Research Questions

Overview

Every Data Analytics SAT starts with a research question. Your research question decides what data you need to collect, what analysis you perform, and what your infographic or dynamic data visualisation needs to show. A well-chosen question makes the rest of the SAT much easier; a poorly chosen one can leave you without enough data, or with nothing interesting to say.

This section covers what the study design expects from a research question, the features of a good research question, and how to refine a broad topic into a question that is ready for the SAT.

Research Questions in the Study Design

Unit 3 Outcome 2

Key Knowledge:

• features of a research question, including:
  • clarity
  • measurability of data requirements
  • feasibility
  • originality

Key Skills:

• propose a research question

The study design also states that “the scope of the research question should be sufficiently detailed to not require a simple yes or no answer, but rather an exploration of the data collected.”

Study design key knowledge and key skills are taken verbatim from the VCE Applied Computing Study Design 2025-2028. © Victorian Curriculum and Assessment Authority. For current versions and related content visit www.vcaa.vic.edu.au

Features of a Good Research Question

Clarity

A clear research question can only be read one way. Anyone reading it should understand exactly what you are investigating, who or what it is about, and over what period or place.

• Unclear: How has the world changed?
• Clear: How has life expectancy changed in Australia, Indonesia and Nigeria between 1960 and 2020?

Avoid vague words such as “better”, “good” or “affect” unless you define what you mean by them.

Measurability of Data Requirements

You must be able to identify the specific data needed to answer the question, and that data must be measurable. When you write your question, you should be able to list the variables (fields) you will need.

For example, “Is there a relationship between GDP per capita and life expectancy?” requires:

• GDP per capita for each country, for each year
• Life expectancy for each country, for each year

If you cannot list the data you need, your question is probably not measurable yet.

Feasibility

A feasible question can actually be answered with the time, skills, software tools and data available to you. Ask yourself:

• Does the data exist, and can I access it (for free)?
• Is there enough data to find trends, relationships and patterns?
• Can I collect the primary data I need ethically and within the time available?
• Can I complete the analysis and the visualisation before the SAT is due?

A question about the spending habits of every Australian teenager might be interesting, but it is not feasible if no such data set is available to you.

Originality

Your question should be your own. It should not simply repeat a question that has already been answered in an article or a sample SAT, and it should give you the opportunity to find something out by exploring the data. Combining data sets in a new way, or focusing on a particular location, group or time period, is a good way to make your question original.

Not a Yes or No Question

A research question that can be answered with “yes” or “no” does not give you enough to explore. Questions that start with how, to what extent or what is the relationship between usually lead to a deeper exploration of the data.

• Yes/No: Do richer countries have longer life expectancy?
• Better: To what extent does GDP per capita explain differences in life expectancy between countries, and has this relationship changed since 1960?

Refining a Research Question

Most good research questions start out as a broad topic. Refining the question means narrowing it down until it meets each of the features above.

Worked Example 1: Health and Wealth

Step	Question	What’s wrong with it?
1. Topic	Health around the world	This is a topic, not a question
2. First question	Are people healthier in rich countries?	Yes/no answer. “Healthier” and “rich” are not defined
3. Measurable	Is there a relationship between GDP per capita and life expectancy?	Measurable, but very broad and widely answered already
4. SAT-ready	How has the relationship between GDP per capita and life expectancy changed in South-East Asian countries between 1980 and 2020?	Clear, measurable, feasible (Gapminder data) and more original

Worked Example 2: Student Wellbeing

Step	Question	What’s wrong with it?
1. Topic	Students and sleep	This is a topic, not a question
2. First question	Do students get enough sleep?	Yes/no answer. “Enough” is not defined
3. Measurable	How many hours of sleep do Year 12 students get?	Measurable, but only describes the data. There is no relationship to explore
4. SAT-ready	What is the relationship between hours of sleep, screen time before bed and self-reported concentration among Year 11 and 12 students at our school?	Clear and measurable. Uses primary data (a survey) that can be collected ethically with consent

The second example relies on primary data. If you collect data from people, make sure you follow the procedures on the Legal and Ethical Considerations page, such as using consent forms and de-identifying personal data.

Checklist

Before you submit your research question, check that you can answer “yes” to each of the following:

• Could someone else read my question and understand exactly what I am investigating?
• Can I list the specific data (variables) I need?
• Have I found the data, or do I have a realistic plan to collect it?
• Is there enough data to find trends, relationships or patterns?
• Does my question require an exploration of the data, not a yes/no answer?
• Is my question my own, and not one that has already been answered?

Learning Resources

• Developing Research Questions
• Research Questions: Examples and Samples
• SAT Overview - your research question is the first part of the Unit 3 SAT
• Data Collection Methods - primary and secondary data collection

Data Sources

When developing your research question, check that suitable data is available to answer it. Some good places to look:

• Gapminder
• Maven Analytics Data Playground
• Australian Bureau of Statistics (ABS)
• data.gov.au

Topic Questions

Question 1

List the four features of a research question described in the study design. For each feature, explain in one sentence why it is important.

Question 2

Explain why “Do people in cities earn more than people in the country?” is not a suitable research question for the SAT. Rewrite it so that it is.

Question 3

For the research question “How has the proportion of renewable energy in electricity generation changed across Australian states since 2005?”, list the data (variables) you would need to collect, and suggest a source for that data.

Question 4

A student proposes the research question “What is the relationship between the weather and the mood of every secondary student in Victoria?” Evaluate the feasibility of this question and suggest how it could be made more feasible.

Question 5

Start with a broad topic that interests you. Using the worked examples above as a guide, refine it in at least three steps into a SAT-ready research question. At each step, explain what you changed and why.