week 3

3.3 Data taxonomy

What is a Taxonomy?

Think of a taxonomy like a family tree, but for data. It’s a way of splitting things into groups so we know what type of data we’re dealing with.

3.3.1 Definition of qualitative and quantitative, its purpose, and how data is categorised

Quantitative

quantitative data – which basically means numbers. If you can count it or measure it, it’s quantitative.

Two Types of Quantitative Data can be

Discrete Data

• Discrete means things you can count in whole numbers.

• You can’t have half of one, it’s either 1, 2, 3… but not 2.5.

• In IT support/security:

  • How many times a student typed the wrong password.

  • The number of emails flagged as spam.

  • How many viruses an antivirus tool finds.

 “How many login attempts failed this morning?” and you answer “7”, that’s discrete data.

Continuous Data

• Continuous means measurements – and you can have decimals.

• In IT support/security:

  • The server room temperature (22.3°C, 22.4°C, etc.).

  • Bandwidth speed during an esports match (245.6 Mbps).

  • CPU load (%) on a computer.

“What’s the server temperature right now?” and it says “23.5°C” – that’s continuous data.

Both are useful, but in different ways:

• Discrete data is great for counting events – like how many people tried to hack into your system.

• Continuous data is better for monitoring performance – like spotting if your server is overheating or slowing down.

Take Amazon Web Services (AWS) they’re running thousands of servers worldwide they use discrete data to count login attempts and block suspicious ones. At the same time, they use continuous data to monitor server performance. If both types spike at once, they know something is wrong.

Qualitative.

What is Qualitative Data?

Qualitative data is about descriptions, opinions, and categories rather than numbers.

Types of Qualitative Data;

Categorical (or Nominal) Data

Data that can be sorted into groups, but the groups don’t have a natural order.

• In Digital Support & Security:

  • Type of cyberattack: phishing, malware, ransomware, brute force.

  • Operating system: Windows, macOS, Linux.

  • User role: student, staff, admin.

It’s like labels – they tell you what “type” something is, but not which one is bigger or better.

Ordinal Data

Data that can be put in a ranked order, but the gaps between them aren’t necessarily equal.

• In Digital Support & Security:

  • Student feedback on password security training (Poor, Okay, Good, Excellent).

  • Cybersecurity risk ratings: Low, Medium, High, Critical.

  • Priority of support tickets: Urgent, Medium, Low.

So ordinal data has a sense of order, but it’s not really about numbers. “High risk” is more serious than “Low risk,” but we can’t say it’s exactly “two times” more serious.

Quantitative data is great for spotting patterns in numbers – but qualitative data adds the human side:

• What people think

• How people feel

• Why something is happening

NCSC (National Cyber Security Centre, UK):

They collect quantitative data about how many phishing emails are reported but they also collect qualitative data from feedback surveys asking staff how confident they feel spotting phishing emails, by combining the two, they can judge not just how many phishing attempts are happening, but also how well people are prepared to deal with them.

Case Study: College Cybersecurity Awareness

Your college has recently run a campaign to improve cybersecurity awareness among students and staff. The IT support and security team collected both quantitative and qualitative data to see if it worked.

Data Collected:
• Quantitative (numbers):
   - 1,200 phishing emails reported in Term 1.
   - Only 450 phishing emails reported in Term 2.
   - 95% of students logged in successfully without needing password resets.

• Qualitative (opinions/descriptions):
   - “I feel more confident spotting phishing emails now.”
   - “The password rules are still too complicated.”
   - “Training was useful but too short.”
   - Risk ratings given by IT staff: Low, Medium, High.

Task Part 1 – Analysis (20 mins, group work)

Work in small groups and:
1. Identify the quantitative data in the case study.
2. Identify the qualitative data in the case study.
3. Explain how each type of data helps the IT team understand the effectiveness of the campaign.
4. Make a judgement: Do the numbers and opinions show the campaign was successful? Why or why not?

Task Part 2 – Research (Homework or 30 mins independent task)

Each group must research a real-world cybersecurity awareness campaign. Examples:
- NCSC “Cyber Aware” (UK)
- Google Security Checkup
- StaySafeOnline.org (US)
- OR another campaign you find.

For your chosen case:
- Find one example of quantitative data they collected.
- Find one example of qualitative data they used.
- Explain how combining both types of data made their campaign stronger.

Task Part 3 – Group Presentation (15 mins prep + delivery in next lesson)

Prepare a 5-minute presentation to share with the class. Your presentation should include:
1. A short explanation of the difference between quantitative and qualitative data.
2. An analysis of the college case study – was the awareness campaign effective?
3. Findings from your research case study.
4. A recommendation: If you were the IT manager, what would you do next to improve cybersecurity awareness?

Tip: Use visuals like graphs (for quantitative data) and word clouds or quotes (for qualitative data).

Extension / Stretch Task
Design your own mini research survey that collects both quantitative and qualitative data about how safe students feel online. Share 3–5 questions (mix of numerical scales and open-ended questions).

