Analysing Data

Data and Information

🎯 (CT5-DAT-01, CT5-DAT-02)

📌 Differentiate between data and information

Data is a set of raw facts, such as numbers, words or pictures. On its own, data often means very little. Look at this list: 14, 9, 17, 12, 9.

Information is what you get when data is presented in a way people can understand. NESA’s glossary describes it as “the presentation of data in a manner that is readily understood”.

Say the list is the quiz scores from a made-up Year 8 class. Sorted from lowest to highest, the scores are 9, 9, 12, 14, 17. Now you can say: “Two students scored 9, the most common score.” That sentence is information.

Inside a computer, all data is stored as number codes. For example, the capital letter A is stored as the code 65 in ASCII, a common code for letters.

The Purpose of Data Analysis

🎯 (CT5-DAT-02)

📌 Describe the purpose of analysing data

Analysing data means examining data to find out what it shows. You look for patterns (things that repeat), trends (changes that build up over time) and outliers (values that sit far away from the rest). Then you use what you found to answer a question or make a decision.

People analyse data for several reasons:

• To answer a question, such as which snack sells best.
• To spot a problem early, such as a drop in attendance.
• To make a decision, such as which days to open the canteen for longer.
• To explain something to others, such as with a chart at a school assembly.

She did not need every name on the list. She needed the pattern. Finding the pattern is the purpose of the analysis.

Inputs, Storage, Transmission and Outputs

🎯 (CT5-THI-01, CT5-DAT-01)

📌 Describe inputs, storage, transmission, processes and outputs in data analysis

Many data analyses use five stages. Not every analysis uses all five, but these are the five to know: input, storage, transmission, process and output.

• Input is the data that goes into the system, such as answers typed into a form.
• Storage is where the data is kept, such as a spreadsheet file or a database (an organised collection of data).
• Transmission is sending data from one place to another over a network. The transmission media are the means it travels through, such as Wi-Fi or a cable.
• Process is what the system does with the data, such as sorting it, adding it up or working out averages.
• Output is the result, such as a chart, a table or a report.

Functional Requirements

🎯 (CT5-DPM-01, CT5-THI-01)

📌 Specify the functional requirements of a data analysis, including stating the purpose of a solution, describing use cases and developing test cases of inputs and expected outputs

A functional requirement says what the data tool must do. Each one starts with “The system must”. You should be able to test each one, so keep it specific.

To specify functional requirements, follow four steps:

1. State the purpose. Write one sentence that says why the tool exists.
2. Describe a use case. This is a short story of one user doing one task.
3. List the requirements. Number each one.
4. Write test cases. Each test case has an input (what goes in) and an expected output (what should come out).

Write each expected output before you run the test. If you decide what should happen after you see the result, you may accept a wrong answer.

Non-Functional Requirements

🎯 (CT5-DPM-01, CT5-DAT-02)

📌 Specify the non-functional requirements of a data analysis

• Cost of access to data.
• Social benefit or fraud prevention.
• Scaling efficiently to large datasets.

A non-functional requirement says how well a data analysis must work, or what it must be like. It covers qualities such as cost, speed and how useful the results are to people. Where you can, write it so that you can check it.

NESA gives three examples. Here is how each one could look for the youth club’s survey tool.

Cost of access to data

The club has a small budget. A possible requirement: the total cost of the data and tools must stay within the club’s budget. A possible test: add up every cost and compare the total with the committee’s budget.

Social benefit or fraud prevention

A social benefit is a good effect on people. Fraud is cheating to get money or other things. Suppose the club runs a made-up raffle. A possible requirement: the tool must flag any ticket number that appears twice. A possible test: enter the same ticket number twice and check that the tool flags it.

Scaling efficiently to large datasets

A tool that works well on 30 rows may slow down badly on 300,000 rows. A possible requirement: the tool must count 300,000 made-up rows within two minutes. A possible test: create a made-up file with that many rows, run the count and time it.

Social, Ethical and Legal Responsibilities

🎯 (CT5-SAF-01)

📌 Consider the social impacts and ethical and legal responsibilities in analysing data

When you analyse data, the results affect real people. Your choices about what to collect, how to show it and who sees it all matter. To consider these effects, sort them into three groups: social, ethical and legal.

