CELPIP Speaking Task 4: How to Predict What Will Happen Next
By David · Speaking · Published · 3 min read
Task 4 shows you the same picture you described in Task 3 and asks what will probably happen next. You get 30 seconds to prepare and 60 seconds to speak. Many people repeat their Task 3 description or invent one dramatic story. The answers that score well do something simpler: they make several reasonable predictions, each based on something you can see.
How do I predict what will happen in CELPIP Speaking Task 4?
Choose three or four people or objects that are in the middle of an action, and for each one say what will probably happen next and why, based on a clue in the picture. A clue plus a prediction plus a result is the whole structure.
For example: a waiter is carrying a tall stack of plates and a child is running toward him, so he will probably have to stop suddenly, and some plates might fall. The clue (child running), the prediction (stop suddenly) and the result (plates fall) make the answer logical and easy to follow.
What should I do in the 30 seconds of preparation?
- Find three or four 'unfinished actions': someone about to do something, or something about to go wrong.
- Write one or two keywords for each, in the order you will talk about them.
- Pick one that could connect to another, so you can link two predictions if you have time.
You already know the picture from Task 3, so use the preparation time only for what will happen, not for describing again.
What grammar and phrases work in Task 4?
Use the future with different degrees of certainty, and connect each prediction to its reason. Variety here shows range without any rare vocabulary.
| Certainty | Phrases |
|---|---|
| Very likely | is going to, will almost certainly, I'm sure that |
| Likely | will probably, is likely to, I expect that |
| Possible | might, may, could, there's a chance that |
| Linking a reason | because, since, judging by, it looks as if |
| Linking a result | so, as a result, which means, and then |
What are common mistakes in Speaking Task 4?
- Describing the picture again instead of predicting.
- Telling one long invented story that loses the connection to the picture.
- Using only will, which sounds repetitive and certain about everything.
- Making a prediction without a reason from the picture.
- Spending 50 seconds on one person and leaving no time for others.
What does a good Task 4 answer sound like?
Original CELPass example (a park on a windy afternoon): "I think a few things are about to happen in this park. First, the man reading on the bench has left his newspaper open beside him, and since it's so windy, it's probably going to blow away, so he'll have to chase it across the grass. Near the pond, the little girl is leaning over the edge to feed the ducks. Her mother is looking at her phone, so the girl might lean too far, and her mother will probably rush over to pull her back. Finally, the dark clouds on the right suggest it's going to rain soon, which means the family having a picnic will likely pack up quickly and run for the shelter."
Frequently asked questions
Is Task 4 the same picture as Task 3?
Yes. Task 4 uses the picture from Task 3, so what you noticed while describing it is your material for the predictions.
Can my predictions be creative?
Yes, as long as each one is reasonable and linked to something in the picture. Logical and specific beats dramatic.
Keep reading
CELPass is an independent practice platform and is not affiliated with Paragon Testing Enterprises. CELPIP is a trademark of Paragon Testing Enterprises. Check the official CELPIP website and IRCC for current test rules and requirements.