Test Security and the Examiner's Own Obligations
You have a test kit in a cupboard in a room in your house. Nobody is going to inspect it. There is no district assessment coordinator, no secure testing suite, no colleague to sign protocols in and out. Test security and the examiner's own obligations are, for most educational therapists, a set of rules written for somebody else's building.
That is not a complaint about the rules. It is a description of them. We read the current user agreements and security policies from four assessment publishers, the ethics code that binds educational therapists, the APA's own standard, and the position papers the neuropsychology bodies have written on disclosure. The rules are real, they are specific in places, and they are almost entirely addressed to institutions.
The short version
Test security is two separate obligations and they are usually confused. One protects the client's information. The other protects the publisher's materials. They have different sources, different rules, and different consequences.
The AET Code of Ethics regulates the first in considerable detail and does not mention the second at all. There is no clause on test security in it.
Three of the four publishers we read say the same thing about copies and disagree about who supervises a viewing. Pearson's rule names a representative of the school, college or institution. A solo practice is none of those.
FERPA appears in three publishers' policies, and each time it appears in order to say what the statute does not require. We report that and nothing more.
The only distinction that actually does any work was drawn by a neuropsychology body in 2003, not by a publisher and not by a code of ethics.
Nobody specifies a lock. The governing phrase, across every source we read, is "reasonable efforts."
The newest clause in two of these agreements is about artificial intelligence, and it is written broadly enough to cover ordinary professional curiosity.
What this post is not
Nothing here is legal advice. We are not lawyers and this is not a review of any statute. Where a question is about what the law entitles a parent to receive, what a subpoena obliges you to produce, or what your records-retention duty is in your state, that question belongs with an attorney licensed where you practise, and we say so again at the end.
What we can do is report, with links, what each publisher's own agreement says and what each professional body's own code says. Every requirement below is attributed to the body that issued it. Where a source could not be read, we say so rather than filling the gap.
Report what the publisher wrote and what the code says. Route the entitlement question to counsel. Those are two different jobs and only one of them is ours.
Test security is two obligations, and your code of ethics covers one of them
The phrase "test security" gets used for two things that have almost nothing in common.
The first is confidentiality of the client's information — scores, responses, reports, the fact of the evaluation at all. Its source is your professional code, your consent agreement, and privacy law. The harm from a breach lands on the client.
The second is security of the publisher's materials — items, stimulus books, protocols, scoring keys, manuals. Its source is a contract you signed when you bought the kit and a copyright claim that sits underneath it. The harm from a breach lands on the instrument, and through it on every future examinee.
These two obligations can pull in opposite directions in the same meeting. A parent who asks to see what their child was asked is making a request the first obligation would ordinarily favour and the second obligation restricts.
What the AET Code actually says, clause by clause
The AET Code of Ethics and Standards for the Professional Practice of Educational Therapy is Revision #9, 2025, reissued September 2025. On the client-information side it is detailed and current.
Section Two, I.I — educational therapists "maintain confidentiality of information except where information is released under specific conditions," those conditions being written consent or a legal requirement.
Section Two, I.J — they "establish and maintain confidentiality policies and procedures consistent with relevant statutes."
Section Two, I.K — they "provide adequate security and security controls for client information and data within information systems."
Section Two, I.K.1 — adequate security, the Code says, citing the Committee on National Security Systems, "is commensurate with the risk" and the magnitude of potential harm.
Section Two, I.L — they "recognize the possibility that any electronic communication can have a high risk of public discovery."
Section Two, I.M.2 — cloud storage must "adhere to privacy and security standards in compliance with HIPAA regulations and other relevant federal" rules.
Section Two, I.M.3 — after a breach of unencrypted data they "notify clients and other appropriate individuals/organizations as soon as possible."
Section Two, I.M.4 — they "follow current environmental protection guidelines and relevant statutes and regulations related to record retention" and disposal.
That is a serious set of obligations about data. It maps cleanly onto the consent conversation and onto your file system. If you have already worked through what an informed consent document has to carry and what a defensible session note looks like, these clauses will be familiar.
