ICEQC-R-2013-09 — Learning Outcomes and Curriculum Coverage: Interpreting Evidence without Overstatement

The relevant evidence concerns the authorised aims, content, sequence and expected depth for a stated stage. The principal risk is that curriculum labels can conceal differences in scope, emphasis and time. Authorities should describe the operative curriculum and the level of analysis.[REF-05]

For intended curriculum, interpretation depends on the authorised aims, content, sequence and expected depth for a stated stage. The principal risk is that curriculum labels can conceal differences in scope, emphasis and time. Evidence producers should therefore describe the operative curriculum and the level of analysis.

The definition test asks which concept, unit and level of curriculum or outcome is being described. Analysts should state terms beside the result. The governing proposition is that familiar terminology does not ensure a common measure. Every claim should identify population, domain, period, source and material limitation.

Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured.

Evidence source is material because the authorised aims, content, sequence and expected depth for a stated stage cannot be inferred from a score, curriculum title or coverage percentage. In this domain, curriculum labels can conceal differences in scope, emphasis and time. The immediate safeguard is to describe the operative curriculum and the level of analysis.

A defensible account should establish which assessment, record, observation or report supports the claim and should identify provenance, date, coverage and collection method, recognising that different sources may complement one another without measuring the same object. Original categories and missing evidence should remain visible.

Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result.

The public-interest meaning of intended curriculum concerns the authorised aims, content, sequence and expected depth for a stated stage. If curriculum labels can conceal differences in scope, emphasis and time, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to describe the operative curriculum and the level of analysis.

Under coverage, review concerns which domains, learners, institutions and periods are represented. Statistical and education bodies should publish what remains outside the evidence, subject to the rule that unobserved content or learners should not be treated as average. Difference caused by method should be distinguished from plausible educational difference.

Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support.

Analysis of intended curriculum should begin with the authorised aims, content, sequence and expected depth for a stated stage. A foreseeable failure arises where curriculum labels can conceal differences in scope, emphasis and time. Authorities should describe the operative curriculum and the level of analysis.

The alignment standard requires consideration of how intended curriculum, teaching and assessment relate. Producers should map content and cognitive demand without assuming complete correspondence, because correlation of labels is not evidence of substantive alignment. Revisions and breaks in series should be adjacent to the affected comparison.

The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism.

Intended curriculum requires explicit metadata because it concerns the authorised aims, content, sequence and expected depth for a stated stage. The risk is that curriculum labels can conceal differences in scope, emphasis and time. A competent producer should describe the operative curriculum and the level of analysis.

Review of comparability should determine which populations, constructs and procedures are sufficiently common. It should compare metadata before values and acknowledge that a shared scale does not remove material contextual difference. Personal information should remain protected while material inequality remains visible.

The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty.

The public purpose of evidence on intended curriculum is a responsible account of the authorised aims, content, sequence and expected depth for a stated stage. Where curriculum labels can conceal differences in scope, emphasis and time, results can overstate both weakness and certainty. Analysts should describe the operative curriculum and the level of analysis.

For uncertainty, bodies should identify which sampling, measurement, linking and classification limits apply and should publish appropriate intervals and qualitative limitations. Any conclusion must respect that precision of display should not exceed evidence precision.

The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions.

Applied to intended curriculum, this requirement has a distinct interpretive consequence. For intended curriculum, interpretation depends on the authorised aims, content, sequence and expected depth for a stated stage. The principal risk is that curriculum labels can conceal differences in scope, emphasis and time. Evidence producers should therefore describe the operative curriculum and the level of analysis. The implication for comparison and policy use should be recorded.

The equity test asks which learners had different opportunity, access or assessment conditions. Analysts should analyse relevant distributions and missing populations. The governing proposition is that equal testing does not establish equal opportunity to learn. Every claim should identify population, domain, period, source and material limitation.

Applied to intended curriculum, this requirement has a distinct interpretive consequence. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The implication for comparison and policy use should be recorded.

Interpretive boundary is material because the authorised aims, content, sequence and expected depth for a stated stage cannot be inferred from a score, curriculum title or coverage percentage. In this domain, curriculum labels can conceal differences in scope, emphasis and time. The immediate safeguard is to describe the operative curriculum and the level of analysis.

A defensible account should establish which inference is supported and which remains unresolved and should separate observation, association and causal explanation, recognising that a result should not be extended from tested evidence to the whole education system. Original categories and missing evidence should remain visible.

Applied to intended curriculum, this requirement has a distinct interpretive consequence. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The implication for comparison and policy use should be recorded.

Applied to intended curriculum, this requirement has a distinct interpretive consequence. The public-interest meaning of intended curriculum concerns the authorised aims, content, sequence and expected depth for a stated stage. If curriculum labels can conceal differences in scope, emphasis and time, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to describe the operative curriculum and the level of analysis. The implication for comparison and policy use should be recorded.

Under decision use, review concerns which curriculum, teaching, support or accountability decision the evidence informs. Statistical and education bodies should state authority, consequence and additional evidence needed, subject to the rule that data should be proportionate to the stakes of the decision. Difference caused by method should be distinguished from plausible educational difference.

Applied to intended curriculum, this requirement has a distinct interpretive consequence. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The implication for comparison and policy use should be recorded.

Applied to intended curriculum, this requirement has a distinct interpretive consequence. Analysis of intended curriculum should begin with the authorised aims, content, sequence and expected depth for a stated stage. A foreseeable failure arises where curriculum labels can conceal differences in scope, emphasis and time. Authorities should describe the operative curriculum and the level of analysis. The implication for comparison and policy use should be recorded.

The review and correction standard requires consideration of how definitions, data and claims are challenged or revised. Producers should retain versions, reasons and limitations, because revisions should distinguish methodological change from educational change. Revisions and breaks in series should be adjacent to the affected comparison.

Applied to intended curriculum, this requirement has a distinct interpretive consequence. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The implication for comparison and policy use should be recorded.

The relevant evidence concerns the content and learning activity teachers and learners actually undertake. The principal risk is that plans and official syllabuses can be represented as delivered learning. Authorities should use classroom, teacher and learner evidence with stated limitations.[REF-10]

Implemented curriculum requires explicit metadata because it concerns the content and learning activity teachers and learners actually undertake. The risk is that plans and official syllabuses can be represented as delivered learning. A competent producer should use classroom, teacher and learner evidence with stated limitations.

Review of definition should determine which concept, unit and level of curriculum or outcome is being described. It should state terms beside the result and acknowledge that familiar terminology does not ensure a common measure. Personal information should remain protected while material inequality remains visible.

Applied to implemented curriculum, this requirement has a distinct interpretive consequence. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The implication for comparison and policy use should be recorded.

The public purpose of evidence on implemented curriculum is a responsible account of the content and learning activity teachers and learners actually undertake. Where plans and official syllabuses can be represented as delivered learning, results can overstate both weakness and certainty. Analysts should use classroom, teacher and learner evidence with stated limitations.

For evidence source, bodies should identify which assessment, record, observation or report supports the claim and should identify provenance, date, coverage and collection method. Any conclusion must respect that different sources may complement one another without measuring the same object.

Applied to implemented curriculum, this requirement has a distinct interpretive consequence. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The implication for comparison and policy use should be recorded.

For implemented curriculum, interpretation depends on the content and learning activity teachers and learners actually undertake. The principal risk is that plans and official syllabuses can be represented as delivered learning. Evidence producers should therefore use classroom, teacher and learner evidence with stated limitations.

The coverage test asks which domains, learners, institutions and periods are represented. Analysts should publish what remains outside the evidence. The governing proposition is that unobserved content or learners should not be treated as average. Every claim should identify population, domain, period, source and material limitation.

disclosed, likely, effect, interpretation require ### 15. Coverage to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. Any departure should be disclosed with its likely effect on interpretation.

Alignment is material because the content and learning activity teachers and learners actually undertake cannot be inferred from a score, curriculum title or coverage percentage. In this domain, plans and official syllabuses can be represented as delivered learning. The immediate safeguard is to use classroom, teacher and learner evidence with stated limitations.

A defensible account should establish how intended curriculum, teaching and assessment relate and should map content and cognitive demand without assuming complete correspondence, recognising that correlation of labels is not evidence of substantive alignment. Original categories and missing evidence should remain visible.

disclosed, likely, effect, interpretation require ### 16. Alignment to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. Any departure should be disclosed with its likely effect on interpretation.

The public-interest meaning of implemented curriculum concerns the content and learning activity teachers and learners actually undertake. If plans and official syllabuses can be represented as delivered learning, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to use classroom, teacher and learner evidence with stated limitations.

Under comparability, review concerns which populations, constructs and procedures are sufficiently common. Statistical and education bodies should compare metadata before values, subject to the rule that a shared scale does not remove material contextual difference. Difference caused by method should be distinguished from plausible educational difference.

disclosed, likely, effect, interpretation require ### 17. Comparability to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. Any departure should be disclosed with its likely effect on interpretation.

Analysis of implemented curriculum should begin with the content and learning activity teachers and learners actually undertake. A foreseeable failure arises where plans and official syllabuses can be represented as delivered learning. Authorities should use classroom, teacher and learner evidence with stated limitations.

The uncertainty standard requires consideration of which sampling, measurement, linking and classification limits apply. Producers should publish appropriate intervals and qualitative limitations, because precision of display should not exceed evidence precision. Revisions and breaks in series should be adjacent to the affected comparison.

disclosed, likely, effect, interpretation require ### 18. Uncertainty to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. Any departure should be disclosed with its likely effect on interpretation.

Applied to implemented curriculum, this requirement has a distinct interpretive consequence. Implemented curriculum requires explicit metadata because it concerns the content and learning activity teachers and learners actually undertake. The risk is that plans and official syllabuses can be represented as delivered learning. A competent producer should use classroom, teacher and learner evidence with stated limitations. The implication for comparison and policy use should be recorded.

Review of equity should determine which learners had different opportunity, access or assessment conditions. It should analyse relevant distributions and missing populations and acknowledge that equal testing does not establish equal opportunity to learn. Personal information should remain protected while material inequality remains visible.

disclosed, likely, effect, interpretation require ### 19. Equity to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. Any departure should be disclosed with its likely effect on interpretation.

Applied to implemented curriculum, this requirement has a distinct interpretive consequence. The public purpose of evidence on implemented curriculum is a responsible account of the content and learning activity teachers and learners actually undertake. Where plans and official syllabuses can be represented as delivered learning, results can overstate both weakness and certainty. Analysts should use classroom, teacher and learner evidence with stated limitations. The implication for comparison and policy use should be recorded.

For interpretive boundary, bodies should identify which inference is supported and which remains unresolved and should separate observation, association and causal explanation. Any conclusion must respect that a result should not be extended from tested evidence to the whole education system.

disclosed, likely, effect, interpretation require ### 20. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. Any departure should be disclosed with its likely effect on interpretation.

Applied to implemented curriculum, this requirement has a distinct interpretive consequence. For implemented curriculum, interpretation depends on the content and learning activity teachers and learners actually undertake. The principal risk is that plans and official syllabuses can be represented as delivered learning. Evidence producers should therefore use classroom, teacher and learner evidence with stated limitations. The implication for comparison and policy use should be recorded.

The decision use test asks which curriculum, teaching, support or accountability decision the evidence informs. Analysts should state authority, consequence and additional evidence needed. The governing proposition is that data should be proportionate to the stakes of the decision. Every claim should identify population, domain, period, source and material limitation.

For implemented curriculum, analysts should test this safeguard against the construct and population. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The conclusion should retain the material limitation and denominator.

Review and correction is material because the content and learning activity teachers and learners actually undertake cannot be inferred from a score, curriculum title or coverage percentage. In this domain, plans and official syllabuses can be represented as delivered learning. The immediate safeguard is to use classroom, teacher and learner evidence with stated limitations.

A defensible account should establish how definitions, data and claims are challenged or revised and should retain versions, reasons and limitations, recognising that revisions should distinguish methodological change from educational change. Original categories and missing evidence should remain visible.

For implemented curriculum, analysts should test this safeguard against the construct and population. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The conclusion should retain the material limitation and denominator.

The relevant evidence concerns the scheduled, delivered and engaged time available for learning. The principal risk is that calendar time can overstate exposure where interruption, absence or shortened sessions occur. Authorities should distinguish planned, delivered and usable time.[REF-12]

The public-interest meaning of instructional time concerns the scheduled, delivered and engaged time available for learning. If calendar time can overstate exposure where interruption, absence or shortened sessions occur, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to distinguish planned, delivered and usable time.

Under definition, review concerns which concept, unit and level of curriculum or outcome is being described. Statistical and education bodies should state terms beside the result, subject to the rule that familiar terminology does not ensure a common measure. Difference caused by method should be distinguished from plausible educational difference.

retain, material, limitation, denominator require ### 24. Definition to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The conclusion should retain the material limitation and denominator.

Analysis of instructional time should begin with the scheduled, delivered and engaged time available for learning. A foreseeable failure arises where calendar time can overstate exposure where interruption, absence or shortened sessions occur. Authorities should distinguish planned, delivered and usable time.

The evidence source standard requires consideration of which assessment, record, observation or report supports the claim. Producers should identify provenance, date, coverage and collection method, because different sources may complement one another without measuring the same object. Revisions and breaks in series should be adjacent to the affected comparison.

retain, material, limitation, denominator require ### 25. Evidence source to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The conclusion should retain the material limitation and denominator.

Instructional time requires explicit metadata because it concerns the scheduled, delivered and engaged time available for learning. The risk is that calendar time can overstate exposure where interruption, absence or shortened sessions occur. A competent producer should distinguish planned, delivered and usable time.

Review of coverage should determine which domains, learners, institutions and periods are represented. It should publish what remains outside the evidence and acknowledge that unobserved content or learners should not be treated as average. Personal information should remain protected while material inequality remains visible.

retain, material, limitation, denominator require ### 26. Coverage to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The conclusion should retain the material limitation and denominator.

The public purpose of evidence on instructional time is a responsible account of the scheduled, delivered and engaged time available for learning. Where calendar time can overstate exposure where interruption, absence or shortened sessions occur, results can overstate both weakness and certainty. Analysts should distinguish planned, delivered and usable time.

For alignment, bodies should identify how intended curriculum, teaching and assessment relate and should map content and cognitive demand without assuming complete correspondence. Any conclusion must respect that correlation of labels is not evidence of substantive alignment.

retain, material, limitation, denominator require ### 27. Alignment to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The conclusion should retain the material limitation and denominator.

For instructional time, interpretation depends on the scheduled, delivered and engaged time available for learning. The principal risk is that calendar time can overstate exposure where interruption, absence or shortened sessions occur. Evidence producers should therefore distinguish planned, delivered and usable time.

The comparability test asks which populations, constructs and procedures are sufficiently common. Analysts should compare metadata before values. The governing proposition is that a shared scale does not remove material contextual difference. Every claim should identify population, domain, period, source and material limitation.

Applied to instructional time, this requirement has a distinct interpretive consequence. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The implication for comparison and policy use should be recorded.

