Recovery Casebook | Compare Decisions, Evidence and Results

A case contains a person, a difficulty, a choice and something that happened afterwards. Those ingredients make it readable. They do not, by themselves, make its explanation correct. A learner may improve after receiving a new notebook, but the notebook could coincide with clearer teaching, easier questions, more available time or ordinary variation in performance.

The first useful question is therefore precise: What can this case actually show us? It might show that a question was misunderstood, that a timetable was impossible, or that the evidence collected did not match the intended goal. It might show a promising way to ask the next question. It rarely licenses a prediction about every learner facing a broadly similar setback.

Read the cases here as small reasoning exercises. Their facts are supplied so that we can practise examining a record. In real life, we would also need to verify whether those facts were reported accurately. Fiction removes that practical difficulty while keeping the intellectual one: the conclusion still has to fit the evidence we have been given.

For example, “she completed more work” is different from “she learned the topic”. The first statement concerns output. The second concerns capability, which requires an appropriate check. “He worked independently on Tuesday” is also narrower than “he can now work independently”. A single occasion does not tell us what will happen with a different question next week.

This need for restraint does not make case reading pessimistic. It makes a useful change easier to recognise. A small improvement in a specific condition can be worth protecting even when a larger claim remains unproven. We can say that a learner now explains one important step without declaring the entire subject repaired.

Keep three sentences available as you read: “The record shows…”, “One possible explanation is…”, and “The next useful check would be…”. Each performs a different job. When all three collapse into “this method worked”, the reader loses the details needed to make a sound decision.

Yale’s teaching guidance presents case-based learning as applying knowledge to scenarios and examining possible responses. That is the educational purpose of this casebook. Yale Poorvu Center: Case-Based and Problem-Based Learning.

Contents · Next chapter

An evidence card is a short description of the conditions surrounding a result. It prevents a memorable final score from swallowing the rest of the story. You do not need a special form. A notebook page with the following fields is enough.

Notice the order. We define the capability before deciding whether the result looks good. If the aim is to select a method independently, a page of correct answers completed after method prompts is relevant practice evidence, but it does not answer the independence question. If the aim is to finish within a time limit, an accurate answer with unlimited time establishes a different capability.

Record assistance in ordinary language. “A parent said that this was a ratio question” is more useful than “some help”. “The teacher supplied the opening sentence” tells us more than “supported writing”. Specificity lets another reader understand what the learner contributed and what the environment supplied.

Also record what was absent. A case without delayed work cannot show retention over that delay. A case containing only one topic cannot demonstrate performance across the subject. Missing information is a boundary around the conclusion, not permission to assume the most favourable version.

Do not turn this into surveillance. For an everyday learning review, two short work samples and a few honest notes can be sufficient to choose a next step. A record that takes longer to maintain than the learning itself is probably asking too much. Collect the information needed for the current decision, then stop.

The next four chapters use this evidence card in different ways. Their outcomes deliberately differ. A casebook made entirely of successful endings would leave us poorly prepared to recognise partial progress, mistaken interpretation and a plan that should be changed.

Contents · Previous chapter · Next chapter

Fictional teaching case. Elena can carry out multiplication and division, but she struggles to choose when a proportional calculation is appropriate. Her aim is to decide on a method from the question, then explain why that method fits. Her family initially records only whether her final answer is correct.

On a short independent set containing ten straightforward proportional questions, she answers four correctly. During later practice, an adult points out the relationship in three questions and reminds her to calculate the amount for one unit. She answers eight of ten questions correctly. The family celebrates the apparent move from four to eight.

The worked practice has value, but the comparison needs its conditions. The first set was independent. The later set included three method prompts. The questions were different, and their difficulty was not formally matched. The record supports the statement that Elena completed more of the later set correctly under those conditions. It does not isolate the reason.