Social impacts

• Labels. A chart can make a group look good or bad. If a club only shows its most frequent attenders, quieter members may feel left out.
• Trust. People are often less willing to share data when they are unsure how it will be used.

Ethical responsibilities

• Honesty. Do not hide results that do not match what you hoped to find.
• Consent. Ask people before you use information about them, and say what it is for.
• Credit. If you use someone else’s dataset, say where it came from.

Legal responsibilities

• Privacy. The Privacy Act 1988 (Cth) and the Australian Privacy Principles set rules for how personal information is collected, used and kept safe. The Office of the Australian Information Commissioner (OAIC) explains these rules.
• Copyright. The Copyright Act 1968 can protect some datasets and other creative works. Check the licence or permission before you copy or share them.

Perspectives of Diverse Groups

🎯 (CT5-DES-01, CT5-SAF-01)

📌 Explore data analysis considering the perspectives of diverse groups, including Aboriginal and Torres Strait Islander Peoples, culturally and linguistically diverse people, people of different ages and gender, and people with disability

• Sampling bias for underrepresented groups in machine learning.
• Subtitles or closed captions for people who are d/Deaf or hard of hearing.
• Indigenous Cultural and Intellectual Property (ICIP), such as recording Aboriginal languages.
• World Wide Web Consortium (W3C) standards for machine-readable text.

A perspective is a point of view. The same data can look different to different groups of people. Good data analysis thinks about the groups it affects. The syllabus names four groups: Aboriginal and Torres Strait Islander Peoples; culturally and linguistically diverse people; people of different ages and genders; and people with disability. Culturally and linguistically diverse people come from many cultural backgrounds, and some speak another language at home.

Sampling bias

A sample is the group of people or items you collect data from. Sampling bias happens when a sample leaves out some groups or has too many of others. In machine learning, a computer model learns from examples. If most examples come from one group, the model may work less well for the others. A made-up example: a voice app trained mostly on one accent may struggle with other accents.

Captions and accessible text

Closed captions show spoken words as text on the screen. They help people who are d/Deaf or hard of hearing. The World Wide Web Consortium (W3C) writes web standards. Good standards mark up text so that a screen reader (software that reads the screen aloud) can read it.

Indigenous Cultural and Intellectual Property (ICIP)

ICIP covers the cultural heritage of Aboriginal and Torres Strait Islander Peoples, such as their knowledge, stories, art and languages. It also covers the rights communities hold over them. Recording an Aboriginal language is a data task, and a cultural one too. The community should decide who records it, where it is stored and who can use it.

Data Compression

🎯 (CT5-DAT-01, CT5-DAT-02)

📌 Explain simple compression of data and types of compression

Compression makes a file smaller. A smaller file takes up less storage and can be sent faster over a network. Think of packing a suitcase. Rolling and folding clothes lets more fit inside, but you can still unpack them.

There are two main types. They work in different ways.

• Lossless compression keeps enough information to rebuild the original exactly. Nothing is lost. It suits text documents, program code and files that must stay exact. Examples: ZIP files and PNG images.
• Lossy compression throws away details that people are unlikely to notice. The file becomes much smaller, but the original cannot be rebuilt exactly. It suits photos, music and video, especially when they are streamed. Examples: JPEG photos and MP3 music.

Small and Big Datasets

🎯 (CT5-DAT-02, CT5-EVL-01)

📌 Explore the applications of small and big datasets

A dataset is a collection of data gathered for a specific purpose. Datasets come in different sizes. Size changes what you can do with the data, and how you do it.

A small dataset is one you can check by eye. It might be a class’s birthdays or a team’s scores for one season. A spreadsheet holds it easily, and you can read every row yourself.

A big dataset is too large to check by hand. It may hold very many records, or it may come from many sources at once. Ordinary spreadsheets struggle with it, so people use special tools and storage.

