Formative Assessment¶

Principle · All principles
Evidence · 35 claims (20 for, 15 mixed) · 45 studies (12 quant-synthesis, 11 causal, 11 review, 4 theoretical, 3 associational, 2 qualitative, 2 design), q1–q4 · 15 of 45 report an effect size · 23 claims rest on one study

Description¶

A conditional model relating learner state, activity and subsequent observation; the linked pattern specifies a reusable design policy.

Conditional relationship¶

When an aligned response supplies interpretable evidence and is used to select a relevant next action, assessment can support a learning transition; the response does not directly reveal the underlying state, and an unused checkpoint does not instantiate this mechanism. This is a conditional explanatory model, not a universal individual forecast. The formative assessment pattern supplies a reusable observation and design policy. Specific local examples below illustrate it rather than define its scope.

Learner state, objective and valued goal¶

This is a reusable design model. A study result supplies a bounded observation; it does not reveal the state or future outcome of a new learner.

Before selecting a configuration, record the target task and representation, task-specific prior experience, access and language constraints, time available, observed responses and the exact help provided. Preserve an unknown rather than replacing it with a generic label such as novice. Ask the learner what they want to accomplish and why it matters. Record that valued goal separately from the designer's learning objective; agreement is an observation to elicit, not an assumption. If they diverge, negotiate the task or purpose and retain the unresolved difference.

Define the intended capability, a representative assessment task and scoring instrument, the criterion chosen for this design, and the assessment horizon. Agree what response would warrant changing support and how to check later independent performance. Criteria are local decisions, not universal mastery thresholds. If the brief only asks for better learning, elicit these fields before selecting an exact dose or forecasting a gain.

Keep the record in this form: response under stated conditions → uncertain state hypotheses → discriminating observation → next activity → response at the stated horizon. Difficulty, accuracy, speed, confidence and engagement are different observations. None alone proves learning. Record a learner's changed purpose or constraints as well as a changed performance.

Observation and adaptation¶

This diagnostic policy is a design proposal; synthesis-level effects do not validate its specific branches. Elicit an unassisted explanation or application aligned with the intended capability, plus the learner's reasoning and confidence. Record the prompt, response mode, language demand, stakes and assistance. Interpret the response using explicit rival hypotheses before selecting feedback; a checkpoint without a usable next action is not this pattern.

Observation	Rival interpretations and discriminating observation	Proposed next action
An explanation omits an important relation	A conceptual misconception and an expression/language barrier can produce this. Compare two cases varying one relevant quantity, ask for a prediction and justification, and allow a diagram or accessible oral response.	If the relation still fails across response modes, give a targeted contrast and invite revision. If reasoning is coherent in another mode, address expression/access rather than reteach the concept wholesale.
A revised answer becomes correct	Answer copying and revised understanding both fit. Ask for a new unassisted case and justification, then repeat at the intended delay.	If only the feedback example succeeds, retain targeted support and another independent attempt. If the new case succeeds, reduce that support provisionally and check later.
A learner stays silent or guesses	Uncertainty about the concept and evaluative threat/disengagement are different possibilities. Ask privately about purpose and constraints and compare a low-stakes accessible attempt with the original response.	Adjust the conditions when they obstruct expression; negotiate purpose when the task lacks learner significance. Do not interpret participation alone as competence.

For repeated force-only explanations, ask how acceleration changes when mass varies at the same force, and ask why. Connect the task to a learner-valued aim such as explaining bicycle braking while separating the teacher's conceptual target. Agree a reasoning rubric for the relation, an unassisted new braking example after feedback today and another next week. Those are local design choices; the source does not establish their optimality. If the target, purpose or starting explanation is missing, elicit it before declaring a misconception or choosing feedback.

Evidence-bounded expectation¶

Kingston and Nash's corrected synthesis reports a science mean of 0.19 with 95% CI 0.06–0.31 across 17 effects; the erratum supersedes the original 0.09 science estimate: assessment-for-learning observation and comparison frame [+M]. It also removes an inappropriate comparison between two formative approaches. The synthesis supports a bounded program-level expectation and a need for better research; it cannot forecast this learner's response, distinguish these diagnostic branches or establish next-week transfer.

Before comparing another synthesis or applying an effect-size label, record assignment and baseline comparability, learner/context characteristics, exact treatment and control, outcome alignment and independence, score-range/standardization constraints, horizon and uncertainty. Preserve unknown fields. A correction removes one known error; it does not establish that all included estimates are comparable. Report raw magnitude alongside a justified context-sensitive interpretation; do not rank interventions or infer individual probabilities from these pooled values. The formative loop's causal pathway remains a hypothesis to test against subsequent responses.

