Education Needs More: helping people find a responsible next step Historical research scope. This September 19, 2026 report examines the earlier Education Needs More guidance proposition. The current entities are editorial-first publications: Education Needs More is the editorial umbrella within Education Angel Group, and the individual Need More publications sit beneath it. This report does not evaluate the newer editorial model. The findings, proposed mechanisms and prior tagline below belong to the earlier research scope, not a claim about current product effectiveness or availability. Prepared by Education Angel Group (EAG), with AI-assisted research and drafting. Research date: September 19, 2026. EAG is developing Education Needs More and has an interest in its success. This is a research-led examination of the proposed service, not an independent evaluation, a record of user interviews or evidence of product impact. 1. Proposition, audience and overview In the earlier proposal, Education Needs More was the proposed explanation, listening, publication and resource-guidance hub for education questions that cross roles. A person might know what is troubling them without knowing whether the next conversation belongs with a teacher, a school office, another family member or a specialist. The hub is intended to help clarify that question, explain relevant sources and options, and support a deliberate next step. The earlier proposal's tagline was “Tell us what education is missing. Help shape what comes next.” EAG product authority, A0 The distinction between orientation and service delivery is essential. Education Needs More would explain and route. Parents Need More, Students Need More, Teachers Need More, Schools Need More and Coaches Need More own their respective audience jobs. Runstr owns separately approved identity, supporter and resource behavior. A referral is not a grant, an entitlement, an accepted school purchase or a completed service. The parent pattern is intended to be proved with Teachers Need More before other children follow. This report neither verifies that implementation nor changes that sequence. The research conclusion is conditional: a small, sourced explanation could help when a visitor cannot identify the right question or destination. Another compulsory navigation layer could make an already clear task harder. The proposition is strongest when it helps someone act outside the EAG ecosystem and remains useful without registration, private storytelling or a product conversion. Its value must be demonstrated against direct access to the same official source. The primary audience is therefore a visitor facing an ambiguous education question, rather than every person seeking any educational information. A reader arriving for an article is a second audience. An editor maintaining sources and handling corrections is a third. Their success conditions differ: orientation requires an understood next step; reading requires trustworthy explanation; maintenance requires ownership and correction capacity. Combining these jobs without distinguishing them could produce an attractive publication that fails as guidance. 2. Questions and the standard of evidence The central question is whether a shared orientation layer helps people find appropriate education guidance without adding navigation burden. Five subsidiary questions make that test concrete: 1. Which problems genuinely span roles, and which should go directly to an existing provider? 2. Does guided clarification improve comprehension and destination choice relative to a plain annotated link? 3. Can readers distinguish research findings, official guidance, EAG interpretation and untested product proposals? 4. Does listening change editorial priorities without exposing private stories or implying individual case management? 5. Can EAG keep a narrow guide accurate, accessible and useful at a sustainable maintenance cost? These questions require different evidence. An official page establishes what an institution currently publishes. An experiment tests a specific intervention in its setting. A usability observation could reveal a confusing route without demonstrating learning gains. A maintained correction record could establish operational responsiveness without proving public trust. This report uses original research, official guidance and first-party alternatives. It does not convert the presence of citations into a general effectiveness rating. No retained study evaluates Education Needs More, its sibling network or its proposed listening process. The direct product question remains open. In particular, the existence of an education problem is insufficient evidence that a new intermediary is needed. 3. Need and context: uncertainty can concern the question itself The U.S. Department of Education describes a process for school and district staff that begins with local needs, develops useful search terms, and considers whether evidence fits the local context. Its clearinghouse guidance also warns that ratings from different repositories may not be interchangeable. This is professional guidance, not an experiment demonstrating the benefits of a family-facing hub. U.S. Department of Education, S1 (https://www.ed.gov/teaching-and-administration/lead-and-manage-my-school/state-support-network/ssn-resources/selecting-evidence-based-practices-for-tiers-1-2-and-3-navigating-clearinghouses-and-databases) The relevant inference for EAG is modest: clarification deserves attention before recommendation. For example, “we need more reading help” could mean a family wants an explanation of classroom expectations, a teacher needs instructional evidence, or a school needs to decide how to allocate