• Small: a canteen tallies the snacks sold each day for two weeks, to decide what to order next.
• Big: a city council combines traffic sensor readings from many roads over a year, to plan where to put bus lanes.
• Small: a family compares the costs of three phone plans before choosing one.
• Big: a weather service combines readings from many stations to study long-term change.

Each size has trade-offs. A small dataset is easy to check, but it may be too small to show a reliable pattern. A big dataset can show more, but it is harder to store, protect and check.

How Data Analysis Has Evolved

🎯 (CT5-EVL-01)

📌 Describe how data analysis has evolved in response to people's needs and opportunities

People have always counted things. At first that meant tally marks on paper, or lists kept in a notebook. As people needed to count more, and more quickly, they found better tools. Each change came from a need or a new opportunity, such as a new invention.

This table shows a simplified story. The stages overlap, and not everyone used each tool at the same time.

Each stage changed what people could ask. A spreadsheet lets you change one number and see the result straight away. A database lets many people use the same records. Sensors and online data let people study things that were too big or too fast to track by hand.

Design Principles and Data Ownership

🎯 (CT5-DES-01, CT5-DAT-02, CT5-SAF-01)

📌 Explore design principles and issues relevant to analysing data, including visualisation principles, data trails and ownership of data

• Accessing and restricting your historical data or tracking users to customise advertising with browser cookies.
• Indigenous Cultural and Intellectual Property (ICIP) or onselling of databases to third parties.

A design principle is a guideline that helps a solution work well. For data analysis, three areas matter: visualisation principles, data trails and ownership of data.

Visualisation principles

A visualisation is a chart, graph or picture that helps an audience understand data. Good ones follow simple rules:

• Give the chart a clear title and labelled axes (the lines along the edges of a graph).
• Start a bar chart’s scale at zero, so the bar lengths are honest.
• Do not rely on colour alone. Some people cannot tell certain colours apart.
• Use a chart type that fits the data. A bar chart compares categories. A line graph shows change over time.

Data trails

A data trail is the record of where data came from, who changed it and where it went. A digital footprint is the traces a person leaves behind when they use digital systems.

Ownership of data

Ownership asks who has the right to control data. It may be the person it describes, the business that collected it or the community it came from. Two NESA examples show the issues:

• Browser cookies. A cookie is a small file a website stores in your browser. It can remember you, and it can track what you visit so that adverts can be customised.
• Onselling databases. This means a business sells its database to a third party, which is any organisation outside the original agreement. Community knowledge, such as Indigenous Cultural and Intellectual Property (ICIP), should not be sold on without the community’s agreement.

Collecting and Interpreting Data

🎯 (CT5-SAF-01, CT5-DAT-02)

📌 Collect and interpret data adhering to privacy and cybersecurity principles

Collecting data is more than gathering it. You must respect the people the data is about, and keep it safe. Two ideas guide this work:

• Privacy is the right of people to control their personal information. Personal information is anything that can identify a person.
• Cybersecurity is the protection of information technology elements, including hardware, software, data and network services.

Use this checklist when you collect data:

1. Collect only what you need. If a survey only needs favourite sports, do not ask for a birthday.
2. Tell people why. Say what the data is for, and who will see it.
3. Ask for permission. Let people choose whether to take part, and tell them they can stop.
4. Protect it. Use a password on the file, and share it only with people who need it.
5. Remove names when you can. Many results make sense without them.

When you interpret data, be careful about what it can prove. A survey of 20 students does not show what a whole school thinks.

Data Types and Structuring Data

🎯 (CT5-DAT-02, CT5-DAT-01)

📌 Represent and store data to facilitate computation, including selecting appropriate data types, understanding data type limitations and structuring data systematically

A data type tells a computer what kind of value it is storing. Common types are:

• Text (also called a string): letters and symbols, such as names. Text cannot be added up.
• Whole number (integer): a number with no decimal point, such as 14 goals.
• Decimal: a number with a decimal point, such as 3.5 seconds.
• Date: a day, month and year, such as a training date.
• True or false (also called Boolean): a yes or no answer, such as “paid: true”.

Each type has limits. A whole number cannot store 3.5, so use a decimal for times.

Structure data in a systematic way. Use one table. Each row is a record, such as one student. Each column is a field, such as a name. Each field holds one data type.

