CELPIP Listening Part 5: Discussion | Complete Guide

CELPIP Listening Part 5 is the longest single audio clip in the Listening section and it has the most questions. Three or four people are discussing a topic. They have different perspectives. They agree on some things, disagree on others, and sometimes offer compromises. Your job is to track every speaker accurately and answer 8 questions about who said what and what was decided.

This is where many students’ scores drop a band. Not because the English is harder - it is not - but because the cognitive load is heavier. You need sustained focus, disciplined note-taking, and a clear system for attributing opinions to specific speakers.

This guide explains the format, walks through a realistic scenario, lists the traps, and gives you a step-by-step approach.

Part 5 at a glance

What is Part 5?

Part 5 is called Listening to a Discussion. You hear a group of people - usually three or four - discussing a topic where each one has a different view. It could be a meeting, a panel, or an informal group conversation. The discussion is not a debate. It is a working-through of perspectives.

The topics are accessible and contextual. A team deciding how to allocate a small budget. Neighbours discussing a community garden proposal. Colleagues working out how to handle a scheduling conflict. The vocabulary is everyday workplace and community English.

The structure of the discussion usually has someone introducing the topic, then each speaker contributing their view, then some back-and-forth where speakers respond to each other. By the end, there may or may not be a clear decision - sometimes the discussion just maps out the positions.

The exact format with a worked example

Imagine you hear this discussion between four colleagues - Mike, Sarah, Dev, and Lisa - about whether to move their team to a four-day work week:

Mike: Okay, so the proposal is to move our team to a four-day work week. Same hours per week, just compressed. I want to hear what everyone thinks. Sarah: I am in favour. Studies show people are more productive in shorter weeks. I think we should pilot it for three months. Dev: I am not sure. We have clients who expect five-day coverage. If we are all off on Fridays, who handles Friday calls? Lisa: That is a real issue. But what if we staggered it? Half the team off Monday, half off Friday. We still get coverage. Mike: That is interesting. Sarah, would you be okay with a staggered version instead of everyone off the same day? Sarah: I prefer everyone off the same day for team cohesion, but staggered is better than no pilot. I could live with it. Dev: With staggering, I am more comfortable. The coverage issue was my main concern. Lisa: Great. So we could pilot the staggered version for three months and review. Mike: Let me draft a proposal and circulate it. We will need HR approval, but I think we can move on this.

Typical Part 5 questions might include:

What is Dev’s main concern about the proposal? A. The cost of the pilot B. Coverage for clients on the off-day C. The length of the pilot period D. Team cohesion

The correct answer is B. Dev raises the coverage issue explicitly. D is a trap because Sarah mentions team cohesion - not Dev. A and C are not raised at all.

What compromise does the group reach? A. Cancel the proposal B. Move to a four-day week with everyone off Friday C. Pilot a staggered version with split off-days D. Increase weekly hours to make up for the shorter week

The correct answer is C. The group settles on the staggered pilot. B is a trap because that was the original proposal but not the agreed version.

What CELPIP raters and the algorithm test

A CLB 9 listener on Part 5 keeps a clean mental map of all speakers and their evolving views. A CLB 6 to 7 listener typically conflates two speakers or loses track when the discussion shifts.

Step-by-step approach for Part 5

Step 1: Memorise speaker names in the preview window

Before the audio plays, you can see the speakers’ names or roles. Spend the preview reading every name aloud in your head. If there are four speakers - Mike, Sarah, Dev, Lisa - say their names twice. This primes your memory so that when each voice appears, you have a slot ready.

Step 2: Anchor each voice in the first 10 seconds

The moment each speaker first speaks, anchor their voice to a feature you can remember:

• Mike: deeper male voice, leads the discussion
• Sarah: higher female voice, speaks first
• Dev: softer male voice, raises concerns
• Lisa: confident female voice, offers compromise

Write these anchors on your scratch pad. Voice anchoring is the single most effective Part 5 technique.

Step 3: Set up a speaker grid

On your scratch pad, draw a quick grid with one column per speaker:

As the discussion progresses, add one or two keywords under each name. For the example above:

This grid is your map of the discussion. When a question asks about a specific speaker, you go straight to their column.

