PracticeStudy GuidesBlogFlashcardsEspañol
All Practice Exams

100+ Free CDEP Practice Questions

Pass your CertNexus Certified Data Ethics Professional (CDEP) exam on the first try — instant access, no signup required.

✓ No registration✓ No credit card✓ No hidden fees✓ Start practicing immediately
~65-75% Pass Rate
100+ Questions
100% Free
1 / 10
Question 1
Score: 0/0

The Belmont Report identifies which three core ethical principles for research with human subjects?

A
B
C
D
to track
2026 Statistics

Key Facts: CDEP Exam

60

Exam Questions

CertNexus

90 min

Exam Duration

CertNexus

60-70%

Passing Score

CertNexus (scaled)

$250

Exam Fee

CertNexus

3 years

Validity

CEC renewal

Vendor-neutral

Format

Cross-industry

The CDEP exam has 60 questions in 90 minutes with a 60-70% scaled passing score. Domains cover ethical frameworks (Belmont, OECD, EU AI Act, NIST AI RMF), bias types (sampling, historical, label, algorithmic, automation), fairness metrics (demographic parity, equalized odds, calibration), privacy law (GDPR, CCPA/CPRA, HIPAA, FERPA, COPPA, k-anonymity, differential privacy), informed consent and dark patterns, governance (RBAC/ABAC, lineage, retention), accountability (model cards, AIAs), explainability (SHAP, LIME, counterfactuals), domain cases (healthcare, hiring, lending, justice), and professional codes. Exam fee is $250. Valid 3 years.

Sample CDEP Practice Questions

Try these sample questions to test your CDEP exam readiness. Each question includes a detailed explanation. Start the interactive quiz above for the full 100+ question experience with AI tutoring.

1The Belmont Report identifies which three core ethical principles for research with human subjects?
A.Speed, Scale, Sustainability
B.Respect for Persons, Beneficence, Justice
C.Liberty, Equality, Fraternity
D.Privacy, Profit, Performance
Explanation: The 1979 Belmont Report (US National Commission) defines Respect for Persons (autonomy, informed consent), Beneficence (do no harm, maximize benefits), and Justice (fair distribution of risks and benefits). It underpins modern IRB review and data ethics frameworks.
2The Menlo Report extends Belmont principles for which domain?
A.Biomedical research
B.Information and communications technology research, especially security and networking
C.Industrial design
D.Aerospace engineering
Explanation: The Menlo Report (DHS, 2012) adapts Belmont for ICT research, adding Respect for Law and Public Interest. It is a key reference for cybersecurity and data-driven research ethics.
3The OECD AI Principles emphasize:
A.Maximum surveillance
B.Inclusive growth, human-centered values, transparency, robustness, and accountability
C.Profit only
D.Banning all AI
Explanation: OECD AI Principles (2019, updated 2024) cover inclusive growth, human-centered values and fairness, transparency and explainability, robustness/security, and accountability. Endorsed by 47+ countries.
4The EU AI Act's risk-based classification includes which highest-risk category that is prohibited?
A.Limited risk
B.Unacceptable risk (e.g., social scoring by governments, manipulative techniques exploiting vulnerabilities)
C.Minimal risk
D.High risk
Explanation: EU AI Act categories: Unacceptable (banned — social scoring, real-time biometric ID with narrow exceptions, manipulative AI, certain emotion recognition), High (heavy obligations), Limited (transparency obligations), Minimal.
5NIST AI Risk Management Framework (AI RMF) defines four core functions:
A.Plan, Build, Deploy, Retire
B.Govern, Map, Measure, Manage
C.Identify, Protect, Detect, Respond
D.Encrypt, Train, Test, Audit
Explanation: NIST AI RMF 1.0 (2023) functions: GOVERN (org accountability), MAP (context and risks), MEASURE (quantify), and MANAGE (prioritize and mitigate). Voluntary, sector-agnostic, broadly adopted.
6FATML stands for:
A.Frequency, Analysis, Testing of ML
B.Fairness, Accountability, Transparency in Machine Learning
C.Federated, Anonymized, Trusted ML
D.Foundation, Audit, Trust, ML
Explanation: FAT/FATML/FAccT is a research community focused on Fairness, Accountability, and Transparency in ML. The annual ACM FAccT conference is a leading venue for AI ethics research.
7Sampling bias occurs when:
A.Data is too large
B.The data collected is not representative of the target population, skewing model conclusions
C.Models train too quickly
D.GPUs are too fast
Explanation: Sampling bias arises from non-representative data: convenience sampling, missing demographics, regional concentration. Mitigations: stratified sampling, oversampling underrepresented groups, and explicit data audits.
8Historical bias is best described as:
A.Bias from old hardware
B.Bias reflecting past social inequities baked into historical data, even if data is collected accurately
C.Bias from old algorithms
D.Bias from operating systems
Explanation: Even perfectly collected historical data encodes social inequities (e.g., arrest records reflecting policing patterns, hiring records reflecting past discrimination). Models trained on such data perpetuate inequities. Causal analysis and fairness constraints help.
9Label bias arises when:
A.Labels are stored efficiently
B.Ground-truth labels are themselves biased due to labeler subjectivity, proxies, or systemic measurement issues
C.Labels are too large
D.Labels use UTF-8
Explanation: Label bias occurs when targets are biased proxies (arrests as proxy for crime), labelers carry implicit bias, or annotation guidelines are flawed. Even unbiased models can replicate label bias. Audit label quality across subgroups.
10Algorithmic bias refers to:
A.Bias introduced by hardware
B.Bias arising from algorithmic choices (objective function, sampling strategy, optimization) that produce systematically worse outcomes for some groups
C.Bias of the user
D.Bias of the encoding
Explanation: Algorithmic bias can arise even with unbiased data — through loss functions that favor majority groups, sampling that ignores subgroups, or threshold choices that disparately impact protected classes.