One question asks for the flour needed for ten equal servings when six servings require 150 grams. Elena writes 150 ÷ 6 = 25, then 25 × 10 = 250 grams. Asked why, she explains that each serving requires the same amount. That explanation is stronger evidence of understanding than the final number alone, although it remains one example.

The next set mixes proportional questions with questions involving a fixed charge. A delivery costs a fixed $4 plus $2 for each parcel. For five parcels, Elena multiplies the one-parcel total of $6 by five and gives $30. The correct total is $4 + 5 × $2 = $14. She has extended a successful proportional procedure into a situation where part of the cost stays fixed.

Her independent score on this mixed set is four out of ten. Calling the entire week wasted would be inaccurate. Calling the earlier eight out of ten proof of independent selection would also be inaccurate. The mixed set has revealed a condition that the earlier practice did not adequately examine.

Between the two mixed sets, Elena compares examples and explains what changes with each extra unit. The later seven out of ten is encouraging. It is not a controlled estimate of the effect of that comparison activity. The new questions differ, classroom teaching continues, and only one later occasion is available.

The next decision is narrow: retain examples in which she explains the relationship, then check another mixed set after a delay. The aim is to discover whether she selects appropriately without a familiar layout or a reminder. Adding more questions of the already recognisable type would not resolve that uncertainty.

What transfers from Elena’s case is the question about selection. The exact ratio exercise, number of questions and weekly schedule need not transfer. A learner who cannot yet divide 150 by six accurately needs a different investigation. A learner who selects correctly but runs out of time needs another.

For the method behind using a skill in an unfamiliar task, see How to Translate Knowledge Into Action. For subject-specific work, use the Mathematics Tuition and Learning Hub.

Contents · Previous chapter · Next chapter

Fictional teaching case. Safiya is writing a short email about a school reading session. The invented task asks her to invite a classmate, explain one practical benefit, mention the meeting place and keep the message within 90 words. Her initial draft contains the invitation but leaves out the other three requirements.

An adult asks questions, supplies a benefit and rewrites several sentences. Together they produce a polished response that meets all four requirements. If the only record is “first draft: one requirement; final draft: four”, the case looks like a complete recovery. The final artefact is better, but it contains two people’s decisions.

Here is a short extract from the original invented draft: “Please come to the reading session. I think it will be good.” The adult’s revision adds: “We can practise explaining difficult passages before Friday’s discussion. Meet me outside the library after lunch.” The added sentences solve genuine problems in the draft. They do not show that Safiya independently noticed or solved those problems.

A week later, a new task asks for a message about a poster-making session. Safiya again meets only one of four requirements independently. This does not prove that feedback is ineffective. It tells us that the jointly produced paragraph did not establish the capability the family believed it had established.

The next practice keeps ownership visible. Safiya reads the new task and tells the adult which decisions she needs to make. The adult can ask where an idea came from, but does not supply the idea. When Safiya cannot explain a requirement, they pause for teaching rather than quietly repairing the response for her.

On a later independent task, she meets the invitation, benefit and meeting-place requirements, but exceeds the stated word limit. That is partial progress in this fictional record. It is not a reason to pretend that the omitted constraint has become unimportant. It is also not a reason to deny the decisions she now makes for herself.

The useful comparison is between independent performances, with the difference in tasks acknowledged. The jointly revised draft belongs in the record as evidence of what supported practice produced. Keeping it separate preserves its value without turning it into an independence claim.

This case also changes the conversation at home. “You did it perfectly last week” would be unfair: last week’s perfection included substantial adult contribution. A more accurate sentence is, “Last week we completed those decisions together; today we can see which ones you can make alone.” That wording directs attention to the work instead of assigning blame.

Before borrowing the lesson, check the learner’s starting point. Someone who cannot understand the task vocabulary may need clarification before independent planning becomes possible. Someone who understands every requirement but cannot formulate sentences needs language teaching. Removing all help indiscriminately could conceal the very difficulty we need to identify.