Uncertainty is material because the scheduled, delivered and engaged time available for learning cannot be inferred from a score, curriculum title or coverage percentage. In this domain, calendar time can overstate exposure where interruption, absence or shortened sessions occur. The immediate safeguard is to distinguish planned, delivered and usable time.

A defensible account should establish which sampling, measurement, linking and classification limits apply and should publish appropriate intervals and qualitative limitations, recognising that precision of display should not exceed evidence precision. Original categories and missing evidence should remain visible.

Applied to instructional time, this requirement has a distinct interpretive consequence. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 30. Equity to distinguish authority, delivery, distribution and remedy. The public-interest meaning of instructional time concerns the scheduled, delivered and engaged time available for learning. If calendar time can overstate exposure where interruption, absence or shortened sessions occur, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to distinguish planned, delivered and usable time. The implication for comparison and policy use should be recorded.

Under equity, review concerns which learners had different opportunity, access or assessment conditions. Statistical and education bodies should analyse relevant distributions and missing populations, subject to the rule that equal testing does not establish equal opportunity to learn. Difference caused by method should be distinguished from plausible educational difference.

Applied to instructional time, this requirement has a distinct interpretive consequence. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The implication for comparison and policy use should be recorded.

Applied to instructional time, this requirement has a distinct interpretive consequence. Analysis of instructional time should begin with the scheduled, delivered and engaged time available for learning. A foreseeable failure arises where calendar time can overstate exposure where interruption, absence or shortened sessions occur. Authorities should distinguish planned, delivered and usable time. The implication for comparison and policy use should be recorded.

The interpretive boundary standard requires consideration of which inference is supported and which remains unresolved. Producers should separate observation, association and causal explanation, because a result should not be extended from tested evidence to the whole education system. Revisions and breaks in series should be adjacent to the affected comparison.

comparison, policy, should, recorded require ### 31. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The implication for comparison and policy use should be recorded.

Applied to instructional time, this requirement has a distinct interpretive consequence. Instructional time requires explicit metadata because it concerns the scheduled, delivered and engaged time available for learning. The risk is that calendar time can overstate exposure where interruption, absence or shortened sessions occur. A competent producer should distinguish planned, delivered and usable time. The implication for comparison and policy use should be recorded.

Review of decision use should determine which curriculum, teaching, support or accountability decision the evidence informs. It should state authority, consequence and additional evidence needed and acknowledge that data should be proportionate to the stakes of the decision. Personal information should remain protected while material inequality remains visible.

comparison, policy, should, recorded require ### 32. Decision use to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 33. Review and correction to distinguish authority, delivery, distribution and remedy. The public purpose of evidence on instructional time is a responsible account of the scheduled, delivered and engaged time available for learning. Where calendar time can overstate exposure where interruption, absence or shortened sessions occur, results can overstate both weakness and certainty. Analysts should distinguish planned, delivered and usable time. The implication for comparison and policy use should be recorded.

For review and correction, bodies should identify how definitions, data and claims are challenged or revised and should retain versions, reasons and limitations. Any conclusion must respect that revisions should distinguish methodological change from educational change.

comparison, policy, should, recorded require ### 33. Review and correction to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The implication for comparison and policy use should be recorded.

The relevant evidence concerns the resources, teaching and prior content needed to encounter assessed knowledge. The principal risk is that low performance can be attributed to learners without examining whether the content was available. Authorities should measure exposure and conditions without treating exposure as mastery.[REF-15]

For opportunity to learn, interpretation depends on the resources, teaching and prior content needed to encounter assessed knowledge. The principal risk is that low performance can be attributed to learners without examining whether the content was available. Evidence producers should therefore measure exposure and conditions without treating exposure as mastery.

comparison, policy, should, recorded require ### 35. Definition to distinguish authority, delivery, distribution and remedy. The definition test asks which concept, unit and level of curriculum or outcome is being described. Analysts should state terms beside the result. The governing proposition is that familiar terminology does not ensure a common measure. Every claim should identify population, domain, period, source and material limitation. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 35. Definition to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. Any departure should be disclosed with its likely effect on interpretation.

Evidence source is material because the resources, teaching and prior content needed to encounter assessed knowledge cannot be inferred from a score, curriculum title or coverage percentage. In this domain, low performance can be attributed to learners without examining whether the content was available. The immediate safeguard is to measure exposure and conditions without treating exposure as mastery.

comparison, policy, should, recorded require ### 36. Evidence source to distinguish authority, delivery, distribution and remedy. A defensible account should establish which assessment, record, observation or report supports the claim and should identify provenance, date, coverage and collection method, recognising that different sources may complement one another without measuring the same object. Original categories and missing evidence should remain visible. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 36. Evidence source to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. Any departure should be disclosed with its likely effect on interpretation.

The public-interest meaning of opportunity to learn concerns the resources, teaching and prior content needed to encounter assessed knowledge. If low performance can be attributed to learners without examining whether the content was available, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to measure exposure and conditions without treating exposure as mastery.

comparison, policy, should, recorded require ### 37. Coverage to distinguish authority, delivery, distribution and remedy. Under coverage, review concerns which domains, learners, institutions and periods are represented. Statistical and education bodies should publish what remains outside the evidence, subject to the rule that unobserved content or learners should not be treated as average. Difference caused by method should be distinguished from plausible educational difference. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 37. Coverage to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. Any departure should be disclosed with its likely effect on interpretation.

Analysis of opportunity to learn should begin with the resources, teaching and prior content needed to encounter assessed knowledge. A foreseeable failure arises where low performance can be attributed to learners without examining whether the content was available. Authorities should measure exposure and conditions without treating exposure as mastery.

comparison, policy, should, recorded require ### 38. Alignment to distinguish authority, delivery, distribution and remedy. The alignment standard requires consideration of how intended curriculum, teaching and assessment relate. Producers should map content and cognitive demand without assuming complete correspondence, because correlation of labels is not evidence of substantive alignment. Revisions and breaks in series should be adjacent to the affected comparison. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 38. Alignment to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. Any departure should be disclosed with its likely effect on interpretation.

Opportunity to learn requires explicit metadata because it concerns the resources, teaching and prior content needed to encounter assessed knowledge. The risk is that low performance can be attributed to learners without examining whether the content was available. A competent producer should measure exposure and conditions without treating exposure as mastery.

comparison, policy, should, recorded require ### 39. Comparability to distinguish authority, delivery, distribution and remedy. Review of comparability should determine which populations, constructs and procedures are sufficiently common. It should compare metadata before values and acknowledge that a shared scale does not remove material contextual difference. Personal information should remain protected while material inequality remains visible. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 39. Comparability to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. Any departure should be disclosed with its likely effect on interpretation.

The public purpose of evidence on opportunity to learn is a responsible account of the resources, teaching and prior content needed to encounter assessed knowledge. Where low performance can be attributed to learners without examining whether the content was available, results can overstate both weakness and certainty. Analysts should measure exposure and conditions without treating exposure as mastery.

comparison, policy, should, recorded require ### 40. Uncertainty to distinguish authority, delivery, distribution and remedy. For uncertainty, bodies should identify which sampling, measurement, linking and classification limits apply and should publish appropriate intervals and qualitative limitations. Any conclusion must respect that precision of display should not exceed evidence precision. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 40. Uncertainty to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 41. Equity to distinguish authority, delivery, distribution and remedy. For opportunity to learn, interpretation depends on the resources, teaching and prior content needed to encounter assessed knowledge. The principal risk is that low performance can be attributed to learners without examining whether the content was available. Evidence producers should therefore measure exposure and conditions without treating exposure as mastery. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 41. Equity to distinguish authority, delivery, distribution and remedy. The equity test asks which learners had different opportunity, access or assessment conditions. Analysts should analyse relevant distributions and missing populations. The governing proposition is that equal testing does not establish equal opportunity to learn. Every claim should identify population, domain, period, source and material limitation. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 41. Equity to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The conclusion should retain the material limitation and denominator.

Interpretive boundary is material because the resources, teaching and prior content needed to encounter assessed knowledge cannot be inferred from a score, curriculum title or coverage percentage. In this domain, low performance can be attributed to learners without examining whether the content was available. The immediate safeguard is to measure exposure and conditions without treating exposure as mastery.

comparison, policy, should, recorded require ### 42. Interpretive boundary to distinguish authority, delivery, distribution and remedy. A defensible account should establish which inference is supported and which remains unresolved and should separate observation, association and causal explanation, recognising that a result should not be extended from tested evidence to the whole education system. Original categories and missing evidence should remain visible. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 42. Interpretive boundary to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 43. Decision use to distinguish authority, delivery, distribution and remedy. The public-interest meaning of opportunity to learn concerns the resources, teaching and prior content needed to encounter assessed knowledge. If low performance can be attributed to learners without examining whether the content was available, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to measure exposure and conditions without treating exposure as mastery. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 43. Decision use to distinguish authority, delivery, distribution and remedy. Under decision use, review concerns which curriculum, teaching, support or accountability decision the evidence informs. Statistical and education bodies should state authority, consequence and additional evidence needed, subject to the rule that data should be proportionate to the stakes of the decision. Difference caused by method should be distinguished from plausible educational difference. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 43. Decision use to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 44. Review and correction to distinguish authority, delivery, distribution and remedy. Analysis of opportunity to learn should begin with the resources, teaching and prior content needed to encounter assessed knowledge. A foreseeable failure arises where low performance can be attributed to learners without examining whether the content was available. Authorities should measure exposure and conditions without treating exposure as mastery. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 44. Review and correction to distinguish authority, delivery, distribution and remedy. The review and correction standard requires consideration of how definitions, data and claims are challenged or revised. Producers should retain versions, reasons and limitations, because revisions should distinguish methodological change from educational change. Revisions and breaks in series should be adjacent to the affected comparison. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 44. Review and correction to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The conclusion should retain the material limitation and denominator.

The relevant evidence concerns the order and prerequisite relationships through which learning develops. The principal risk is that coverage counts can give equal weight to isolated mention and sustained cumulative instruction. Authorities should examine sequence, depth and prerequisite opportunity.[REF-05]

Curriculum sequence requires explicit metadata because it concerns the order and prerequisite relationships through which learning develops. The risk is that coverage counts can give equal weight to isolated mention and sustained cumulative instruction. A competent producer should examine sequence, depth and prerequisite opportunity.

Applied to curriculum sequence, this requirement has a distinct interpretive consequence. Review of definition should determine which concept, unit and level of curriculum or outcome is being described. It should state terms beside the result and acknowledge that familiar terminology does not ensure a common measure. Personal information should remain protected while material inequality remains visible. The implication for comparison and policy use should be recorded.

For curriculum sequence, analysts should test this safeguard against the construct and population. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The conclusion should retain the material limitation and denominator.

The public purpose of evidence on curriculum sequence is a responsible account of the order and prerequisite relationships through which learning develops. Where coverage counts can give equal weight to isolated mention and sustained cumulative instruction, results can overstate both weakness and certainty. Analysts should examine sequence, depth and prerequisite opportunity.

Applied to curriculum sequence, this requirement has a distinct interpretive consequence. For evidence source, bodies should identify which assessment, record, observation or report supports the claim and should identify provenance, date, coverage and collection method. Any conclusion must respect that different sources may complement one another without measuring the same object. The implication for comparison and policy use should be recorded.

For curriculum sequence, analysts should test this safeguard against the construct and population. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The conclusion should retain the material limitation and denominator.

For curriculum sequence, interpretation depends on the order and prerequisite relationships through which learning develops. The principal risk is that coverage counts can give equal weight to isolated mention and sustained cumulative instruction. Evidence producers should therefore examine sequence, depth and prerequisite opportunity.

comparison, policy, should, recorded require ### 48. Coverage to distinguish authority, delivery, distribution and remedy. The coverage test asks which domains, learners, institutions and periods are represented. Analysts should publish what remains outside the evidence. The governing proposition is that unobserved content or learners should not be treated as average. Every claim should identify population, domain, period, source and material limitation. The implication for comparison and policy use should be recorded.

Applied to curriculum sequence, this requirement has a distinct interpretive consequence. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The implication for comparison and policy use should be recorded.

Alignment is material because the order and prerequisite relationships through which learning develops cannot be inferred from a score, curriculum title or coverage percentage. In this domain, coverage counts can give equal weight to isolated mention and sustained cumulative instruction. The immediate safeguard is to examine sequence, depth and prerequisite opportunity.

Applied to curriculum sequence, this requirement has a distinct interpretive consequence. A defensible account should establish how intended curriculum, teaching and assessment relate and should map content and cognitive demand without assuming complete correspondence, recognising that correlation of labels is not evidence of substantive alignment. Original categories and missing evidence should remain visible. The implication for comparison and policy use should be recorded.

Applied to curriculum sequence, this requirement has a distinct interpretive consequence. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The implication for comparison and policy use should be recorded.

The public-interest meaning of curriculum sequence concerns the order and prerequisite relationships through which learning develops. If coverage counts can give equal weight to isolated mention and sustained cumulative instruction, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to examine sequence, depth and prerequisite opportunity.

Applied to curriculum sequence, this requirement has a distinct interpretive consequence. Under comparability, review concerns which populations, constructs and procedures are sufficiently common. Statistical and education bodies should compare metadata before values, subject to the rule that a shared scale does not remove material contextual difference. Difference caused by method should be distinguished from plausible educational difference. The implication for comparison and policy use should be recorded.

Applied to curriculum sequence, this requirement has a distinct interpretive consequence. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The implication for comparison and policy use should be recorded.

Analysis of curriculum sequence should begin with the order and prerequisite relationships through which learning develops. A foreseeable failure arises where coverage counts can give equal weight to isolated mention and sustained cumulative instruction. Authorities should examine sequence, depth and prerequisite opportunity.

comparison, policy, should, recorded require ### 51. Uncertainty to distinguish authority, delivery, distribution and remedy. The uncertainty standard requires consideration of which sampling, measurement, linking and classification limits apply. Producers should publish appropriate intervals and qualitative limitations, because precision of display should not exceed evidence precision. Revisions and breaks in series should be adjacent to the affected comparison. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 51. Uncertainty to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The implication for comparison and policy use should be recorded.

Applied to curriculum sequence, this requirement has a distinct interpretive consequence. Curriculum sequence requires explicit metadata because it concerns the order and prerequisite relationships through which learning develops. The risk is that coverage counts can give equal weight to isolated mention and sustained cumulative instruction. A competent producer should examine sequence, depth and prerequisite opportunity. The implication for comparison and policy use should be recorded.

Applied to curriculum sequence, this requirement has a distinct interpretive consequence. Review of equity should determine which learners had different opportunity, access or assessment conditions. It should analyse relevant distributions and missing populations and acknowledge that equal testing does not establish equal opportunity to learn. Personal information should remain protected while material inequality remains visible. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 52. Equity to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 53. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The public purpose of evidence on curriculum sequence is a responsible account of the order and prerequisite relationships through which learning develops. Where coverage counts can give equal weight to isolated mention and sustained cumulative instruction, results can overstate both weakness and certainty. Analysts should examine sequence, depth and prerequisite opportunity. The implication for comparison and policy use should be recorded.