The clause that is missing, and why it is missing
There is no clause in the AET Code about protecting test content. Not about restricting access to instruments, not about storage of a kit, not about what may be shown to a parent, not about copying a protocol. We searched the current revision for it.
The nearest two clauses are about competence and about use, not about security:
Section One, II.D.2 — educational therapists "use published materials ethically and only for the purposes intended."
Section Two, I.C — they "use only those assessment instruments for which they have been adequately trained."
Section One, II.D.2 is the clause doing the work, and it is doing it by implication. Using a published instrument "only for the purposes intended" is a reasonable place to hang a security duty, but the Code does not hang one there in terms, and an implied duty is a weak thing to rely on when a parent is sitting across from you asking a direct question.
This is not an accusation. It is the same structural observation we made about what the four-item "DO NOT" list everybody quotes actually binds and about how six bodies define a supervised hour six different ways. A code can be detailed in one area and silent in the one next to it, and the silence is worth knowing about before somebody asks.
The code that governs the person most likely to be holding a test kit in a spare room regulates the client's data in eight clauses and the publisher's materials in none.
Pearson's six principles, and the one that presumes an institution
Pearson's Terms and Conditions of Sale and Use carries a numbered section 7 headed "Test security." It opens: "Customer agrees to comply with the following basic principles of minimum test security." The six bullets, verbatim:
"Test takers must not be coached using the test or receive test answers before beginning the test"
"Test materials (text, graphic images, or the oral reading of items) may not be displayed, reproduced, or performed (for example, filming an administration) in any manner, electronically or otherwise, including posting on any mass media or social media site, without the prior written permission of Pearson"
"Access to test materials must be limited to qualified persons with a responsible, professional interest who agree to safeguard their use"
"Test materials and scores may be released only to persons qualified to interpret and use them properly"
"Test materials must not be resold, re-licensed, transferred, or otherwise redistributed for any purpose without prior written permission from Pearson"
And the sixth, which is the one worth reading twice:
If a test taker or the parent of a minor child who has taken a test wishes to examine test responses or results, the parent or test taker may be permitted to review the test and the test answers in the presence of a representative of the school, college, or institution that administered the test
A private educational therapy practice is not a school, a college, or an institution. Pearson's only provision for a family reviewing a protocol is written in terms that do not describe a solo practitioner. Read strictly, the clause does not forbid the viewing and does not authorise it either; it describes a setting you are not in.
Two further Pearson clauses matter. Section 6 states that software outputs "including but not limited to reports, are protected as trade secrets," and that trade secrets "are exempt from disclosure in response to requests made pursuant to HIPAA or to any other data disclosure law that exempts disclosure of information or documents protected as trade secrets."
Section 13 states that purchase "does not grant Customer a right to reproduce additional copies of materials or content." It then allows you to excerpt "portions of output reports, limited to the minimum text necessary to accurately describe the Qualified User's significant core conclusions" in a written evaluation.
That excerpt permission is narrow and specific and it is the clause most often exceeded in practice. It covers minimum necessary text from a software-generated report. It does not cover reproducing a subtest, an item, or a scoring table.
PRO-ED: no copies to parents, and FERPA named to say what it does not require
PRO-ED's Terms of Use, effective 6 October 2025, carry a section headed "Test Security and Nondisclosure." It is blunter than Pearson's and it addresses parents directly:
For Schools, PRO-ED does not permit the copying or disclosure of its tests including any items, questions, subtests, or other components, to any parents, guardians or students.
Then, in the same paragraph, the concession:
We understand that schools may need to provide a student or their parent or guardian responses to their test and that there may be a need to allow a parent/guardian/student to view a specific test item; however, this information should not be copied.
Note the construction. The prohibition is categorical and the concession is narrow: one item, viewed, not copied. And note the addressee again — "For Schools." PRO-ED's security section, like Pearson's family-review bullet, is written to an institution.
