Detecting Overfitting from Training and Test Performance
A financial institution uses a regression model to predict credit risk. The model has a high R-squared on training data but low R-squared on test data. Which of the following is the most likely cause?
Quick Answer
The answer is overfitting, because a high R-squared on training data paired with a low R-squared on test data is the definitive signature of a model that has memorized the training set rather than learning generalizable patterns. This performance gap occurs when the model captures noise and specific quirks in the training data, causing it to fail on unseen examples—exactly what a regression model predicting credit risk would do if it overfits. On the CompTIA AI+ AI0-001 exam, this scenario tests your ability to detect overfitting from train/test performance gap, often appearing as a trap where candidates mistakenly blame data leakage or insufficient features. The key is to remember that a large gap between training and test metrics always signals overfitting, not underfitting. Memory tip: think of it as the “gap rule”—the wider the gap, the more the model is cheating on the training data.
⚠ Common exam trap
CompTIA often tests the distinction between overfitting and underfitting by presenting a high training metric with a low test metric, tempting candidates to think the model is 'too good' or that data preprocessing (like standardization) is the fix.
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
✓
The model is overfitting the training data.
A high R-squared on training data combined with a low R-squared on test data is the classic symptom of overfitting. The model has memorized noise and specific patterns in the training set rather than learning generalizable relationships, causing poor performance on unseen data.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
The features were not standardized before training.
Why it's wrong here
Unstandardised features affect gradient descent convergence and regularisation scaling, but they do not by themselves produce a large train-test R-squared gap. Standardisation matters when features differ vastly in scale, such as mixing income in dollars with age in years.
- ✓
The model is overfitting the training data.
Why this is correct
High training R-squared with low test R-squared is the defining symptom of variance error: the model has fitted noise and idiosyncrasies specific to the training set, so it fails to generalise to unseen credit-risk data.
- ✗
The model is underfitting the training data.
Why it's wrong here
Underfitting produces low R-squared on both training and test data, so it cannot explain a high training score. It is tempting because poor test performance often signals insufficient model capacity, and underfitting would be the answer if both scores were low.
- ✗
There is multicollinearity among the input features.
Why it's wrong here
Multicollinearity inflates coefficient variance but leaves overall R-squared largely intact on both sets, so it cannot produce this train-test gap. It is tempting because it is a genuine regression pitfall, and would be the answer if coefficients were unstable while predictive fit stayed similar.
About these practice questions
One of 962 original AI0-001 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
Same concept, more angles
1 more way this is tested on AI0-001
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. A company deploys an AI model to predict equipment failure. The model performs well on historical data but fails to generalize to new data from a different factory. Which concept best describes this issue?
easy- A.Transfer learning
- B.Underfitting
- ✓ C.Overfitting
- D.Bias-variance tradeoff
Why C: (Overfitting) is correct because the model learned patterns specific to the historical data from the original factory, including noise and factory-specific nuances, rather than generalizable features. When applied to new data from a different factory, those learned patterns do not hold, causing poor performance. This is the classic symptom of overfitting: high accuracy on training data but low accuracy on unseen data.
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI0-001 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 AI0-001 exam.