Applied to curriculum sequence, this requirement has a distinct interpretive consequence. For interpretive boundary, bodies should identify which inference is supported and which remains unresolved and should separate observation, association and causal explanation. Any conclusion must respect that a result should not be extended from tested evidence to the whole education system. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 53. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The implication for comparison and policy use should be recorded.

Applied to curriculum sequence, this requirement has a distinct interpretive consequence. For curriculum sequence, interpretation depends on the order and prerequisite relationships through which learning develops. The principal risk is that coverage counts can give equal weight to isolated mention and sustained cumulative instruction. Evidence producers should therefore examine sequence, depth and prerequisite opportunity. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 54. Decision use to distinguish authority, delivery, distribution and remedy. The decision use test asks which curriculum, teaching, support or accountability decision the evidence informs. Analysts should state authority, consequence and additional evidence needed. The governing proposition is that data should be proportionate to the stakes of the decision. Every claim should identify population, domain, period, source and material limitation. The implication for comparison and policy use should be recorded.

Within curriculum sequence, the principle bears on the evidence described in this part. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. Any departure should be disclosed with its likely effect on interpretation.

Review and correction is material because the order and prerequisite relationships through which learning develops cannot be inferred from a score, curriculum title or coverage percentage. In this domain, coverage counts can give equal weight to isolated mention and sustained cumulative instruction. The immediate safeguard is to examine sequence, depth and prerequisite opportunity.

Applied to curriculum sequence, this requirement has a distinct interpretive consequence. A defensible account should establish how definitions, data and claims are challenged or revised and should retain versions, reasons and limitations, recognising that revisions should distinguish methodological change from educational change. Original categories and missing evidence should remain visible. The implication for comparison and policy use should be recorded.

Within curriculum sequence, the principle bears on the evidence described in this part. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. Any departure should be disclosed with its likely effect on interpretation.

The relevant evidence concerns the construct, domain and purpose represented by an assessment. The principal risk is that a result can be extended to the complete curriculum although only a selected domain was tested. Authorities should publish construct, blueprint and intended uses.[REF-62]

The public-interest meaning of assessment framework concerns the construct, domain and purpose represented by an assessment. If a result can be extended to the complete curriculum although only a selected domain was tested, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to publish construct, blueprint and intended uses.

Applied to assessment framework, this requirement has a distinct interpretive consequence. Under definition, review concerns which concept, unit and level of curriculum or outcome is being described. Statistical and education bodies should state terms beside the result, subject to the rule that familiar terminology does not ensure a common measure. Difference caused by method should be distinguished from plausible educational difference. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 57. Definition to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. Any departure should be disclosed with its likely effect on interpretation.

Analysis of assessment framework should begin with the construct, domain and purpose represented by an assessment. A foreseeable failure arises where a result can be extended to the complete curriculum although only a selected domain was tested. Authorities should publish construct, blueprint and intended uses.

comparison, policy, should, recorded require ### 58. Evidence source to distinguish authority, delivery, distribution and remedy. The evidence source standard requires consideration of which assessment, record, observation or report supports the claim. Producers should identify provenance, date, coverage and collection method, because different sources may complement one another without measuring the same object. Revisions and breaks in series should be adjacent to the affected comparison. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 58. Evidence source to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. Any departure should be disclosed with its likely effect on interpretation.

Assessment framework requires explicit metadata because it concerns the construct, domain and purpose represented by an assessment. The risk is that a result can be extended to the complete curriculum although only a selected domain was tested. A competent producer should publish construct, blueprint and intended uses.

Applied to assessment framework, this requirement has a distinct interpretive consequence. Review of coverage should determine which domains, learners, institutions and periods are represented. It should publish what remains outside the evidence and acknowledge that unobserved content or learners should not be treated as average. Personal information should remain protected while material inequality remains visible. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 59. Coverage to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. Any departure should be disclosed with its likely effect on interpretation.

The public purpose of evidence on assessment framework is a responsible account of the construct, domain and purpose represented by an assessment. Where a result can be extended to the complete curriculum although only a selected domain was tested, results can overstate both weakness and certainty. Analysts should publish construct, blueprint and intended uses.

Applied to assessment framework, this requirement has a distinct interpretive consequence. For alignment, bodies should identify how intended curriculum, teaching and assessment relate and should map content and cognitive demand without assuming complete correspondence. Any conclusion must respect that correlation of labels is not evidence of substantive alignment. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 60. Alignment to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. Any departure should be disclosed with its likely effect on interpretation.

For assessment framework, interpretation depends on the construct, domain and purpose represented by an assessment. The principal risk is that a result can be extended to the complete curriculum although only a selected domain was tested. Evidence producers should therefore publish construct, blueprint and intended uses.

comparison, policy, should, recorded require ### 61. Comparability to distinguish authority, delivery, distribution and remedy. The comparability test asks which populations, constructs and procedures are sufficiently common. Analysts should compare metadata before values. The governing proposition is that a shared scale does not remove material contextual difference. Every claim should identify population, domain, period, source and material limitation. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 61. Comparability to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The conclusion should retain the material limitation and denominator.

Uncertainty is material because the construct, domain and purpose represented by an assessment cannot be inferred from a score, curriculum title or coverage percentage. In this domain, a result can be extended to the complete curriculum although only a selected domain was tested. The immediate safeguard is to publish construct, blueprint and intended uses.

Applied to assessment framework, this requirement has a distinct interpretive consequence. A defensible account should establish which sampling, measurement, linking and classification limits apply and should publish appropriate intervals and qualitative limitations, recognising that precision of display should not exceed evidence precision. Original categories and missing evidence should remain visible. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 62. Uncertainty to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 63. Equity to distinguish authority, delivery, distribution and remedy. The public-interest meaning of assessment framework concerns the construct, domain and purpose represented by an assessment. If a result can be extended to the complete curriculum although only a selected domain was tested, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to publish construct, blueprint and intended uses. The implication for comparison and policy use should be recorded.

Applied to assessment framework, this requirement has a distinct interpretive consequence. Under equity, review concerns which learners had different opportunity, access or assessment conditions. Statistical and education bodies should analyse relevant distributions and missing populations, subject to the rule that equal testing does not establish equal opportunity to learn. Difference caused by method should be distinguished from plausible educational difference. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 63. Equity to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The conclusion should retain the material limitation and denominator.

Applied to assessment framework, this requirement has a distinct interpretive consequence. Analysis of assessment framework should begin with the construct, domain and purpose represented by an assessment. A foreseeable failure arises where a result can be extended to the complete curriculum although only a selected domain was tested. Authorities should publish construct, blueprint and intended uses. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 64. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The interpretive boundary standard requires consideration of which inference is supported and which remains unresolved. Producers should separate observation, association and causal explanation, because a result should not be extended from tested evidence to the whole education system. Revisions and breaks in series should be adjacent to the affected comparison. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 64. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The conclusion should retain the material limitation and denominator.

Applied to assessment framework, this requirement has a distinct interpretive consequence. Assessment framework requires explicit metadata because it concerns the construct, domain and purpose represented by an assessment. The risk is that a result can be extended to the complete curriculum although only a selected domain was tested. A competent producer should publish construct, blueprint and intended uses. The implication for comparison and policy use should be recorded.

Applied to assessment framework, this requirement has a distinct interpretive consequence. Review of decision use should determine which curriculum, teaching, support or accountability decision the evidence informs. It should state authority, consequence and additional evidence needed and acknowledge that data should be proportionate to the stakes of the decision. Personal information should remain protected while material inequality remains visible. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 65. Decision use to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 66. Review and correction to distinguish authority, delivery, distribution and remedy. The public purpose of evidence on assessment framework is a responsible account of the construct, domain and purpose represented by an assessment. Where a result can be extended to the complete curriculum although only a selected domain was tested, results can overstate both weakness and certainty. Analysts should publish construct, blueprint and intended uses. The implication for comparison and policy use should be recorded.

Applied to assessment framework, this requirement has a distinct interpretive consequence. For review and correction, bodies should identify how definitions, data and claims are challenged or revised and should retain versions, reasons and limitations. Any conclusion must respect that revisions should distinguish methodological change from educational change. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 66. Review and correction to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The conclusion should retain the material limitation and denominator.

The relevant evidence concerns which content, cognitive demand and response modes appear in the instrument. The principal risk is that a limited sample of tasks can create unstable conclusions about broad curriculum domains. Authorities should report sampling and uncertainty and avoid item-level generalisation.[REF-63]

For task and item coverage, interpretation depends on which content, cognitive demand and response modes appear in the instrument. The principal risk is that a limited sample of tasks can create unstable conclusions about broad curriculum domains. Evidence producers should therefore report sampling and uncertainty and avoid item-level generalisation.

disclosed, likely, effect, interpretation require ### 68. Definition to distinguish authority, delivery, distribution and remedy. The definition test asks which concept, unit and level of curriculum or outcome is being described. Analysts should state terms beside the result. The governing proposition is that familiar terminology does not ensure a common measure. Every claim should identify population, domain, period, source and material limitation. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 68. Definition to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The implication for comparison and policy use should be recorded.

Evidence source is material because which content, cognitive demand and response modes appear in the instrument cannot be inferred from a score, curriculum title or coverage percentage. In this domain, a limited sample of tasks can create unstable conclusions about broad curriculum domains. The immediate safeguard is to report sampling and uncertainty and avoid item-level generalisation.

disclosed, likely, effect, interpretation require ### 69. Evidence source to distinguish authority, delivery, distribution and remedy. A defensible account should establish which assessment, record, observation or report supports the claim and should identify provenance, date, coverage and collection method, recognising that different sources may complement one another without measuring the same object. Original categories and missing evidence should remain visible. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 69. Evidence source to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The implication for comparison and policy use should be recorded.

The public-interest meaning of task and item coverage concerns which content, cognitive demand and response modes appear in the instrument. If a limited sample of tasks can create unstable conclusions about broad curriculum domains, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to report sampling and uncertainty and avoid item-level generalisation.

disclosed, likely, effect, interpretation require ### 70. Coverage to distinguish authority, delivery, distribution and remedy. Under coverage, review concerns which domains, learners, institutions and periods are represented. Statistical and education bodies should publish what remains outside the evidence, subject to the rule that unobserved content or learners should not be treated as average. Difference caused by method should be distinguished from plausible educational difference. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 70. Coverage to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The implication for comparison and policy use should be recorded.

Analysis of task and item coverage should begin with which content, cognitive demand and response modes appear in the instrument. A foreseeable failure arises where a limited sample of tasks can create unstable conclusions about broad curriculum domains. Authorities should report sampling and uncertainty and avoid item-level generalisation.

disclosed, likely, effect, interpretation require ### 71. Alignment to distinguish authority, delivery, distribution and remedy. The alignment standard requires consideration of how intended curriculum, teaching and assessment relate. Producers should map content and cognitive demand without assuming complete correspondence, because correlation of labels is not evidence of substantive alignment. Revisions and breaks in series should be adjacent to the affected comparison. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 71. Alignment to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The implication for comparison and policy use should be recorded.

Task and item coverage requires explicit metadata because it concerns which content, cognitive demand and response modes appear in the instrument. The risk is that a limited sample of tasks can create unstable conclusions about broad curriculum domains. A competent producer should report sampling and uncertainty and avoid item-level generalisation.

disclosed, likely, effect, interpretation require ### 72. Comparability to distinguish authority, delivery, distribution and remedy. Review of comparability should determine which populations, constructs and procedures are sufficiently common. It should compare metadata before values and acknowledge that a shared scale does not remove material contextual difference. Personal information should remain protected while material inequality remains visible. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 72. Comparability to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The implication for comparison and policy use should be recorded.

The public purpose of evidence on task and item coverage is a responsible account of which content, cognitive demand and response modes appear in the instrument. Where a limited sample of tasks can create unstable conclusions about broad curriculum domains, results can overstate both weakness and certainty. Analysts should report sampling and uncertainty and avoid item-level generalisation.

disclosed, likely, effect, interpretation require ### 73. Uncertainty to distinguish authority, delivery, distribution and remedy. For uncertainty, bodies should identify which sampling, measurement, linking and classification limits apply and should publish appropriate intervals and qualitative limitations. Any conclusion must respect that precision of display should not exceed evidence precision. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 73. Uncertainty to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 74. Equity to distinguish authority, delivery, distribution and remedy. For task and item coverage, interpretation depends on which content, cognitive demand and response modes appear in the instrument. The principal risk is that a limited sample of tasks can create unstable conclusions about broad curriculum domains. Evidence producers should therefore report sampling and uncertainty and avoid item-level generalisation. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 74. Equity to distinguish authority, delivery, distribution and remedy. The equity test asks which learners had different opportunity, access or assessment conditions. Analysts should analyse relevant distributions and missing populations. The governing proposition is that equal testing does not establish equal opportunity to learn. Every claim should identify population, domain, period, source and material limitation. Any departure should be disclosed with its likely effect on interpretation.

disclosed, likely, effect, interpretation require ### 74. Equity to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. Any departure should be disclosed with its likely effect on interpretation.

Interpretive boundary is material because which content, cognitive demand and response modes appear in the instrument cannot be inferred from a score, curriculum title or coverage percentage. In this domain, a limited sample of tasks can create unstable conclusions about broad curriculum domains. The immediate safeguard is to report sampling and uncertainty and avoid item-level generalisation.

disclosed, likely, effect, interpretation require ### 75. Interpretive boundary to distinguish authority, delivery, distribution and remedy. A defensible account should establish which inference is supported and which remains unresolved and should separate observation, association and causal explanation, recognising that a result should not be extended from tested evidence to the whole education system. Original categories and missing evidence should remain visible. Any departure should be disclosed with its likely effect on interpretation.

disclosed, likely, effect, interpretation require ### 75. Interpretive boundary to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 76. Decision use to distinguish authority, delivery, distribution and remedy. The public-interest meaning of task and item coverage concerns which content, cognitive demand and response modes appear in the instrument. If a limited sample of tasks can create unstable conclusions about broad curriculum domains, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to report sampling and uncertainty and avoid item-level generalisation. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 76. Decision use to distinguish authority, delivery, distribution and remedy. Under decision use, review concerns which curriculum, teaching, support or accountability decision the evidence informs. Statistical and education bodies should state authority, consequence and additional evidence needed, subject to the rule that data should be proportionate to the stakes of the decision. Difference caused by method should be distinguished from plausible educational difference. Any departure should be disclosed with its likely effect on interpretation.

disclosed, likely, effect, interpretation require ### 76. Decision use to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 77. Review and correction to distinguish authority, delivery, distribution and remedy. Analysis of task and item coverage should begin with which content, cognitive demand and response modes appear in the instrument. A foreseeable failure arises where a limited sample of tasks can create unstable conclusions about broad curriculum domains. Authorities should report sampling and uncertainty and avoid item-level generalisation. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 77. Review and correction to distinguish authority, delivery, distribution and remedy. The review and correction standard requires consideration of how definitions, data and claims are challenged or revised. Producers should retain versions, reasons and limitations, because revisions should distinguish methodological change from educational change. Revisions and breaks in series should be adjacent to the affected comparison. Any departure should be disclosed with its likely effect on interpretation.

disclosed, likely, effect, interpretation require ### 77. Review and correction to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. Any departure should be disclosed with its likely effect on interpretation.