PRO-ED also states that it "does NOT allow the sharing of its test products with persons who are not involved in diagnostic testing," and that resale is not permitted on consumer platforms "in which the buyer is not vetted under PRO-ED's test qualification standards."
On copying it goes further than the others. The copying of its tests, PRO-ED says, "does not fall under the 'fair use' exception of copyright law because, among other reasons, copying has a significant detrimental effect on the potential market for the copyrighted works."
If you are unclear which of your instruments PRO-ED would even sell you, that is a separate question with a separate answer, and we have mapped the qualification letters across six publishers.
Riverside: access is not copying
Riverside Insights' Test Disclosure Policy, which sits alongside its general terms of use, draws the line in one phrase. FERPA grants parents the right to inspect and review their children's educational records, and the policy's position is that this right "does not include the right to make copies of the materials provided."
If you want the statute's own framing rather than a publisher's, the Department of Education publishes a parent guide to FERPA. Treat it as a starting point for a conversation with an attorney, not as a substitute for one.
From that it follows, in the policy's own words, that there is "no obligation to provide copies or to permit the parent/legal guardian to make copies," and that the access right is satisfied where "the test protocols are made available to the parent/legal guardian or a representative thereof for review."
Riverside's provision is the most usable of the four for a private practitioner, because it names a representative of the parent rather than a representative of an institution. It does not say the review must happen on your premises, and it does not say who must be present.
Riverside also states plainly that it does not consent to copying test materials or distributing them to people not professionally qualified to obtain them, and that non-professional note-taking and audio or video recording during test administrations are not permitted.
PAR: the only publisher that names a court going the other way
PAR's position on photocopying test materials and its intellectual property position reach the same destination as the others by a more careful route, and they are the only publisher documents we read that acknowledge a contrary legal outcome.
PAR "strongly opposes the release of copies of protocols" and states that FERPA gives parents a right to inspect and review but that there is "nothing in the statute that requires the release of copies of test questions or protocols." It notes that professionals may show protocols or items to parents during discussion.
And then it concedes the other side. PAR acknowledges a 2005 federal ruling in California that supported the right of parents to receive a copy of test items and protocols. It does not agree with that outcome, but it reports it.
PAR also offers the one piece of practical machinery in the set: where school counsel recommends releasing copies, "the school should obtain a confidentiality agreement signed by the parents."
A signed confidentiality agreement is a device a private practitioner could actually use, and PAR is the only publisher we read that names it. We are reporting it as a thing PAR recommends to schools. Whether it would protect you is a question for your own attorney.
Four publishers, four answers to one question
Put the question plainly — may a parent see the protocol? — and the four documents give four different shapes of answer.
Pearson — yes, in the presence of a representative of the school, college or institution that administered the test. No copies.
PRO-ED — no disclosure of items to parents, with a narrow allowance for viewing a specific item. No copies.
Riverside — yes, made available to the parent or a representative thereof for review. No copies, and no obligation to provide them.
PAR — items and protocols may be shown during discussion. No copies released. A confidentiality agreement is suggested where counsel advises release.
All four agree about copies. None of them agrees with another about who supervises the viewing. Three of the four write the rule to a school. One writes it to a parent's representative. This is the same finding we reached about grade equivalents and about what the clearinghouse bubbles require: the practitioner is asked to follow a single rule that does not exist as a single rule.
The document you actually signed does not mention security
Here is the part that is easy to miss. The security obligations above live in terms of sale and in policy pages. The form you filled in to be allowed to buy the kit is a different document, and at least one publisher's version of it says nothing about security at all.
PRO-ED's User Acceptance Form is the document a purchaser signs to qualify at Level B or C. Its certifications are these:
"I am qualified to properly use any test and assessment products I order."
"Any test and assessment products purchased under my account will be used by me and/or under my supervision."
"Any test and assessment products purchased under my account will be used in accordance with all applicable legal and ethical guidelines."
"I agree I will not resell any test or assessment products."
"I have read and hereby agree to PRO-ED/LinguiSystems's Qualification Terms and Conditions of Sale and Use."