The transferable lesson is to distinguish the quality of an artefact from the independence of its author. The non-transferable detail is the four-item checklist: a different writing task needs its own criteria. Use How to Translate Feedback Into Change for the feedback process, and the English Tuition and Learning Hub for subject routes.

Contents · Previous chapter · Next chapter

Fictional teaching case. Arun’s family wants him to improve his explanations in Science. They assemble two short learning checks with the same broad structure: fifteen factual items and five explanation items, each carrying one point in this invented exercise. These are practice checks, not official examination papers or marking schemes.

Arun’s first result is ten correct factual items and one correct explanation item: eleven out of twenty. After several practice sessions, he answers fourteen factual items and one explanation item correctly on the second check: fifteen out of twenty. The overall percentage has risen from 55% to 75%, a difference of 20 percentage points.

That headline is arithmetically correct. It is incomplete as an answer to the family’s question. Every additional point is in the factual component. The explanation component remains one out of five. Furthermore, the two checks contain different questions, so even the component comparison should be treated as a classroom signal rather than a precise measurement of growth.

One explanation asks why a conclusion cannot be supported by an experiment that changes two conditions at once. Arun names both conditions correctly but does not explain that their separate contributions cannot be distinguished from that comparison alone. His factual knowledge is relevant. The missing step lies in connecting the experimental arrangement to what can be concluded.

Reading the answer matters because the score cannot tell us which part is missing. A second learner might receive the same explanation score because they misunderstand the scientific vocabulary. The same total would then point towards a different next question.

Arun’s family later prepares a check containing ten factual items and ten explanation items. He scores eight and four respectively, producing twelve out of twenty, or 60%. A quick reading says that he has fallen from 75% to 60%. Yet half the later check now concerns the area that previously represented only one quarter. The changed composition prevents a clean comparison of the totals.

We should not reverse the mistake and announce that everything improved. Four out of ten explanation points is a higher percentage than one out of five, but the tasks differ and the sample is small. Inspecting the answers could reveal better reasoning, easier questions or both. The appropriate response is a more careful comparison, not a more optimistic headline.

The next decision is to preserve factual knowledge while examining a small set of explanations under consistent conditions. That choice respects what still appears useful. It avoids sacrificing an established strength merely because it does not solve the whole problem. The Defence guide develops that principle.

What transfers is the habit of opening up a total into relevant components. The fifteen-to-five split is only a feature of this fictional exercise. Real assessments have their own purposes, formats and weighting. For subject pathways, use the Science Tuition and Learning Hub; for reading an actual returned paper, use the examination-result repair guide.

Contents · Previous chapter · Next chapter

Fictional teaching case. Jun wants to recall and explain several geography ideas without looking at notes. He starts a routine of copying highlighted paragraphs into a second notebook. He plans four fifteen-minute sessions each week for five weeks: twenty planned sessions in total.

He completes eighteen sessions, or 90% of those planned. Eighteen sessions multiplied by fifteen minutes gives 270 minutes, equivalent to four and a half hours. The notebook becomes tidy and substantial. His family’s tracking sheet records every completed session, and the routine appears successful.

The intended capability, however, was recall and explanation without notes. Before the routine, Jun could answer two of eight short questions independently. After the five weeks, he answers two of eight different questions from the same topic area. These small checks are not perfectly matched, but they do not show the hoped-for improvement in independent performance.

One later answer repeats a phrase about a river carrying material without explaining where the material comes from or why its movement changes. During copying, Jun rarely had to produce that explanation. He could complete the activity by locating the next highlighted line and reproducing it accurately.

We cannot conclude that all copying is useless or that Jun learned nothing. He may have become more familiar with the wording, noticed spellings or organised information. Those possibilities would need their own evidence. We can conclude that the routine record measures completion, while the selected outcome check raises concern about the intended capability.

The unsuccessful result is part of the case. It should not be quietly removed to keep the story encouraging. Jun invested time, followed the plan and still lacks the evidence he hoped to obtain. The next decision is to reconsider the task inside the routine before demanding more sessions of it.

