CELPIP Speaking Task 4: Making Predictions

Predict what will happen next in a scene.

~2 min each

0 of 28 sets done

Practice sets

1 Predict the Chaotic Classroom Outcome FREE 2 Predict the Downtown Crosswalk Incident 3 Predict the Busy Cafe Mishap 4 Predict the Lakeside Campsite Moment 5 Predict the Crowded Mall Mishap 6 Predict the Ball-Hockey Goal Outcome 7 Predict the Farmers' Market Moment 8 Predict the Airplane Cabin Chaos 9 Predict the Sea-Lion Show Splash 10 Predict the Hotel Lobby Check-In 11 Predict the Lakefront Water-Sports Outcome 12 Predict the School Playground Moment 13 Predict the Intersection Fender-Bender Outcome 14 Predict the Family Baking Mishap 15 Predict the Lakeside Fishing-Trip Moment 16 Predict the Sports-Day Field Outcome 17 Predict the Family Dinner Gathering 18 Predict the Moving-Day Mishap 19 Predict the Daycare Playroom Mess 20 Predict the Park Field-Day Finish 21 Predict the Community Kitchen Outcome 22 Predict the Working Farm Moment 23 Predict the Break-Room Slip 24 Predict the Parking Lot Mishap 25 Predict the Amusement Park Moment 26 Predict the Big-Top Circus Mishap 27 Predict the Construction Site Lift 28 Predict the Carnival Midway Win

Up next in Speaking Task 5 · Comparing and Persuading Preview

Guide

Task 4 gives you back the picture from Task 3 and asks what happens next. Because the image is familiar, the gravitational pull is to describe it again, and describing it again scores nothing.

What is CELPIP Speaking Task 4?

Paragon's description is unambiguous: "Look at the same illustration from Task 3 and describe what you think will happen next." You get 30 seconds to prepare and 60 seconds to speak. The on-screen instruction is simply: "In the picture below, what do you think will happen next?"

The reuse of the Task 3 image is official and it is the most exploitable fact about this task. It means your Task 4 preparation can begin during Task 3.

How CELPIP Speaking Task 4 is scored

Every CELPIP Speaking response is rated by trained human raters (Paragon says each performance is seen by three to five of them, working independently) on the same four dimensions: Content/Coherence, Vocabulary, Listenability and Task Fulfillment. Your Speaking level is a product of your scores across all eight tasks, so no single task can sink you, and no single task can carry you.

Paragon does not publish a per-task weighting, and you should be sceptical of any site that claims one. What the official rater commentary does show is which dimension tends to get flagged on which task, and that is what the notes below describe.

Content/Coherence is where this task is won and lost. The official rater notes on a Level 8 response were blunt: the quality and number of ideas "could be improved and developed with examples", and the organisation "could be improved with more effective transitions and linking words." Under-developed, listy predictions are exactly what this task exposes.

There is also a pure relevance risk under Task Fulfillment: anything you say that is already visible in the picture earns you nothing here. Description was the previous task.

CELPIP Speaking Task 4 strategies

Use your imagination, but stay realistic. Official wording. Your predictions must connect to what the people are actually doing.
Ask what the people will do next and what they will feel as a result. The feelings half is officially named and most candidates skip it.
Mine the surrounding objects for ideas. The props in the background are prediction fuel.
Vary your future forms deliberately. Sixty seconds of "will" shows a narrow range; mix in "is going to", "is about to", "is likely to", "I imagine".
Chain your ideas: prediction, then reason, then consequence. Two or three developed predictions beat six one-liners.

Common CELPIP Speaking Task 4 mistakes

Describing the picture again. Same image, strong pull, direct relevance penalty.
Recycling your Task 3 sentences. They felt good the first time; they score nothing the second.
One flat, obvious prediction ("they will watch the movie") stretched across 60 seconds, or the opposite failure: predictions so wild they lose the thread of what the people are doing.

Useful phrases for CELPIP Speaking Task 4

Bridge from scene to future: "Judging by the way they're queuing up, I'd say they're about to..."
Vary the forms: "They're going to..." / "She'll probably..." / "He's likely to..." / "I imagine that..." / "Chances are..."
Hedge (cheap vocabulary range): "Most likely..." / "I wouldn't be surprised if..."
Consequences and feelings: "As a result, she'll probably feel..." / "Once that happens, I expect..."

Tip: A prediction is only worth saying if it has a "because" and a "so". "They'll head inside because it's starting to rain, so the vendors will probably start packing up." That is one idea doing the work of three.

CELPIP Speaking Task 4: frequently asked questions

Is CELPIP Speaking Task 4 the same picture as Task 3?

Yes. Paragon's official study pack defines Task 4 as looking at "the same illustration from Task 3". Bank a few unused details while you are describing it in Task 3 and use them here.

What if my prediction is wrong?

There is no correct prediction. The scene is imaginary. What is required officially is that your predictions be realistic and connected to what the people in the picture are doing.

What tense should I use in CELPIP Speaking Task 4?

Future forms, and deliberately more than one: "will", "is going to", "is about to", "is likely to", plus hedges like "probably" and "I imagine". Range is part of the score.

Can I just describe the picture again?

No. Description was Task 3. This task is scored on what happens next, so anything already visible in the image earns nothing.

How many predictions should I make?

Typically two or three, each with a reason and a likely consequence. The official rater note on the sample response was that the ideas needed developing with examples, not that there were too few of them.

How long is CELPIP Speaking Task 4?

30 seconds to prepare, 60 seconds to speak.