Modelling Entities and Attributes

🎯 (CT5-DAT-02, CT5-THI-01)

📌 Model entities, events and their attributes using structured data

An entity is a person, place, thing or idea that you want to store data about. An attribute is a detail that describes an entity. An event is something that happens at a time and place, such as a match or a loan.

Think about a school library. The entities are books and members. A book has attributes such as title, author and shelf code. A member has attributes such as member number, name and year level. A loan is an event. It records when a book was borrowed and when it is due back.

To model an entity, follow three steps: name it, list its attributes and choose a data type for each one.

Modelling Relationships in Data

🎯 (CT5-DAT-01, CT5-DAT-02)

📌 Model the relationships between entities and events using relational data

A relational database stores data in several linked tables. Each table holds one kind of entity or event. Tables are linked by IDs that appear in more than one table.

Two key words describe the links:

• A primary key is an attribute that uniquely identifies each row in its own table, such as MemberID.
• A foreign key is an attribute in one table that holds the primary key of another table. In the Loan table, MemberID is a foreign key.

The most common relationship is one-to-many. One member can have many loans, but each loan belongs to one member. One book can be loaned many times, but each loan is for one book.

Spreadsheets and Databases

🎯 (CT5-DAT-02, CT5-THI-01)

📌 Compare the usability of data using a spreadsheet or database to analyse the same dataset

A spreadsheet stores data in cells arranged in rows and columns. A database stores data in tables, and a database program manages them. Both can hold the same dataset, so you can compare how easy each one is to use for the same job.

Cyber Safety and Privacy Issues

🎯 (CT5-SAF-01, CT5-DAT-01)

📌 Investigate issues with the use of data, including cyber safety, security, privacy and ethics

• Breach of copyright.
• On-selling of databases to third parties.
• Storing online payment details, such as credit card numbers.
• Sharing personal identifier details, such as age, address, gender, geolocation, MAC address.

Cyber safety is about the behaviour and precautions you take when you share personal information online. Security protects data from people who should not see or change it. Privacy is about who is allowed to know personal details. Ethics asks whether an action is fair and responsible, even when it is legal.

To investigate an issue, ask three questions: who is affected, what could go wrong, and what would reduce the risk? Here are four issues from the NESA examples.

• Breach of copyright. This happens when someone copies or uses a work, such as a dataset or a chart, without permission. Use only data you are allowed to use, and say where it came from.
• Onselling databases to third parties. A business might sell a list of customers’ details to another business. People who shared their details for one reason may not expect that.
• Storing online payment details. A credit card number is very sensitive. Do not save card numbers in a spreadsheet or a school file. Payments should go through the trusted payment service itself.
• Sharing personal identifiers. A personal identifier is a detail that can point to one person. Examples include age, address, gender, geolocation (where a device is) and a MAC address. A MAC address is a code built into a device’s network card. Shared together, these details can reveal who someone is.

Flat-File and Relational Databases

🎯 (CT5-DAT-02, CT5-COM-01)

📌 Analyse data in both a flat-file and relational database using queries and reports

A flat-file database keeps all its data in one table. A relational database keeps data in several linked tables. You can ask either kind of database questions. A query is a question, such as “which books are overdue?” A report is a neat, readable summary of the answers.

Queries use keywords. This page uses a small set: SELECT, FROM, WHERE, ORDER BY, LIMIT, AND, INSERT INTO and UPDATE.

-- Flat file: one table called Loans
SELECT Title, Borrower, DateDue
FROM Loans
WHERE DaysLate > 0
ORDER BY DateDue
LIMIT 5;

Read it line by line. SELECT picks the columns. FROM names the table. WHERE keeps only the rows that match. ORDER BY sorts them. LIMIT shows only the first five.

In a flat file, the borrower’s name is repeated in every row. In a relational database, the Loans table holds only a MemberID, and the name lives in the Members table. A question about both tables can be answered in two simple steps. First, find the member. Then, list that member’s loans.