Qualification, supervision, lawful use, no resale, and a reference to another document. No storage undertaking. No non-disclosure clause. No statement about what may be shown to a parent. The security duty exists, but it reaches you by incorporation rather than by anything you read and initialled.
The form's own level definitions are worth noticing too. Level B names "a master's degree in psychology, education, occupational therapy, social work, counseling" with formal assessment training, or "certification by, or full active membership in, a professional organization" requiring relevant training. Level C names a doctorate, state licensure, or membership of an organisation "such as APA, NASP, NAN, or INS."
At Level C the qualifying memberships named are psychology bodies. That is a reasonable thing for a publisher to decide, and it is also the first hint of a tension that gets explicit further down this page.
The pattern repeats at a publisher whose programs educational therapists buy constantly. Gander Publishing's qualifications policy, which governs the Lindamood-Bell instruments, accepts orders "from individuals when a User Acceptance Form has been submitted and accepted," sets Level B at a master's degree or professional-organisation membership, and says that for Level A "there are no special qualifications to purchase these products."
It does not address test security, confidentiality or non-disclosure anywhere. We looked. A practitioner can qualify, buy, and administer without ever encountering a sentence about protecting the content — which is the cleanest illustration of the gap this post is about.
The distinction that does the work came from neuropsychology, not from a publisher
The useful line in this whole area is between test data and test materials, and the clearest published statement of it is in the National Academy of Neuropsychology's test security update, approved by its Board on 13 October 2003. The original position was approved 5 October 1999 and published in Archives of Clinical Neuropsychology, volume 15, number 5, pages 383 to 386.
The two definitions, as NAN states them:
Test data refers to raw and scaled scores, client/patient responses to test questions or stimuli.
Test materials refers to manuals, instruments, protocols, and test questions or stimuli and does not include test data.
The refinement is the part that matters in practice. NAN notes that data and materials overlap, so releasing data can amount to releasing materials, and that data which reveal or easily imply the test questions should be treated as materials. Conversely, raw or calculated scores "that do not reveal test questions" do not require the same protection.
That gives you a workable test for a request you are about to answer. Does what I am about to hand over tell the reader what was asked? If it does, it is functionally test material whatever the file is called. If it does not, it is a score.
NAN's recommendations also include offering to send the material to another qualified professional once that professional commits to protecting it, and treating embedded or inferred questions conservatively.
The APA standard is one sentence and specifies nothing
The APA Ethics Code's assessment section contains the standard everybody cites. As reproduced on the APA's own FAQ on maintaining test security in the age of technology, last updated July 2026, Standard 9.11 reads that psychologists "make reasonable efforts to maintain the integrity and security of test materials and other assessment techniques consistent with law and contractual obligations in a manner that permits adherence to this Ethics Code."
"Reasonable efforts" is the entire specification. There is no list, no threshold, no inventory requirement. The companion standard, 9.04, supplies the data definition: those portions of test materials that include client responses are included in the definition of test data.
Two more APA documents are worth knowing. The Statement on the Use of Secure Psychological Tests in Education says access "should be granted only to qualified psychologists or other professionals," and that departments "must store all psychological test materials under conditions that prevent access by unauthorized individuals."
It also says that unauthorised modification of a test for a research project violates copyright. It is marked revised 1994 with a page creation date of 2015, and it is addressed to universities rather than to practices.
The APA resolution on protecting psychological test security, dated February 2025, is the most recent statement in the area. It defines test security as protection of the content and key operational characteristics of a test, and states that coaching "decreases the validity of test results, contaminates test data, and compromises public safety."
It also resolves that APA "supports state and federal statutes, regulations, and policies that enhance test security." We report the resolution. We do not characterise any statute it refers to.
The APA's FAQ on disclosure of test data and test materials, dated 11 April 2019, adds the practical warning: disclosing secure materials to unqualified persons may "render the test invalid for any future use with that individual." That is a measurement consequence, not just a contractual one, and it connects directly to what the manuals permit about re-testing intervals and practice effects.
