VCE Applied Computing (Year 11) · Unit 1 — Data Analytics · 16m read

Data — Types, Collection and Organisation

Ever wondered how Netflix knows what shows you'll love, or how your school tracks attendance? That's all about collecting the right information and organizing it smartly. In this topic, you'll learn to spot different types of data (like numbers versus text), figure out the best ways to gather it, and arrange it so it actually makes sense. These skills are crucial for VCE Applied Computing and any job where you need to turn messy information into useful insights.

Learning objective: Classify data types and describe methods for collecting and organising data for analysis.

By the end of this lesson you can:

• I can classify data as quantitative/qualitative and continuous/discrete.
• I can describe primary and secondary data collection methods.
• I can organise data in a structure suitable for analysis.

Data is raw ore. It must be mined (collected), cleaned and organised before it becomes valuable information.

Data types and collection

### Types of data

Quantitative data: Numerical — can be measured and calculated.
- *Continuous:* Any value in a range (height, temperature, time). Measured.
- *Discrete:* Countable, whole numbers (number of students, car sales). Counted.

Qualitative data: Descriptive, categorical.
- *Nominal:* Named categories with no order (colour, suburb, gender).
- *Ordinal:* Categories with meaningful order (Low/Medium/High, Strongly Agree→Disagree).

Derived data: Calculated from other data (average, percentage, rate of change).

### Data collection methods

Primary data: Collected directly by the analyst for a specific purpose.
- Surveys/questionnaires (structured or unstructured)
- Observations and measurements
- Experiments and controlled trials
- Interviews

Secondary data: Collected by others, reused for analysis.
- Government databases (ABS, AIHW)
- Research publications
- Commercial datasets
- IoT sensor logs

### Data organisation

Spreadsheets: Simple, flexible. Rows = records; Columns = fields/attributes.

Relational databases: Tables linked by keys.
- *Primary key:* Unique identifier for each record.
- *Foreign key:* Links records in different tables.
- SQL queries extract data: SELECT, FROM, WHERE, GROUP BY.

Data quality issues:
- Missing values (NULL)
- Duplicate records
- Inconsistent formats (date: 01/04/2024 vs April 1, 2024)
- Outliers (data entry errors vs genuine extreme values)

AARI TIP: Ask Aari to explain the difference between structured and unstructured data.

Data type identification

• Continuous quantitative: Can take any value in a range. Examples: temperature, weight, height.
• Discrete quantitative: Whole numbers only — countable. Examples: children per family, website visits.
• Nominal qualitative: Named categories, no order. Examples: car colour, country, blood type.
• Ordinal qualitative: Ordered categories. Examples: survey ratings, education level.

Quick reference

| Data type | Example |
|---|---|
| Integer | 42 |
| Float/real | 3.14 |
| String | "Hello" |
| Boolean | True/False |
| Date | 2026-10-05 |
| Qualitative vs quantitative | Words vs numbers |

Worked example (Easy)

Classify each of the following: (a) monthly rainfall (b) number of goals scored (c) customer satisfaction rating (1-5) (d) suburb name.

1. (a) Monthly rainfall. Measured in mm; can be any value → continuous quantitative.
2. (b) Number of goals. Counted in whole numbers → discrete quantitative.
3. (c) Satisfaction rating 1-5. Ordered categories; the categories have meaning (3 > 2) → ordinal qualitative.
4. (d) Suburb name. Named categories with no inherent order → nominal qualitative.

Answer: (a) Continuous quantitative. (b) Discrete quantitative. (c) Ordinal qualitative. (d) Nominal qualitative.

Worked example (Medium)

A school survey collects: name, year level, hours of homework per week, and a 1–5 rating of subject difficulty. Classify each field by data type.

1. Name. Categorical (nominal) text.
2. Year level. Categorical (ordinal) — has order but not interval.
3. Hours of homework. Numerical, ratio (true zero, can divide).
4. Difficulty rating 1–5. Categorical (ordinal) or numerical interval depending on treatment.

Answer: Identifying type is the first step before choosing a chart or summary statistic — wrong type leads to misleading analysis.

Common mistakes

• Classifying Likert scale data as quantitative Students see numbers (1-5) and assume the data is numerical, but a satisfaction rating represents ordered categories. Fix: Likert scales (Strongly Agree to Strongly Disagree, or 1-5 ratings) are ordinal qualitative — the numbers represent ordered categories, not measurements. The gap between 1 and 2 may not equal the gap between 4 and 5.
• Confusing primary and secondary data based on format Students think digital data is always secondary or that survey data is always primary. Fix: The distinction is about WHO collected the data and WHY. Data you collect for your specific purpose = primary. Data collected by someone else that you reuse = secondary. Format is irrelevant.

Structured vs unstructured data

Data comes in different levels of organisation, and understanding these levels is increasingly important as data analytics expands beyond traditional spreadsheets.

Structured data fits neatly into rows and columns — spreadsheets, relational databases, CSV files. Each field has a defined data type, and relationships between tables are explicit. Structured data is easy to search, sort, filter, and analyse with SQL or spreadsheet formulas. Examples: student enrolment records, sales transactions, sensor readings.

Semi-structured data has some organisational properties but doesn't fit perfectly into tables. JSON (JavaScript Object Notation) and XML are common formats — they use tags or key-value pairs to label data, but individual records can have different fields. APIs (Application Programming Interfaces) typically return data in JSON format.

Unstructured data has no predefined format — text documents, emails, social media posts, images, videos, audio recordings. An estimated 80-90% of the world's data is unstructured. Analysing it requires specialised techniques: natural language processing (NLP) for text, computer vision for images, speech recognition for audio.

For Unit 1 Applied Computing, you'll primarily work with structured data. However, understanding that most real-world data is semi-structured or unstructured helps explain why data science has become such an important field — the challenge is converting messy, unstructured information into structured formats that can be analysed.

Practice questions

1. A researcher records the languages spoken by survey respondents. This data is:

   1. Continuous quantitative
   2. Discrete quantitative
   3. Nominal qualitative
   4. Ordinal qualitative

   (1 marks)

2. Explain three data quality issues that may affect a dataset collected by an online survey.

   (3 marks)

Summary

• Quantitative: measured/counted. Continuous (any value) or discrete (whole numbers).
• Qualitative: described. Nominal (no order) or ordinal (ordered categories).
• Primary: you collect. Secondary: already exists (ABS, research databases).
• Data quality: missing values, duplicates, inconsistent formats, outliers.
• Organisation: spreadsheet (rows=records, columns=fields) or relational database (tables + keys).