A useful conversation begins with the distinction between commitment and method: “You completed most of the planned sessions. Let us look at what those sessions asked you to do.” That statement acknowledges a real accomplishment without converting it into an unsupported learning claim. It also gives Jun a reason to keep participating in the review.

The next version might ask him to attempt a short explanation before consulting the notes, then inspect the gap. This is a proposal to test, not the successful ending of the case. No later result is provided. A responsible reader should leave the outcome open rather than supplying an imagined improvement.

The transfer question is whether your own routine requires the capability you want to develop. Jun’s session length and completion percentage are not prescriptions. Another learner may have an appropriate task but an impossible schedule; changing the task would miss their difficulty. Another may need initial teaching before any independent explanation is reasonable.

For maintaining a useful action across actual weeks, use Habits and Behaviour Change. For deciding how the wider learning arrangement should change, use Rebuild. Jun’s case makes the handoff clear: first establish that the action serves the goal, then work on making it repeatable.

Contents · Previous chapter · Next chapter

Elena’s case feels hopeful because a later independent score rises. Safiya’s feels uncomfortable because adult help has been mistaken for independent competence. Arun’s contains both an encouraging total and an unresolved weakness. Jun’s ends without the desired result. Their emotional tones differ, but tone is a poor basis for deciding which lesson applies.

Compare the structure of the evidence instead. Each case contains a gap between what was recorded and what someone wanted to infer. Once that gap becomes visible, a more useful next decision becomes possible.

Suppose your child’s homework is consistently correct but classwork is not. Elena and Safiya may both be relevant, although the subject could be different. The shared question concerns what the home setting supplies: a method cue, a model, extra time or someone else’s wording. The surface label “mathematics” or “English” is less useful than the condition that changes.

Suppose instead that the learner works independently in both places but struggles only on unfamiliar question types. Elena’s selection problem is a closer comparison. If the work is unfamiliar because the vocabulary is inaccessible, even that comparison may be premature. A case is a candidate explanation to examine, not a label to attach permanently to a learner.

When comparing two cases, write one similarity that could matter and one difference that could break the comparison. For Elena and Safiya, both later checks remove assistance. But selecting a mathematical relationship and composing a message require different knowledge. The independence question can travel; the instructional content cannot simply be copied.

This method of comparison also prevents unfair ranking. Jun is not the least capable learner because his case lacks a successful ending. The cases contain different tasks, timescales and evidence. They were written to expose different reasoning problems, not to compare the value or potential of their fictional learners.

If several difficulties genuinely interact, move to Academic Recovery, which deals with coordinating priorities and available support. A case comparison can help identify the questions that plan needs to answer.

Contents · Previous chapter · Next chapter

An event can happen after a new practice without being caused entirely by that practice. In everyday learning, several things often change together. A learner receives new teaching, grows more familiar with a topic, changes their available study time and encounters a different assessment. A simple before-and-after story cannot separate all those influences.

This does not mean we must suspend every ordinary decision until a research study is available. We still need to choose tomorrow’s practice. It means the strength of our language should match the strength of our evidence. “The later answer was more complete” is an observation. “This activity caused the improvement” is a stronger explanation requiring more support.

The What Works Clearinghouse publishes formal standards for reviewing educational research and characterising findings. Informal learning records should not be presented as equivalent evidence. WWC Handbooks and Reviewer Resources.

For a practical case review, consider three alternative explanations before settling on a preferred one. The later task may have been easier. The learner may have received unrecorded help. Another teaching experience may have contributed. These are questions to investigate, not accusations that invalidate every improvement.

In Elena’s case, a later mixed set contains different questions. Inspecting those questions helps us decide whether the observed rise is worth following up. In Safiya’s case, the recorded adult contribution directly explains why the jointly revised paragraph cannot stand as evidence of independent authorship. The uncertainty is different in each case; we should not paste the same disclaimer onto both.