Nobody names a lock
Here is a gap you can verify yourself in about ten minutes. Not one of the four publisher agreements we read contains a storage clause for a physical test kit. Pearson's Terms have no storage provision. PRO-ED's security section addresses copying, disclosure and resale and not storage. Riverside and PAR address disclosure.
The nearest specifications come from outside the agreements. The APA FAQ advises "using encryption or passcodes when storing data," says test materials "are best maintained separately from patient records," and cites the International Test Commission guidelines, table 1.4.1: "Ensure secure storage of and control access to test materials."
The Iowa Testing Programs' guidance, which is written for a district rather than a practice, gets closer to a concrete rule than any publisher does: test booklets "should not be readily accessible to teachers, other district staff members, or students other than during actual test administration sessions," and material retained for later use needs to be "stored securely with limited access."
There is a great deal of published writing about who may buy a test and almost none about where it should sleep.
What follows is not that storage does not matter. It is that you are setting your own threshold, and you should be able to say what it is.
A lockable drawer, a kit that does not travel in a car boot, and a rule about who else in the household can open which cupboard is more than most of these documents ask for. It is also easy to describe to a parent who asks.
Recording and remote administration have the most specific rules in the whole area
This is the one corner where a publisher has written genuinely detailed guidance. Pearson's page on appropriate use of its clinical assessment content covers recording and screen capture item by item.
Permission for recording "must be established between the clinician and the examinee/caregiver."
Clinical recording "is allowed only for scoring purposes by qualified users as outlined in the test manual," and recordings should be permanently deleted once scored.
Completed response pages or written responses "may be screen captured on secure, organization-authorized devices for scoring purposes only," then deleted after scoring.
Professionals may not save, store or share recordings or captures containing test content anywhere unqualified users can reach them — and this applies "even if the examinee/caregiver has given approval."
Parents and caregivers may not record test content in any form, and the guidance is not to "Ask a parent, caregiver, or untrained/non-professional facilitator to participate in the item response," recommending a facilitator who is "not a family member or caregiver."
Recordings and captures may not be posted on "social media or any public-facing location (e.g., YouTube)."
The clause that will catch people is the consent one. Caregiver approval does not unlock storage of test content, because the obligation being protected is not the family's. Riverside's parallel rule is shorter and points the same way: non-professional note-taking and recording during administrations are not permitted.
If you are administering anything remotely, read these alongside what the publishers actually authorise for remote administration, because the two sets of rules are issued by the same companies and are not always in the same document.
The newest clause in every agreement is about artificial intelligence
Both publishers whose full terms we could read have added one, and they are broad.
Pearson, section 6, verbatim:
No part of the Products, including but not limited to any test content and materials, may be collected, used or reproduced in any manner or for any purpose in connection with the development, deployment, or use of artificial intelligence technologies or similar technologies.
PRO-ED's Terms prohibit "any use the Site or any content for the development of any software program, model, algorithm, or generative AI tool."
Read the Pearson clause carefully, because "use" is in it alongside "development." The clause is not limited to training a model. A practitioner pasting item text into a general-purpose assistant to ask how to score something is using test content in connection with the use of an artificial intelligence technology, on the plain words.
The APA's July 2026 FAQ reaches the same subject from the security side, noting that "Generative AI and related tools can introduce new test-security breaches" and that programs should consider defining prohibited AI-assisted responding. It also lists what counts as secure: live item content, complete test forms, scoring keys, and access credentials.
This is the fastest-moving clause in the area and the one most likely to have changed since you signed. Check the current terms rather than the ones you remember.
Litigation is the only thing all four publishers agree about
When a matter becomes contested, the four documents converge almost word for word.
Riverside — copies only where "a court of competent jurisdiction issues a valid order requiring production of a copy," with a protective order "prohibiting parties from copying and distributing the materials," and with the materials and all references to their content kept out of the public record.
PRO-ED — "copies of PRO-ED's test materials should not be disseminated to parties or their counsel or made available to the public," litigation records concerning test materials should be subject to a protective order, and the materials "should also not be disclosed in any public records request."
