Key Takeaways
- Q4 answers demonstrate empathy through specific actions and language — not by simply stating 'I would be empathetic'
- Q2 answers tend to jump straight to problem-solving without acknowledging the human element
- Raters spend 30–90 seconds per response, so your opening sentence is critical
- Acknowledging complexity and normalising difficulty are hallmarks of top-scoring responses
- The same content scored Q2 can become Q4 with different framing and emotional depth
Understanding the Q1–Q4 scoring scale
Every response in the GP SJT is scored on a quartile scale from Q1 (lowest) to Q4 (highest). Each of your 22 answers is marked by a different independent human rater who typically reviews 40–70 questions per hour. This means raters form their impression quickly — often within the first few sentences of your response.
| Quartile | What it signals | Typical characteristics |
|---|---|---|
| Q4 (top 25%) | Outstanding judgement and empathy | Leads with genuine empathy, acknowledges complexity, normalises difficulty, shows nuance before practical steps |
| Q3 (50–75th) | Competent but missing depth | Shows some empathy but quickly moves to solutions, decent structure, lacks emotional depth |
| Q2 (25–50th) | Adequate but formulaic | Uses buzzwords without demonstrating them, jumps to action, may sound rehearsed or robotic |
| Q1 (bottom 25%) | Poor judgement or awareness | Dismissive of emotions, inappropriate responses, misses the human element entirely |
Example scenario
Scenario
You are a GP registrar. A colleague, Dr Patel, confides that she has been making small clinical errors recently because she is going through a difficult divorce and not sleeping well. She asks you not to tell anyone. What would you do?
A typical Q2 response
"I would listen to Dr Patel in a non-judgmental way and show empathy. I would advise her to speak to her supervisor about her situation. Patient safety is the priority, so I would encourage her to take some time off if needed. If she refused, I would need to escalate to the practice principal because patient safety comes first."
This response is not wrong — it covers the right themes. But notice how it tells the rater about empathy ('non-judgmental', 'show empathy') without actually demonstrating it. It jumps quickly to advice and escalation. A rater reading this at speed sees a checklist, not a thoughtful human response.
A Q4 response to the same scenario
"This must be incredibly difficult for Dr Patel — managing a divorce while maintaining the demands of clinical work takes an enormous toll, and I appreciate her trust in sharing this with me. I would start by letting her know that it is completely understandable to struggle in these circumstances and that many doctors face similar challenges without feeling safe to speak up. I would gently ask how she is feeling right now and what support she currently has in place. Regarding the clinical errors, I would want to explore this with her — not as an accusation but as a shared concern. I might say something like 'I can see you care deeply about your patients, and I wonder if we could think together about how to keep things safe for you and for them while you navigate this.' I would explore whether she has considered speaking to her supervisor or accessing the Doctors' Health Advisory Service, framing this not as an obligation but as something that could genuinely help. If the clinical errors posed an immediate safety risk and she was unwilling to seek support, I would explain that I have a professional obligation to ensure patient safety, but I would do this transparently with her — not behind her back."
What makes the Q4 response different?
The difference is not about length — it is about depth and authenticity. Here are the specific patterns that separate Q4 from Q2:
- Opens with genuine empathy — 'This must be incredibly difficult' vs 'I would show empathy'
- Normalises the situation — 'many doctors face similar challenges' removes shame and builds trust
- Asks before advising — 'I would gently ask how she is feeling' vs jumping straight to solutions
- Uses specific language — quotes what you might actually say ('I can see you care deeply...') rather than describing actions abstractly
- Acknowledges the tension — explicitly holds the tension between supporting a colleague and protecting patients
- Acts transparently — if escalation is needed, it happens with the colleague, not behind their back
The golden rule
Show, don't tell. Instead of writing 'I would be empathetic', write what an empathetic person would actually say. Raters are trained to spot the difference — and they see hundreds of responses that claim empathy without demonstrating it.
Common patterns in Q2 answers
- Buzzword stacking — 'I would be empathetic, non-judgmental, and supportive' without showing what that looks like
- Premature action — jumping to 'I would refer them to...' before acknowledging the person's experience
- Robotic structure — 'Firstly... Secondly... Thirdly...' that feels rehearsed rather than authentic
- Binary thinking — treating the scenario as having one correct answer rather than holding multiple valid perspectives
- Missing the human — focusing entirely on procedures and policies while ignoring how the person feels
How to practise moving from Q2 to Q4
The good news: this is a learnable skill. You do not need to be naturally eloquent — you need to train yourself to pause before problem-solving and genuinely engage with the human situation first. Here is a practical framework:
- Read the scenario twice — first for the facts, second for the emotions
- Identify the person's emotional state — what are they feeling? Scared? Ashamed? Overwhelmed?
- Write your opening sentence — it should acknowledge that emotion specifically
- Normalise before advising — one sentence showing this is an understandable human response
- Then and only then — offer practical support, framed collaboratively rather than directively
Try this framework on our free practice scenarios. Each one includes a model Q4 answer so you can compare your response and identify specific areas for improvement. Our AI feedback evaluates your response across all 9 competencies and highlights exactly where you gained or lost marks.