Also distinguish uncertainty about the result from uncertainty about its cause. We can know that a learner produced a correct answer under specified conditions while remaining unsure which teaching activity helped most. We can preserve that achievement and continue investigating the explanation. Careful reasoning need not erase an observable success.

For a low-cost, reversible classroom choice, promising evidence may justify another carefully observed attempt. For a large commitment of time or money, we would want more than one striking story. The practical standard rises with the consequence of the decision. A fictional teaching case cannot supply the outcome evidence needed to choose a paid programme.

Keep the next check close to the disputed claim. If the uncertainty concerns independence, record assistance. If it concerns persistence, look after a delay. If it concerns question selection, change the mix. Collecting more of an irrelevant measure will not settle the question, however impressive the resulting spreadsheet looks.

Contents · Previous chapter · Next chapter

Transfer begins with a relationship, not a resemblance. Two learners can share an age, subject and recent score while facing different difficulties. Two learners in different subjects can face the same dependence on an adult’s decision. A case becomes useful when we identify the relevant relationship and then check whether it exists in the new situation.

The National Research Council distinguishes remembering taught material from extending learning to new contexts. Success on a familiar task does not settle the transfer question. How People Learn: Learning and Transfer.

Use three sentences when borrowing a lesson. First: “The condition I think is shared is…”. Second: “A difference that could make the lesson fail is…”. Third: “A small check that would help decide is…”. Those sentences force the comparison to become specific enough to challenge.

Take Safiya’s case into a Science explanation. The shared condition might be that an adult supplies the missing connection while the learner copies it. A relevant difference is that the learner may lack the scientific concept, rather than the process for checking a writing task. A small check is to ask the learner to explain the relationship orally in a fresh example before deciding whether sentence support is the main need.

Now take Jun’s case into vocabulary learning. A learner repeatedly reads a word list but wants to use the words accurately in sentences. The shared concern is a mismatch between the activity and intended performance. The difference is that recognising meanings may be a legitimate intermediate goal. We should inspect which goal the current session is serving before replacing the activity.

A useful lesson should survive an honest counterexample. “More independent work is always better” fails when a learner has not been taught the necessary idea. “Higher scores mean recovery” fails when the assessment becomes easier or samples a different component. “A completed routine proves discipline will solve the problem” fails when the repeated task does not address the difficulty.

Rewrite those broad statements conditionally. Independent checks help reveal what the learner can currently do without assistance. A higher score is more informative when task demands and support are comparable. A routine is worth maintaining when its activity serves the intended capability and its cost is manageable.

Conditional lessons are less memorable than slogans, but they travel more responsibly. They include the circumstances under which another reader should try them and the circumstances under which they should stop. That is how a case becomes a useful question in a new setting instead of an instruction copied out of context.

Contents · Previous chapter · Next chapter

A case should record more than the result we hoped to obtain. It should also tell us what the attempt required. Time, attention, access to help and interference with other responsibilities can change whether an apparently useful approach fits another learner’s life.

Jun’s eighteen sessions required four and a half hours. That calculation does not prove the time was wasted: the case does not measure every possible benefit. It does make the cost visible. Repeating the same five-week plan without examining its task would be a decision to commit more time despite unresolved evidence about the main goal.

Safiya’s polished paragraph also has a hidden cost if an adult must keep making the important decisions. The issue is not that adult assistance is bad. It is that a plan relying on that level of assistance may not be available during independent work. A case reader needs to know what the support accomplishes and what happens when it is absent.

Ask the learner for a prediction before a check: “Which part do you expect to manage, and where might you need help?” Then compare that prediction with the work. A learner who predicts no success but completes a relevant step has something concrete to reconsider. A learner who predicts complete independence but requests repeated prompts has a different discrepancy to examine.

