Vocabulary Assessment | How to Measure Word Knowledge

THE MASTERY CLUB · VOCABULARY ASSESSMENT · MEASURE THE RIGHT THING → FIND THE GAP → TEACH THE GAP → RETEST TRANSFER

Vocabulary assessment is the process of finding out what a learner really knows about words, how much vocabulary the learner can recognise, how much can be retrieved and used, how deep that word knowledge is, and whether the vocabulary survives real reading, listening, speaking and writing. A good vocabulary test therefore does more than ask for definitions. It can measure vocabulary size, vocabulary breadth, vocabulary depth, receptive vocabulary, productive vocabulary, lexical access, word meanings, collocations, morphology, pronunciation, spelling, context, transfer and the ability to choose the right word under realistic conditions. The central assessment question is not “Can the learner pass this quiz?” but “What claim about vocabulary can this evidence actually support?”

For teachers, parents and learners searching for how to assess vocabulary, vocabulary level test, vocabulary size test, receptive and productive vocabulary assessment, vocabulary depth test, word knowledge assessment, vocabulary quizzes or vocabulary diagnostics, the most important principle is alignment. The assessment must match the learning problem. A multiple-choice definition test can be efficient for broad recognition, but it cannot by itself prove that a learner can retrieve the word in writing. A writing sample can reveal productive vocabulary, but it may miss words the learner knows receptively. A single total score can be convenient, but it can hide the exact component that needs repair.

This article is the canonical Vocabulary Assessment owner inside The Mastery Club Vocabulary apex. It is additive and leaves existing eduKateSG articles unchanged. For quantity, use Vocabulary | Vocabulary Size. For breadth versus depth, use What Is Vocabulary | Vocabulary Breadth and Vocabulary Depth. For detailed dimensions of one word, use Vocabulary | Word Knowledge. For retrieval, use Vocabulary | Lexical Access. This page owns the measurement problem: how to gather evidence that is valid enough to decide what vocabulary work should happen next.

The 60-Second Answer

A vocabulary assessment is useful when it measures the vocabulary construct you actually care about. If the question is how many words a learner recognises, use a breadth or size measure. If the question is how well important words are known, test depth. If the question is whether the learner can use vocabulary independently, use productive tasks. If the question is whether vocabulary survives reading, listening, speaking or writing, assess transfer in those modes.

Every vocabulary score is a sample, not a complete inventory of the learner’s mental lexicon. The score depends on which words were sampled, how “knowing” was defined, what response format was used, whether guessing was possible, whether the words were presented in context, and what population the test was designed for. Good assessment makes those assumptions visible.

1. Start With the Claim, Not the Quiz

The strongest vocabulary assessment begins by stating the claim you want to make. “This learner knows 8,000 words” is one kind of claim. “This learner can use high-utility academic vocabulary accurately in writing” is another. “This learner recognises spoken subject vocabulary quickly enough for classroom listening” is another. These claims require different evidence.

Starting with a familiar test format is tempting because worksheets and quizzes are easy to administer. The danger is measurement by convenience. A convenient test can produce a neat score while answering the wrong question.

Write the claim first, then ask what performance would count as evidence. If the claim concerns independent production, the target word cannot always be supplied in the question. If the claim concerns listening, written recognition is insufficient. If the claim concerns depth, one form–meaning match is too narrow.

Assessment becomes more useful when every item has a reason for being there.

2. Vocabulary Is Multidimensional

Knowing a word includes multiple components: spoken form, written form, meaning, senses, morphology, grammar, collocation, register, associations and the ability to retrieve and use the item. No short test measures all of these equally well.

This multidimensionality explains why learners can look strong on one vocabulary measure and weak on another. A learner may recognise many words but have limited productive access. Another may know a smaller vocabulary deeply and use it precisely. A third may have strong subject vocabulary but weaker general academic language.

Assessment should therefore avoid turning one narrow measure into a total description of the learner. The test result belongs to the construct it sampled.

The practical question is always: which dimension matters for the next educational decision?

3. Breadth and Vocabulary Size

Vocabulary breadth or size estimates how many lexical items a learner knows under a stated counting system. Size measures are useful for broad profiling, placement, reading research and identifying where knowledge becomes weaker across frequency bands.

These tests usually sample words rather than inspect the learner’s entire vocabulary. The sample is then used to estimate performance across a larger lexical population. The quality of the estimate depends on the sampling framework and scoring method.