Step 4: Track shifts in position

Speakers in Part 5 often change their views during the discussion. Sarah starts wanting everyone off the same day, then accepts a stagger. Dev starts opposed, then comes around with the compromise. Mark these shifts with arrows or notes:

• Sarah: same day → ok with stagger
• Dev: against → ok with stagger

The test loves to ask about the final position, not the initial one.

Step 5: Identify the final decision

The decision in Part 5 is usually in the last 30 seconds. It may be a compromise, a deferred decision, or a clear choice. Note exactly what was decided - if it was a staggered pilot, not just a pilot. Detail matters in the decision question.

Common traps in Part 5

CELPIP Part 5 vs IELTS GT Listening vs PTE Core

Part 5 maps most closely to IELTS Listening Section 3, which is also a multi-speaker discussion in an academic or work context. The speaker-attribution skill transfers directly. The main difference is the accent (Canadian for CELPIP, mixed for IELTS) and the position-shift element, which CELPIP includes more often than IELTS.

For PTE-trained students, Part 5 is a step up in complexity. PTE tasks are usually shorter and have fewer simultaneous voices to track. The Part 5 sustained 3-4 minute discussion with 4 speakers is more demanding.

How to practise Part 5

Multi-speaker podcast practice. Listen to Canadian podcasts that feature roundtable discussions or panel conversations. CBC has several of these. After each segment, write down each speaker’s position. The skill of holding three or four positions in mind simultaneously is the core Part 5 skill.

Voice anchoring drills. Practise identifying speakers by voice contrast. Listen to short clips with multiple speakers and try to name them within the first 10 seconds based on voice features alone. Build this reflex so it happens automatically in the test.

Speaker grid practice. Take any 3-minute discussion clip. Draw a speaker grid on paper, fill it in as you listen once only, then check against a transcript. Over time, your grid will become faster and more accurate.

Endurance practice. Listen to multi-speaker audio for at least 30 minutes a day. Part 5 comes after 30+ minutes of test, so mental fatigue is a factor. The students who score CLB 9 have trained their listening stamina, not just their listening accuracy.

Full Part 5 sets under timed conditions. Sit through complete Part 5 sets without pausing. Use the virtual scratch pad. Review every wrong answer and figure out which speaker you misattributed or which position shift you missed.

The Marvel Education CELPIP practice software gives you multi-speaker Canadian-accent discussions with the exact Part 5 format, a virtual scratch pad, and detailed answer review that flags speaker-attribution errors. Start your free trial here.

Quick checklist before test day

• I can identify 3 or 4 speakers by voice within the first 10 seconds of a discussion
• I can keep a speaker grid on a scratch pad while listening continuously
• I track position shifts when speakers change their view
• I can listen to 3 to 4 minutes of multi-speaker audio without losing focus
• I have done at least 10 full-format Part 5 sets under timed conditions
• I have reviewed every wrong answer and identified which speaker I confused

Part 5 is the part where preparation discipline shows up clearly. The students who built voice-anchoring and speaker-grid habits do well. The students who tried to wing it usually drop marks. After Part 5 you face Part 6, which is shorter but more nuanced in its opinion analysis.

A deeper look at Part 5 question types

The 8 questions in Part 5 are the most of any CELPIP Listening part, which makes the part heavily weighted in your overall score. The questions follow a predictable mix.

The topic question

Usually the first question. What is the group discussing? The answer is in the opening exchange and is usually clear. The trap is picking an option that focuses on one speaker’s framing rather than the overall topic.

The speaker-view questions

The bulk of Part 5 questions ask about specific speakers. What does Mike think? What is Sarah’s main concern? What does Dev suggest? These reward your speaker grid. Go to the right column, find the position, match it to an option.

The traps are wrong-speaker options (the view belongs to a different speaker) and initial-position options (the speaker changed their view during the discussion).

The agreement question

Some Part 5 sets ask which speakers agreed with each other or which speakers held opposing views. This tests whether you tracked agreements as they emerged.

The signal phrases are “I agree with that”, “exactly”, “that is fair”, “I see what you mean”. When you hear these, note the agreement on your scratch pad - “Sarah agrees with Lisa on stagger”.