PAR — "copies of test materials should not be disseminated to parties or their counsel."
Pearson — section 6 directs a customer required to disclose secure test materials under a subpoena or court order to Pearson's legal policies rather than stating a rule inline.
What all four are describing is a protective order, which is a thing only a lawyer can obtain. That is the practical content of the unanimity: the publishers' shared position on litigation is that you should not be handling it yourself.
If a demand arrives, the sequence the documents imply is to not produce anything, tell the publisher, and get counsel. We are not telling you what any order obliges you to do.
A 2024 position paper that would exclude you from the exchange
Two documents from the neuropsychology bodies deserve reporting precisely because of what they imply about educational therapists, and it is not comfortable.
The American Academy of Clinical Neuropsychology's official position on test security was published in The Clinical Neuropsychologist in April 2022, volume 36, issue 3, pages 523 to 545.
Its stated objective was to provide education on "the critical importance of test security for neuropsychological and psychological tests" and to establish best-practice recommendations, on the explicit ground that "previous test security guidelines were not adequately specified." Its conclusion is that clinical neuropsychologists "must commit to protecting sensitive neuropsychological and psychological test information from exposure to nonpsychologists."
The 2024 interorganizational position paper is sharper. Boone, Kaufmann, Sweet, Leatherberry, Beattey, Silva, Victor, Boone, Spector, Hebben, Hanks and James published "Attorney demands for protected psychological test information: Is access necessary for cross examination or does it lead to misinformation?" in The Clinical Neuropsychologist in May 2024, volume 38, issue 4, pages 889 to 906.
Its argument is that releasing protected test information to attorneys introduces inaccurate information and jeopardises future test use, since non-psychologists are not ethically bound to protect test content. Its recommendation is that protected test information be exchanged only between clinical psychologist or neuropsychologist experts.
Taken at its word, that recommendation excludes an educational therapist from the exchange. NAN's 2003 recommendation has the same shape: send the material to another qualified neuropsychologist. The paper's bibliographic record is also held by one of the author institutions if you want the citation without the paywall.
The framing is put even more plainly in a December 2024 article by Jerry J. Sweet, Kyle Brauer Boone and Robert A. Beattey in the American Board of Professional Psychology's newsletter, Protecting Test Secure Information. Its definition is one sentence: "Test security involves protection of test materials from release to non-psychologists."
On that definition, you are the non-psychologist. The article's argument is that "threats to test security have the potential to affect specialists in all of psychology's specialties," that "by far the greatest attention and energetic defense of protected test information has come from within Clinical Neuropsychology," and that attorneys "have no obligation to protect the scientific validity of empirical assessment measures."
Every word of that is defensible on its own terms, and it is also written from inside one profession about a class of document that several professions legitimately hold. Educational therapists qualify to buy at Level B under every publisher policy we read. Nobody has written the corresponding guidance for them, which is exactly the hole this post is reporting rather than filling.
We are reporting the position, not endorsing it and not disputing it. It is worth knowing for two reasons. It explains why a psychologist may decline to send you a protocol you have a professional reason to want.
And it is a live example of the thing we keep finding on this site: a rule written by one profession about a document another profession also holds. The clean way to handle it is the referral discipline we set out in where the referral line sits between educational therapy, speech and occupational therapy.
What the Standards say, and why we are not going to tell you
Every document above defers to the same volume. Pearson's definitions clause 15(g) names "applicable professional guidelines" as including "the Standards for Educational and Psychological Testing, published by the American Educational Research Association (AERA), American Psychological Association (APA), and the National Council on Measurement in Education (NCME)." PRO-ED defers to the professional ethical standards its purchasers are subject to.
We could not read the 2014 Standards this run. It is a purchased volume and is not published in full as a free document, and we are not going to summarise numbered standards we did not read. That is the same discipline we applied to an organisation whose continuing-education figures sit behind a member login.