-- Step 1: find the MemberID
SELECT MemberID, Name
FROM Members
WHERE Name = 'Student B';

-- Step 2: list that member's loans
SELECT BookID, DateDue
FROM Loans
WHERE MemberID = 1002
ORDER BY DateDue;

A report turns results like these into a neat page. The club might print the five most overdue loans, with a title and the date, for the coordinator to read at a meeting.

Classifiers and Machine Learning

🎯 (CT5-DAT-02, CT5-EVL-01)

📌 Explore how a classifier uses data analysis for machine learning

A classifier is a program or model that sorts items into groups, called classes. It decides which group an item belongs to by looking at its features (its details). A spam filter that sorts email into “spam” and “not spam” is one example.

Machine learning is a way of building a model from examples. The model finds patterns in past examples, then uses them to sort new items.

A classifier usually follows four steps:

1. Collect examples. Gather data where the right answer is known. This is labelled data.
2. Train. The model looks for patterns that link the features to the labels.
3. Test. Give the model new examples it has not seen, and count how many it sorts correctly.
4. Use and check. Let the model sort new items, and keep checking its results.

Real-World Problems and Users

🎯 (CT5-DPM-01, CT5-THI-01, CT5-DES-01)

📌 Define a real-world problem or question that can be solved by analysing data, including breaking it down into manageable parts and describing the users of the solution

A problem is a gap between what is and what should be. A good data project starts with a question that data can answer. “Why is everyone unhappy?” is too big to answer with data. “Which lunch options do Year 9 students choose most often?” is specific enough to analyse.

Follow three steps:

1. Write the question in one or two sentences.
2. Decompose it. To decompose a problem means to break a complex problem into parts that are easier to understand, design and implement. Each part should be a smaller question you can answer.
3. Describe the users. A user is someone who uses the solution. A stakeholder is anyone affected by it, even if they never use it.

Alternative Designs and Choices

🎯 (CT5-DPM-01, CT5-THI-01)

📌 Generate alternative designs and evaluate them against the requirements to select a preferred design

A design is one possible way to solve a problem. The first idea is rarely the best one. Good designers generate alternative designs (two or more different options), check each one against the requirements, and choose one with a reason.

Follow four steps:

1. Generate at least two different designs.
2. Check each one against the same functional and non-functional requirements.
3. Record the results in a table, using Yes, Partly or No.
4. Choose the preferred design, and give a reason.

Design B is the preferred choice. It meets every requirement and is easy for beginners. Design C is more powerful, but it fails the beginner requirement. Design A is slow to check for 200 members.

Digital Solutions from Data

🎯 (CT5-DPM-01, CT5-DAT-02)

📌 Develop a digital solution using a range of software to interpret and represent data to create information for a real-world scenario

A digital solution uses software to turn raw data into something people can understand. Data is raw facts, such as numbers or words. Information is data that has been organised so a reader can make sense of it.

You will build a small solution in free spreadsheet software. The data comes from a made-up youth club bake sale. The stall sold food and drinks on a Friday and a Saturday. Each row is one item, and each column holds one kind of data.

Before you share the solution, check the totals a second way. The Friday column adds to 165 and the Saturday column adds to 200. Together they make 365. The Total column also adds to 365, so the two methods agree.

Data Collection, Ownership and Protection

🎯 (CT5-SAF-01, CT5-DAT-01)

📌 Specify what data is collected, who owns it, and how it will be protected

Before you collect any data, write a data plan. It answers the questions people often forget: what is collected, why, who owns it, how it is protected and when it is deleted.

Data ownership means who has the right to decide how the data is used. It is not always obvious, so write down who you think owns the data before you share it.

Here is a made-up plan for a class survey about favourite canteen lunches. It collects no names.

If the data is about Aboriginal and Torres Strait Islander people or communities, ask the community first. This connects to Indigenous Data Sovereignty, the right to decide how data is collected, stored and used. Indigenous Cultural and Intellectual Property (ICIP) also matters, such as recordings of Aboriginal languages. Communities have their own rules, so ask for their guidance.

If anyone asks you for personal details online, tell a trusted adult.