3.3.2 Know the definition for structured data, understand its purpose, and understand that quantitative data is structured.

Structured data is data that is organised and stored in a defined format, usually within tables, rows, and columns, such as in databases or spreadsheets. It follows strict rules, which make it easier to enter, store, search, and analyse. Because it is predictable and consistent, structured data can be processed quickly by machines and used to support decision-making

The purpose of structured data is to:

• Enable fast access and retrieval – information is easily searchable with SQL queries or filters.

• Support accurate analysis – data can be aggregated, compared, and visualised (charts, dashboards, reports).

• Improve reliability – stored in databases with validation rules, ensuring accuracy and reducing errors.

• Aid security and compliance – structured systems can apply access controls and encryption consistently.

Quantitative data is numerical data that can be measured and counted. It is structured by:

• Discrete values – whole numbers, e.g. number of employees.

• Continuous values – measured data, e.g. temperature, sales revenue.

• Categorical values – numerical codes representing groups, e.g. “1 = Male, 2 = Female” in a HR database.

This data fits neatly into tables where each record (row) contains values across defined fields (columns).

Case Studies

1. Tesco (Retail)
   Tesco uses structured data in their loyalty programme (Clubcard). Customer transactions are stored in databases: product IDs, time of purchase, cost, and store location. Structured quantitative data allows Tesco to identify buying patterns, target promotions, and forecast stock demand.

2. NHS (Healthcare)
   The NHS uses structured patient data – age, blood pressure readings, appointment times – stored in Electronic Health Records. This ensures doctors can quickly retrieve accurate medical histories, track quantitative health measures, and comply with legal standards such as GDPR.

3. Airlines (British Airways)
   Airlines store structured data for bookings: passenger details, flight numbers, seat allocations, ticket prices. Quantitative data (ticket sales, baggage weight, passenger counts) helps them optimise scheduling, revenue management, and compliance with aviation regulations.

Spot the Structure (15 minutes – group task)

Your Challenge

In this task, you will work in small groups to explore different types of data and decide which ones are structured and which are not. Then, you’ll look at how organisations use numbers (quantitative data) to make decisions.

Step 1 – Sort the Data (5 minutes)
You will be given a sheet with different examples of data:

- A shopping list with prices
- A short blog post
- Patient heart rate readings
- A set of photos
- Flight booking details

With your group, sort the data into two piles:
- Structured data (fits into a table, rows, or columns)
- Unstructured data (free text, images, videos, anything without a clear format)

Step 2 – Find the Numbers (5 minutes)
From your structured data pile, highlight or circle the quantitative values (numbers, measurements, statistics).
Example: prices on the shopping list, heart rate readings, ticket sales.
Then, discuss:
How could an organisation use these numbers?
What decisions could they make based on them?

Step 3 – Share Your Findings (5 minutes)
Choose one example from your group
Be ready to tell the class:
1. Is it structured or unstructured?
2. What numbers did you find?
3. How could a business or organisation use that information?

What You’ll Learn
By the end of this activity, you should be able to:
- Spot the difference between structured and unstructured data.
- Identify where numbers (quantitative data) appear in structured data.
- Explain how organisations can use structured data to make decisions.

Task Inforamtion Download file

Example 1 Shopping List
Example 2 Flight Booking
Example 3 Patient Readings 
Example 4 Sales Report

3.3.3 Know the definition for unstructured data, understand its purpose, and understand that qualitative data is unstructured.

Unstructured data is information that does not have a predefined format or structure. It does not fit neatly into tables of rows and columns, and it is often text-heavy, image-based, or multimedia. Examples include emails, social media posts, documents, photos, audio, and video files.

The purpose of unstructured data is to:

• Capture rich, descriptive detail – allows organisations to understand opinions, behaviours, and context.

• Support decision-making beyond numbers – text, images, and speech can provide meaning that numbers alone cannot.

• Enable qualitative analysis – helps to identify themes, trends, or insights in customer feedback, medical notes, or research interviews.

• Drive innovation – unstructured data can reveal opportunities for product design, marketing, or service improvement.

Qualitative Data and Unstructured Data

Qualitative data is descriptive, non-numerical data – such as feelings, opinions, and experiences. It is usually unstructured because it cannot be easily measured or placed into rows and columns.

• Example: A customer saying “The product was too difficult to set up” in a feedback survey.

• Unlike quantitative data (numbers), qualitative data focuses on meaning, reasons, and motivations.

Case Studies

1. BBC (Media)
   The BBC analyses unstructured social media comments, audience feedback emails, and video views to understand what viewers like or dislike. This qualitative data helps shape programme schedules and digital content.

2. Amazon (E-commerce)
   Amazon uses unstructured product reviews and customer questions to improve product recommendations. Sentiment analysis (positive/negative reviews) gives insight into customer satisfaction beyond raw sales numbers.