The relevant evidence concerns the criteria or thresholds used to describe levels of learning. The principal risk is that standards may be treated as natural facts although they depend on purpose and judgement. Authorities should state method, authority, consequence and uncertainty around classification.[REF-64]

Performance standards requires explicit metadata because it concerns the criteria or thresholds used to describe levels of learning. The risk is that standards may be treated as natural facts although they depend on purpose and judgement. A competent producer should state method, authority, consequence and uncertainty around classification.

disclosed, likely, effect, interpretation require ### 79. Definition to distinguish authority, delivery, distribution and remedy. Review of definition should determine which concept, unit and level of curriculum or outcome is being described. It should state terms beside the result and acknowledge that familiar terminology does not ensure a common measure. Personal information should remain protected while material inequality remains visible. Any departure should be disclosed with its likely effect on interpretation.

disclosed, likely, effect, interpretation require ### 79. Definition to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. Any departure should be disclosed with its likely effect on interpretation.

The public purpose of evidence on performance standards is a responsible account of the criteria or thresholds used to describe levels of learning. Where standards may be treated as natural facts although they depend on purpose and judgement, results can overstate both weakness and certainty. Analysts should state method, authority, consequence and uncertainty around classification.

disclosed, likely, effect, interpretation require ### 80. Evidence source to distinguish authority, delivery, distribution and remedy. For evidence source, bodies should identify which assessment, record, observation or report supports the claim and should identify provenance, date, coverage and collection method. Any conclusion must respect that different sources may complement one another without measuring the same object. Any departure should be disclosed with its likely effect on interpretation.

disclosed, likely, effect, interpretation require ### 80. Evidence source to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. Any departure should be disclosed with its likely effect on interpretation.

For performance standards, interpretation depends on the criteria or thresholds used to describe levels of learning. The principal risk is that standards may be treated as natural facts although they depend on purpose and judgement. Evidence producers should therefore state method, authority, consequence and uncertainty around classification.

disclosed, likely, effect, interpretation require ### 81. Coverage to distinguish authority, delivery, distribution and remedy. The coverage test asks which domains, learners, institutions and periods are represented. Analysts should publish what remains outside the evidence. The governing proposition is that unobserved content or learners should not be treated as average. Every claim should identify population, domain, period, source and material limitation. Any departure should be disclosed with its likely effect on interpretation.

retain, material, limitation, denominator require ### 81. Coverage to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The conclusion should retain the material limitation and denominator.

Alignment is material because the criteria or thresholds used to describe levels of learning cannot be inferred from a score, curriculum title or coverage percentage. In this domain, standards may be treated as natural facts although they depend on purpose and judgement. The immediate safeguard is to state method, authority, consequence and uncertainty around classification.

disclosed, likely, effect, interpretation require ### 82. Alignment to distinguish authority, delivery, distribution and remedy. A defensible account should establish how intended curriculum, teaching and assessment relate and should map content and cognitive demand without assuming complete correspondence, recognising that correlation of labels is not evidence of substantive alignment. Original categories and missing evidence should remain visible. Any departure should be disclosed with its likely effect on interpretation.

retain, material, limitation, denominator require ### 82. Alignment to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The conclusion should retain the material limitation and denominator.

The public-interest meaning of performance standards concerns the criteria or thresholds used to describe levels of learning. If standards may be treated as natural facts although they depend on purpose and judgement, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to state method, authority, consequence and uncertainty around classification.

disclosed, likely, effect, interpretation require ### 83. Comparability to distinguish authority, delivery, distribution and remedy. Under comparability, review concerns which populations, constructs and procedures are sufficiently common. Statistical and education bodies should compare metadata before values, subject to the rule that a shared scale does not remove material contextual difference. Difference caused by method should be distinguished from plausible educational difference. Any departure should be disclosed with its likely effect on interpretation.

retain, material, limitation, denominator require ### 83. Comparability to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The conclusion should retain the material limitation and denominator.

Analysis of performance standards should begin with the criteria or thresholds used to describe levels of learning. A foreseeable failure arises where standards may be treated as natural facts although they depend on purpose and judgement. Authorities should state method, authority, consequence and uncertainty around classification.

disclosed, likely, effect, interpretation require ### 84. Uncertainty to distinguish authority, delivery, distribution and remedy. The uncertainty standard requires consideration of which sampling, measurement, linking and classification limits apply. Producers should publish appropriate intervals and qualitative limitations, because precision of display should not exceed evidence precision. Revisions and breaks in series should be adjacent to the affected comparison. Any departure should be disclosed with its likely effect on interpretation.

retain, material, limitation, denominator require ### 84. Uncertainty to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The conclusion should retain the material limitation and denominator.

Applied to performance standards, this requirement has a distinct interpretive consequence. Performance standards requires explicit metadata because it concerns the criteria or thresholds used to describe levels of learning. The risk is that standards may be treated as natural facts although they depend on purpose and judgement. A competent producer should state method, authority, consequence and uncertainty around classification. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 85. Equity to distinguish authority, delivery, distribution and remedy. Review of equity should determine which learners had different opportunity, access or assessment conditions. It should analyse relevant distributions and missing populations and acknowledge that equal testing does not establish equal opportunity to learn. Personal information should remain protected while material inequality remains visible. Any departure should be disclosed with its likely effect on interpretation.

retain, material, limitation, denominator require ### 85. Equity to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The conclusion should retain the material limitation and denominator.

Applied to performance standards, this requirement has a distinct interpretive consequence. The public purpose of evidence on performance standards is a responsible account of the criteria or thresholds used to describe levels of learning. Where standards may be treated as natural facts although they depend on purpose and judgement, results can overstate both weakness and certainty. Analysts should state method, authority, consequence and uncertainty around classification. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 86. Interpretive boundary to distinguish authority, delivery, distribution and remedy. For interpretive boundary, bodies should identify which inference is supported and which remains unresolved and should separate observation, association and causal explanation. Any conclusion must respect that a result should not be extended from tested evidence to the whole education system. Any departure should be disclosed with its likely effect on interpretation.

retain, material, limitation, denominator require ### 86. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The conclusion should retain the material limitation and denominator.

Applied to performance standards, this requirement has a distinct interpretive consequence. For performance standards, interpretation depends on the criteria or thresholds used to describe levels of learning. The principal risk is that standards may be treated as natural facts although they depend on purpose and judgement. Evidence producers should therefore state method, authority, consequence and uncertainty around classification. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 87. Decision use to distinguish authority, delivery, distribution and remedy. The decision use test asks which curriculum, teaching, support or accountability decision the evidence informs. Analysts should state authority, consequence and additional evidence needed. The governing proposition is that data should be proportionate to the stakes of the decision. Every claim should identify population, domain, period, source and material limitation. Any departure should be disclosed with its likely effect on interpretation.

Applied to performance standards, this requirement has a distinct interpretive consequence. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The implication for comparison and policy use should be recorded.

Review and correction is material because the criteria or thresholds used to describe levels of learning cannot be inferred from a score, curriculum title or coverage percentage. In this domain, standards may be treated as natural facts although they depend on purpose and judgement. The immediate safeguard is to state method, authority, consequence and uncertainty around classification.

disclosed, likely, effect, interpretation require ### 88. Review and correction to distinguish authority, delivery, distribution and remedy. A defensible account should establish how definitions, data and claims are challenged or revised and should retain versions, reasons and limitations, recognising that revisions should distinguish methodological change from educational change. Original categories and missing evidence should remain visible. Any departure should be disclosed with its likely effect on interpretation.

Applied to performance standards, this requirement has a distinct interpretive consequence. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The implication for comparison and policy use should be recorded.

The relevant evidence concerns which learners were eligible, sampled, present, excluded or unobserved. The principal risk is that absence and exclusion can make results unrepresentative of the enrolled or age population. Authorities should publish denominators, participation and characteristics of missing learners.[REF-65]

The public-interest meaning of population and participation concerns which learners were eligible, sampled, present, excluded or unobserved. If absence and exclusion can make results unrepresentative of the enrolled or age population, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to publish denominators, participation and characteristics of missing learners.

disclosed, likely, effect, interpretation require ### 90. Definition to distinguish authority, delivery, distribution and remedy. Under definition, review concerns which concept, unit and level of curriculum or outcome is being described. Statistical and education bodies should state terms beside the result, subject to the rule that familiar terminology does not ensure a common measure. Difference caused by method should be distinguished from plausible educational difference. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 90. Definition to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The implication for comparison and policy use should be recorded.

Analysis of population and participation should begin with which learners were eligible, sampled, present, excluded or unobserved. A foreseeable failure arises where absence and exclusion can make results unrepresentative of the enrolled or age population. Authorities should publish denominators, participation and characteristics of missing learners.

disclosed, likely, effect, interpretation require ### 91. Evidence source to distinguish authority, delivery, distribution and remedy. The evidence source standard requires consideration of which assessment, record, observation or report supports the claim. Producers should identify provenance, date, coverage and collection method, because different sources may complement one another without measuring the same object. Revisions and breaks in series should be adjacent to the affected comparison. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 91. Evidence source to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The implication for comparison and policy use should be recorded.

Population and participation requires explicit metadata because it concerns which learners were eligible, sampled, present, excluded or unobserved. The risk is that absence and exclusion can make results unrepresentative of the enrolled or age population. A competent producer should publish denominators, participation and characteristics of missing learners.

disclosed, likely, effect, interpretation require ### 92. Coverage to distinguish authority, delivery, distribution and remedy. Review of coverage should determine which domains, learners, institutions and periods are represented. It should publish what remains outside the evidence and acknowledge that unobserved content or learners should not be treated as average. Personal information should remain protected while material inequality remains visible. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 92. Coverage to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The implication for comparison and policy use should be recorded.

The public purpose of evidence on population and participation is a responsible account of which learners were eligible, sampled, present, excluded or unobserved. Where absence and exclusion can make results unrepresentative of the enrolled or age population, results can overstate both weakness and certainty. Analysts should publish denominators, participation and characteristics of missing learners.

disclosed, likely, effect, interpretation require ### 93. Alignment to distinguish authority, delivery, distribution and remedy. For alignment, bodies should identify how intended curriculum, teaching and assessment relate and should map content and cognitive demand without assuming complete correspondence. Any conclusion must respect that correlation of labels is not evidence of substantive alignment. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 93. Alignment to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The implication for comparison and policy use should be recorded.

For population and participation, interpretation depends on which learners were eligible, sampled, present, excluded or unobserved. The principal risk is that absence and exclusion can make results unrepresentative of the enrolled or age population. Evidence producers should therefore publish denominators, participation and characteristics of missing learners.

disclosed, likely, effect, interpretation require ### 94. Comparability to distinguish authority, delivery, distribution and remedy. The comparability test asks which populations, constructs and procedures are sufficiently common. Analysts should compare metadata before values. The governing proposition is that a shared scale does not remove material contextual difference. Every claim should identify population, domain, period, source and material limitation. Any departure should be disclosed with its likely effect on interpretation.

disclosed, likely, effect, interpretation require ### 94. Comparability to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. Any departure should be disclosed with its likely effect on interpretation.

Uncertainty is material because which learners were eligible, sampled, present, excluded or unobserved cannot be inferred from a score, curriculum title or coverage percentage. In this domain, absence and exclusion can make results unrepresentative of the enrolled or age population. The immediate safeguard is to publish denominators, participation and characteristics of missing learners.

disclosed, likely, effect, interpretation require ### 95. Uncertainty to distinguish authority, delivery, distribution and remedy. A defensible account should establish which sampling, measurement, linking and classification limits apply and should publish appropriate intervals and qualitative limitations, recognising that precision of display should not exceed evidence precision. Original categories and missing evidence should remain visible. Any departure should be disclosed with its likely effect on interpretation.

disclosed, likely, effect, interpretation require ### 95. Uncertainty to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 96. Equity to distinguish authority, delivery, distribution and remedy. The public-interest meaning of population and participation concerns which learners were eligible, sampled, present, excluded or unobserved. If absence and exclusion can make results unrepresentative of the enrolled or age population, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to publish denominators, participation and characteristics of missing learners. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 96. Equity to distinguish authority, delivery, distribution and remedy. Under equity, review concerns which learners had different opportunity, access or assessment conditions. Statistical and education bodies should analyse relevant distributions and missing populations, subject to the rule that equal testing does not establish equal opportunity to learn. Difference caused by method should be distinguished from plausible educational difference. Any departure should be disclosed with its likely effect on interpretation.

disclosed, likely, effect, interpretation require ### 96. Equity to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 97. Interpretive boundary to distinguish authority, delivery, distribution and remedy. Analysis of population and participation should begin with which learners were eligible, sampled, present, excluded or unobserved. A foreseeable failure arises where absence and exclusion can make results unrepresentative of the enrolled or age population. Authorities should publish denominators, participation and characteristics of missing learners. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 97. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The interpretive boundary standard requires consideration of which inference is supported and which remains unresolved. Producers should separate observation, association and causal explanation, because a result should not be extended from tested evidence to the whole education system. Revisions and breaks in series should be adjacent to the affected comparison. Any departure should be disclosed with its likely effect on interpretation.

disclosed, likely, effect, interpretation require ### 97. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 98. Decision use to distinguish authority, delivery, distribution and remedy. Population and participation requires explicit metadata because it concerns which learners were eligible, sampled, present, excluded or unobserved. The risk is that absence and exclusion can make results unrepresentative of the enrolled or age population. A competent producer should publish denominators, participation and characteristics of missing learners. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 98. Decision use to distinguish authority, delivery, distribution and remedy. Review of decision use should determine which curriculum, teaching, support or accountability decision the evidence informs. It should state authority, consequence and additional evidence needed and acknowledge that data should be proportionate to the stakes of the decision. Personal information should remain protected while material inequality remains visible. Any departure should be disclosed with its likely effect on interpretation.

disclosed, likely, effect, interpretation require ### 98. Decision use to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 99. Review and correction to distinguish authority, delivery, distribution and remedy. The public purpose of evidence on population and participation is a responsible account of which learners were eligible, sampled, present, excluded or unobserved. Where absence and exclusion can make results unrepresentative of the enrolled or age population, results can overstate both weakness and certainty. Analysts should publish denominators, participation and characteristics of missing learners. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 99. Review and correction to distinguish authority, delivery, distribution and remedy. For review and correction, bodies should identify how definitions, data and claims are challenged or revised and should retain versions, reasons and limitations. Any conclusion must respect that revisions should distinguish methodological change from educational change. Any departure should be disclosed with its likely effect on interpretation.

disclosed, likely, effect, interpretation require ### 99. Review and correction to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. Any departure should be disclosed with its likely effect on interpretation.