Interactive Online Data Sharing

🎯 (CT5-COM-01, CT5-DES-01, CT5-DAT-02)

📌 Create interactive solutions for sharing information online with a visualisation library

Sometimes you want other people to see your data online. An interactive chart lets the viewer do something, such as hover over a bar to see its value. A visualisation library is a set of ready-made chart tools that a program can use.

A library usually has a licence. This is the set of rules that says how you may use it. Choose a free library with a clear licence and a tutorial. Then follow these steps.

Project Notebooks

🎯 (CT5-COL-01, CT5-DPM-01)

📌 Document the design and implementation of the solution in a project notebook

A project notebook is a written record of your project. It explains what you built, why you built it that way and what happened when you tested it. Someone who has never seen your project should be able to follow it from start to finish.

Start a new, dated entry each time you make a change. Keep the entries in order. Each entry should include these parts:

• Goal: what the work is for.
• Steps: what you did, in order.
• Tools and data: the software and the data you used.
• Problems: what went wrong, and how you fixed it.
• Result: what the solution now does.
• Next: what you will check or change next.

Collecting and Validating Data

🎯 (CT5-DAT-02, CT5-DAT-01, CT5-SAF-01)

📌 Use appropriate methods to collect, store, validate and verify qualitative and quantitative data, considering data integrity considering privacy and personally identifying information (PII)

Data comes in two main kinds. Qualitative data describes qualities, such as opinions or favourite flavours. Quantitative data is measured or counted, such as sales or the number of sports played.

Before data is saved, check it. Validation checks that data is sensible and follows the rules, such as “a whole number from 0 to 10”. It cannot tell you whether the data is true. Verification checks that data is correct by comparing it with its original source, such as a paper form. Keeping data accurate and complete is called data integrity.

This made-up club survey, about sport and lunch, uses four rules:

1. Presence check: the lunch choice must not be left blank.
2. Range check: the number of sports played must be a whole number from 0 to 10.
3. List check: the lunch choice must be one of the five listed options.
4. Type check: the number box must not accept letters.

Verification happens too. Each day, check five surveys at random against the paper copies.

Collect only what you need, and remove personal details before you share data. This is called anonymising. Personally identifying information (PII) is any information that can identify a person.

Summary Formulas and Functions

🎯 (CT5-DAT-02, CT5-THI-01)

📌 Summarise data using formulas, functions and features of a spreadsheet, including complex formulas, aggregate functions and lookup functions

A formula is a calculation you type into a cell. It starts with an equals sign. A function is a named formula that the spreadsheet already knows, such as SUM. Formulas and functions turn a long list into a few useful numbers.

Some functions combine many values into one answer. These are aggregate functions, such as SUM and AVERAGE. Others find a value in a table. These are lookup functions, such as VLOOKUP. Here is the bake-sale data with a Total column (E) added.

Type each formula below into a blank cell, then check one result by hand.

• =SUM(E2:E6) adds the five totals. Result: 365.
• =AVERAGE(E2:E6) finds the mean of the totals. Result: 73.
• =MAX(E2:E6) and =MIN(E2:E6) find the largest and smallest totals. Results: 120 and 25.
• =COUNTIF(E2:E6,">=60") counts the totals of 60 or more. Result: 3.
• =SUMPRODUCT(B2:B6,E2:E6) multiplies each price by its total, then adds the results. Result: $1017.50.

Check your ranges. A range that stops one row too early misses a row, so the answer is wrong.

Spreadsheet Filters and Sorting

🎯 (CT5-DAT-02)

📌 Filter, group and sort data using a spreadsheet, including using filters and sorting, using conditional formatting and grouping and aggregating data

A filter hides rows that do not match a rule. It does not delete them. A sort puts rows in order, such as from largest to smallest. Conditional formatting changes how a cell looks, such as its colour, when a rule is true. Grouping and aggregating means putting rows into groups, then calculating a total or average for each group.

Dashboards, Predictions and Decisions

🎯 (CT5-COM-01, CT5-DAT-02, CT5-THI-01)

📌 Present data and make predictions and decisions using a spreadsheet, including creating a data dashboard or report in a spreadsheet, decision formulas and optimisation

A dashboard is one page that shows the most important numbers and a chart, so a reader can decide quickly. A report is a set of results with a short explanation. Your readers might be a committee, a teacher or the people who run the stall.