Do not turn that comparison into a judgment about personality. “You overestimated this task” is a bounded observation; “you are overconfident” can become a sweeping label. Similarly, a cautious prediction after a setback does not tell us everything about a learner’s confidence elsewhere.

The dedicated guide to Rebuilding Confidence develops that task-specific comparison. Here it helps us read cases more fully: what the learner expected, what they actually did, what the attempt cost and what remains reasonable to try next. A story with an improved score but an impossible workload is not automatically a usable model for another family.

Contents · Previous chapter · Next chapter

The following records form an independent exercise. They are fictional, separate from the four earlier cases and deliberately incomplete. Your job is to make a defensible comparison, not to discover an unreported perfect explanation.

Record P: Leo’s explanation practice. Leo initially answers three of six short explanation questions correctly without help. After practice, he answers five of six different questions correctly, but an adult supplies the key connection in two of those five answers. The remaining three correct answers are completed without that assistance. No delayed check is available.

Record Q: Mina’s mixed review. Mina initially answers six of eight definition questions and two of four application questions correctly. On a later independent check with the same broad composition but different items, she answers seven of eight definitions and three of four applications correctly. Both checks allow twenty minutes. No claim is made that the items are precisely equal in difficulty.

Record R: Ravi’s new schedule. Ravi plans three ten-minute sessions per week for four weeks and completes all twelve. Before the schedule, he answers four of ten questions correctly without notes. Afterward, he answers seven of ten correctly while consulting his summary sheet. The sheet is allowed in the later practice, but the eventual task he is preparing for does not permit it.

Answer these questions before moving to the explanation.

1. For each record, state one observation that the facts support without adding an explanation.
2. Which record offers the clearest indication of improvement across comparable independent conditions, and what uncertainty remains?
3. Can Leo’s five out of six be used as an independent score? Explain how the assistance changes the claim.
4. Calculate Mina’s totals and the change in her overall percentage. Then identify the component changes.
5. Calculate Ravi’s planned study time. Does completing it establish readiness for the eventual task?
6. Choose a useful next check for each learner. Specify the condition the check needs to reveal.
7. Write one lesson that could transfer between two records and one detail that should not be transferred.

Make your answer short enough that another person can inspect its reasoning. Avoid broad descriptions such as “Leo is weak” or “Mina’s method is best”. The records contain no basis for ranking their general abilities or deciding which teaching method caused the results.

If you are discussing the exercise with someone else, compare the evidence each person used before comparing recommendations. Two people may choose different sensible next checks because they prioritise different remaining uncertainties. That is productive disagreement when both can explain the connection between the record, the question and the proposed check.

Contents · Previous chapter · Next chapter

For Leo, a supported observation is that five of six later answers were correct with the recorded assistance. Only three of those correct answers were completed without the adult supplying the key connection. It would be misleading to report five independent successes. It would also be premature to treat the other answers as proof that he could never complete them independently under a fresh, properly observed check.

For Mina, the first total is six plus two, giving eight out of twelve. The later total is seven plus three, giving ten out of twelve. These are approximately 66.7% and 83.3%; the exact difference is two twelfths, or sixteen and two-thirds percentage points. Rounded to one decimal place, that change is 16.7 percentage points.

Both of Mina’s components rise by one correct answer. Definitions move from six out of eight to seven out of eight; applications move from two out of four to three out of four. This differs from Arun’s earlier fictional case, where the target explanation component stayed unchanged while the total rose.

Mina provides the clearest indication among these records because both checks are independent, share the broad question composition and use the same time allowance. The indication remains limited. Different items may vary in difficulty, the numbers are small, and no later retention check appears. “More encouraging evidence under comparable stated conditions” is justified; “the practice caused lasting recovery” is not.

Ravi’s time calculation is three sessions × ten minutes × four weeks = 120 minutes, or two hours. Twelve completed sessions show that he followed the planned schedule in this record. Seven correct answers with a summary sheet show performance with that resource. Neither observation establishes readiness for a task where the sheet is unavailable.