The relevant evidence concerns the conditions, language, accessibility and procedures under which evidence was obtained. The principal risk is that different conditions can affect performance independently of intended learning. Authorities should verify material variation and qualify comparisons.[REF-10]

For administration and comparability, interpretation depends on the conditions, language, accessibility and procedures under which evidence was obtained. The principal risk is that different conditions can affect performance independently of intended learning. Evidence producers should therefore verify material variation and qualify comparisons.

retain, material, limitation, denominator require ### 101. Definition to distinguish authority, delivery, distribution and remedy. The definition test asks which concept, unit and level of curriculum or outcome is being described. Analysts should state terms beside the result. The governing proposition is that familiar terminology does not ensure a common measure. Every claim should identify population, domain, period, source and material limitation. The conclusion should retain the material limitation and denominator.

retain, material, limitation, denominator require ### 101. Definition to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The conclusion should retain the material limitation and denominator.

Evidence source is material because the conditions, language, accessibility and procedures under which evidence was obtained cannot be inferred from a score, curriculum title or coverage percentage. In this domain, different conditions can affect performance independently of intended learning. The immediate safeguard is to verify material variation and qualify comparisons.

retain, material, limitation, denominator require ### 102. Evidence source to distinguish authority, delivery, distribution and remedy. A defensible account should establish which assessment, record, observation or report supports the claim and should identify provenance, date, coverage and collection method, recognising that different sources may complement one another without measuring the same object. Original categories and missing evidence should remain visible. The conclusion should retain the material limitation and denominator.

retain, material, limitation, denominator require ### 102. Evidence source to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The conclusion should retain the material limitation and denominator.

The public-interest meaning of administration and comparability concerns the conditions, language, accessibility and procedures under which evidence was obtained. If different conditions can affect performance independently of intended learning, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to verify material variation and qualify comparisons.

retain, material, limitation, denominator require ### 103. Coverage to distinguish authority, delivery, distribution and remedy. Under coverage, review concerns which domains, learners, institutions and periods are represented. Statistical and education bodies should publish what remains outside the evidence, subject to the rule that unobserved content or learners should not be treated as average. Difference caused by method should be distinguished from plausible educational difference. The conclusion should retain the material limitation and denominator.

retain, material, limitation, denominator require ### 103. Coverage to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The conclusion should retain the material limitation and denominator.

Analysis of administration and comparability should begin with the conditions, language, accessibility and procedures under which evidence was obtained. A foreseeable failure arises where different conditions can affect performance independently of intended learning. Authorities should verify material variation and qualify comparisons.

retain, material, limitation, denominator require ### 104. Alignment to distinguish authority, delivery, distribution and remedy. The alignment standard requires consideration of how intended curriculum, teaching and assessment relate. Producers should map content and cognitive demand without assuming complete correspondence, because correlation of labels is not evidence of substantive alignment. Revisions and breaks in series should be adjacent to the affected comparison. The conclusion should retain the material limitation and denominator.

retain, material, limitation, denominator require ### 104. Alignment to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The conclusion should retain the material limitation and denominator.

Administration and comparability requires explicit metadata because it concerns the conditions, language, accessibility and procedures under which evidence was obtained. The risk is that different conditions can affect performance independently of intended learning. A competent producer should verify material variation and qualify comparisons.

retain, material, limitation, denominator require ### 105. Comparability to distinguish authority, delivery, distribution and remedy. Review of comparability should determine which populations, constructs and procedures are sufficiently common. It should compare metadata before values and acknowledge that a shared scale does not remove material contextual difference. Personal information should remain protected while material inequality remains visible. The conclusion should retain the material limitation and denominator.

retain, material, limitation, denominator require ### 105. Comparability to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The conclusion should retain the material limitation and denominator.

The public purpose of evidence on administration and comparability is a responsible account of the conditions, language, accessibility and procedures under which evidence was obtained. Where different conditions can affect performance independently of intended learning, results can overstate both weakness and certainty. Analysts should verify material variation and qualify comparisons.

retain, material, limitation, denominator require ### 106. Uncertainty to distinguish authority, delivery, distribution and remedy. For uncertainty, bodies should identify which sampling, measurement, linking and classification limits apply and should publish appropriate intervals and qualitative limitations. Any conclusion must respect that precision of display should not exceed evidence precision. The conclusion should retain the material limitation and denominator.

retain, material, limitation, denominator require ### 106. Uncertainty to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 107. Equity to distinguish authority, delivery, distribution and remedy. For administration and comparability, interpretation depends on the conditions, language, accessibility and procedures under which evidence was obtained. The principal risk is that different conditions can affect performance independently of intended learning. Evidence producers should therefore verify material variation and qualify comparisons. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 107. Equity to distinguish authority, delivery, distribution and remedy. The equity test asks which learners had different opportunity, access or assessment conditions. Analysts should analyse relevant distributions and missing populations. The governing proposition is that equal testing does not establish equal opportunity to learn. Every claim should identify population, domain, period, source and material limitation. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 107. Equity to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The implication for comparison and policy use should be recorded.

Interpretive boundary is material because the conditions, language, accessibility and procedures under which evidence was obtained cannot be inferred from a score, curriculum title or coverage percentage. In this domain, different conditions can affect performance independently of intended learning. The immediate safeguard is to verify material variation and qualify comparisons.

retain, material, limitation, denominator require ### 108. Interpretive boundary to distinguish authority, delivery, distribution and remedy. A defensible account should establish which inference is supported and which remains unresolved and should separate observation, association and causal explanation, recognising that a result should not be extended from tested evidence to the whole education system. Original categories and missing evidence should remain visible. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 108. Interpretive boundary to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 109. Decision use to distinguish authority, delivery, distribution and remedy. The public-interest meaning of administration and comparability concerns the conditions, language, accessibility and procedures under which evidence was obtained. If different conditions can affect performance independently of intended learning, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to verify material variation and qualify comparisons. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 109. Decision use to distinguish authority, delivery, distribution and remedy. Under decision use, review concerns which curriculum, teaching, support or accountability decision the evidence informs. Statistical and education bodies should state authority, consequence and additional evidence needed, subject to the rule that data should be proportionate to the stakes of the decision. Difference caused by method should be distinguished from plausible educational difference. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 109. Decision use to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 110. Review and correction to distinguish authority, delivery, distribution and remedy. Analysis of administration and comparability should begin with the conditions, language, accessibility and procedures under which evidence was obtained. A foreseeable failure arises where different conditions can affect performance independently of intended learning. Authorities should verify material variation and qualify comparisons. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 110. Review and correction to distinguish authority, delivery, distribution and remedy. The review and correction standard requires consideration of how definitions, data and claims are challenged or revised. Producers should retain versions, reasons and limitations, because revisions should distinguish methodological change from educational change. Revisions and breaks in series should be adjacent to the affected comparison. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 110. Review and correction to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The implication for comparison and policy use should be recorded.

The relevant evidence concerns professional evidence from tasks, observation and ongoing assessment. The principal risk is that teacher judgement can be dismissed as subjective or used without common criteria and moderation. Authorities should use documented criteria, multiple evidence and proportionate moderation.[REF-67]

Teacher judgement and classroom evidence requires explicit metadata because it concerns professional evidence from tasks, observation and ongoing assessment. The risk is that teacher judgement can be dismissed as subjective or used without common criteria and moderation. A competent producer should use documented criteria, multiple evidence and proportionate moderation.

retain, material, limitation, denominator require ### 112. Definition to distinguish authority, delivery, distribution and remedy. Review of definition should determine which concept, unit and level of curriculum or outcome is being described. It should state terms beside the result and acknowledge that familiar terminology does not ensure a common measure. Personal information should remain protected while material inequality remains visible. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 112. Definition to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The implication for comparison and policy use should be recorded.

The public purpose of evidence on teacher judgement and classroom evidence is a responsible account of professional evidence from tasks, observation and ongoing assessment. Where teacher judgement can be dismissed as subjective or used without common criteria and moderation, results can overstate both weakness and certainty. Analysts should use documented criteria, multiple evidence and proportionate moderation.

retain, material, limitation, denominator require ### 113. Evidence source to distinguish authority, delivery, distribution and remedy. For evidence source, bodies should identify which assessment, record, observation or report supports the claim and should identify provenance, date, coverage and collection method. Any conclusion must respect that different sources may complement one another without measuring the same object. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 113. Evidence source to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The implication for comparison and policy use should be recorded.

For teacher judgement and classroom evidence, interpretation depends on professional evidence from tasks, observation and ongoing assessment. The principal risk is that teacher judgement can be dismissed as subjective or used without common criteria and moderation. Evidence producers should therefore use documented criteria, multiple evidence and proportionate moderation.

retain, material, limitation, denominator require ### 114. Coverage to distinguish authority, delivery, distribution and remedy. The coverage test asks which domains, learners, institutions and periods are represented. Analysts should publish what remains outside the evidence. The governing proposition is that unobserved content or learners should not be treated as average. Every claim should identify population, domain, period, source and material limitation. The conclusion should retain the material limitation and denominator.

disclosed, likely, effect, interpretation require ### 114. Coverage to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. Any departure should be disclosed with its likely effect on interpretation.

Alignment is material because professional evidence from tasks, observation and ongoing assessment cannot be inferred from a score, curriculum title or coverage percentage. In this domain, teacher judgement can be dismissed as subjective or used without common criteria and moderation. The immediate safeguard is to use documented criteria, multiple evidence and proportionate moderation.

retain, material, limitation, denominator require ### 115. Alignment to distinguish authority, delivery, distribution and remedy. A defensible account should establish how intended curriculum, teaching and assessment relate and should map content and cognitive demand without assuming complete correspondence, recognising that correlation of labels is not evidence of substantive alignment. Original categories and missing evidence should remain visible. The conclusion should retain the material limitation and denominator.

disclosed, likely, effect, interpretation require ### 115. Alignment to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. Any departure should be disclosed with its likely effect on interpretation.

The public-interest meaning of teacher judgement and classroom evidence concerns professional evidence from tasks, observation and ongoing assessment. If teacher judgement can be dismissed as subjective or used without common criteria and moderation, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to use documented criteria, multiple evidence and proportionate moderation.

retain, material, limitation, denominator require ### 116. Comparability to distinguish authority, delivery, distribution and remedy. Under comparability, review concerns which populations, constructs and procedures are sufficiently common. Statistical and education bodies should compare metadata before values, subject to the rule that a shared scale does not remove material contextual difference. Difference caused by method should be distinguished from plausible educational difference. The conclusion should retain the material limitation and denominator.

disclosed, likely, effect, interpretation require ### 116. Comparability to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. Any departure should be disclosed with its likely effect on interpretation.

Analysis of teacher judgement and classroom evidence should begin with professional evidence from tasks, observation and ongoing assessment. A foreseeable failure arises where teacher judgement can be dismissed as subjective or used without common criteria and moderation. Authorities should use documented criteria, multiple evidence and proportionate moderation.

retain, material, limitation, denominator require ### 117. Uncertainty to distinguish authority, delivery, distribution and remedy. The uncertainty standard requires consideration of which sampling, measurement, linking and classification limits apply. Producers should publish appropriate intervals and qualitative limitations, because precision of display should not exceed evidence precision. Revisions and breaks in series should be adjacent to the affected comparison. The conclusion should retain the material limitation and denominator.

disclosed, likely, effect, interpretation require ### 117. Uncertainty to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 118. Equity to distinguish authority, delivery, distribution and remedy. Teacher judgement and classroom evidence requires explicit metadata because it concerns professional evidence from tasks, observation and ongoing assessment. The risk is that teacher judgement can be dismissed as subjective or used without common criteria and moderation. A competent producer should use documented criteria, multiple evidence and proportionate moderation. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 118. Equity to distinguish authority, delivery, distribution and remedy. Review of equity should determine which learners had different opportunity, access or assessment conditions. It should analyse relevant distributions and missing populations and acknowledge that equal testing does not establish equal opportunity to learn. Personal information should remain protected while material inequality remains visible. The conclusion should retain the material limitation and denominator.

disclosed, likely, effect, interpretation require ### 118. Equity to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 119. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The public purpose of evidence on teacher judgement and classroom evidence is a responsible account of professional evidence from tasks, observation and ongoing assessment. Where teacher judgement can be dismissed as subjective or used without common criteria and moderation, results can overstate both weakness and certainty. Analysts should use documented criteria, multiple evidence and proportionate moderation. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 119. Interpretive boundary to distinguish authority, delivery, distribution and remedy. For interpretive boundary, bodies should identify which inference is supported and which remains unresolved and should separate observation, association and causal explanation. Any conclusion must respect that a result should not be extended from tested evidence to the whole education system. The conclusion should retain the material limitation and denominator.

disclosed, likely, effect, interpretation require ### 119. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 120. Decision use to distinguish authority, delivery, distribution and remedy. For teacher judgement and classroom evidence, interpretation depends on professional evidence from tasks, observation and ongoing assessment. The principal risk is that teacher judgement can be dismissed as subjective or used without common criteria and moderation. Evidence producers should therefore use documented criteria, multiple evidence and proportionate moderation. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 120. Decision use to distinguish authority, delivery, distribution and remedy. The decision use test asks which curriculum, teaching, support or accountability decision the evidence informs. Analysts should state authority, consequence and additional evidence needed. The governing proposition is that data should be proportionate to the stakes of the decision. Every claim should identify population, domain, period, source and material limitation. The conclusion should retain the material limitation and denominator.

retain, material, limitation, denominator require ### 120. Decision use to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The conclusion should retain the material limitation and denominator.

Review and correction is material because professional evidence from tasks, observation and ongoing assessment cannot be inferred from a score, curriculum title or coverage percentage. In this domain, teacher judgement can be dismissed as subjective or used without common criteria and moderation. The immediate safeguard is to use documented criteria, multiple evidence and proportionate moderation.

retain, material, limitation, denominator require ### 121. Review and correction to distinguish authority, delivery, distribution and remedy. A defensible account should establish how definitions, data and claims are challenged or revised and should retain versions, reasons and limitations, recognising that revisions should distinguish methodological change from educational change. Original categories and missing evidence should remain visible. The conclusion should retain the material limitation and denominator.

retain, material, limitation, denominator require ### 121. Review and correction to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The conclusion should retain the material limitation and denominator.

The relevant evidence concerns variation across learners, institutions, places and relevant population groups. The principal risk is that means can conceal both severe disadvantage and high performance within the same system. Authorities should report distributions and responsible disaggregation with uncertainty.[REF-12]

The public-interest meaning of subgroup and distribution concerns variation across learners, institutions, places and relevant population groups. If means can conceal both severe disadvantage and high performance within the same system, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to report distributions and responsible disaggregation with uncertainty.

retain, material, limitation, denominator require ### 123. Definition to distinguish authority, delivery, distribution and remedy. Under definition, review concerns which concept, unit and level of curriculum or outcome is being described. Statistical and education bodies should state terms beside the result, subject to the rule that familiar terminology does not ensure a common measure. Difference caused by method should be distinguished from plausible educational difference. The conclusion should retain the material limitation and denominator.

retain, material, limitation, denominator require ### 123. Definition to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The conclusion should retain the material limitation and denominator.