What we can report is one standard as quoted in a technical report that cites it. The Joint Commission on National Dental Examinations' National Board Dental Examination technical report quotes Standard 7.9: where test security is critical to interpreting scores, the documentation "should explain the steps necessary to protect test materials and to prevent inappropriate exchange of information during the test administration session."
Note where that standard points. It is in the chapter on test documentation and it places the duty to explain the protective steps on the test's own documentation, which is the publisher's job rather than the examiner's.
We are not claiming that is all the Standards say about security. We are saying that is the standard we could verify, and that the volume itself is the authority. If a security question is load-bearing for you, read your own copy.
A decision sequence, not a decision
We are not going to tell you whether to show a particular protocol to a particular parent. Here is the order in which the published record suggests the question gets answered.
Identify which obligation is in play. Is the request for the client's data, or for the publisher's material, or for something that is both? NAN's test is the practical one: does the thing I am about to hand over reveal what was asked?
Find the agreement for that specific instrument. The four publishers do not say the same thing, and the instrument on your desk is governed by its own publisher's terms, not by a general rule.
Separate viewing from copying. All four publishers prohibit copies. Three of the four contemplate some form of viewing. That distinction survives every disagreement between them.
Ask who is supposed to be present, and notice when the answer does not describe you. If the clause names a school representative and you are not one, you are making a judgement the clause does not make for you — so make it deliberately and write down what you did.
Check whether the request arrived in a contested context. If it did, stop, notify the publisher, and get counsel. Every publisher we read points this way.
Do not resolve an entitlement question yourself. What FERPA, a state statute or an IDEA process obliges anybody to produce is a legal question. Publishers tell you what the statutes do not require of them. That is not the same as advice about what is required of you.
Write down your own standard and apply it the same way every time. Since no agreement specifies storage, your consistency is the only evidence of reasonable effort you will ever have.
When several organisations share a vocabulary, check whether they share a definition
This post is the latest instance of a pattern that has turned up in every area of this field we have examined, and the prior posts each name their own version of it.
Qualification letters are a purchasing rule for one storefront rather than an industry scale, which is why the letters run A, B, C at one publisher and A, B, C, N at another.
"Supervised hours" means hours of supervision in one of six credentialing systems and something else in the other five.
A continuing-education hour is not a portable currency between bodies.
"Orton-Gillingham based" is a claim no organisation adjudicates.
A norm sample has a shelf life the publishers describe differently.
The standard error of measurement is reported in units that do not convert cleanly between instruments.
Handwriting publishers each teach the letters in an order no two of them share.
And the roles themselves are defined differently by every body that defines them, which is the subject of the comparison of educational therapists, learning specialists and executive function coaches.
Test security is the version of this pattern with the sharpest consequence, because the disagreement is not between definitions in two documents you can read side by side. It is between a contract you signed, a code that is silent, and a family in front of you waiting for an answer.
Where the published record runs out
In the interest of not overstating what we checked:
We did not read the 2014 Standards for Educational and Psychological Testing. It is a purchased volume. Standard 7.9 above is reported as quoted in a third-party technical report, and we make no claim about what any other numbered standard says.
We did not locate full, readable terms of sale and use for WPS or MHS this run. MHS publishes a qualification-level checker and WPS publishes qualification information, but neither yielded a readable security or disclosure policy to us. Their policies may say something different from the four above, and we are not going to guess at them. If an instrument from either is central to your practice, request the current terms directly.
We did not verify the 2005 California ruling PAR refers to. PAR names it and does not cite it, and we report PAR's reference rather than the case.
We did not read the full text of the AACN 2022 position or the 2024 interorganizational paper. Both are behind publisher paywalls. The citations and the abstract-level findings above come from the bibliographic records, and the specific recommendations in each paper may be more nuanced than the abstracts convey.
We do not know whether any publisher has ever enforced these clauses against an independent practitioner. No enforcement record was located, and absence of a record is not evidence either way.
We make no claim about what any statute requires of you. Three publishers say FERPA does not require disclosure of test materials. That is their reading of their own obligations, reported as such.