The compromise or decision question

The final question often asks what the group decided or what compromise was reached. This is where the modification trap appears - the original proposal versus the final agreed version.

The signal phrases for decisions are “so we will”, “okay let us go with”, “I will draft a proposal for”, “we agree to”. Note the final decision with all modifications on your scratch pad.

The specific-detail question

Within a discussion, one speaker often mentions a fact - a number, a date, a place. A question may ask about that specific fact. This requires catching detail while still tracking the conversation.

The trick is to capture both speaker positions and specific facts. Your scratch pad needs to accommodate both - a position line per speaker plus a separate notes area for facts.

The reasoning question

Sometimes a question asks why a speaker takes their position. Why is Dev concerned? Why does Sarah prefer the same day? The answer is in the reasoning the speaker offers.

Listen for “because”, “since”, “the reason is”, “I think this because”. Note reasons next to positions on your scratch pad.

Score-level differences in Part 5

If your Part 5 score is stuck below CLB 9, the most common cause is endurance. The audio is 3 to 4 minutes plus 8 questions, all coming after 30 minutes of prior test. Build listening stamina with full-length practice sessions, not just isolated Part 5 sets.

A 7-day pre-test preparation plan for Part 5

Day 1 to 2 - Voice anchoring drills

Find multi-speaker podcast clips - panel discussions, roundtables. Practise identifying each speaker by voice within the first 10 seconds. Build a mental anchor for each one. Repeat with at least 10 different clips.

Day 3 to 4 - Speaker grid practice

Take 5 multi-speaker clips (3 to 5 minutes each). Build a speaker grid on paper as you listen once only. Capture each speaker’s position with 2 to 3 keywords. Check against a transcript.

Day 5 to 6 - Full Part 5 sets

Do 2 full Part 5 sets per day under timed conditions. Use the virtual scratch pad. Answer all 8 questions and review wrong answers. Categorise errors - wrong speaker, initial position, missed agreement, missed modification.

Day 7 - Endurance simulation

Sit through a full-length Listening section all in one go - all 6 parts. This builds the focus and stamina you need for Part 5 to land cleanly after 30 minutes of prior audio.

Final tips for test day

Memorise speaker names in the preview window. This is the single most important pre-audio task. If you do not know the speakers when the audio starts, you will spend the first 30 seconds figuring them out instead of listening.

Anchor each voice as soon as it appears. Within the first sentence of each speaker, lock their voice to their name. Write a one-word anchor on your scratch pad. This is the foundation of everything else in Part 5.

Use a clear grid layout on your scratch pad. Even four columns labelled with names is enough. Add 2 to 3 keywords under each name as the discussion progresses. The visual layout makes it easier to retrieve information for each question.

Track position shifts with arrows. Speakers change their minds in Part 5. Mark the initial position and the final position separately. The test will often ask about both.

Stay focused all the way to the end. Many students drift in the last 30 seconds because they think they have caught everything. The final 30 seconds usually contain the decision or compromise. Stay locked in.

Part 5 is a stamina test as much as a listening test. Treat it that way. Build endurance, build speaker-tracking habits, and the 8 questions become a steady bank of marks.

A second worked example - community garden discussion

Here is another Part 5 discussion to walk through. Four neighbours - Tom, Aisha, Karen, and Devon - are discussing a proposed community garden in their condo building.

Tom: So the proposal is to convert the rooftop space into a community garden. Each unit could rent a small plot for the season. Aisha: I think it is a great idea. I have been wanting to grow vegetables for years. Count me in. Karen: I am supportive in principle, but I am worried about maintenance. Who waters when people are away? Who fixes the irrigation if it breaks? Devon: Same concern. And what about cost? If we install garden boxes and irrigation, we are talking real money. Tom: The proposal includes a small monthly fee from participants to cover maintenance and a shared maintenance schedule. Boxes are funded by a one-time setup fee. Aisha: That sounds reasonable. I would pay the setup fee. Karen: If maintenance is covered by participants, I am much more open to it. Could we start with a smaller pilot - maybe 10 boxes - to see how it goes? Devon: A pilot makes sense. I would join in year two if the pilot works. Tom: Okay, so a 10-box pilot for one season, funded by participants. Karen, can you help me put the formal proposal together? Karen: Sure. I will draft the maintenance schedule.

