Ch : 2

what is business driver in general

In general, a business driver is any internal or external factor that has a major impact
on the performance, direction, or success of a business.

One-liner definition:

A business driver is a key element that influences business outcomes and helps guide
strategic decisions.

Examples of general business drivers:

 Market demand
 Customer satisfaction
 Technological innovation
 Cost efficiency
 Employee productivity
 Competitive pressure
 Government regulations
 Brand reputation

Why business drivers matter:

They help leaders answer questions like:

 What should we prioritize?

 Where should we invest time and money?
 How do we measure success?

So in essence, business drivers are the “why” behind business actions and
strategies.

P a g e 1 | 10
Ch : 2

Summary of Chapter 2: Data Handling Ethics

Definition:
Data Handling Ethics are the principles guiding how organizations collect, manage, use, and
dispose of data while respecting fairness, privacy, integrity, and human dignity.

1. Introduction

Ethics means doing what's right, even when no one is watching.

 Ethical data handling helps organizations build trust, reduce risk, and comply with laws.
 Data represents real people (customers, employees, patients), so mishandling it can hurt them.
 Managing data quality, security, and privacy is both a legal and moral responsibility.
 Ethical data practices must apply throughout the entire data lifecycle — from creation to
deletion.

2. Business Drivers

 Ethical data handling builds trust, reputation, and competitive advantage.

 It's part of organizational culture and governance.
 Roles like Chief Data Officer or Chief Privacy Officer lead this effort, but everyone is
responsible.
 Ethical practices reduce misuse risks by insiders and outsiders.

3. Essential Concepts

3.1 Ethical Principles for Data (Inspired by Medical/Bioethics)

 Respect for Persons – Treat people with dignity and honor their rights.
 Beneficence – Do no harm; maximize benefits and reduce risks.
 Justice – Treat people fairly and avoid discrimination.
 Respect for Law and Public Interest – Follow laws and act for the good of society.

3.2 Principles Behind Data Privacy Laws

a. GDPR (EU) – 7 Key Principles:

Principle Meaning
Fairness, Lawfulness,
Data should be processed fairly and openly
Transparency
Use data only for the reason it was
Purpose Limitation
collected
Data Minimization Don’t collect more data than necessary

P a g e 2 | 10
Ch : 2

Principle Meaning
Accuracy Keep data correct and updated
Storage Limitation Don’t keep data longer than needed
Integrity & Confidentiality Protect data from loss or misuse
Accountability Be responsible and prove compliance

b. PIPEDA (Canada):

Covers consent, limiting data use, accuracy, safeguards, and user access rights.

c. FTC Guidelines (US):

Focus on privacy by design, choice, transparency, enforcement, and redress.

3.3 Online Data Ethics

Ownership: Who controls your personal data?

Right to be Forgotten: Request removal of personal data online.

Identity & Speech: Respect for identity and freedom of expression.

3.4 Risks of Unethical Practices

Timing Manipulation: Hiding or distorting facts by selective timing.

Misleading Visuals: Distorting charts to change meaning.

Invalid Comparisons: Comparing unrelated data to mislead.

Overuse of Statistics: Using statistical tricks to show fake significance.

Goals of Ethical Data Handling

Define ethical practices.

Educate staff about risks.

Create ethical culture and behavior.

Track regulations and adjust policies accordingly.

Deliverables

Gap analysis

Strategy

P a g e 3 | 10
Ch : 2

Training & communication plans

Ethical data policies

KPIs and monitoring reports

Tools & Techniques

Wikis, microblogs, intranet

Ethics checklists and affirmations

Training metrics and reporting dashboards

Metrics

Training completion rates

Incident counts (ethical breaches)

Executive participation

Summary of Chapter 2: Data Handling Ethics

1. Introduction

Ethics in data handling involves acting fairly, transparently, and responsibly when collecting,
using, storing, and discarding data. This is essential not only for legal compliance, but also for
building trust, reputation, and social responsibility in a data-driven world.

Ethical data practices help avoid misuse, maintain quality,

and protect individuals’ rights.

2. Goals of Data Handling Ethics

Define what ethical data handling looks like.

Educate employees about risks of unethical practices.

Instill ethical culture across the organization.

Continuously monitor and align with evolving regulations.

P a g e 4 | 10
Ch : 2

3. Deliverables & Stakeholders

Deliverables include training programs, updated policies, ethical strategies, and reporting
mechanisms.

Key stakeholders:

Executives

Data Stewards

Regulators

Governance bodies

4. Activities

Phase Activities
P Review practices & define principles
D Create strategy, educate staff
C Monitor alignment and compliance
O Execute ethical practices

5. Essential Concepts

5.1 Ethical Principles

Adapted from Belmont Principles and EDPS, include:

Respect for Persons: Value individual dignity & autonomy.

Beneficence: Maximize benefits; minimize harm.

Justice: Ensure fair and equal treatment.

Respect for Law and Public Interest: Uphold legal and moral standards.