A useful next check for Leo would use fresh explanation questions and record exactly what assistance is given. The aim is to see which connections he can supply himself. For Mina, a later independent sample with similar demands would help investigate whether the observed capability persists. For Ravi, an initial attempt without the sheet would reveal the gap between supported and required performance; subsequent teaching can respond to the actual gap.

The lesson that travels between Leo and Ravi is to record what the environment supplies before making an independence claim. The form of support does not travel unchanged: a spoken connection and a written summary are different resources. Removing either may expose different knowledge needs.

You may have proposed a different next check. Test it against one question: would its result help resolve the uncertainty you identified? More supported practice could be appropriate teaching, but it would not by itself answer an independence question. Another completion tally could describe Ravi’s consistency, but his consistency was not the missing information.

The purpose of the answer is to make these distinctions visible. A careful reader leaves some conclusions open because the records leave them open. That is a correct response to incomplete evidence, not a failure to solve the exercise.

Contents · Previous chapter · Next chapter

You can create a private learning case in a few paragraphs. Begin with the question you were trying to answer at the time. Include the original uncertainty before describing the outcome. Otherwise, hindsight can make an uncertain decision look obvious and remove the part another reader most needs to understand.

Use an actual work sample where possible. A sentence, a calculation or a description of an observable action is more informative than “things were terrible”. Keep the conditions alongside it: whether the task was familiar, which materials were available and what help was given. Record the same kinds of information for the later work.

Then distinguish the intended change from the change that actually occurred. A plan might call for independent attempts followed by feedback, while the record shows that most sessions became guided demonstrations. That difference is not a reason to edit the account into compliance. It is information about what the plan demanded and what the setting could support.

Include an awkward detail if it affects interpretation. Perhaps the improvement disappeared on a different topic. Perhaps the learner completed the task accurately but needed far more time than expected. Perhaps the new routine displaced an activity that was already useful. These details can determine whether another person should borrow the approach.

End with the next decision and the reason for it. A case can finish with “we need a clearer check” or “we stopped this activity”. A useful account does not require a triumphant ending. It requires an honest connection between the available evidence and the choice that follows.

Keep private notes private unless there is a clear reason and appropriate permission to share them. If a public teaching example is invented or combines features for illustration, label that plainly. Do not attach precise improvement figures to a fictional learner in a way that suggests an actual programme outcome. The label should be visible where the reader encounters the case.

For a chosen strategy that is still too vague to observe, use How to Translate Strategy Into Behaviour. It helps specify the actions a future case would need to record. This casebook then supplies the separate discipline of comparing what happened with what the record can reasonably support.

Contents · Previous chapter · Next chapter

After reading a case, write one useful question for your own situation. Keep it smaller than “how do we fix everything?” A question such as “does the learner select the method without a topic heading?” points towards an observable check. “Why is this learner unsuccessful?” is so broad that almost any story could seem to answer it.

If you already have useful actions but cannot tell whether the learner can perform them independently, begin by improving the evidence record. State the help provided and inspect a fresh attempt. If the difficulty spans several subjects and available time is limited, the academic recovery guide can help coordinate the decisions rather than treating every subject as an isolated emergency.

Keep support decisions connected to the actual question. A person helping with learning should be able to explain what they are observing, what they are trying and how they will judge the next step. The tuition and support decision guide offers a route for considering suitable help. The case-reading tools here can be used independently, whether or not you seek additional support.

The strongest thing to take from a recovery case is often a better distinction: correct versus independent, completed versus learned, higher total versus relevant improvement, promising change versus established cause. Those distinctions prevent us from discarding real progress or promising more than the work shows.

Return to your chosen question after the next relevant piece of work. Preserve what the evidence supports, name what remains unresolved and choose a proportionate next action. The case becomes useful at the point where it improves that decision.

Contents · Previous chapter · Continue through the learning library