What to notice:

• Aisha is enthusiastic from the start
• Karen is cautious but moves to supportive when maintenance is addressed
• Devon is cautious and commits to year two, not year one
• Tom is the facilitator throughout
• The final decision is a 10-box pilot, not the full rooftop conversion

Typical questions and answers:

What is Devon’s position by the end of the discussion? A. Fully opposed to the garden B. Joining the pilot in year one C. Open to joining in year two if the pilot works D. Volunteering to help organise

The answer is C. Devon never commits to year one and explicitly says “year two if the pilot works”. A is too extreme. B contradicts his cautious approach. D is Karen’s role, not Devon’s.

What did the group decide? A. To convert the entire rooftop into a garden B. To run a 10-box pilot funded by participants C. To cancel the garden idea D. To wait until next year to decide

The answer is B. The original proposal (full rooftop) was modified to a pilot. A is the unmodified original. C and D were not the final outcome.

Common Part 5 scenarios you should be ready for

Part 5 discussions fall into a small set of recurring scenarios. Recognising the scenario type helps you anticipate the kind of discussion structure and the likely decision pattern.

Workplace decision scenarios

A team meeting where the group is deciding something - a new policy, a project approach, a budget allocation, a hiring approach. Each speaker brings a different angle - cost, risk, fairness, efficiency. Vocabulary - proposal, pilot, review, budget, team, schedule, policy.

Community planning scenarios

A neighbourhood or community group discussing a local initiative - a park improvement, a community event, a building project. Speakers have varied stakes and perspectives. Vocabulary - residents, council, proposal, funding, location, schedule, community.

Event organising scenarios

A group planning an event - a conference, a celebration, a workshop. Decisions involve venue, timing, format, budget. Vocabulary - venue, schedule, programme, attendees, budget, logistics, RSVP.

Workplace conflict resolution scenarios

A group working out a disagreement or difficult situation - how to handle a workload imbalance, how to address a process issue, how to support a struggling team member. Vocabulary - workload, support, performance, communication, resolution.

Policy or rule discussion scenarios

A group debating a rule or policy - workplace dress code, scheduling rules, communication standards. Different speakers advocate for different approaches. Vocabulary - policy, expectation, flexibility, standard, accountability.

Resource allocation scenarios

A group deciding how to use limited resources - time, budget, staff. Each speaker pulls for different priorities. Vocabulary - allocation, priority, budget, resources, capacity, investment.

Recovery strategies if you lose track

Losing track of speakers is the most common Part 5 failure. Recovery techniques:

If you lost track in the middle of the discussion, anchor on the next speaker who is clearly identifiable. Then work backwards in your head - “that was the person who was just talking, so the previous one must have been”. Reset your grid based on the speaker you just identified.

If you missed a position, listen for the speaker to reference their view again later. People in discussions often restate their positions, especially when responding to others. “As I said before, my main concern is” tells you what their position was.

If you missed the final decision, check your grid for who agreed with what at the end of the discussion. The decision usually aligns with the convergence point - the option that the most speakers supported.

If you missed an entire segment, use the question stems for clues. If a question asks about a speaker who barely featured in your notes, you can usually narrow the options using what you did catch and educated elimination.

What CELPIP Part 5 is really testing about your English

Part 5 measures whether you can function in Canadian meetings and group discussions. Most professional Canadian work involves meetings - team huddles, project reviews, decision-making sessions. The skill of tracking multiple speakers, attributing views, and following compromises is daily currency in Canadian workplaces.

Beyond work, this skill matters in community life. Strata meetings, parent-teacher discussions, volunteer organisations, social planning. Whenever a group is figuring something out together, you need the Part 5 skillset.

This is why Part 5 has 8 questions - the most of any Listening part. Paragon weights it heavily because it tests the kind of listening that matters most in real Canadian life. Master it, and you are ready not just for the test but for the conversations you will have in your first month in Canada.