Breadth tests are powerful because they are efficient. They are limited because they usually reduce “knowing” to a form–meaning relationship. A learner may pass the item without knowing collocation, register or productive use.

Use size scores for size decisions, not for claims the test was never designed to support.

4. Vocabulary Depth

Depth asks how well a learner knows a word. This can include several senses, word families, semantic relations, collocations, grammatical behaviour, register, pronunciation, spelling and contextual flexibility.

Depth is difficult to compress into one score because its components are heterogeneous. One assessment may focus on associations, another on morphology, another on collocations or definitions.

For educational diagnosis, depth is often best sampled selectively on high-value words rather than measured exhaustively across thousands of items.

The learner who knows many words shallowly may need different instruction from the learner whose breadth is small but whose known words are rich and stable.

5. Receptive Vocabulary Assessment

Receptive tests present a word or phrase and ask the learner to recognise or select its meaning. They can be efficient and allow large samples.

The supplied form reduces retrieval demand, which is appropriate when the construct is recognition. It also means the result should not be treated as evidence of independent production.

Reading-based receptive tests measure written access. Listening-based receptive tests are needed when spoken vocabulary matters because sound recognition creates different demands.

A complete receptive profile can therefore distinguish written from aural vocabulary when that distinction is educationally important.

6. Productive Vocabulary Assessment

Productive assessment requires the learner to generate language rather than select from visible options. Tasks can range from supplying a missing word to producing vocabulary in sentences, speaking or extended writing.

Productive tasks are closer to the demands of real communication, but scoring can become more complex. Several answers may be semantically acceptable, and grammar or spelling errors may complicate interpretation.

The assessment must decide whether it cares about exact target retrieval or adequate communicative alternatives. These are different claims.

Productive vocabulary is usually smaller than receptive vocabulary, so the two should not be treated as interchangeable measures.

7. Recognition vs Recall

Recognition gives the target as a cue. Recall removes it. This simple distinction changes the difficulty and meaning of the task.

Multiple-choice and matching items mainly test recognition. Fill-in-the-blank, translation into the target language, naming and open production place stronger demands on recall.

A learner who succeeds at recognition but fails recall may not need basic reteaching. The more precise diagnosis is weak retrieval.

Use both formats when you need to distinguish stored knowledge from accessible knowledge.

8. Form–Meaning Knowledge

The core of many vocabulary tests is the ability to connect a lexical form with a meaning. This is foundational because without the link, the learner cannot use the item in comprehension.

However, form–meaning knowledge can be partial. A learner may know an approximate translation but not the exact semantic boundary. Another may know one dominant sense but not a less common academic sense.

Assessment can therefore vary in strictness. A broad size test may accept a rough meaning, while a precision task may require a more exact distinction.

Scoring rules should reflect the intended construct rather than being strict or generous by habit.

9. Written Form Assessment

Spelling is one part of lexical knowledge. Dictation, word completion and free writing can reveal whether orthographic forms are accessible.

A spelling error does not always mean the word is semantically unknown. The learner may understand and pronounce it correctly while the written representation remains weak.

This distinction matters for intervention. Meaning drills will not repair an orthographic problem efficiently.

Assess form separately when spelling accuracy matters to the learner’s task.

10. Spoken Form Assessment

Spoken form includes pronunciation, stress and the ability to recognise the word in connected speech. A written vocabulary test cannot reveal whether the learner can hear the word accurately.

Listening items can present the target in phrases or sentences and ask for meaning, identification or transcription. Speaking tasks can test production.

Pronunciation scoring should focus on intelligibility and relevant contrasts rather than accent conformity unless a specialised task requires otherwise.

Spoken vocabulary becomes assessable when the mode matches the real listening or speaking demand.

11. Meaning Precision

Definition matching may show that a learner has the general idea. Precision assessment asks whether the learner can distinguish near-neighbours.

Pairs such as economic/economical, assume/infer or describe/explain reveal whether semantic boundaries are stable.

Use contrastive scenarios where one word fits and the other does not. Ask the learner to justify the choice.

This produces diagnostic evidence that a generic synonym question cannot provide.

12. Multiple Senses

A learner can know one sense of a word and fail another. Assessment should sample relevant senses when polysemy is important to the curriculum.