staff time. Those are illustrative possibilities, not observed visitor categories. A single recommended product would prematurely collapse different questions. Arkansas already supplies practical starting points. Its Academic Playbook family page organizes resources around questions about reading, mathematics, learning readiness and disability, and lists parent guides in English and Spanish. Its special-education family page provides a guide, FAQs, rights materials, contacts and an explanation of ARTASC support. These retrieved pages establish available entry points; this review did not verify each linked document, service response time or a family's eligibility. Arkansas DESE Academic Playbook, S5 (https://dese.ade.arkansas.gov/Offices/special-projects/addressing-unfinished-learning/academic-playbook--family-and-community-engagement), Arkansas DESE family resources, S6 (https://dese.ade.arkansas.gov/Offices/special-education/early-childhood-special-education/parent-and-family-resources) Consequently, EAG cannot credibly begin with “there is nowhere to turn.” A more defensible question is whether a particular visitor can recognize the relevant existing route, understand its purpose and use it without avoidable disclosure. That is a narrower proposition with a fairer comparator. The local demand is unmeasured. We have no interviews showing which cross-role questions Arkansas families bring, no estimate of how often people fail to find these resources, and no evidence that EAG is the intermediary they would trust. Nor does this report estimate the prevalence of information difficulty from national attainment statistics. Such numbers would describe an educational outcome, not the navigation mechanism under examination. 4. Existing alternatives and the possible gap The What Works Clearinghouse describes its practice guides as recommendations for educators informed by research review, practitioner experience and expert panels. Its index includes dated releases and revisions. It is an existing evidence-navigation alternative, especially for professional instructional questions; an individual recommendation still requires its own reading and appraisal. What Works Clearinghouse, S2 (https://ies.ed.gov/ncee/wwc/practiceguides) The comparison below distinguishes documented alternatives from proposed EAG functions. It is not a ranking of effectiveness. Starting situation | Existing or simpler route | What an EAG layer would have to add A teacher knows the instructional topic | WWC topic search and relevant practice guide | Clarification of fit without replacing the guide A family needs an Arkansas learning resource | DESE Academic Playbook | Help recognizing the relevant question and source A family asks about special education | DESE family resources and named support channels | A clear boundary and direct route, without deciding rights or eligibility A visitor already knows the responsible school contact | That contact and the school's current information | Often nothing; offer a direct exit A visitor cannot identify the responsible role | A plain explanation or a conversation with a trusted person | A tested way to narrow the problem without forcing a story Existing sources answer different jobs First-party resource comparison, not a service-quality ranking. Landing pages were checked September 19, 2026; linked services and documents were not exhaustively appraised. These equivalents accompany the reader figures. Source IDs refer to the report references. A0 identifies EAG's product definition reviewed for the original report, not an external study or proof of availability. Other source IDs resolve to this report's reference and access record. Figure 1. Existing sources answer different jobs First-party resource comparison, not a service-quality ranking. Landing pages were checked September 19, 2026; linked services and documents were not exhaustively appraised. School evidence selection | S1: Start from a local need. Official guidance addresses staff selecting evidence that fits their context. Instructional questions | S2: WWC practice guides. Existing recommendations for educators; inspect the particular guide and its evidence. Family learning questions | S5: Arkansas Academic Playbook. Question-oriented resource links and parent guide language options. Special-education questions | S6: Arkansas family resources. Guides, FAQs, contacts and ARTASC information. No eligibility determination. Sources: S1, S2, S5, S6. S1: https://www.ed.gov/teaching-and-administration/lead-and-manage-my-school/state-support-network/ssn-resources/selecting-evidence-based-practices-for-tiers-1-2-and-3-navigating-clearinghouses-and-databases S2: https://ies.ed.gov/ncee/wwc/practiceguides S5: https://dese.ade.arkansas.gov/Offices/special-projects/addressing-unfinished-learning/academic-playbook--family-and-community-engagement S6: https://dese.ade.arkansas.gov/Offices/special-education/early-childhood-special-education/parent-and-family-resources Figure 1. First-party source comparison, not a service-quality ranking. The proposed hub must add comprehension or routing value beyond these entry points. Text and data alternative (#figure-text-and-data-alternatives). Several competing approaches remain credible. A maintained article with a short “where to go next” paragraph might be enough. A small issue index could let visitors scan across audiences without stepping through questions. A guided sequence might help someone who cannot yet phrase the issue. A human conversation may be preferable when the question is sensitive, contested or unusually specific. The costs are different. Articles require editorial upkeep but little interaction infrastructure. An index can be easy to scan, although