Testing ethics on the NCE: test selection, administration, interpretation and fairness under Section E.
One test, one client, a long list of duties. Here is what Section E asks of you before, during and after an assessment, and the answer patterns the exam rewards.
By Dr. Andrew Wichterman · October 11, 2026 · 6 min read
AI-narrated overview made with Google NotebookLM. The written article below is by Dr. Andrew Wichterman.
Testing ethics on the NCE is not a statistics question in disguise. It is an ethics question about a tool. Section E of the ACA Code governs assessment, and the items built on it ask whether you are competent to use a test, whether the client consented, who gets to see the results, and whether the test was fair to the person who took it.
The good news: the answers follow a small set of rules, and the distractors break those rules in predictable ways. For the instruments themselves, see assessment and testing on the NCE. This post is about the duties that come with using them.
The practice item in this article was written for this site in the style of the NCE. It is not a question from the actual exam. NBCC does not release its items, and candidates sign a nondisclosure agreement.
Competence comes first: Levels A, B and C
You give and interpret only the tests you are trained and competent to use. That duty does not go away when a computer does the scoring. If you hand a client a computer-generated report, you are still responsible for how those results are used.
Test publishers sort instruments by the training a user needs. Learn the three levels with an example for each:
- Level A: minimal training. Achievement tests sit here.
- Level B: graduate coursework in testing. Interest inventories and many personality inventories.
- Level C: advanced training plus supervised experience. The Wechsler scales, the MMPI and projective tests.
Two reference documents back all of this. The RUST, Responsibilities of Users of Standardized Tests, comes from AACE. The Standards for Educational and Psychological Testing come jointly from AERA, APA and NCME. Expect them as answer options when a stem asks which document sets expectations for test users.
Exam tip. Wechsler or MMPI in the stem means Level C. A counselor without advanced training and supervised experience should not be giving it.
Choosing the test and getting consent
Selection is an ethical act. The instrument needs evidence of reliability and validity for this purpose and for people like this client. Language, culture, age and disability all bear on that choice. A well-known test normed on a different population is not automatically the right test. For the measurement side, see reliability and validity types for the NCE.
Before any testing starts, the client hears three things in words they can follow: what the assessment is, why it is being given, and how the results will be used. Then you get consent. Clients also have a right to their results and to an explanation of them. If a third party, such as a court or an employer, will receive the results, the client learns that before testing, not after.
Forensic evaluation adds two rules. You do not evaluate someone you are counseling. And you tell the person being evaluated what the evaluation is for and that it is not a counseling relationship.
Administration and scoring: standard conditions
Norms only mean something when the test is given the way it was designed. Administer under standardized conditions, and if anything departs from them, note it in the report.
- Do not use obsolete tests or outdated data.
- Protect test security. Do not copy copyrighted materials.
- Make disability accommodations that keep the construct intact, and document them.
- Check your scoring. Scoring errors are a common way results become invalid.
Computer-adaptive testing may show up too. The test adjusts item difficulty to how the examinee is answering, using item response theory, and reaches comparable precision with fewer items.
Interpretation and release of results
A score is never read alone. You interpret it in light of the client's characteristics, the norms used, the standard error of measurement and the testing situation. Any option that makes a decision from one number is the trap.
Release follows a fixed pattern. Results go to people qualified to interpret them, or to others with the client's consent, and an accurate interpretation goes with them. When you explain results to the client, use plain language and consider their cultural context.
Diagnosis counts as interpretation. The Code lets you refrain from diagnosing when a diagnosis would harm the client, and it asks you to watch for cultural bias when you do diagnose.
A few item analysis terms can appear. Item difficulty is the proportion of examinees who answer correctly, ideally close to .50. Item discrimination is how well an item separates high scorers from low scorers. Ceiling and floor effects happen when a test is too easy or too hard to spread people out at the top or bottom.
Exam tip. Release questions: qualified persons or client consent, with interpretation. Single-score questions: the standard error of measurement.
Test bias, fairness and adverse impact
These three terms sound alike and the exam uses that. Keep them apart:
- Test bias is systematic error that puts a group at a disadvantage for reasons unrelated to the construct. Heavy language load, culturally specific content and differential item functioning are the usual sources.
- Fairness is broader: whether the test is used and interpreted equitably.
- Adverse impact is a legal concept about group differences in selection rates.
Culture-fair tests reduce cultural loading. They do not remove it, so an option claiming a test is free of cultural influence is wrong.
With a client from a different background, ask four questions. Do the norms include people like this client? Did translation keep the meaning? Does the construct mean the same thing in the client's culture? Did stereotype threat or test anxiety affect the performance?
One case to know by name: Larry P. v. Riles. It restricted the use of IQ tests to place Black children in special education in California, and it is the standard legal citation for test bias.
How this shows up on the exam
Most testing ethics items describe a counselor doing something slightly off and ask what should happen instead. Read the stem for the missing duty: no training for the test level, no consent, results sent to someone unqualified, one score treated as the whole story, or norms that do not fit the client.
The tempting distractor usually sounds efficient or helpful. Mailing the full report to a school because the parent asked. Placing a student based on one IQ score. Using a familiar test because it is quick. Each skips a step the Code requires, and that skipped step is your answer.
A counselor gives a client an interest inventory as part of career counseling. The client's new employer calls and asks for a copy of the results. The client has not discussed this with the counselor. What should the counselor do?
- Send the score report, since interest inventories are Level B and carry little risk
- Release the results only with the client's consent, along with an accurate interpretation
- Send the raw scores without interpretation so the employer can draw its own conclusions
- Refuse to release the results under any circumstances
Show the answer
B. Results go to qualified persons or to others with the client's consent, and an accurate interpretation goes with any release. A is tempting because the test level sounds like a reason, but test level governs who may give a test, not who may receive results without consent. C strips out the interpretation the Code requires, and D ignores that the client can consent to a release.
Check the current handbook. Exam logistics, fees and policies change. Confirm specifics against the NBCC candidate handbook and your state board's current rules. Aligned to the current NCE blueprint, valid for exams through June 30, 2027.
Find the duty the counselor skipped: competence, consent, release, the SEM or fit to the client. Then answer.