🧠 AI Research Assistant: Defining 'Good' Questions

Kia ora, Namaskaram 🙏🏾

Before we dive into gathering evidence, we need ‘good’ research questions.

AI can think like a reliable PhD-level research assistant.

But only if we train it well — to ask sharper questions, and to seek your feedback as the senior researcher.

Here’s a prompt for your next project.

👉🏾 Check out my prompt to ask ‘good’ questions directly on ChatGPT

Sample ChatGPT Input:

Act as my curious junior researcher with a PhD in behavioural science. Ask me, one-by-one, the ‘5Ws for Behaviour Change’ questions to help clarify the objectives of our research project:

What behaviours are we seeking to change?
Who are the primary and secondary audiences?
Why do we believe this change will have a positive impact?
When do we need this change to happen?
Where can we reach our audiences?

After I share this, ask for feedback on this:
A. Reflect to me all the mental models, assumptions, and worldviews that we are introducing with this research project.
B. Use the COM-B model of behaviour change to frame 12 research questions on what we don’t know and what we need to find out.
(P.S. Ask your senior researcher for feedback.)

⏭️ Don’t provide evidence. Only ask questions.

My initial research question :

How might we support young New Zealanders (18–30) to reduce their use of Buy Now, Pay Later (BNPL) services for non-essential purchases?

📚 Reference (COM-B Model for Behaviour Change)

Michie, S., van Stralen, M. M., & West, R. (2011).The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation Science, 6(1), 42. https://doi.org/10.1186/1748-5908-6-42

Sample ChatGPT Output:

COM-B: 12 Research Questions on what we don’t know yet

Capability

1. Do young people fully understand how BNPL works and its potential long-term consequences?

2. What financial literacy gaps exist around BNPL services among 18–30 year olds?

3. How confident do users feel about managing spending without relying on BNPL?

Opportunity

4. How heavily are young people targeted by BNPL promotions across different channels (e.g., social media, fintech apps)?

5. What alternative opportunities exist for making purchases without BNPL (e.g., savings, low-cost credit options)?

6. What social norms currently surround BNPL use among peer groups or online communities?

Motivation

7. What emotions (e.g., excitement, shame, fear) are driving BNPL use for non-essential purchases?

8. What immediate rewards are people seeking when they choose BNPL instead of saving or delaying purchase?

9. How much does “fear of missing out” (FOMO) contribute to impulsive BNPL usage?

Behavioural context (additional framing)

10. What life events or moments (e.g., sales, birthdays, back-to-school) most strongly trigger BNPL usage?

11. What role does perceived affordability play in decision-making around BNPL vs. traditional credit options?

12. How do lower-income young people perceive BNPL compared to higher-income peers — necessity, luxury, or risk?

Shhh...

Early access to my course is now open !

Thank you for supporting my work 💚 

With love and gratitude, Vishal