labels may assume knowledge the visitor lacks. A guided sequence can explain distinctions, but every step creates another opportunity for misunderstanding or abandonment. Human help can interpret context, but requires actual staffing and a clear service promise. No comparative trial among these approaches was identified for this EAG use case. The proposed gap is therefore not “more resources.” It is an observable failure to connect a person's question with a suitable existing action. EAG should first establish where that failure occurs. If visitors already choose the correct official destination, a new hub may be unnecessary for that task. 5. Evidence, counterevidence and transfer limits Verification behavior is different from correct judgment Breakstone and colleagues studied four one-hour source-evaluation modules embedded in two sections of a university nutrition course in summer 2020. Among 87 completers, lateral reading on one task rose from 3 students before instruction to 67 after; correct credibility challenges rose from 2 to 36. The study had counterbalanced assessments but no untreated comparison group. It did not establish durability or everyday transfer. Breakstone et al., S3 (https://misinforeview.hks.harvard.edu/article/lateral-reading-college-students-learn-to-critically-evaluate-internet-sources-in-an-online-course/) Checking sources and judging credibility differ Task 3 counts among 87 university-course completers in summer 2020. Before and after four modules; no untreated comparison. This is not an EAG effect estimate. Figure 2. Checking sources and judging credibility differ Task 3 counts among 87 university-course completers in summer 2020. Before and after four modules; no untreated comparison. This is not an EAG effect estimate. Measure | Before instruction | After instruction Read laterally | 3 students | 67 students Correct credibility challenge | 2 students | 36 students 87 undergraduate completers across two sections of one university nutrition course; each count uses the same 87 participants. Descriptive counts; no intervals displayed. No untreated group. Strategy use and correct judgment overlap and must not be added. Source counts unchanged; shared zero-based 0-87 scale. Sources: S3. S3: https://misinforeview.hks.harvard.edu/article/lateral-reading-college-students-learn-to-critically-evaluate-internet-sources-in-an-online-course/ Figure 2. Observed counts among 87 university-course completers in 2020, before and after instruction. Different assessment forms were counterbalanced. These are descriptive counts, not a controlled product-effect estimate or a forecast for families. Text and data alternative (#figure-text-and-data-alternatives). For EAG, the distinction suggests an evaluation choice: measure what visitors conclude, rather than only whether they open a citation. A source panel could make the author and evidence visible but still leave a mistaken impression. A visitor who confidently attributes EAG's opinion to a government agency has not been well served, even if every link works. This is a proposed design response, not a finding from the university study. Information can matter when a usable option exists The NBER abstract for Hastings and Weinstein's two school-choice experiments reports that information increased selection of higher-performing schools and that nearby alternatives mattered. The retained version is Working Paper 13623, revised February 2008. A PDF endpoint initially returned indexed text but subsequent retrieval failed; this report uses the institutional abstract only and retains no numerical effect. The result concerns a specific school-choice program, not broad education guidance. Hastings and Weinstein, S7 (https://www.nber.org/papers/w13623) The proposed hub should distinguish knowing the next step from being able to take it. A destination may require an appointment, an institutional decision or a service that is unavailable locally. A polished explanation cannot remove those constraints. An honest “this source explains the process; it cannot confirm availability” can be more useful than an apparently decisive recommendation. Collective action differs from a self-service page Pandey's 2023 paper reports a cluster randomized trial conducted in 2006-2009 across 610 local-government village clusters in three Indian states. Repeated public meetings explained community roles in school management. Teacher effort and mathematics outcomes improved in some settings, with narrower language gains and substantial contextual differences. The discussion describes limits from unequal participation, authority and teaching capacity. The intervention involved organized, repeated community activity; it was not a test of posting information on a website. Pandey, S4 (https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0280803) For EAG, this is reason to avoid treating a listening box as participation itself. A submission might inform an editorial decision, but it does not necessarily give the contributor influence over a school policy. The hub should say which decisions it can change and which institution retains authority. It should also distinguish acknowledging a question from promising a personal response. Fewer choices are not automatically better The publisher abstract of Scheibehenne, Greifeneder and Todd's 2010 meta-analysis reports a near-zero mean choice-overload effect across 63 conditions from 50 experiments, with substantial variation. Full-text retrieval was unsuccessful; detailed moderator claims are not retained here. Consumer-choice experiments also differ from navigating education services. The abstract nevertheless contradicts a blanket claim that reducing options always improves