The hardest student to teach in a mixed-level speaking class is not the advanced student who dominates every activity. It's the A2 student sitting next to a B2 partner, who has stopped talking because they don't have the vocabulary to say what they want to say.
Teachers often respond by giving lower-level students simpler activities. That's understandable, but it's usually the wrong solution. The problem isn't that the topic is too complex. It's that the structural support isn't there. Given the right scaffolding, an A2 student can discuss the same topic as a C1 student - just differently.
YapYapGo is a classroom speaking practice tool for ESL and EFL teachers that filters questions by CEFR level, so lower-level students in a mixed class get accessible versions of the same topic. But scaffolding goes beyond the question itself.What scaffolding actually means
Scaffolding is borrowed from construction: a temporary structure that supports the real building until it can stand on its own. In language teaching, it means temporary language support that allows students to communicate at a slightly higher level than they could manage independently - and then gradually removes that support as ability grows.
The key word is temporary. Scaffolding that's never removed becomes a crutch. Good scaffolding is designed to become unnecessary.
Six scaffolding strategies that work in speaking
1. Sentence frames and starters
The most practical and underused tool in lower-level speaking classes. Before a discussion activity, put four to six sentence frames on the board:
- "I think... because..."
- "In my opinion, ... is important/difficult/interesting because..."
- "I agree/disagree because..."
- "One example of this is..."
- "In my country, ..."
Lower-level students use these frames as launching pads. They're not copying sentences - they're borrowing a structure that gets them started. The content is their own.
Remove them after three or four lessons on the same topic. Students who needed them initially will have internalised the patterns.
2. Pre-teach five words, not fifty
Vocabulary overload is the single most common reason lower-level students go silent during speaking activities. They have something to say but hit a word wall and stop.
Pre-teaching works best when it's targeted: five specific words that will come up in this activity, in this context. Not a vocabulary list - five words with example sentences and one practice use each.
"Today we're discussing healthcare. Here are five words you'll probably need: prevention, access, treatment, affordable, public. Can you use one in a sentence right now?"
Five words is manageable. Fifty creates more anxiety than it resolves.
3. Thinking time before speaking
This is the scaffolding most teachers skip because it feels like wasted time. It isn't.
Research by Foster and Skehan found that even 60 to 90 seconds of silent preparation before a speaking task produces significant improvements in fluency, complexity, and accuracy. Lower-level students benefit more than higher-level ones because their cognitive load is higher - they're doing more work to retrieve language in real time.
Before every pair discussion, give students 60 seconds of silent preparation. They can write key words (not sentences) if they want. Then pairs start. The difference in output is immediately noticeable.
Tool tip: YapYapGo's IELTS mode includes a built-in prep timer that mirrors the real exam - one minute to prepare before the student speaks. The same principle applies to any speaking activity. A standalone classroom countdown timer and speech timer visible to the whole class makes thinking time feel structured rather than awkward.
4. Model the extended answer
Lower-level students often don't know what "speak for two minutes" actually sounds like. They've been trained by grammar exercises to give one correct short answer.
Model it yourself before the activity. Answer the question: "What's the best meal you've ever eaten?" and show students the structure: give the direct answer, add a reason, add a detail, extend with a memory or example. Do it at normal conversational speed.
Then ask students to try the same structure. "Your answer, your reason, one specific detail." Three elements, not one.
5. Reduce the language demand without reducing the topic
Open-ended questions can be answered at any level. "What would you do if you had a million pounds?" can produce:
- A2: "I would buy a house and travel."
- C1: "I'd probably divide it between ethical investments and property, keeping a small amount liquid for immediate experiences while the rest compounds..."
Both are valid answers. Both students are on the same topic. The A2 student is not excluded - they just need the question phrased in a way that gives them a foothold.
Concrete personal questions work better than abstract hypotheticals for lower levels. "What do you spend most of your money on?" instead of "How do economic systems create inequality?" Same general area, very different cognitive demand.
6. Pairs of similar level for timed fluency work
Stretch pairing (mixing levels) works well for open discussion where the natural interaction scaffolds the lower-level student. But for timed fluency activities - the 4/3/2 technique, timed talks, rapid pair debates - matched pairs work better.
When an A2 student is timed against a C1 partner, the pace mismatch produces anxiety rather than development. Matched pairs allow lower-level students to work at their natural pace and feel the benefit of timed practice rather than just feeling outrun.
A random student picker is useful for calling on individuals to model good output. YapYapGo supports matched pairing based on the level you set for each student in your class settings. For the full framework on when to use which pairing approach, see our post on stretch pairing vs matched pairing. For managing the bigger picture of mixed-level classes, see speaking activities for mixed-level ESL classrooms.
The confidence loop
Every scaffolding strategy above serves the same underlying goal: getting lower-level students to actually speak. Not perfectly. Just consistently.
When lower-level students speak more - even imperfectly, even with frames, even with five pre-taught words - they build the confidence that makes them willing to take risks next time. That confidence is what accelerates acquisition. Students who are too afraid to speak are stuck. Students who speak imperfectly are moving.
The scaffold is temporary. The speaking habit it builds is not.
Sources:
- Foster, P. & Skehan, P. (1996). The Influence of Planning and Task Type on Second Language Performance. Studies in Second Language Acquisition. - Planning time significantly improves lower-level output.
- Vygotsky, L.S. (1978). Mind in Society. Harvard University Press. - Scaffolding and the Zone of Proximal Development.
- Krashen, S. (1982). Principles and Practice in Second Language Acquisition. - The Affective Filter: anxiety blocks production more at lower levels.
- Wood, D., Bruner, J., & Ross, G. (1976). The Role of Tutoring in Problem Solving. Journal of Child Psychology and Psychiatry. - The original scaffolding framework.