Analysis of subgroup and distribution should begin with variation across learners, institutions, places and relevant population groups. A foreseeable failure arises where means can conceal both severe disadvantage and high performance within the same system. Authorities should report distributions and responsible disaggregation with uncertainty.

retain, material, limitation, denominator require ### 124. Evidence source to distinguish authority, delivery, distribution and remedy. The evidence source standard requires consideration of which assessment, record, observation or report supports the claim. Producers should identify provenance, date, coverage and collection method, because different sources may complement one another without measuring the same object. Revisions and breaks in series should be adjacent to the affected comparison. The conclusion should retain the material limitation and denominator.

retain, material, limitation, denominator require ### 124. Evidence source to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The conclusion should retain the material limitation and denominator.

Subgroup and distribution requires explicit metadata because it concerns variation across learners, institutions, places and relevant population groups. The risk is that means can conceal both severe disadvantage and high performance within the same system. A competent producer should report distributions and responsible disaggregation with uncertainty.

retain, material, limitation, denominator require ### 125. Coverage to distinguish authority, delivery, distribution and remedy. Review of coverage should determine which domains, learners, institutions and periods are represented. It should publish what remains outside the evidence and acknowledge that unobserved content or learners should not be treated as average. Personal information should remain protected while material inequality remains visible. The conclusion should retain the material limitation and denominator.

retain, material, limitation, denominator require ### 125. Coverage to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The conclusion should retain the material limitation and denominator.

The public purpose of evidence on subgroup and distribution is a responsible account of variation across learners, institutions, places and relevant population groups. Where means can conceal both severe disadvantage and high performance within the same system, results can overstate both weakness and certainty. Analysts should report distributions and responsible disaggregation with uncertainty.

retain, material, limitation, denominator require ### 126. Alignment to distinguish authority, delivery, distribution and remedy. For alignment, bodies should identify how intended curriculum, teaching and assessment relate and should map content and cognitive demand without assuming complete correspondence. Any conclusion must respect that correlation of labels is not evidence of substantive alignment. The conclusion should retain the material limitation and denominator.

retain, material, limitation, denominator require ### 126. Alignment to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The conclusion should retain the material limitation and denominator.

For subgroup and distribution, interpretation depends on variation across learners, institutions, places and relevant population groups. The principal risk is that means can conceal both severe disadvantage and high performance within the same system. Evidence producers should therefore report distributions and responsible disaggregation with uncertainty.

retain, material, limitation, denominator require ### 127. Comparability to distinguish authority, delivery, distribution and remedy. The comparability test asks which populations, constructs and procedures are sufficiently common. Analysts should compare metadata before values. The governing proposition is that a shared scale does not remove material contextual difference. Every claim should identify population, domain, period, source and material limitation. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 127. Comparability to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The implication for comparison and policy use should be recorded.

Uncertainty is material because variation across learners, institutions, places and relevant population groups cannot be inferred from a score, curriculum title or coverage percentage. In this domain, means can conceal both severe disadvantage and high performance within the same system. The immediate safeguard is to report distributions and responsible disaggregation with uncertainty.

retain, material, limitation, denominator require ### 128. Uncertainty to distinguish authority, delivery, distribution and remedy. A defensible account should establish which sampling, measurement, linking and classification limits apply and should publish appropriate intervals and qualitative limitations, recognising that precision of display should not exceed evidence precision. Original categories and missing evidence should remain visible. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 128. Uncertainty to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 129. Equity to distinguish authority, delivery, distribution and remedy. The public-interest meaning of subgroup and distribution concerns variation across learners, institutions, places and relevant population groups. If means can conceal both severe disadvantage and high performance within the same system, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to report distributions and responsible disaggregation with uncertainty. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 129. Equity to distinguish authority, delivery, distribution and remedy. Under equity, review concerns which learners had different opportunity, access or assessment conditions. Statistical and education bodies should analyse relevant distributions and missing populations, subject to the rule that equal testing does not establish equal opportunity to learn. Difference caused by method should be distinguished from plausible educational difference. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 129. Equity to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 130. Interpretive boundary to distinguish authority, delivery, distribution and remedy. Analysis of subgroup and distribution should begin with variation across learners, institutions, places and relevant population groups. A foreseeable failure arises where means can conceal both severe disadvantage and high performance within the same system. Authorities should report distributions and responsible disaggregation with uncertainty. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 130. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The interpretive boundary standard requires consideration of which inference is supported and which remains unresolved. Producers should separate observation, association and causal explanation, because a result should not be extended from tested evidence to the whole education system. Revisions and breaks in series should be adjacent to the affected comparison. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 130. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 131. Decision use to distinguish authority, delivery, distribution and remedy. Subgroup and distribution requires explicit metadata because it concerns variation across learners, institutions, places and relevant population groups. The risk is that means can conceal both severe disadvantage and high performance within the same system. A competent producer should report distributions and responsible disaggregation with uncertainty. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 131. Decision use to distinguish authority, delivery, distribution and remedy. Review of decision use should determine which curriculum, teaching, support or accountability decision the evidence informs. It should state authority, consequence and additional evidence needed and acknowledge that data should be proportionate to the stakes of the decision. Personal information should remain protected while material inequality remains visible. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 131. Decision use to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 132. Review and correction to distinguish authority, delivery, distribution and remedy. The public purpose of evidence on subgroup and distribution is a responsible account of variation across learners, institutions, places and relevant population groups. Where means can conceal both severe disadvantage and high performance within the same system, results can overstate both weakness and certainty. Analysts should report distributions and responsible disaggregation with uncertainty. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 132. Review and correction to distinguish authority, delivery, distribution and remedy. For review and correction, bodies should identify how definitions, data and claims are challenged or revised and should retain versions, reasons and limitations. Any conclusion must respect that revisions should distinguish methodological change from educational change. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 132. Review and correction to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The implication for comparison and policy use should be recorded.

The relevant evidence concerns the extent to which constructs, populations, languages and curricula support comparison. The principal risk is that common scales can create an appearance of curricular equivalence that does not exist. Authorities should compare metadata and limit claims to shared constructs.[REF-70]

For cross-system comparison, interpretation depends on the extent to which constructs, populations, languages and curricula support comparison. The principal risk is that common scales can create an appearance of curricular equivalence that does not exist. Evidence producers should therefore compare metadata and limit claims to shared constructs.

comparison, policy, should, recorded require ### 134. Definition to distinguish authority, delivery, distribution and remedy. The definition test asks which concept, unit and level of curriculum or outcome is being described. Analysts should state terms beside the result. The governing proposition is that familiar terminology does not ensure a common measure. Every claim should identify population, domain, period, source and material limitation. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 134. Definition to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. Any departure should be disclosed with its likely effect on interpretation.

Evidence source is material because the extent to which constructs, populations, languages and curricula support comparison cannot be inferred from a score, curriculum title or coverage percentage. In this domain, common scales can create an appearance of curricular equivalence that does not exist. The immediate safeguard is to compare metadata and limit claims to shared constructs.

comparison, policy, should, recorded require ### 135. Evidence source to distinguish authority, delivery, distribution and remedy. A defensible account should establish which assessment, record, observation or report supports the claim and should identify provenance, date, coverage and collection method, recognising that different sources may complement one another without measuring the same object. Original categories and missing evidence should remain visible. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 135. Evidence source to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. Any departure should be disclosed with its likely effect on interpretation.

The public-interest meaning of cross-system comparison concerns the extent to which constructs, populations, languages and curricula support comparison. If common scales can create an appearance of curricular equivalence that does not exist, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to compare metadata and limit claims to shared constructs.

comparison, policy, should, recorded require ### 136. Coverage to distinguish authority, delivery, distribution and remedy. Under coverage, review concerns which domains, learners, institutions and periods are represented. Statistical and education bodies should publish what remains outside the evidence, subject to the rule that unobserved content or learners should not be treated as average. Difference caused by method should be distinguished from plausible educational difference. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 136. Coverage to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. Any departure should be disclosed with its likely effect on interpretation.

Analysis of cross-system comparison should begin with the extent to which constructs, populations, languages and curricula support comparison. A foreseeable failure arises where common scales can create an appearance of curricular equivalence that does not exist. Authorities should compare metadata and limit claims to shared constructs.

comparison, policy, should, recorded require ### 137. Alignment to distinguish authority, delivery, distribution and remedy. The alignment standard requires consideration of how intended curriculum, teaching and assessment relate. Producers should map content and cognitive demand without assuming complete correspondence, because correlation of labels is not evidence of substantive alignment. Revisions and breaks in series should be adjacent to the affected comparison. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 137. Alignment to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. Any departure should be disclosed with its likely effect on interpretation.

Cross-system comparison requires explicit metadata because it concerns the extent to which constructs, populations, languages and curricula support comparison. The risk is that common scales can create an appearance of curricular equivalence that does not exist. A competent producer should compare metadata and limit claims to shared constructs.

comparison, policy, should, recorded require ### 138. Comparability to distinguish authority, delivery, distribution and remedy. Review of comparability should determine which populations, constructs and procedures are sufficiently common. It should compare metadata before values and acknowledge that a shared scale does not remove material contextual difference. Personal information should remain protected while material inequality remains visible. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 138. Comparability to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. Any departure should be disclosed with its likely effect on interpretation.

The public purpose of evidence on cross-system comparison is a responsible account of the extent to which constructs, populations, languages and curricula support comparison. Where common scales can create an appearance of curricular equivalence that does not exist, results can overstate both weakness and certainty. Analysts should compare metadata and limit claims to shared constructs.

comparison, policy, should, recorded require ### 139. Uncertainty to distinguish authority, delivery, distribution and remedy. For uncertainty, bodies should identify which sampling, measurement, linking and classification limits apply and should publish appropriate intervals and qualitative limitations. Any conclusion must respect that precision of display should not exceed evidence precision. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 139. Uncertainty to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. Any departure should be disclosed with its likely effect on interpretation.

comparison, policy, should, recorded require ### 140. Equity to distinguish authority, delivery, distribution and remedy. For cross-system comparison, interpretation depends on the extent to which constructs, populations, languages and curricula support comparison. The principal risk is that common scales can create an appearance of curricular equivalence that does not exist. Evidence producers should therefore compare metadata and limit claims to shared constructs. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 140. Equity to distinguish authority, delivery, distribution and remedy. The equity test asks which learners had different opportunity, access or assessment conditions. Analysts should analyse relevant distributions and missing populations. The governing proposition is that equal testing does not establish equal opportunity to learn. Every claim should identify population, domain, period, source and material limitation. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 140. Equity to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The conclusion should retain the material limitation and denominator.

Interpretive boundary is material because the extent to which constructs, populations, languages and curricula support comparison cannot be inferred from a score, curriculum title or coverage percentage. In this domain, common scales can create an appearance of curricular equivalence that does not exist. The immediate safeguard is to compare metadata and limit claims to shared constructs.

comparison, policy, should, recorded require ### 141. Interpretive boundary to distinguish authority, delivery, distribution and remedy. A defensible account should establish which inference is supported and which remains unresolved and should separate observation, association and causal explanation, recognising that a result should not be extended from tested evidence to the whole education system. Original categories and missing evidence should remain visible. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 141. Interpretive boundary to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 142. Decision use to distinguish authority, delivery, distribution and remedy. The public-interest meaning of cross-system comparison concerns the extent to which constructs, populations, languages and curricula support comparison. If common scales can create an appearance of curricular equivalence that does not exist, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to compare metadata and limit claims to shared constructs. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 142. Decision use to distinguish authority, delivery, distribution and remedy. Under decision use, review concerns which curriculum, teaching, support or accountability decision the evidence informs. Statistical and education bodies should state authority, consequence and additional evidence needed, subject to the rule that data should be proportionate to the stakes of the decision. Difference caused by method should be distinguished from plausible educational difference. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 142. Decision use to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 143. Review and correction to distinguish authority, delivery, distribution and remedy. Analysis of cross-system comparison should begin with the extent to which constructs, populations, languages and curricula support comparison. A foreseeable failure arises where common scales can create an appearance of curricular equivalence that does not exist. Authorities should compare metadata and limit claims to shared constructs. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 143. Review and correction to distinguish authority, delivery, distribution and remedy. The review and correction standard requires consideration of how definitions, data and claims are challenged or revised. Producers should retain versions, reasons and limitations, because revisions should distinguish methodological change from educational change. Revisions and breaks in series should be adjacent to the affected comparison. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 143. Review and correction to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The conclusion should retain the material limitation and denominator.

The relevant evidence concerns whether results over time represent comparable populations, instruments and conditions. The principal risk is that reform, disruption, participation or scaling changes can be mistaken for learning change. Authorities should document breaks, linking uncertainty and alternative explanations.[REF-71]

Trend and change requires explicit metadata because it concerns whether results over time represent comparable populations, instruments and conditions. The risk is that reform, disruption, participation or scaling changes can be mistaken for learning change. A competent producer should document breaks, linking uncertainty and alternative explanations.

comparison, policy, should, recorded require ### 145. Definition to distinguish authority, delivery, distribution and remedy. Review of definition should determine which concept, unit and level of curriculum or outcome is being described. It should state terms beside the result and acknowledge that familiar terminology does not ensure a common measure. Personal information should remain protected while material inequality remains visible. The implication for comparison and policy use should be recorded.

For trend and change, analysts should test this safeguard against the construct and population. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The conclusion should retain the material limitation and denominator.

The public purpose of evidence on trend and change is a responsible account of whether results over time represent comparable populations, instruments and conditions. Where reform, disruption, participation or scaling changes can be mistaken for learning change, results can overstate both weakness and certainty. Analysts should document breaks, linking uncertainty and alternative explanations.

comparison, policy, should, recorded require ### 146. Evidence source to distinguish authority, delivery, distribution and remedy. For evidence source, bodies should identify which assessment, record, observation or report supports the claim and should identify provenance, date, coverage and collection method. Any conclusion must respect that different sources may complement one another without measuring the same object. The implication for comparison and policy use should be recorded.

For trend and change, analysts should test this safeguard against the construct and population. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The conclusion should retain the material limitation and denominator.

For trend and change, interpretation depends on whether results over time represent comparable populations, instruments and conditions. The principal risk is that reform, disruption, participation or scaling changes can be mistaken for learning change. Evidence producers should therefore document breaks, linking uncertainty and alternative explanations.

comparison, policy, should, recorded require ### 147. Coverage to distinguish authority, delivery, distribution and remedy. The coverage test asks which domains, learners, institutions and periods are represented. Analysts should publish what remains outside the evidence. The governing proposition is that unobserved content or learners should not be treated as average. Every claim should identify population, domain, period, source and material limitation. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 147. Coverage to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The implication for comparison and policy use should be recorded.

Alignment is material because whether results over time represent comparable populations, instruments and conditions cannot be inferred from a score, curriculum title or coverage percentage. In this domain, reform, disruption, participation or scaling changes can be mistaken for learning change. The immediate safeguard is to document breaks, linking uncertainty and alternative explanations.

comparison, policy, should, recorded require ### 148. Alignment to distinguish authority, delivery, distribution and remedy. A defensible account should establish how intended curriculum, teaching and assessment relate and should map content and cognitive demand without assuming complete correspondence, recognising that correlation of labels is not evidence of substantive alignment. Original categories and missing evidence should remain visible. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 148. Alignment to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The implication for comparison and policy use should be recorded.