3. NHS (Healthcare)
   Doctors’ notes, medical scans, and patient feedback are unstructured but essential for care. Analysing this qualitative data helps identify patterns in patient experiences and improve treatment plans.

Supporting Activity (15 minutes – Small Groups)

Title: “Unpack the Unstructured”

Your Challenge
In this task, you will explore different types of unstructured data and think about how organisations can use them to understand people’s experiences and opinions

Step 1 – Identify the Unstructured Data (5 minutes)
You will be given a sheet with examples of data:
- A tweet from a customer about poor service
- A product review from Amazon
- A doctor’s note about a patient’s symptoms
- A video clip description from YouTube
- A company sales report

With your group, decide which examples are unstructured and which (if any) are structured.

Step 2 – Spot the Qualitative Information (5 minutes)
From the unstructured examples, highlight or underline the qualitative details (opinions, descriptions, experiences).

Example: “The app keeps crashing and is frustrating to use.”

Then discuss:
- How could an organisation use this type of feedback?
- What changes or improvements could it lead to?

Step 3 – Share Your Insights (5 minutes)

Pick one example and be ready to share:
1.Why is it unstructured?
2. What qualitative information did you find?
3. How could an organisation act on this information?


What You’ll Learn
By the end of this activity, you should be able to:

Recognise examples of unstructured data.
Understand how qualitative data provides meaning and context.
Explain how organisations use unstructured data to improve services or products.

Download file
Example 5 Tweet
Example 6 Amazon Review
Example 7 Doctors Note
Example 8 Youtube Description
Example 9 Exit Notes

3.3.4 Know the definition for each representation and understand the representations of quantitative data:

When working with data, it’s important to understand how numbers can be represented and organised. Quantitative data is data that deals with numbers and measurements  it tells us how many, how much, or how often.

However, not all numbers behave in the same way. Some numbers are easy to count, some are measured on a scale, and others are used to represent categories or groups. To make sense of this, quantitative data is usually represented in three main forms:

Discrete values

Values you count which can only take certain distinct (usually whole number) values. There are gaps between possible values. Examples: number of students in a class; number of defects in a product; count of hospital visits.

Continuous values

Values you measure, which can take any value within a (possibly infinite) range, including decimals/fractions. There are no gaps between possible values in theory. Examples: height, weight, temperature, time, distance

Categorical values.

Values that represent categories or groups rather than numerical amounts. Sometimes further divided into nominal (no inherent order) and ordinal (order matters, but distances between categories are not necessarily equal). Examples: blood type; customer rating (poor / fair / good / excellent); brand; gender.

Examples

To clarify, here are some concrete examples in organisational / real-world settings, showing each type in action, plus mixed use, and evaluation.

1. Retail company / inventory management

   • Discrete: Number of units of each product in stock (e.g. 15 chairs, 200 mugs).

   • Continuous: The weight of a shipment in kg; the size of product packaging (volume).

   • Categorical: Type of product (furniture vs kitchenware vs electronics); category of suppliers; product color.

   Benefit: Using discrete counts allows quick decisions about restocking. Continuous values help with logistics (weight, volume) for shipping. Categorical helps in analysing patterns (which product categories sell best).

   Drawback: If continuous measures are too precise (e.g. milligrams) that don’t affect business decisions, they add complexity for little benefit. If categories are too many or poorly defined, comparisons become messy.

2. Healthcare / Hospitals

   • Discrete: Number of patients admitted per day; number of surgeries done.

   • Continuous: Patients' temperature; blood pressure; time taken in surgery; length of hospital stay (in hours).

   • Categorical: Disease type; severity category (mild/moderate/severe); patient sex; patient risk group.

   Benefit: Continuous values allow detecting small changes in vital signs; discrete counts help with capacity planning; categorical allows grouping by disease or risk for policy decisions.

   Drawback: Measurement error in continuous values can mislead (e.g. inaccurate blood pressure readings). Discrete counts fluctuate daily and may be influenced by external factors. Some categorical groupings (severity) are subjective.

3. Education / Schools

   • Discrete: Number of students in a class; number of books borrowed; number of discipline incidents.

   • Continuous: Test scores (if measured on continuous scale); time students spend on tasks; percentage marks.

   • Categorical: Grades (A / B / C / Fail); subject area (Math, English, Science); level of satisfaction in surveys.

   Benefit: Discrete and continuous data enable quantitative tracking (progress, comparison). Categorical help with grouping and reporting to stakeholders.

   Drawback: Grades (categorical) may hide wide variation in actual performance. Continuous scores may vary slightly due to test difficulty but may not represent real learning. Also, privacy/ethical issues when dealing with precise student data.

3.3.5 Know and understand the properties of qualitative data:

• stored and retrieved only as a single object

• codified into structured data.

3.3.6 Understand the interrelationships between data categories data structure and transformation and make judgements about the suitability of data categories, data structure and transformation in digital support and security.

Last Updated
2025-10-09 21:44:22