The AET Code's silence on test security is a finding about the current revision as published, Revision #9, 2025, read on 12 November 2026. If a future revision adds a clause, this post is out of date and the Code is the authority.
Frequently asked questions
Is there a clause in the AET Code of Ethics about test security?
No. The current revision — Revision #9, 2025, reissued September 2025 — has detailed clauses on confidentiality of client information, electronic security, cloud storage, breach notification and record retention, and no clause addressing the security of test content or instruments.
The nearest provisions are Section One, II.D.2, requiring that published materials be used "ethically and only for the purposes intended," and Section Two, I.C, limiting use to instruments the therapist has been adequately trained for.
Can I let a parent look at a completed protocol?
The four publishers we read do not agree. Pearson's terms contemplate review "in the presence of a representative of the school, college, or institution that administered the test." Riverside's policy is satisfied where protocols are "made available to the parent/legal guardian or a representative thereof for review."
PRO-ED does not permit disclosure of items to parents but allows that there "may be a need to allow a parent/guardian/student to view a specific test item." PAR notes that professionals may show protocols or items to parents during discussion. Find the agreement for the instrument in front of you.
Can I give a parent a copy?
Every publisher we read says no. Pearson prohibits reproduction without prior written permission. Riverside states that the FERPA access right "does not include the right to make copies of the materials provided" and that there is "no obligation to provide copies."
PRO-ED says item-level information "should not be copied." PAR "prohibits the release of copies of protocols." Whether any law nonetheless entitles a particular parent to a copy is a legal question for an attorney, not something we can answer.
What is the difference between test data and test materials?
The clearest published definitions are the National Academy of Neuropsychology's. Test data are "raw and scaled scores, client/patient responses to test questions or stimuli." Test materials are "manuals, instruments, protocols, and test questions or stimuli," and the definition explicitly "does not include test data."
NAN's refinement is the useful part: because the two overlap, data that reveal or easily imply the test questions should be treated as materials, while scores that do not reveal questions need not be.
How am I supposed to store a test kit?
No publisher agreement we read specifies physical storage at all. The nearest published guidance is the APA's, which advises encryption or passcodes for stored data and says test materials "are best maintained separately from patient records," and the International Test Commission guideline it cites, "Ensure secure storage of and control access to test materials."
The APA Ethics Code's own standard requires only "reasonable efforts." In practice that means you are setting the threshold, so set one you can describe.
Can I use an AI tool to help me score or interpret a protocol?
Both publishers whose full terms we could read prohibit it in broad language. Pearson's section 6 states that no part of its products, including test content and materials, "may be collected, used or reproduced in any manner or for any purpose in connection with the development, deployment, or use of artificial intelligence technologies or similar technologies."
PRO-ED prohibits use of its content "for the development of any software program, model, algorithm, or generative AI tool." Pearson's wording covers use as well as development. Check the current terms for your instrument before assuming otherwise.
A lawyer has asked me for a protocol. What do the publishers say?
This is the one point on which all four converge, and what they describe is a protective order. Riverside requires a valid court order plus an order prohibiting copying and distribution and keeping the content out of the public record.
PRO-ED says copies should not be disseminated to parties or their counsel and that litigation records should be subject to a protective order. PAR says the same about counsel. Pearson refers the question to its legal policies. The practical implication is that this is not a request to answer yourself.
Can I record a session for scoring?
Pearson's appropriate-use guidance allows recording "only for scoring purposes by qualified users as outlined in the test manual," with permanent deletion once scored, and prohibits storing or sharing recordings containing test content where unqualified users could reach them — expressly "even if the examinee/caregiver has given approval."
Parents and caregivers may not record test content in any form. Riverside does not permit non-professional note-taking or audio and video recording during administrations. Caregiver consent does not unlock storage, because the obligation in question is not the family's.
Keep the practitioner-facing work coming
This is the kind of question that gets asked in supervision and almost never gets written up, because the honest answer is a map of four documents rather than a rule. If that is the kind of thing you want more of, join the educator list and we will send the new work as it goes up.





Comments