Contextual items are useful because they require sense selection rather than isolated definition recall.

Do not assume that one successful headword item proves full knowledge of the lexical entry.

Sense-specific assessment is especially important for common words used technically in school subjects.

13. Morphology

Morphological assessment can test recognition of roots and affixes, interpretation of derivatives, or productive selection of the correct family member.

A learner who knows analyse may still fail analysis or analytical in writing. Family structure is therefore a separate dimension from base-word recognition.

Sentence transformations are useful because they test morphology inside grammar.

The companion Word Forms, Lemmas and Word Families owner provides the structural framework behind these assessments.

14. Collocation

Collocation assessment asks whether the learner knows natural word partnerships. This is important because definitions alone do not guarantee natural use.

Matching can test recognition efficiently, while sentence completion or writing can test production.

Scoring should distinguish a truly unnatural combination from a less frequent but acceptable one. Corpus-informed examples can help.

Collocation assessment is especially valuable for advanced writing and speaking.

15. Grammar and Word Behaviour

Words carry grammatical requirements. Verbs take particular complements, adjectives prefer particular prepositions, and nouns participate in count patterns.

A vocabulary assessment that ignores grammar can overestimate productive control.

Use sentence frames that force the learner to select the right lexical item and construction.

This reveals whether the learner knows how the word behaves, not merely what it roughly means.

16. Register

Register assessment checks whether vocabulary fits the situation, audience and level of formality.

A semantically correct word can still be pragmatically wrong in an academic essay, casual conversation or professional email.

Give the learner several contexts and ask which expression belongs in each.

Register is part of usable vocabulary because communication always occurs somewhere.

17. Connotation and Tone

Connotation adds evaluative and emotional loading. Two words can refer to similar events while creating different attitudes.

Assessment should use contextual choices and ask what changes when one candidate replaces another.

This is especially important in persuasive writing, literature and media analysis.

A learner who knows only denotation may still misread tone or produce unintended effects.

18. Lexical Access

Assessment should sometimes ask not only whether retrieval succeeds, but how readily it succeeds. A word that arrives after a long search is known differently from one that is immediately available.

Classroom assessment need not measure milliseconds. Qualitative categories such as immediate, delayed, cued and failed can be informative.

Speed should be evaluated only after accuracy is secure.

The Lexical Access owner explains this dimension in depth.

19. Automaticity

Automaticity concerns efficient access with limited conscious effort. It matters for fluent reading, listening, speaking and writing.

A learner can score well on untimed vocabulary tasks while still processing language too slowly for real communication.

Timed tasks can add useful information when designed carefully, but time pressure should not distort the construct by creating anxiety or careless errors.

Use speed as a secondary dimension of established knowledge, not as the first gate.

20. Transfer

Transfer asks whether vocabulary learned in one setting appears in another. This is one of the strongest tests of usable knowledge.

A learner may pass a worksheet and still fail to use the word in a fresh paragraph, unfamiliar reading passage or spontaneous discussion.

Transfer tasks should change topic, format or modality while preserving the underlying lexical demand.

Assessment is incomplete when success depends on reproducing the original learning environment.

21. Vocabulary in Reading

Reading assessment can reveal how vocabulary functions inside comprehension. Unknown-word density, vocabulary-in-context questions and interpretation of academic terms all provide evidence.

The task should distinguish lexical difficulty from syntax, background knowledge and inferencing where possible.

A reader who knows most words but still struggles may need a different intervention.

Vocabulary assessment becomes most educationally useful when it helps separate these possibilities.

22. Vocabulary in Listening

Listening assessment tests whether spoken words are recognised quickly enough in continuous speech.

Written follow-up questions can hide a listening vocabulary problem if the key term is displayed after the audio.

Use audio-only recognition, short transcription or meaning tasks when spoken access is the construct.

Compare with print performance to determine whether the weakness is lexical meaning or auditory recognition.

23. Vocabulary in Speaking

Speaking samples productive vocabulary under time pressure. It can reveal retrieval, collocation, register and flexibility.

Open speaking tasks also introduce many other variables such as confidence, pronunciation and discourse skill, so vocabulary scores should be interpreted cautiously.

Focused prompts can make lexical evidence clearer by requiring description, comparison or explanation around a known topic.

Rubrics should separate vocabulary from unrelated speaking dimensions when the goal is diagnosis.