decisions. Scheibehenne et al., S8 (https://doi.org/10.1086/651235) A short menu may hide the only appropriate route. A longer menu may expose useful distinctions. The useful design question is whether a visitor can understand and compare the options relevant to their task. Counting menu items alone is a poor substitute for observing that decision. Information evidence has different boundaries Comparison of research settings and access limits. None evaluates Education Needs More. No pooled effects or product forecast. Figure 3. Information evidence has different boundaries Comparison of research settings and access limits. None evaluates Education Needs More. No pooled effects or product forecast. School choice | S7: Options must be usable. Historical experiments, institutional abstract only. Not a broad guidance-hub evaluation. Community accountability | S4: Meetings and formal roles. India trial, 2006-2009. Organized repeated activity differs from a self-service page. Choice overload | S8: No universal menu rule. Publisher abstract only. Average result hides variation; consumer decisions differ from education navigation. Sources: S4, S7, S8. S4: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0280803 S7: https://www.nber.org/papers/w13623 S8: https://doi.org/10.1086/651235 Figure 3. Research comparison with access and transfer limits. None tests Education Needs More, and no effects are pooled across settings. Text and data alternative (#figure-text-and-data-alternatives). 6. Proposed mechanism and where it could fail Separate visitor and editorial decisions EAG conceptual model only. Proposed links are untested; reading and offline action require no account. Figure 4. Separate visitor and editorial decisions EAG conceptual model only. Proposed links are untested; reading and offline action require no account. Visitor path. Recognize the question → Read bounded explanation → Compare source and destination → Choose action, pause or exit. Unknown: understanding, burden and destination usefulness. Editorial path. Receive minimal optional feedback → Check sources and publication permission → Revise or withdraw explanation. A private submission is not permission to publish a story. Destination boundary. Name the receiving owner → Visitor chooses whether to proceed → Receiving service applies its own rules. No default private-data transfer, entitlement, school acceptance or completed service. Sources: A0. A0. EAG product definition: /projects/education-needs-more Figure 4. Conceptual model only. Comprehension, destination usefulness and editorial response remain untested. A private interaction is not public content, and a referral does not grant permission or entitlement. Text and data alternative (#figure-text-and-data-alternatives). The proposed mechanism is an explicit hypothesis: a visitor recognizes a problem, clarifies the question, reads a bounded explanation, compares suitable sources or options, and chooses an action they understand. The action may be an official resource, a conversation, a sibling service, a pause or a decision that no listed route fits. All of these can be legitimate outcomes. The first causal assumption is that uncertainty concerns orientation. If the visitor already knows what to do but lacks time, money or institutional permission, the hub may contribute little. The second is that the explanation improves understanding enough to justify its reading burden. The third is that the destination can actually address the question. A broken or mismatched destination invalidates the route even if the preceding explanation is excellent. A separate editorial loop could turn recurring, low-sensitivity questions into better explanations. That requires a responsible editor to assess the question, check sources, decide what can be published and correct errors. The loop does not require publishing the original submission. An identifying story should never become an article simply because it arrived through a listening surface. The practical value would lie in the explanation between the question and the destination. For instance, a hypothetical family question about learning support could receive a short account of which source explains general expectations and which contact can discuss a particular child's situation. The page would not need the child's name, records or diagnosis to make that distinction. This example is an author-created scenario, not advice about a real case or a demonstrated product feature. 7. Design concepts worth testing These are research-derived concepts for later design consideration, not approved interface specifications. A question-first explainer. Start with a recognizable situation and state what the page can help clarify. Let visitors change audience without losing their place. An experienced teacher should not have to pass through a family introduction, while a visitor who does not understand institutional terminology should not have to guess a specialist label. An annotated destination. Each proposed next step would identify its owner, explain why it is relevant, name what it cannot decide, and show when EAG last checked it. The source's publication date and EAG's check date answer different questions. A page that still loads can contain outdated guidance; a recent check date alone cannot establish substantive currency. A short preparation aid. The visitor could leave with a question to ask or a small list of facts to check privately. The aid should be useful without sending information to EAG. For the first research prototype, a readable paragraph or printable note may be enough. Saving a personal case history would add a different purpose and risk. A bounded listening option. A person might select “the