The public-interest meaning of trend and change concerns whether results over time represent comparable populations, instruments and conditions. If reform, disruption, participation or scaling changes can be mistaken for learning change, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to document breaks, linking uncertainty and alternative explanations.

comparison, policy, should, recorded require ### 149. Comparability to distinguish authority, delivery, distribution and remedy. Under comparability, review concerns which populations, constructs and procedures are sufficiently common. Statistical and education bodies should compare metadata before values, subject to the rule that a shared scale does not remove material contextual difference. Difference caused by method should be distinguished from plausible educational difference. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 149. Comparability to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The implication for comparison and policy use should be recorded.

Analysis of trend and change should begin with whether results over time represent comparable populations, instruments and conditions. A foreseeable failure arises where reform, disruption, participation or scaling changes can be mistaken for learning change. Authorities should document breaks, linking uncertainty and alternative explanations.

comparison, policy, should, recorded require ### 150. Uncertainty to distinguish authority, delivery, distribution and remedy. The uncertainty standard requires consideration of which sampling, measurement, linking and classification limits apply. Producers should publish appropriate intervals and qualitative limitations, because precision of display should not exceed evidence precision. Revisions and breaks in series should be adjacent to the affected comparison. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 150. Uncertainty to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 151. Equity to distinguish authority, delivery, distribution and remedy. Trend and change requires explicit metadata because it concerns whether results over time represent comparable populations, instruments and conditions. The risk is that reform, disruption, participation or scaling changes can be mistaken for learning change. A competent producer should document breaks, linking uncertainty and alternative explanations. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 151. Equity to distinguish authority, delivery, distribution and remedy. Review of equity should determine which learners had different opportunity, access or assessment conditions. It should analyse relevant distributions and missing populations and acknowledge that equal testing does not establish equal opportunity to learn. Personal information should remain protected while material inequality remains visible. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 151. Equity to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 152. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The public purpose of evidence on trend and change is a responsible account of whether results over time represent comparable populations, instruments and conditions. Where reform, disruption, participation or scaling changes can be mistaken for learning change, results can overstate both weakness and certainty. Analysts should document breaks, linking uncertainty and alternative explanations. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 152. Interpretive boundary to distinguish authority, delivery, distribution and remedy. For interpretive boundary, bodies should identify which inference is supported and which remains unresolved and should separate observation, association and causal explanation. Any conclusion must respect that a result should not be extended from tested evidence to the whole education system. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 152. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 153. Decision use to distinguish authority, delivery, distribution and remedy. For trend and change, interpretation depends on whether results over time represent comparable populations, instruments and conditions. The principal risk is that reform, disruption, participation or scaling changes can be mistaken for learning change. Evidence producers should therefore document breaks, linking uncertainty and alternative explanations. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 153. Decision use to distinguish authority, delivery, distribution and remedy. The decision use test asks which curriculum, teaching, support or accountability decision the evidence informs. Analysts should state authority, consequence and additional evidence needed. The governing proposition is that data should be proportionate to the stakes of the decision. Every claim should identify population, domain, period, source and material limitation. The implication for comparison and policy use should be recorded.

Within trend and change, the principle bears on the evidence described in this part. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. Any departure should be disclosed with its likely effect on interpretation.

Review and correction is material because whether results over time represent comparable populations, instruments and conditions cannot be inferred from a score, curriculum title or coverage percentage. In this domain, reform, disruption, participation or scaling changes can be mistaken for learning change. The immediate safeguard is to document breaks, linking uncertainty and alternative explanations.

comparison, policy, should, recorded require ### 154. Review and correction to distinguish authority, delivery, distribution and remedy. A defensible account should establish how definitions, data and claims are challenged or revised and should retain versions, reasons and limitations, recognising that revisions should distinguish methodological change from educational change. Original categories and missing evidence should remain visible. The implication for comparison and policy use should be recorded.

Within trend and change, the principle bears on the evidence described in this part. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. Any departure should be disclosed with its likely effect on interpretation.

The relevant evidence concerns which conclusions about teaching, curriculum and policy are supported by the evidence. The principal risk is that association between coverage and outcomes can be presented as proof of one causal mechanism. Authorities should use a bounded account and connect decisions with additional evidence.[REF-15]

The public-interest meaning of causal and policy interpretation concerns which conclusions about teaching, curriculum and policy are supported by the evidence. If association between coverage and outcomes can be presented as proof of one causal mechanism, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to use a bounded account and connect decisions with additional evidence.

comparison, policy, should, recorded require ### 156. Definition to distinguish authority, delivery, distribution and remedy. Under definition, review concerns which concept, unit and level of curriculum or outcome is being described. Statistical and education bodies should state terms beside the result, subject to the rule that familiar terminology does not ensure a common measure. Difference caused by method should be distinguished from plausible educational difference. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 156. Definition to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. Any departure should be disclosed with its likely effect on interpretation.

Analysis of causal and policy interpretation should begin with which conclusions about teaching, curriculum and policy are supported by the evidence. A foreseeable failure arises where association between coverage and outcomes can be presented as proof of one causal mechanism. Authorities should use a bounded account and connect decisions with additional evidence.

comparison, policy, should, recorded require ### 157. Evidence source to distinguish authority, delivery, distribution and remedy. The evidence source standard requires consideration of which assessment, record, observation or report supports the claim. Producers should identify provenance, date, coverage and collection method, because different sources may complement one another without measuring the same object. Revisions and breaks in series should be adjacent to the affected comparison. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 157. Evidence source to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. Any departure should be disclosed with its likely effect on interpretation.

Causal and policy interpretation requires explicit metadata because it concerns which conclusions about teaching, curriculum and policy are supported by the evidence. The risk is that association between coverage and outcomes can be presented as proof of one causal mechanism. A competent producer should use a bounded account and connect decisions with additional evidence.

comparison, policy, should, recorded require ### 158. Coverage to distinguish authority, delivery, distribution and remedy. Review of coverage should determine which domains, learners, institutions and periods are represented. It should publish what remains outside the evidence and acknowledge that unobserved content or learners should not be treated as average. Personal information should remain protected while material inequality remains visible. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 158. Coverage to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. Any departure should be disclosed with its likely effect on interpretation.

The public purpose of evidence on causal and policy interpretation is a responsible account of which conclusions about teaching, curriculum and policy are supported by the evidence. Where association between coverage and outcomes can be presented as proof of one causal mechanism, results can overstate both weakness and certainty. Analysts should use a bounded account and connect decisions with additional evidence.

comparison, policy, should, recorded require ### 159. Alignment to distinguish authority, delivery, distribution and remedy. For alignment, bodies should identify how intended curriculum, teaching and assessment relate and should map content and cognitive demand without assuming complete correspondence. Any conclusion must respect that correlation of labels is not evidence of substantive alignment. The implication for comparison and policy use should be recorded.

disclosed, likely, effect, interpretation require ### 159. Alignment to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. Any departure should be disclosed with its likely effect on interpretation.

For causal and policy interpretation, interpretation depends on which conclusions about teaching, curriculum and policy are supported by the evidence. The principal risk is that association between coverage and outcomes can be presented as proof of one causal mechanism. Evidence producers should therefore use a bounded account and connect decisions with additional evidence.

comparison, policy, should, recorded require ### 160. Comparability to distinguish authority, delivery, distribution and remedy. The comparability test asks which populations, constructs and procedures are sufficiently common. Analysts should compare metadata before values. The governing proposition is that a shared scale does not remove material contextual difference. Every claim should identify population, domain, period, source and material limitation. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 160. Comparability to distinguish authority, delivery, distribution and remedy. Learning outcomes and curriculum coverage are related but not interchangeable. Exposure does not establish mastery, and assessed performance does not establish that every curriculum domain was taught or measured. The conclusion should retain the material limitation and denominator.

Uncertainty is material because which conclusions about teaching, curriculum and policy are supported by the evidence cannot be inferred from a score, curriculum title or coverage percentage. In this domain, association between coverage and outcomes can be presented as proof of one causal mechanism. The immediate safeguard is to use a bounded account and connect decisions with additional evidence.

comparison, policy, should, recorded require ### 161. Uncertainty to distinguish authority, delivery, distribution and remedy. A defensible account should establish which sampling, measurement, linking and classification limits apply and should publish appropriate intervals and qualitative limitations, recognising that precision of display should not exceed evidence precision. Original categories and missing evidence should remain visible. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 161. Uncertainty to distinguish authority, delivery, distribution and remedy. Where evidence supports only a narrow domain or population, the conclusion should remain narrow. Wider policy relevance can be explored without representing it as a measured result. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 162. Equity to distinguish authority, delivery, distribution and remedy. The public-interest meaning of causal and policy interpretation concerns which conclusions about teaching, curriculum and policy are supported by the evidence. If association between coverage and outcomes can be presented as proof of one causal mechanism, an apparently precise indicator can direct attention to the wrong instructional condition. The required course is to use a bounded account and connect decisions with additional evidence. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 162. Equity to distinguish authority, delivery, distribution and remedy. Under equity, review concerns which learners had different opportunity, access or assessment conditions. Statistical and education bodies should analyse relevant distributions and missing populations, subject to the rule that equal testing does not establish equal opportunity to learn. Difference caused by method should be distinguished from plausible educational difference. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 162. Equity to distinguish authority, delivery, distribution and remedy. Subgroup interpretation should retain denominators and uncertainty. A ranking should not give small or weakly measured differences a policy significance they cannot support. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 163. Interpretive boundary to distinguish authority, delivery, distribution and remedy. Analysis of causal and policy interpretation should begin with which conclusions about teaching, curriculum and policy are supported by the evidence. A foreseeable failure arises where association between coverage and outcomes can be presented as proof of one causal mechanism. Authorities should use a bounded account and connect decisions with additional evidence. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 163. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The interpretive boundary standard requires consideration of which inference is supported and which remains unresolved. Producers should separate observation, association and causal explanation, because a result should not be extended from tested evidence to the whole education system. Revisions and breaks in series should be adjacent to the affected comparison. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 163. Interpretive boundary to distinguish authority, delivery, distribution and remedy. The evidence should be connected with instructional time, staff, materials, language, attendance and prior learning where these are relevant. No single indicator can identify the complete mechanism. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 164. Decision use to distinguish authority, delivery, distribution and remedy. Causal and policy interpretation requires explicit metadata because it concerns which conclusions about teaching, curriculum and policy are supported by the evidence. The risk is that association between coverage and outcomes can be presented as proof of one causal mechanism. A competent producer should use a bounded account and connect decisions with additional evidence. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 164. Decision use to distinguish authority, delivery, distribution and remedy. Review of decision use should determine which curriculum, teaching, support or accountability decision the evidence informs. It should state authority, consequence and additional evidence needed and acknowledge that data should be proportionate to the stakes of the decision. Personal information should remain protected while material inequality remains visible. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 164. Decision use to distinguish authority, delivery, distribution and remedy. The narrative should distinguish a descriptive finding, an explanatory hypothesis and a policy judgement. Each has a different evidence requirement and degree of uncertainty. The conclusion should retain the material limitation and denominator.

comparison, policy, should, recorded require ### 165. Review and correction to distinguish authority, delivery, distribution and remedy. The public purpose of evidence on causal and policy interpretation is a responsible account of which conclusions about teaching, curriculum and policy are supported by the evidence. Where association between coverage and outcomes can be presented as proof of one causal mechanism, results can overstate both weakness and certainty. Analysts should use a bounded account and connect decisions with additional evidence. The implication for comparison and policy use should be recorded.

comparison, policy, should, recorded require ### 165. Review and correction to distinguish authority, delivery, distribution and remedy. For review and correction, bodies should identify how definitions, data and claims are challenged or revised and should retain versions, reasons and limitations. Any conclusion must respect that revisions should distinguish methodological change from educational change. The implication for comparison and policy use should be recorded.

retain, material, limitation, denominator require ### 165. Review and correction to distinguish authority, delivery, distribution and remedy. The final account should explain what decision can reasonably follow and what additional evidence is necessary. Measurement should support improvement without converting limited evidence into a complete judgement of learners, teachers or institutions. The conclusion should retain the material limitation and denominator.

In this field, the relevant evidence concerns the resources, formats and tasks through which intended content becomes available. A material risk is that official coverage may be recorded although materials are absent, inaccessible, unsuitable or used only briefly. Authorities should examine learner access, sustained use, language, accessibility and continuity.

The definition test asks which material, task or feedback event is being counted. Analysts should state the unit, threshold and source, recognising that availability is not use and use is not evidence of learning. The account should identify population, curriculum domain, period, source and material limitation.

establishes, complete, learning, outcome require ### 169. Learning materials and task access: Definition to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-75]

implication, interpretation, should, recorded require ### 170. Learning materials and task access: Coverage to distinguish authority, delivery, distribution and remedy. In this field, the relevant evidence concerns the resources, formats and tasks through which intended content becomes available. A material risk is that official coverage may be recorded although materials are absent, inaccessible, unsuitable or used only briefly. Authorities should examine learner access, sustained use, language, accessibility and continuity. The resulting implication for interpretation should be recorded.

The coverage test asks which learners, content and period are represented. Analysts should identify missing resources, inaccessible formats and unobserved activity, recognising that unobserved access should not be assigned the average condition. The account should identify population, curriculum domain, period, source and material limitation.

establishes, complete, learning, outcome require ### 170. Learning materials and task access: Coverage to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-05]

Within learning materials and response evidence, the requirement should be read against the specific curriculum evidence. In this field, the relevant evidence concerns the resources, formats and tasks through which intended content becomes available. A material risk is that official coverage may be recorded although materials are absent, inaccessible, unsuitable or used only briefly. Authorities should examine learner access, sustained use, language, accessibility and continuity. Any departure should retain its source and limitation.

The depth test asks whether engagement involved mention, explanation, practice, application or cumulative work. Analysts should describe intensity and sequence, recognising that a binary coverage flag can conceal substantial differences in opportunity. The account should identify population, curriculum domain, period, source and material limitation.

establishes, complete, learning, outcome require ### 171. Learning materials and task access: Depth to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-15]

For learning materials and response evidence, the same boundary has a separate population and decision consequence. In this field, the relevant evidence concerns the resources, formats and tasks through which intended content becomes available. A material risk is that official coverage may be recorded although materials are absent, inaccessible, unsuitable or used only briefly. Authorities should examine learner access, sustained use, language, accessibility and continuity. The conclusion should remain bounded to the evidence described.

The evidence test asks which record, observation or learner account supports the finding. Analysts should triangulate sources and retain disagreement, recognising that a distribution list or lesson plan is not proof of learner use. The account should identify population, curriculum domain, period, source and material limitation.

establishes, complete, learning, outcome require ### 172. Learning materials and task access: Evidence to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-63]

For learning materials and task access: alignment within learning materials and response evidence, this statement concerns a distinct evidentiary application. Applied to learning materials and response evidence, this statement concerns a distinct evidentiary setting. In this field, the relevant evidence concerns the resources, formats and tasks through which intended content becomes available. A material risk is that official coverage may be recorded although materials are absent, inaccessible, unsuitable or used only briefly. Authorities should examine learner access, sustained use, language, accessibility and continuity. The resulting implication for interpretation should be recorded. The implication should remain bounded to the population, source and decision described here.