24. Vocabulary in Writing

Writing offers rich evidence of productive vocabulary: diversity, precision, collocation, morphology, register and contextual fit.

However, one writing sample underrepresents the learner’s full vocabulary because topic and task restrict which words are likely to appear.

Multiple prompts across time provide a better profile than one essay.

Writing assessment should examine whether vocabulary serves the ideas rather than rewarding rarity for its own sake.

25. Vocabulary Size Tests

Vocabulary size tests estimate breadth by sampling lexical items across frequency levels or another ordered framework.

Their value lies in efficient estimation, not complete enumeration.

Interpretation depends on what counts as a word, what response counts as knowledge, and which lexical list underlies the test.

The Vocabulary Size owner explains these counting issues in detail.

26. Vocabulary Levels Tests

Vocabulary levels approaches sample knowledge at different frequency bands to identify where coverage begins to weaken.

This can support placement and planning because the learner may be secure in high-frequency vocabulary but weaker at later bands.

A levels result should guide word selection rather than become a label of intelligence or general ability.

Frequency profiles are most useful when connected to the learner’s real reading and learning goals.

27. Yes/No Vocabulary Tests

Yes/No formats ask learners to indicate whether they know presented words, sometimes including nonwords to estimate response bias.

They can sample many items quickly but rely partly on self-judgment and recognition.

Inflated claims of knowledge are possible when learners respond liberally, which is why control items and scoring corrections may be used.

Such tests are efficient screening tools, not complete profiles of depth or production.

28. Multiple-Choice Tests

Multiple-choice vocabulary items are easy to score and can sample broadly.

Their limitations include guessing, cueing and the possibility that distractor quality changes item difficulty.

Good distractors should be plausible without introducing irrelevant reading complexity.

Use multiple choice when recognition is the construct, not merely because automated scoring is convenient.

29. Matching Tests

Matching formats pair words with meanings or definitions and can test several items compactly.

They reduce writing demand and can be useful for broad receptive assessment.

Shared option sets create interactions among items because one answer can help eliminate another.

The design should account for that cueing effect when interpreting performance.

30. Fill-in-the-Blank Tests

Gap-fill tasks can test productive vocabulary when the target is not visible.

The context must constrain the answer enough that scoring is fair. Poorly written gaps can allow many equally reasonable words.

If one exact target is required, the sentence should make that lexical choice defensible.

Open gaps are powerful when they measure retrieval rather than the test writer’s preference.

31. Cloze Tests

Cloze tasks remove words from a passage and require learners to reconstruct them. Vocabulary, grammar, cohesion and broader comprehension can all contribute.

This makes cloze useful for integrated language performance but less pure as a vocabulary measure.

For diagnostic vocabulary cloze, choose deletions whose main difficulty is lexical and document the intended evidence.

Integrated tasks are valuable when the goal is integrated performance.

32. Translation Tests

Translation can assess form–meaning connections efficiently for bilingual learners.

Direction matters. Target language to first language is closer to receptive meaning access; first language to target language places stronger productive retrieval demands.

Translation equivalence is not always one-to-one, especially for polysemous or culture-specific vocabulary.

Scoring should allow semantically valid alternatives when the construct is meaning rather than memorised wording.

33. Definition Production

Asking learners to define a word can reveal depth, but definitions are cognitively demanding and rely on metalinguistic skill.

A learner may know a word well in use but struggle to produce a dictionary-style definition.

Accept explanations, examples and distinctions when they provide the evidence needed.

The task should not confuse vocabulary knowledge with expertise in writing formal definitions.

34. Example and Non-Example Tasks

Examples show whether a learner can connect a word to real cases. Non-examples show whether the learner understands boundaries.

This format is excellent for abstract academic vocabulary because it tests concepts, not memorised wording.

Ask the learner to explain why the non-example fails.

Boundary reasoning produces deeper diagnostic evidence than repetition.

35. Semantic Association Tests

Association tasks ask learners to connect a target with related meanings, categories or collocates.

They can reveal lexical organisation and some aspects of depth.

Scoring requires careful design because relationships can be diverse and culturally influenced.

Association evidence should complement, not replace, direct use.

36. Word-Family Tests

Word-family assessments can ask learners to identify bases, interpret affixes or produce the grammatical member required by a sentence.

These tasks are especially useful for academic language where derivational morphology is dense.

Do not assume that recognising the base proves knowledge of all relatives.