About the CDEP Exam

Certified Data Ethics Professional (CDEP) is CertNexus's vendor-neutral data ethics certification. CDEP validates the ability to apply ethical frameworks (Belmont, Menlo, OECD, EU AI Act, NIST AI RMF), recognize and mitigate bias, apply fairness metrics, implement privacy protections (GDPR, CCPA, HIPAA, FERPA, COPPA), design data governance, build accountability mechanisms, and apply professional codes (ACM, IEEE, IAPP).

Questions

60 scored questions

Time Limit

90 minutes

Passing Score

60-70% (scaled)

Exam Fee

$250 USD (CertNexus / Pearson VUE)

CDEP Exam Content Outline

~15%

Ethical Frameworks

Belmont Report (respect for persons, beneficence, justice), Menlo Report, OECD AI Principles, EU AI Act risk tiers, NIST AI RMF (Govern, Map, Measure, Manage), FATML, IEEE Ethically Aligned Design

~20%

Bias and Fairness Metrics

Bias types (sampling, historical, label, algorithmic, automation, survivorship, selection, confirmation, recall, measurement), fairness metrics (demographic parity, equalized odds, equal opportunity, calibration, individual and counterfactual fairness), bias mitigation

~25%

Privacy, Consent, and PETs

PII/PHI, GDPR (lawful bases, DPIA, right to erasure, Article 22), CCPA/CPRA, HIPAA, FERPA, COPPA, k-anonymity/l-diversity/t-closeness, differential privacy, federated learning, SMPC, informed consent, dark patterns, IRB

~15%

Governance and Accountability

Data stewardship, lineage, classification, retention, RBAC/ABAC, audit trails, model cards, datasheets for datasets, AI Impact Assessments / FRIA, ethics review boards, US AI Bill of Rights

~15%

Explainability and Domain Cases

SHAP, LIME, counterfactual explanations, contestability, healthcare AI ethics, hiring/HR (NYC AEDT, Colorado AI Act), lending/credit (ECOA, FCRA, CFPB), criminal justice (COMPAS), case studies (Amazon hiring, Apple Card, Cambridge Analytica, ChatGPT leak)

~10%

Professional Codes and Practice

ACM Code of Ethics, IEEE Code of Ethics, INFORMS, IAPP CIPP, whistleblowing duties, applying ethics in vendor and third-party data review

How to Pass the CDEP Exam

What You Need to Know

  • Passing score: 60-70% (scaled)
  • Exam length: 60 questions
  • Time limit: 90 minutes
  • Exam fee: $250 USD

Keys to Passing

  • Complete 500+ practice questions
  • Score 80%+ consistently before scheduling
  • Focus on highest-weighted sections
  • Use our AI tutor for tough concepts

CDEP Study Tips from Top Performers

1Memorize Belmont's three principles (respect for persons, beneficence, justice) and how each applies to data work
2Know the EU AI Act risk tiers: unacceptable, high, limited, minimal — and what each entails
3Understand the impossibility theorem: you cannot satisfy demographic parity, equalized odds, and calibration simultaneously when base rates differ
4Distinguish anonymization from pseudonymization (and why pseudonymized data is still personal data under GDPR)
5Know GDPR Article 22 rights related to solely automated decision-making
6Memorize famous case studies: COMPAS, Amazon hiring algorithm, Apple Card, Cambridge Analytica, ChatGPT leak
7Practice applying ACM and IEEE codes of ethics to scenario questions

Frequently Asked Questions

What is the CDEP exam?

The Certified Data Ethics Professional (CDEP) is CertNexus's vendor-neutral data ethics certification. It validates the ability to apply ethical frameworks, recognize and mitigate bias, apply fairness metrics, implement privacy protections, design governance, build accountability mechanisms, and apply professional codes when working with data and AI systems.

How many questions are on the CDEP exam?

The CDEP exam has 60 questions to complete in 90 minutes. Questions are multiple-choice and scenario-based, focused on real-world ethical decisions in data and AI work. The passing score is scaled and typically corresponds to roughly 60-70% of items correct.

Who should take the CDEP exam?

CDEP targets data scientists, data engineers, ML engineers, analysts, product managers, privacy and compliance professionals, AI ethics officers, and leaders responsible for AI governance. It is broadly applicable to anyone shaping how data is collected, modeled, and used.

Are there prerequisites for the CDEP exam?

There are no formal prerequisites. Familiarity with how data and AI systems are built and deployed is helpful, but the exam is designed to be accessible to professionals from technical, legal, compliance, and management backgrounds. CertNexus official courseware aligns with the objectives.

How long is CDEP valid?

CDEP certification is valid for 3 years from the date you pass. Renewal requires Continuing Education Credits (CECs) and a CertNexus renewal fee. CECs can be earned through training, conferences, publications, or professional activity in data ethics, privacy, or AI governance.

How should I prepare for CDEP?

Plan for 30-50 hours of study over 4-6 weeks. Read the Belmont Report, OECD AI Principles, NIST AI RMF 1.0, EU AI Act overview, and the OSTP AI Bill of Rights blueprint. Study fairness metrics and the impossibility theorem, GDPR core articles, HIPAA, CCPA/CPRA, k-anonymity, and differential privacy. Review famous cases (COMPAS, Amazon hiring, Apple Card, Cambridge Analytica). Complete 100+ practice questions scoring 80%+ before scheduling.