Formative Assessment in Language Classes: A Feedback Loop That Fits Your Routine

A correct answer on an exercise does not, by itself, show what a learner can do in conversation. An incorrect answer does not always reveal why they struggled, either. To decide what to teach next, teachers need evidence connected to a goal and a clear next step.

That is the purpose of formative assessment: gathering and interpreting signs of learning along the way, then using them to adjust teaching and practice. A quiz can be part of the process, but it is not formative by default. It becomes useful when the result informs a decision: revisit a point, offer a different challenge, provide support or invite another attempt.

There is an operational challenge, too. A 2025 study surveyed 288 university EFL teachers in one Chinese province and interviewed 11 of them; 76% of respondents identified assessment workload as a challenge. That figure describes the study’s sample, not language schools or independent teachers everywhere. It still raises a practical question: how can teachers notice enough to guide learning without turning every class into a pile of marking?

Start with the goal and the evidence, not the grade

Before preparing an assessment, complete this sequence:

Communicative goal: what should the learner be able to do?
Success criteria: what observable behaviour would show they can do it?
Evidence: in which task could that behaviour appear?
Decision: what will you do with what you observe?

Instead of “practise speaking,” for example, set a goal such as: “By the end of class, the learner can explain a simple problem and ask for an alternative.” Criteria might include explaining the problem, making a request and responding to a clarification question. A pair role-play can provide evidence; a separate test may not be necessary.

The Council of Europe recommends using relevant CEFR descriptors to bring planning, teaching and assessment closer together. Descriptors can clarify goals, structure activities and help teachers observe performance; they do not have to become permanent labels for learners.

A short formative-assessment loop

Use the main task in a lesson as an opportunity to observe learning. The sequence below is a planning suggestion, not a universal time requirement.

Step	What to do	Example
1. Choose a focus	Pick one goal and up to three observable criteria	“Explain the problem, ask for an alternative and confirm the next step”
2. Elicit a performance	Choose a task that resembles real communication	A role-play between a customer and a service representative
3. Record minimal evidence	Note an observation linked to a criterion, not every mistake	“Explained the problem clearly; has not asked a follow-up question yet”
4. Give actionable feedback	Name what worked and one action to try next	“You explained the problem. In the next round, ask a question before choosing a solution.”
5. Make room to act	Invite the learner to retry, revise or apply the feedback to a similar task	Repeat the role-play with a different problem

The final step is essential. A comment that a learner cannot understand or use is just more information. Whenever possible, leave time for another attempt and check that the learner understands what to do next.

Give specific feedback without correcting everything

When reviewing speaking or writing, it is tempting to mark every error. That can distract from the goal and multiply marking time. For each task, choose a main focus: message clarity, use of a target structure or ability to sustain an interaction, for example.

A simple feedback frame is:

“You were able to [observed evidence]. For your next step, try [specific action]. Shall we try it once more?”

For example: “You described the problem and asked for a solution. In the next round, add a question to check that you understood the alternative. Let’s change the scenario and try again.”

This does not mean ignoring important errors. It means prioritising what helps learners meet the goal of that activity and saving other patterns for a later lesson or feedback conversation. Cambridge English’s guide to integrating learning and assessment recommends linking feedback to task criteria, including next steps and giving learners a chance to act on the feedback.

Match the evidence to the skill

Skill	Possible evidence	A useful question for choosing the next step
Speaking and interaction	A short pair conversation	Did the learner initiate, respond or ask for clarification?
Writing	A short message with a clear purpose	Did the text achieve its purpose and use the expected register?
Listening	An answer paired with a clue the learner heard	Was the difficulty recognising sounds, vocabulary or the meaning of a phrase?
Reading	An answer with the passage that supports it	Did the learner locate information or infer meaning from context?

Getting a reading or listening question right does not always show how a learner reached the answer; getting it wrong does not automatically identify the cause. Ask for a brief clue about the process — which part helped, which word was unfamiliar or what they needed to hear again — to choose a more precise follow-up.

Keep the process manageable

Assess one focus at a time. You do not need to observe every skill in depth in every lesson. Rotate the focus across a course and keep track of the points that need follow-up.

Use activities that are already part of the lesson. A role-play, exit question or peer review can produce evidence when it is aligned with a clear goal. It does not have to become an additional test.

Record little, but make it useful. A short note such as “needs a model to ask for clarification” can help with the next lesson more than a score without context. In larger classes, combine a short response from everyone with rotating oral observations over several lessons; a single sample should not become a definitive diagnosis.

Invite learners into the process. Share the criteria before the task and ask for a simple self-check: “What could you do without help? What would you like to try next time?” Self-assessment can guide the conversation without replacing the teacher’s judgement.

For schools, align criteria without scripting every lesson. A team can agree on a unit goal and discuss examples of evidence while leaving teachers free to choose activities for their own classes. A brief conversation about two samples of student work can reveal whether the criteria are being understood consistently.

What to avoid

Grading every contribution, even when the task is simply practice.
Correcting many things at once without making the priority clear.
Collecting results without changing the lesson plan or giving the learner an action.
Confusing speed, silence or confidence with language proficiency.
Treating one observation as proof of a learner’s level or lasting progress.
Letting a digital or AI tool decide on its own what a learner has learned.

A platform can help organise lessons, exercises and records; it does not automatically turn data into pedagogical feedback. Slingui can help bring class routines and practice into a more organised workflow. Interpreting the evidence and choosing the next step remain the educator’s responsibility.

One question to ask after every lesson

Before closing your lesson plan, ask: what did I see or hear that helps me choose the next step for this learner or class? If you cannot answer yet, adjust the goal or task for next time. If you can, record the action in one sentence and revisit it later.

Formative assessment is not about giving more tests. It is about bringing goals, evidence, feedback and another opportunity to learn together in a cycle small enough to fit into the routine.