source was unclear,” “the destination did not fit,” or “this topic is missing.” Even these categories are proposals requiring testing. Free text should not become mandatory merely because structured options are imperfect. If a future submission route exists, its audience, response expectations and publication boundary would need to be clear before input. Predictable recovery. W3C's explanation of consistent help says repeated help mechanisms should occur in the same relative order across pages within a set. It does not require human support to exist or be constantly available. This informs a predictable help and correction location; it is not a conformance assessment of an unbuilt hub. W3C WAI, S9 (https://www.w3.org/WAI/WCAG22/Understanding/consistent-help.html) The first comparison should be small enough to interpret: the same issue and official source presented as a direct annotated link, a short explainer and a guided sequence. If all three contain different information, a favorable result cannot tell us whether navigation or content caused it. Keep the core information comparable while allowing each format to perform its intended job. 8. Risks, equity and the burden of maintaining trust The largest credibility risk is appearing to be a neutral public authority while routing toward EAG products. The report's authorship disclosure is therefore part of its substance. A future guide should let readers distinguish an official source from EAG's explanation and should not give an EAG destination preference merely because EAG owns it. A successful external exit should count as useful service. Privacy risk begins before an account is created. A question about a named child can disclose educational, family or health circumstances. The canonical proposition allows anonymous reading and does not require private data to cross sites. The research recommendation is to test whether a useful first step can be achieved using issue and audience choices alone. An optional field does not make collection necessary. Language and disability access affect the whole route. Translating the hub while sending a visitor to an unusable destination may only move the obstacle. EAG should record the destination's relevant formats and test understanding with the people the guide intends to serve. The Arkansas pages' English and Spanish links are useful evidence of options, but this review did not appraise the linked translations. No language coverage or accessibility compliance claim follows from their presence. Listening also creates emotional and operational expectations. A person who describes a difficult situation may reasonably expect someone to respond. A future service must make its actual response capacity clear. If EAG can only collect suggestions for future articles, the language should say so. It should not resemble counseling, urgent support or an individual complaint-resolution service unless those functions are separately established. Maintenance may be the constraint that determines the appropriate scale. Each additional issue adds sources, exceptions and destination relationships to check. A narrow, maintained explanation can be more dependable than a broad set of pages whose owners are unclear. Broken-link detection would help find one class of failure, but it would not detect changed eligibility rules, superseded guidance or a once-useful route that no longer answers the question. Finally, publication can distort listening. A vivid account is not necessarily representative; recurring submissions can reflect who has time and confidence to contribute. EAG should not present anecdotal frequency as population prevalence. Editorial selection should explain why a question merits coverage without inventing consensus, fictional contributors or an independent editorial identity. 9. Evaluation and criteria for changing course Begin with one cross-role problem and one verified official source. The question should be selected through appropriate later research or an explicit product decision; this report does not invent an interview result to choose it. Include a direct-source route as the comparator. Recruiting participants, collecting feedback and processing personal information remain separate activities from this literature review. The initial evaluation should ask people to explain the next step in their own words. Can they identify who owns the destination, what it can help with, and what remains uncertain? Did they mistake the page for official advice? A correct click accompanied by a false expectation should count as a failure of explanation. Proposed measure | What it answers | What would prompt revision Correct destination and reason | Did the visitor understand the route? | Misrouting or unexplained selections Appropriate next-step description | Can the visitor act without a false promise? | Assumed eligibility, funding or guaranteed response Time and perceived effort versus direct source | Does orientation justify its added burden? | More effort with no comprehension benefit Recovery from a stale or unmatched source | Can the visitor stop or change course safely? | Dead ends, forced disclosure or invented recommendations Editor time and correction completion | Can the guide remain trustworthy? | Unowned pages or unresolved material corrections Differences by access needs and familiarity | Who benefits and who struggles? | Consistent disadvantage hidden by the average These are proposed criteria, not numerical results or validated thresholds. A small formative study can expose failures but cannot estimate population-wide effects. Later comparative evaluation should define its sampling, scoring and analysis before results are known, report uncertainty and distinguish