The alignment test asks how resources and tasks relate to intended curriculum and assessed construct. Analysts should map content and cognitive demand, recognising that similar labels do not establish substantive alignment. The account should identify population, curriculum domain, period, source and material limitation.

establishes, complete, learning, outcome require ### 173. Learning materials and task access: Alignment to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-66]

For learning materials and task access: equity within learning materials and response evidence, this statement concerns a distinct evidentiary application. Within learning materials and response evidence, the requirement should be read against the specific curriculum evidence. In this field, the relevant evidence concerns the resources, formats and tasks through which intended content becomes available. A material risk is that official coverage may be recorded although materials are absent, inaccessible, unsuitable or used only briefly. Authorities should examine learner access, sustained use, language, accessibility and continuity. Any departure should retain its source and limitation. The implication should remain bounded to the population, source and decision described here.

The equity test asks which learners received weaker access, fewer tasks or less usable feedback. Analysts should analyse disability, language, geography, absence and resource barriers, recognising that equal distribution does not establish equal opportunity to engage. The account should identify population, curriculum domain, period, source and material limitation.

establishes, complete, learning, outcome require ### 174. Learning materials and task access: Equity to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-70]

For learning materials and task access: comparability within learning materials and response evidence, this statement concerns a distinct evidentiary application. For learning materials and response evidence, the same boundary has a separate population and decision consequence. In this field, the relevant evidence concerns the resources, formats and tasks through which intended content becomes available. A material risk is that official coverage may be recorded although materials are absent, inaccessible, unsuitable or used only briefly. Authorities should examine learner access, sustained use, language, accessibility and continuity. The conclusion should remain bounded to the evidence described. The implication should remain bounded to the population, source and decision described here.

The comparability test asks whether definitions, observation periods and collection methods support comparison. Analysts should review metadata before values, recognising that common percentages can describe different thresholds and sources. The account should identify population, curriculum domain, period, source and material limitation.

establishes, complete, learning, outcome require ### 175. Learning materials and task access: Comparability to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-73]

For learning materials and task access: uncertainty within learning materials and response evidence, this statement concerns a distinct evidentiary application. Applied to learning materials and response evidence, this statement concerns a distinct evidentiary setting. In this field, the relevant evidence concerns the resources, formats and tasks through which intended content becomes available. A material risk is that official coverage may be recorded although materials are absent, inaccessible, unsuitable or used only briefly. Authorities should examine learner access, sustained use, language, accessibility and continuity. The resulting implication for interpretation should be recorded. The implication should remain bounded to the population, source and decision described here.

The uncertainty test asks which missing, reported or sampled evidence limits the estimate. Analysts should state likely direction and decision consequence, recognising that precision should not exceed the strength of observation. The account should identify population, curriculum domain, period, source and material limitation.

establishes, complete, learning, outcome require ### 176. Learning materials and task access: Uncertainty to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-77]

For learning materials and task access: interpretation within learning materials and response evidence, this statement concerns a distinct evidentiary application. Within learning materials and response evidence, the requirement should be read against the specific curriculum evidence. In this field, the relevant evidence concerns the resources, formats and tasks through which intended content becomes available. A material risk is that official coverage may be recorded although materials are absent, inaccessible, unsuitable or used only briefly. Authorities should examine learner access, sustained use, language, accessibility and continuity. Any departure should retain its source and limitation. The implication should remain bounded to the population, source and decision described here.

The interpretation test asks which conclusion about teaching or learning is supported. Analysts should separate resource, opportunity, response and outcome, recognising that one stage should not be used as a proxy for the complete learning process. The account should identify population, curriculum domain, period, source and material limitation.

establishes, complete, learning, outcome require ### 177. Learning materials and task access: Interpretation to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-12]

For learning materials and task access: policy use within learning materials and response evidence, this statement concerns a distinct evidentiary application. For learning materials and response evidence, the same boundary has a separate population and decision consequence. In this field, the relevant evidence concerns the resources, formats and tasks through which intended content becomes available. A material risk is that official coverage may be recorded although materials are absent, inaccessible, unsuitable or used only briefly. Authorities should examine learner access, sustained use, language, accessibility and continuity. The conclusion should remain bounded to the evidence described. The implication should remain bounded to the population, source and decision described here.

The policy use test asks which procurement, teaching, support or assessment decision can follow. Analysts should connect the finding with responsible action and review, recognising that additional inputs alone may not correct inaccessible or weakly used provision. The account should identify population, curriculum domain, period, source and material limitation.

establishes, complete, learning, outcome require ### 178. Learning materials and task access: Policy use to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-62]

In this field, the relevant evidence concerns the opportunities learners receive to attempt work, obtain feedback and correct misunderstanding. A material risk is that content presentation may be counted as coverage without evidence that learners engaged or received a usable response. Authorities should distinguish exposure, practice, feedback, revision and demonstrated learning.

implication, interpretation, should, recorded require ### 179. Practice, response and feedback: Definition to distinguish authority, delivery, distribution and remedy. The definition test asks which material, task or feedback event is being counted. Analysts should state the unit, threshold and source, recognising that availability is not use and use is not evidence of learning. The account should identify population, curriculum domain, period, source and material limitation. The resulting implication for interpretation should be recorded.

establishes, complete, learning, outcome require ### 179. Practice, response and feedback: Definition to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-65]

implication, interpretation, should, recorded require ### 180. Practice, response and feedback: Coverage to distinguish authority, delivery, distribution and remedy. In this field, the relevant evidence concerns the opportunities learners receive to attempt work, obtain feedback and correct misunderstanding. A material risk is that content presentation may be counted as coverage without evidence that learners engaged or received a usable response. Authorities should distinguish exposure, practice, feedback, revision and demonstrated learning. The resulting implication for interpretation should be recorded.

implication, interpretation, should, recorded require ### 180. Practice, response and feedback: Coverage to distinguish authority, delivery, distribution and remedy. The coverage test asks which learners, content and period are represented. Analysts should identify missing resources, inaccessible formats and unobserved activity, recognising that unobserved access should not be assigned the average condition. The account should identify population, curriculum domain, period, source and material limitation. The resulting implication for interpretation should be recorded.

establishes, complete, learning, outcome require ### 180. Practice, response and feedback: Coverage to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-69]

Within learning materials and response evidence, the requirement should be read against the specific curriculum evidence. In this field, the relevant evidence concerns the opportunities learners receive to attempt work, obtain feedback and correct misunderstanding. A material risk is that content presentation may be counted as coverage without evidence that learners engaged or received a usable response. Authorities should distinguish exposure, practice, feedback, revision and demonstrated learning. Any departure should retain its source and limitation.

implication, interpretation, should, recorded require ### 181. Practice, response and feedback: Depth to distinguish authority, delivery, distribution and remedy. The depth test asks whether engagement involved mention, explanation, practice, application or cumulative work. Analysts should describe intensity and sequence, recognising that a binary coverage flag can conceal substantial differences in opportunity. The account should identify population, curriculum domain, period, source and material limitation. The resulting implication for interpretation should be recorded.

establishes, complete, learning, outcome require ### 181. Practice, response and feedback: Depth to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-72]

For learning materials and response evidence, the same boundary has a separate population and decision consequence. In this field, the relevant evidence concerns the opportunities learners receive to attempt work, obtain feedback and correct misunderstanding. A material risk is that content presentation may be counted as coverage without evidence that learners engaged or received a usable response. Authorities should distinguish exposure, practice, feedback, revision and demonstrated learning. The conclusion should remain bounded to the evidence described.

implication, interpretation, should, recorded require ### 182. Practice, response and feedback: Evidence to distinguish authority, delivery, distribution and remedy. The evidence test asks which record, observation or learner account supports the finding. Analysts should triangulate sources and retain disagreement, recognising that a distribution list or lesson plan is not proof of learner use. The account should identify population, curriculum domain, period, source and material limitation. The resulting implication for interpretation should be recorded.

establishes, complete, learning, outcome require ### 182. Practice, response and feedback: Evidence to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-76]

For practice, response and feedback: alignment within learning materials and response evidence, this statement concerns a distinct evidentiary application. Applied to learning materials and response evidence, this statement concerns a distinct evidentiary setting. In this field, the relevant evidence concerns the opportunities learners receive to attempt work, obtain feedback and correct misunderstanding. A material risk is that content presentation may be counted as coverage without evidence that learners engaged or received a usable response. Authorities should distinguish exposure, practice, feedback, revision and demonstrated learning. The resulting implication for interpretation should be recorded. The implication should remain bounded to the population, source and decision described here.

implication, interpretation, should, recorded require ### 183. Practice, response and feedback: Alignment to distinguish authority, delivery, distribution and remedy. The alignment test asks how resources and tasks relate to intended curriculum and assessed construct. Analysts should map content and cognitive demand, recognising that similar labels do not establish substantive alignment. The account should identify population, curriculum domain, period, source and material limitation. The resulting implication for interpretation should be recorded.

establishes, complete, learning, outcome require ### 183. Practice, response and feedback: Alignment to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-10]

For practice, response and feedback: equity within learning materials and response evidence, this statement concerns a distinct evidentiary application. Within learning materials and response evidence, the requirement should be read against the specific curriculum evidence. In this field, the relevant evidence concerns the opportunities learners receive to attempt work, obtain feedback and correct misunderstanding. A material risk is that content presentation may be counted as coverage without evidence that learners engaged or received a usable response. Authorities should distinguish exposure, practice, feedback, revision and demonstrated learning. Any departure should retain its source and limitation. The implication should remain bounded to the population, source and decision described here.

implication, interpretation, should, recorded require ### 184. Practice, response and feedback: Equity to distinguish authority, delivery, distribution and remedy. The equity test asks which learners received weaker access, fewer tasks or less usable feedback. Analysts should analyse disability, language, geography, absence and resource barriers, recognising that equal distribution does not establish equal opportunity to engage. The account should identify population, curriculum domain, period, source and material limitation. The resulting implication for interpretation should be recorded.

establishes, complete, learning, outcome require ### 184. Practice, response and feedback: Equity to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-61]

For practice, response and feedback: comparability within learning materials and response evidence, this statement concerns a distinct evidentiary application. For learning materials and response evidence, the same boundary has a separate population and decision consequence. In this field, the relevant evidence concerns the opportunities learners receive to attempt work, obtain feedback and correct misunderstanding. A material risk is that content presentation may be counted as coverage without evidence that learners engaged or received a usable response. Authorities should distinguish exposure, practice, feedback, revision and demonstrated learning. The conclusion should remain bounded to the evidence described. The implication should remain bounded to the population, source and decision described here.

implication, interpretation, should, recorded require ### 185. Practice, response and feedback: Comparability to distinguish authority, delivery, distribution and remedy. The comparability test asks whether definitions, observation periods and collection methods support comparison. Analysts should review metadata before values, recognising that common percentages can describe different thresholds and sources. The account should identify population, curriculum domain, period, source and material limitation. The resulting implication for interpretation should be recorded.

establishes, complete, learning, outcome require ### 185. Practice, response and feedback: Comparability to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-64]

For practice, response and feedback: uncertainty within learning materials and response evidence, this statement concerns a distinct evidentiary application. Applied to learning materials and response evidence, this statement concerns a distinct evidentiary setting. In this field, the relevant evidence concerns the opportunities learners receive to attempt work, obtain feedback and correct misunderstanding. A material risk is that content presentation may be counted as coverage without evidence that learners engaged or received a usable response. Authorities should distinguish exposure, practice, feedback, revision and demonstrated learning. The resulting implication for interpretation should be recorded. The implication should remain bounded to the population, source and decision described here.

implication, interpretation, should, recorded require ### 186. Practice, response and feedback: Uncertainty to distinguish authority, delivery, distribution and remedy. The uncertainty test asks which missing, reported or sampled evidence limits the estimate. Analysts should state likely direction and decision consequence, recognising that precision should not exceed the strength of observation. The account should identify population, curriculum domain, period, source and material limitation. The resulting implication for interpretation should be recorded.

establishes, complete, learning, outcome require ### 186. Practice, response and feedback: Uncertainty to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-67]

For practice, response and feedback: interpretation within learning materials and response evidence, this statement concerns a distinct evidentiary application. Within learning materials and response evidence, the requirement should be read against the specific curriculum evidence. In this field, the relevant evidence concerns the opportunities learners receive to attempt work, obtain feedback and correct misunderstanding. A material risk is that content presentation may be counted as coverage without evidence that learners engaged or received a usable response. Authorities should distinguish exposure, practice, feedback, revision and demonstrated learning. Any departure should retain its source and limitation. The implication should remain bounded to the population, source and decision described here.

implication, interpretation, should, recorded require ### 187. Practice, response and feedback: Interpretation to distinguish authority, delivery, distribution and remedy. The interpretation test asks which conclusion about teaching or learning is supported. Analysts should separate resource, opportunity, response and outcome, recognising that one stage should not be used as a proxy for the complete learning process. The account should identify population, curriculum domain, period, source and material limitation. The resulting implication for interpretation should be recorded.

establishes, complete, learning, outcome require ### 187. Practice, response and feedback: Interpretation to distinguish authority, delivery, distribution and remedy. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome.[REF-71]

For practice, response and feedback: policy use within learning materials and response evidence, this statement concerns a distinct evidentiary application. For learning materials and response evidence, the same boundary has a separate population and decision consequence. In this field, the relevant evidence concerns the opportunities learners receive to attempt work, obtain feedback and correct misunderstanding. A material risk is that content presentation may be counted as coverage without evidence that learners engaged or received a usable response. Authorities should distinguish exposure, practice, feedback, revision and demonstrated learning. The conclusion should remain bounded to the evidence described. The implication should remain bounded to the population, source and decision described here.

implication, interpretation, should, recorded require ### 188. Practice, response and feedback: Policy use to distinguish authority, delivery, distribution and remedy. The policy use test asks which procurement, teaching, support or assessment decision can follow. Analysts should connect the finding with responsible action and review, recognising that additional inputs alone may not correct inaccessible or weakly used provision. The account should identify population, curriculum domain, period, source and material limitation. The resulting implication for interpretation should be recorded.

implication, interpretation, should, recorded require ### 188. Practice, response and feedback: Policy use to distinguish authority, delivery, distribution and remedy. The finding should remain bounded to the stage observed. Resources can create opportunity, tasks can provide evidence of engagement and feedback can support improvement, but none alone establishes the complete learning outcome. The resulting implication for interpretation should be recorded.[REF-75]

Every material claim should identify curriculum, construct, domain, population, period, source, coverage and uncertainty. It should distinguish intended provision, implemented teaching, opportunity to learn and assessed outcome.

Comparison should follow review of constructs, populations, procedures and context. Common metrics support analysis only within the shared evidence they represent and should not erase material curricular difference.

Learning and curriculum evidence serves improvement when its boundaries remain visible. Responsible interpretation resists both dismissal and overstatement: it uses available findings while preserving uncertainty, unequal opportunity and the additional evidence required for explanation and action.