The evidence by question¶

Each question below links a general claim that gathers the studies behind it into one table, one row per distinct study, with each study's design and effect as recorded and no averaged verdict (eval/synthesis/formative-assessment.json, scripts/synthesize_claim.py). The count is of distinct studies across the claim and the narrower claims that bear on it.

Does using assessment evidence during instruction raise achievement compared with the usual teaching? Assessment for learning improves achievement [~S]: 10 distinct studies.
What should the feedback after a check say, and who or what should give it? Feedback improves learning by a medium amount on average, but its effect varies widely with the information it carries, and a substantial share of feedback effects are negative [~S]: 7 distinct studies.
Should learners assess their own work, and can their self-ratings be trusted? Self Assessment Improves Self Regulated Learning [~M]: 13 distinct studies.
Should learners assess each other's work, and does it need a rubric or script? Peer assessment improves academic performance by a small average amount relative to no assessment or to teacher assessment [~M]: 2 distinct studies.
Does adjusting instruction to each response help, and what do teachers do with the evidence? Contingent scaffolding improves learning more than fixed or absent support. [+S]: 20 distinct studies.
What stops formative assessment working in practice: stakes, grading, time and class size? Effective implementation of formative assessment is hindered by its complexity, summative-assessment pressure, and teachers' negative early experiences [~M]: 7 distinct studies.

Fitting the design to a situation¶

Ask for three facts when the brief omits them: what decision each check is meant to inform, how often the check-and-adjust cycle can run (within a lesson, within or between units), and who will read the evidence: the teacher, the learner, peers or software.

If the brief says…	Then change…	Basis
School classes (`pre-k-12`), with achievement on unit tests as the goal	Run the elicit–interpret–adjust cycle within or between units, not only at the end of a course, and plan for a small gain	Assessment for learning improves achievement [+M]
The check will produce only right/wrong scores or a grade	Add information about the task and how to improve it to what goes back to the learner; a verdict or praise alone carries little	Feedback improves learning, varying with its information [+M]; Feedback works best at task and process levels [~M]
The responses that count will be learners' ratings of their own understanding, and the learners are new to the domain	Do not treat the self-rating as the evidence; pair it with a short aligned task, and show a worked model of self-assessment before relying on it	Self-assessment accuracy is low without training [~M]; Prior knowledge needed for accurate self-assessment [~M]
Learners should take part in judging their own work	Share the criteria or rubric before the attempt and have learners self-assess a draft they can still revise; expect a small effect on strategy use, and watch for added stress	Self-assessment improves self-regulated learning [+M]; Rubric-referenced self-assessment improved work [~M]
A large class with one teacher, and written work	Add peer review so that each learner gets more feedback than the teacher alone can give; a fixed script is optional	Peer assessment improves performance [+M]; Structured criteria are not shown to improve peer assessment [~M]
One-to-one tutoring or small groups	Adjust the help after each response: less after success, more after failure, rather than a fixed level of support	Contingent scaffolding improves learning [+M]
The checks will carry marks, or grading rules reward only final scores	Keep the elicitation ungraded and separate from the grade, so that partial understanding can show	Formative assessment implementation barriers [~W]; OpenSciEd grading expectations constrain formative practice [~W]
Software or AI will score drafts (for example, essays)	Use the scoring as a diagnosis that the teacher reviews with the learner before revision, and keep feedback on reasoning and synthesis with a person	Diagnostic review gave larger writing gains [+W]; High-level feedback should remain human [~W]
The goal is retention over weeks	Make the check a recall attempt and follow it with the correct answer; recheck after a delay	Retrieval practice improves retention [+M]
`online-self-paced`, with no teacher reading responses	Build the checks into the materials with automatic feedback, and give learners a next action for each result	Formative-assessment-based learning analytics show positive effects [+W]
Learners stay silent or guess on checks	Offer a private, low-stakes response channel and compare its answers with the public ones before deciding what the silence means	untested proposal

Further evidence, not yet read against this model¶

Claims this page cited before it was rewritten as a conditional model. They are evidence about the relationship, but none has yet been re-read against the model above; each says how far its own sources have been checked.