The formula gives “Make more” for Cookies, Lemonade and Brownies, and “Check first” for Muffins and Fruit cups. A person still decides what to do with that result.

A prediction is a reasoned guess about the future, based on data. Lemonade rose from 50 to 70 between the two days, so it may sell well again. Two days is not much data, so say how sure you are.

Optimisation means choosing the best option from several, within a limit. Suppose the club can make only 100 items. Brownies sell for $4.00 each and lemonade for $2.50 each, so if every item sells, brownies bring in more money. This assumes buyers want them all, and the table leaves out costs.

Database Reports and Decisions

🎯 (CT5-DAT-02, CT5-COM-01)

📌 Analyse data to make decisions and generate reports using a database

A database stores data in tables. Each row of a table is a record, and each column is a field. A query is a question you ask the database. A query can select some rows, sort them and show the results as a report.

The bake-sale data is stored in a table called stall_sales. It has four fields: item, price, friday and saturday.

SELECT item, saturday
FROM stall_sales
WHERE saturday >= 35
ORDER BY saturday;

A report puts the result into a short table with a heading, such as “Items that sold 35 or more on Saturday”. Then it makes a decision. The results support keeping cookies, lemonade and brownies. Fruit cups and muffins are the two lowest Saturday sellers, so the club could check them next time before it decides.

Try it on the stall table

Adding and Updating Database Records

🎯 (CT5-DAT-02, CT5-DAT-01)

📌 Load, insert and update data in a database

Loading means bringing data into a database table. You can load many rows at once, or add one row at a time. Two commands do most of the work. INSERT INTO adds a new row. UPDATE changes values in rows that already exist.

The club is adding a new item, Scones, which has no sales yet:

INSERT INTO stall_sales (item, price, friday, saturday)
VALUES ('Scones', 2.50, 0, 0);

Next, the club lowers the price of cookies to $2.20:

UPDATE stall_sales
SET price = 2.20
WHERE item = 'Cookies';

Check the type of each value as you load it. A price should be a number, such as 2.50. Text such as ‘$2.50’ may not work as a price.

Data Visualisation: Trends and Outliers

🎯 (CT5-DAT-02, CT5-COM-01)

📌 Generate a data visualisation to identify trends and outliers using a range of tools

A visualisation is a picture of data, such as a chart. Charts help people see patterns that are hard to spot in a table. A trend is a pattern that goes up or down over time or across items. An outlier is a value that is far away from the rest.

Choose the chart to suit the question. A column chart compares items. A line chart shows change over time, such as a series of weeks.

Specialist Terminology in Context

🎯 (CT5-COM-01)

📌 Select and use specialist terminology in context

Specialist terminology is the words that people in a field use with exact meanings. Using them correctly shows that you understand the topic. Choose the term that fits the situation, and use it in a full sentence.

Iterative Development Records

🎯 (CT5-DPM-01, CT5-COL-01)

📌 Create a record of project development demonstrating iterative design and evaluation

An iteration is one repeat of a process. Each cycle should move the work closer to the goal. In design, you plan, build, test and improve, then go round again. A development record shows each cycle, so a reader can see what changed, why, and whether it worked.

Each cycle answers three questions: what changed, did it work, and what comes next?

Project Reviews and Requirements

🎯 (CT5-COL-01, CT5-DPM-01)

📌 Evaluate their own project and that of their peers using predetermined functional and non-functional requirements

To evaluate means to judge something against clear criteria and give a reason. The criteria for a project are its requirements, which you wrote down before you built it. Functional requirements say what the solution must do. Non-functional requirements say how well it must do it, such as being easy to use or quick to load.

You can check your own project and a peer’s project in the same way. Make a checklist with one row for each requirement. Mark each one Met, Partly met or Not met, and write the evidence beside it.

When you judge a peer’s work, be specific and kind. Say what works, what is missing and what evidence you saw. Judge the work, not the person.