6. Data Privacy Laws

GDPR Principles (EU)

Lawfulness, fairness, transparency

Purpose limitation

P a g e 5 | 10
Ch : 2

Data minimization

Accuracy

Storage limitation

Integrity and confidentiality

Accountability

🇨🇦 PIPEDA Principles (Canada)

Consent, accuracy, openness

Limiting collection and use

Individual access & complaints

🇺🇸 FTC Privacy Principles (USA)

Notice, consent, access, security, enforcement

7. Emerging Ethical Topics

Data ownership

Right to be forgotten

Online identity and privacy

Freedom of expression vs. harm online

8. Unethical Practices & Risks

Examples:

Timing manipulation (e.g., end-of-day stock trades)

Misleading visuals (e.g., skewed graphs)

Invalid comparisons (e.g., mismatched stats)

Data mining snooping (random correlations shown as facts)

Ethical data handling means truthful, transparent, and fair representation, avoiding harm or
deception.

P a g e 6 | 10
Ch : 2

Baby Definitions Glossary (Chapter 2)

Term Baby Definition

Ethics Knowing what's right and wrong, and acting right.
Doing the right thing with data – collecting, using, and
Data Ethics
sharing it fairly and safely.
Misuse of Data Using data in a wrong or harmful way.
Data Lifecycle The journey of data: collect → use → store → delete.
Beneficence Try to help and not hurt people with data.
Justice Be fair to everyone when using data.
Respect for Treat everyone with care, and let them decide about their
Persons data.
Data Ownership Knowing who has rights over the data.
Right to Be
Asking for your data to be deleted from the internet.
Forgotten
Data Minimization Don’t collect too much data—just what’s needed.
Purpose Limitation Only use data for what you said you would.
Accountability Taking responsibility if you do something wrong with data.
Transparency Be honest and open about what you're doing with data.
Privacy by Design Build privacy into systems from the start.
Informed Consent Getting clear permission before using someone’s data.
PIPEDA Canada’s rules for protecting personal info.
GDPR Europe’s strong data privacy law.
FTC Guidelines USA’s fair data use rules.
Misleading Visuals Charts or graphs that make data look different than it is.
Statistical
Making rough data look smooth by hiding details.
Smoothing
Data Mining Finding patterns that aren’t real by checking too much
Snooping data.
Change Managers People who help teams adjust to new ethical practices.
Data Stewards People who care for and manage data responsibly.

3.4.4 Bias

Bias in data is any systematic deviation from objectivity, occurring during:

Data collection

Data selection

Analysis methods

Presentation of results

P a g e 7 | 10
Ch : 2

Ethical Impact: Bias can reinforce injustice, especially against historically disadvantaged
groups.

Types of Bias:

Data collection for pre-defined result – results manipulated to match expectations.

Biased use of unbiased data – cherry-picking or discarding data.

Hunch and search – only exploring data that confirms a belief.

Biased sampling – non-random or unrepresentative samples.

Contextual/Cultural bias – interpretation affected by local norms.

Analysts should:

Be aware of their biases.

Document population selection.

Ensure algorithm transparency and fairness, especially in predictive models like

criminal risk scoring or predictive policing.

3.4.5 Transforming and Integrating Data

Ethical issues arise when data is transferred across systems, especially without context or
documentation.

Key Risks:

Unknown origin/lineage – weakens data reliability.

Poor quality – needs defined quality metrics.

Unreliable metadata – causes misinterpretation.

No remediation documentation – illegal/unethical changes.

Data lineage, metadata, and change history must be traceable and auditable.

3.4.6 Obfuscation / Redaction of Data

Making data anonymous does not guarantee privacy, especially with large datasets.

Techniques:

Aggregation – summarizing data to avoid PII.

P a g e 8 | 10
Ch : 2

Data marking – labeling by sensitivity.

Data masking – hiding true values during processes.

Risk: Even anonymized data can re-identify individuals when combined.

Ethical data handling must go beyond obfuscation with strong governance and protection
methods.

3.5 Establishing an Ethical Data Culture

Creating an ethical culture means defining behaviors, policies, and training mechanisms backed by
leadership.

Steps:

Review current state – assess alignment of practices with ethical principles.

Identify principles and risks – match risks with protective practices and controls.

Develop a strategy and roadmap – create values, code of ethics, compliance plans, risk
assessments, training, and monitoring.

Adopt ethical risk models – especially in BI, analytics, and Data Science.

Risk Model Example (Fig 13):

Arrow 1: Population selection

Arrow 2: Data capture methods

Arrow 3: Analytical focus

Arrow 4: Result access and usage

Highlight unintended consequences, such as denying services due to biased models.

3.6 Data Ethics and Governance

Oversight responsibility is shared between:

Legal Counsel

Data Governance Teams

Governance Responsibilities:

Monitor legal changes

P a g e 9 | 10
Ch : 2

Set ethical data policies

Enable whistleblower protection

Review high-risk analytics and AI projects

Data professionals (e.g., CDMP certified) are ethically bound to act in society’s interest, not
just business interest.

P a g e 10 | 10