions test the same concept from different angles. Work through them to make sure you can recognise it however the exam phrases it.
Variation 1. Which TWO techniques are commonly used to handle missing values in a dataset for AI training?
hard- A.L1 regularization
- ✓ B.Deletion of rows with missing values
- C.One-hot encoding
- D.Min-max normalization
- ✓ E.Imputation with mean or median
Why B: Deleting rows with missing values is a straightforward technique to handle missing data, especially when the missingness is random and the dataset is large enough that removing a few rows does not significantly impact model performance. This approach avoids introducing bias from imputation methods but can lead to loss of valuable information if too many rows are removed.
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI Associate practice question is part of Courseiva's free Salesforce 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 AI Associate exam.