people who already understand the system from those who do not. Continuation should depend on a recognizable benefit: improved comprehension, a more appropriate action or less avoidable effort without new privacy costs. If the direct source performs as well, publish the short annotation and avoid adding a compulsory sequence. If only one narrow audience benefits, narrow the hub's claims. If visitors cannot distinguish the hub from its destinations, reconsider the family architecture rather than adding more explanatory pages indefinitely. Broader outcomes need separate evidence. Neither an understood referral nor a completed visit proves improved attainment, stronger institutions, resolved resource needs or family wellbeing. Those may matter in the long term, but attributing them would require a study of the relevant service and causal pathway. The hub should first prove that it helps people make a responsible next move. 10. Methods, limitations, references and corrections This was a targeted narrative review conducted on September 19, 2026. Searches covered education information interventions, source evaluation, choice overload, official evidence navigation and Arkansas family resources. The EAG product definition determined the product scope. Original research and institutional sources were preferred; secondary summaries served only as discovery leads. This was not a registered systematic review or exhaustive search, and it does not establish that no other relevant evidence exists. Full HTML was read for S1-S4 and S9. Arkansas S5-S6 were recovered through direct HTTP after browser retrieval errors; only the landing-page content was appraised. S7 is limited to the institutional abstract because reliable full-PDF appraisal failed. S8 is limited to the publisher abstract; its author-hosted PDF failed retrieval. The report does not retain effect sizes or detailed methods from those abstracts. The largest missing evidence concerns local demand, actual comprehension, service fit, language and disability needs, trust, editorial burden and whether an EAG intermediary improves upon the existing route. No interviews, original trials, product telemetry or independent expert review were performed. Historical experiments are labeled by their intervention period where verified; their publication dates do not make them current Arkansas observations. Reference and access record - A0. Education Angel Group, Education Needs More product definition, reviewed September 19, 2026 for this report. Defines intended scope, not implementation or effectiveness. - S1. U.S. Department of Education, Selecting Evidence-Based Practices for Tiers 1, 2, and 3: Navigating Clearinghouses and Databases, page reviewed January 14, 2025. Process table and database caveat. Full HTML; technical legal tier definitions are outside this report's use. - S2. Institute of Education Sciences, What Works Clearinghouse Practice Guides. Introductory description and dated index, accessed September 19, 2026. First-party capability record; individual guide contents not appraised here. - S3. Breakstone, Smith, Connors, Ortega, Kerr and Wineburg (2021), Lateral reading: College students learn to critically evaluate internet sources in an online course. DOI 10.37016/mr-2020-56. Findings 2, Methods and limitations; full article HTML. - S4. Pandey (2023), Does information improve service delivery? A randomized trial in education in India. DOI 10.1371/journal.pone.0280803. Methods, results and discussion; full article HTML, supplementary datasets not reanalyzed. - S5. Arkansas DESE, Academic Playbook: Family and Community Engagement. Parent question list and resource-guide links, accessed September 19, 2026. Page date not established; linked documents not individually verified. - S6. Arkansas DESE, Parent & Family Resources. Family resources and ARTASC description, accessed September 19, 2026. Page date not established; listed support availability not independently verified. - S7. Hastings and Weinstein, Information, School Choice, and Academic Achievement: Evidence from Two Experiments, NBER Working Paper 13623, November 2007, revised February 2008. DOI 10.3386/w13623. Institutional abstract only for retained claims. - S8. Scheibehenne, Greifeneder and Todd (2010), Can There Ever Be Too Many Options? A Meta-Analytic Review of Choice Overload. DOI 10.1086/651235. Publisher abstract only for retained claims. - S9. W3C WAI, Understanding Success Criterion 3.2.6: Consistent Help, updated March 9, 2026. Intent and limitations; informative explanation, not a product audit. The accessed records were checked for displayed revision or correction information, but this review did not run a comprehensive retraction-database search. A correction should identify the claim and source, explain the change and date the revised account. If a destination changes, review both the link and the explanation that justified it. Preserve uncertainty until the replacement evidence has been appraised. This report does not establish public release or independently verify correction handling. Figure text and data alternatives Complete semantic alternatives accompany each figure in this reader and its text download. Source IDs refer to the references in this report. A0 identifies EAG's product definition reviewed for the original report, not an external study or proof of availability. Other source IDs resolve to this report's reference and access record. Corrections and entity context Readers identifying an error can contact EAG (/contact) with the section, disputed statement and supporting source. Education Needs More (/projects/education-needs-more) provides the public entity context, not an external citation for EAG's product claims.