Assess the specific family members that matter.

37. Collocation Tests

Collocation can be assessed through matching, completion, acceptability judgment or production.

Recognition is easier than production, so task direction should match the claim.

Frequency and register should inform scoring because more than one combination may be grammatical.

The goal is natural phrase knowledge, not arbitrary conformity to one example sentence.

38. Context-Clue Assessment

Context-clue tasks test whether learners can infer a plausible meaning from textual evidence.

They measure more than stored vocabulary because inference, syntax and background knowledge contribute.

A learner can infer successfully without permanently learning the word, so context-clue success should not be treated as complete vocabulary acquisition.

Separate inference skill from later retention when both matter.

39. Dictionary Skills Assessment

Vocabulary competence includes the ability to resolve uncertainty using dictionaries or reliable reference tools.

Assessment can ask learners to choose the correct sense, interpret pronunciation or grammar labels, and apply an entry to a sentence.

This is tool-mediated vocabulary skill rather than unaided lexical knowledge.

Both are valuable, but the score should identify which capability was measured.

40. Vocabulary Self-Assessment

Learners can rate whether words are unknown, familiar, understood or usable. Such judgments are quick and can support metacognition.

Self-ratings are vulnerable to overconfidence, underconfidence and differing personal criteria.

Calibrate self-assessment by comparing ratings with actual retrieval and use.

The goal is not perfect introspection but increasingly accurate monitoring.

41. Diagnostic Vocabulary Assessment

Diagnostic assessment asks why vocabulary performance is weak, not merely how low the score is.

The same incorrect answer can arise from missing meaning, spelling confusion, weak morphology, wrong sense selection, collocation error or retrieval failure.

Follow-up probes should isolate the earliest broken link.

Diagnosis is valuable only when it changes the teaching decision.

42. Formative Vocabulary Assessment

Formative assessment gathers evidence while learning can still change. Short retrieval checks, exit tickets and sentence tasks can show which words need more work.

The purpose is adjustment, not ranking.

Feedback should identify the component to repair and create another retrieval opportunity.

Frequent low-stakes evidence makes vocabulary teaching responsive.

43. Summative Vocabulary Assessment

Summative vocabulary assessment reports achievement after a defined period or programme.

Because stakes are higher, alignment, reliability and fairness become especially important.

A summative test should represent the taught vocabulary construct rather than sample arbitrary difficult words.

High-stakes vocabulary scores should never be built from a narrow or unpredictable item set.

44. Pre-Assessment

Pre-assessment finds what learners already know before instruction begins.

It prevents wasting time reteaching secure vocabulary and reveals prerequisite gaps.

For a new subject unit, sample both key technical terms and general academic words needed to understand the lessons.

Pre-assessment should lead directly to grouping, selection or instructional adaptation.

45. Post-Assessment

Post-assessment checks what changed after instruction.

Using the identical items can inflate apparent learning through item familiarity, while completely different items may make comparison noisy.

Parallel forms or mixed familiar and transfer items can provide stronger evidence.

The most important question is whether gains survive beyond the exact training set.

46. Delayed Assessment

Immediate post-tests often capture temporary accessibility. Delayed tests show whether learning remains retrievable after forgetting has begun.

Vocabulary that matters for education must survive beyond the lesson.

Delay can range from days to weeks depending on the programme and purpose.

A smaller delayed gain can be more educationally meaningful than a perfect same-day score.

47. Transfer Assessment

Transfer items use new contexts, prompts or modalities.

They test whether knowledge generalises rather than remains bound to the practice format.

A learner who succeeds only on rehearsed sentences has learned something, but not yet the full transferable skill.

Transfer should be expected only after learners have had a fair opportunity to practise varied use.

48. Reliability

Reliability concerns the consistency of measurement. If tiny irrelevant changes produce wildly different results, the score is difficult to interpret.

Reliability is affected by test length, item quality, scoring consistency and learner conditions.

A reliable test can still measure the wrong thing, so reliability alone is not validity.

Both matter when decisions depend on the score.

49. Validity

Validity concerns whether the evidence supports the intended interpretation and use of the score.

A spelling-heavy test may be invalid for a claim about oral receptive vocabulary. A multiple-choice test may be weak evidence for productive independence.

Validity is not a permanent property printed on a test. It depends on the claim, population and use.