Security and Responsibility Checks

🎯 (CT5-SAF-01, CT5-DPM-01)

📌 Evaluate whether solutions meet social, ethical and legal responsibilities and cybersecurity principles

Every data solution affects people. Its social effects are how it changes people’s lives and relationships. Its ethical questions are about what is fair and respectful. Its legal responsibilities are rules that must be followed, such as privacy laws.

Cybersecurity principles are basic ideas for keeping data safe. Three common ones are: give access only to people who need it; use strong passwords and keep them private; and keep software up to date. The Australian Cyber Security Centre (ACSC) publishes guidance on these ideas.

Use this checklist to evaluate a solution:

• Does it collect only the data it needs?
• Did the people whose data is collected agree to it?
• Is personal information stored securely, and deleted when it is no longer needed?
• Could it harm, or unfairly leave out, any group of people?
• Does it follow the rules that apply, such as the Privacy Act 1988 (Cth)?

The 3Vs: Volume, Variety and Velocity

🎯 (CT5-DAT-02, CT5-DPM-01)

📌 Evaluate sourced data processed using the 3Vs: volume, variety and velocity

The 3Vs describe large or fast data. They are volume (how much data there is), variety (how many different types and formats it comes in) and velocity (how fast new data arrives or changes).

• Volume: 20 rows of sales have low volume. A file with millions of sensor readings has high volume.
• Variety: one table of numbers has low variety. Text, photos, video and sensor readings together have high variety.
• Velocity: a sales report made once a month has low velocity. A live feed that updates every few seconds has high velocity.

When you evaluate sourced data, you judge whether it suits your purpose. The 3Vs give you a structure for that judgement.

Calculations from Datasets

🎯 (CT5-DPM-01, CT5-DAT-02)

📌 Assess a developed solution based on calculations from datasets

To assess a solution means to judge how well it works, using evidence. For a data solution, one of the most important pieces of evidence is whether its calculations give the right answers.

A simple way to check a calculation is to use a small dataset where you already know the answer. This is a test case. It has an input and an expected output, and you work out the expected output before you run the test.

Verification of Data and Results

🎯 (CT5-DAT-02, CT5-DPM-01)

📌 Perform verification of datasets, calculations and outputs

Verification checks that data, calculations and results are correct. You do not assume they are right. You check them against the source, or by a second method. You met verification when you collected data. Here you apply it to the whole solution.

Four checks are useful:

• Compare with the source. Check each value against the paper form or the original file.
• Use a second method. Work the answer out with a calculator, or with a different formula, and compare.
• Reconcile totals. To reconcile means to check that the parts add up to the whole.
• Check the outputs. Does the chart match the table? Does the report match the query?

Tools and Processes for Validation

🎯 (CT5-DPM-01, CT5-THI-01)

📌 Evaluate tools and processes used in the analysis of data for validation

A tool is the software or device you use, such as a spreadsheet, a database or a chart library. A process is a set of steps you follow, such as checking each row before you save it. To evaluate them, judge whether they help you reach correct and trustworthy results.

Validation processes check that data is sensible and follows the rules, such as a range check. Many spreadsheets have a data validation tool that can stop wrong entries. Test it: type a value that should be rejected, and check that it is.

Judge each tool or process against criteria, such as these:

Careers in Data Analysis

🎯 (CT5-EVL-01)

📌 Explore interests and careers in analysing data

Data analysis is used in many kinds of work. Schools, sports clubs, councils, hospitals and businesses all collect data, and they need people to make sense of it. Some jobs focus on collecting and storing data. Others focus on finding patterns and explaining what they mean.

Here are some example roles. Read job descriptions to find the exact duties of each one, because roles differ between employers.

• Data analyst: finds patterns in data and reports them to the people who make decisions.
• Database administrator: sets up, looks after and protects databases.
• Data visualisation specialist: designs charts and dashboards that explain data clearly.
• Business analyst: works out what an organisation needs, and how data can help it.
• Cyber security specialist: protects systems and data from misuse.

To explore your interests, ask yourself three questions. Do you enjoy finding patterns? Do you like explaining things to other people? Do you enjoy keeping systems safe? Match your answers to the roles above.