Contingent scaffolding improves learning more than fixed or absent support. [+M] — not settled: the abstract available could not confirm the entries
Process goals lead to better skill acquisition for novices than outcome goals. [+M] — not settled: the abstract available could not confirm the entries
High-confidence errors are more likely than low-confidence errors to be corrected on a later test after feedback. [+S] — not settled: the abstract available could not confirm the entries
Llmkt Outperforms Existing Kt Methods On Tutoring Dialogues [~W]
Audio Discussions Increase Perceived Pronunciation Ability [+M]
Decreased Perceived Pronunciation Difficulty [+M]
Three-year formative assessment training improved teachers' eliciting and interpreting skills, leading to better student learning [+W] — attached 2026-10-18 from H. P. A. Boshuizen (2016), which proposed "Use flexible, adaptive formative assessment during learning to provide information on results for regulation".
Deeper processing of target language improves lexical retention and long-term learning [+W]
Desirable difficulty: harder-to-learn conditions yield enhanced long-term retention [+W]
Accumulated deliberate practice is associated with performance, explaining a domain-dependent share of its variance: about a quarter in games, 4% in education and under 1% in professions [+W]
Quality out-of-class language learning combines focus on meaning and form and correlates with higher grades [+W]
Structured repetition gains in fluency depend on repeating the same task [+W]
Does second-language practice carry over to skills and tasks other than the one practised? Practice of grammar rules improves only the practiced skill, showing skill specificity of L2 automatization [+M]
Learning analytics research most often supports the 'where the learner is now' stage of formative assessment (51% of studies) and rarely the 'how to get there' stage (2%) [~W] — attached 2026-10-20 from Seyyed Kazem Banihashem et al. (2025), which proposed "Use learning analytics to deliver scaled, real-time, personalized feedback that relieves teacher workload constraints on formative assessment".

Source verification and open tests¶

Kingston & Nash (2011), 2015 publisher erratum. Corrected content-area table checked; full primary-study assignment and measurement details remain unverified in this run.

The examples are local design instances, not additional observations from those studies. Test the proposed interpretation and activity branches with new learner responses; compare competing designs under matched conditions before claiming a causal or predictive advantage. Reassess the model when the observations disagree with it.

Claims¶

Key Sources¶

Black, P., & Wiliam, D. (1998). Assessment and classroom learning. Assessment in Education: Principles, Policy & Practice, 5(1), 7–74. doi:10.1080/0969595980050102
Kingston, N., & Nash, B. (2011). Formative assessment: A meta-analysis and a call for research. Educational Measurement: Issues and Practice, 30(4), 28–37. doi:10.1111/j.1745-3992.2011.00220.x
Lee, H., Chung, H. Q., Zhang, Y., Abedi, J., & Warschauer, M. (2020). The effectiveness and features of formative assessment in US K-12 education: A systematic review. Applied Measurement in Education, 33(2), 124–140. doi:10.1080/08957347.2020.1732383
Pratt, M. W., & Savoy-Levine, K. M. (1998). Contingent tutoring of long-division skills in fourth and fifth graders: Experimental tests of some hypotheses about scaffolding. Journal of Applied Developmental Psychology, 19(2), 287–304. doi:10.1016/s0193-3973(99)80041-0
van de Pol, J., Volman, M., & Beishuizen, J. (2010). Scaffolding in teacher–student interaction: A decade of research. Educational Psychology Review, 22(3), 271–296. doi:10.1007/s10648-010-9127-6
Zimmerman, B. J., & Kitsantas, A. (1999). Acquiring writing revision skill: Shifting from process to outcome self-regulatory goals. Journal of Educational Psychology, 91(2), 241–252. https://doi.org/10.1037/0022-0663.91.2.241
Metcalfe, J. (2017). Learning from errors. Annual Review of Psychology, 68, 465–489. https://doi.org/10.1146/annurev-psych-010416-044022
Fazio, L. K., & Marsh, E. J. (2009). Surprising feedback improves later memory. Psychonomic Bulletin & Review, 16(1), 88–92. https://doi.org/10.3758/PBR.16.1.88

Page metadata

Field	Value
`type`	`principle`
`id`	`formative-assessment`
`aliases`	`analyze-student-dialogue-turns-as-formative-assessment`, `formative-evaluation-conferences-fricke`, `pair-audio-discussions-with-feedback-and-self-assessment`
`title`	`Formative Assessment`
`description`	`A conditional model relating learner state, activity and subsequent observation; the linked pattern specifies a reusable design policy.`
`status`	`review`
`moved_from`	`pattern/three-phase-dp-pp-practice-protocol`
`canonical`	`True`
`generated`	by `codex/unspecified` on `2026-04-07`
`trust`	unverified — no one has confirmed this page's content
`sources`	7 — `clark-2012`, `hawe-2017`, `cross-2015`, `alexander-scarlatos-2024`, `fricke-1976`, `lepore-2014`, `joshua-wedlock-christopher-binnie-2025`

Frontmatter is the page's machine-readable record — see the Schema & Guide.