Ask what inference the score is being used to justify.

50. Construct Underrepresentation

Construct underrepresentation occurs when a test samples too little of the capability it claims to measure.

Calling one definition quiz “vocabulary mastery” underrepresents pronunciation, collocation, retrieval, grammar, senses and transfer.

Broaden evidence when the claim is broad.

Narrow tests are not bad; broad claims from narrow tests are the problem.

51. Construct-Irrelevant Variance

A vocabulary item can become difficult for reasons unrelated to vocabulary, such as complicated instructions, obscure cultural knowledge or excessive reading load.

These irrelevant barriers contaminate the score.

Simplify non-target demands unless they are part of the construct.

Fair assessment isolates the vocabulary capability as much as the purpose allows.

52. Guessing and Chance

Selected-response items permit some correct answers by chance.

The impact depends on the number and quality of options and the learner’s partial knowledge.

Do not interpret one correct multiple-choice answer as definitive mastery.

Repeated evidence across formats provides stronger support.

53. Scoring Partial Knowledge

Vocabulary knowledge is graded, so all-or-nothing scoring can lose useful information.

A learner may know the core meaning but not the collocation, or know the spoken form but miss the spelling.

Analytic scoring can preserve these distinctions when diagnosis matters.

For large-scale screening, simpler scoring may be appropriate; the scoring model should fit the purpose.

54. Rubrics for Productive Vocabulary

Productive tasks need clear rubrics. Useful dimensions can include precision, appropriateness, range, collocation, grammatical fit and retrieval independence.

Avoid rewarding rare vocabulary automatically. A common precise word is better evidence than an obscure misused one.

Rubrics should be small enough for consistent scoring.

Every criterion should correspond to an instructionally meaningful aspect of vocabulary.

55. Vocabulary Assessment for Children

Children may have strong oral knowledge but limited ability to explain words formally or read complex test instructions.

Assessment should reduce unnecessary metalinguistic and literacy demands when those are not the target.

Use pictures, oral prompts, examples and age-appropriate production tasks.

Interpret results developmentally and avoid turning one score into a fixed label.

56. Vocabulary Assessment for Secondary Learners

Secondary learners require assessment of general vocabulary, academic language, subject terms and productive precision.

A balanced profile can combine a broad receptive sample with targeted depth and authentic writing or speaking evidence.

Command words and cross-curricular academic vocabulary deserve special attention because they affect multiple subjects.

Assessment should identify whether the learner needs more breadth, more depth or faster access.

57. Vocabulary Assessment for Advanced Learners

Advanced learners often need finer measures of collocation, register, lexical sophistication, multiple senses and phraseology.

Broad recognition tests can reach a ceiling and stop differentiating meaningful strengths and weaknesses.

Use corpus-informed tasks, writing samples and precise semantic contrasts.

Advanced assessment shifts from “Do you know the word?” toward “Can you use the right lexical option under realistic constraints?”

58. Vocabulary Assessment for Multilingual Learners

Multilingual learners may distribute vocabulary differently across languages and domains.

Translation can be a useful assessment route, but direct concept access should also be sampled when target-language independence matters.

Avoid treating first-language influence as a simple deficit; it can support learning and also create specific interference patterns.

Assessment should describe the language route being measured.

59. Building a Vocabulary Assessment Battery

A battery combines several small measures to cover complementary dimensions.

For example: a receptive size sample, a productive recall task, a depth probe on key words, an aural recognition sample and one authentic writing or speaking task.

The battery does not need to be long if each component has a clear job.

Triangulation produces a stronger profile than stretching one test format beyond its limits.

60. The Mastery Club Rule for Vocabulary Assessment

Assessment should make the next teaching decision better.

Measure the smallest construct that explains the visible problem, then broaden only when necessary.

Do not confuse recognition with retrieval, size with depth, a score with a diagnosis, or same-task success with transfer.

The final standard is independent language use: can the learner understand, retrieve and choose vocabulary accurately when the world stops looking like the worksheet?

20 Diagnostic Vocabulary Profiles

These profiles show how assessment changes when the visible symptom is traced to a more precise construct.

1. High recognition, weak production

Visible pattern: The learner performs strongly on multiple-choice vocabulary tests but uses a narrow range in writing.

Assessment move: Add meaning-to-word recall, sentence generation and delayed productive tasks. Do not spend most time reteaching definitions.