PrepSolution

Speech-Language Pathology · Study journal

SLP Praxis Assessment Questions: Standard Scores, Percentiles and Test Choice

Eleven original SLP Praxis assessment questions on norm- and criterion-referenced tests, z-scores, percentiles, confidence bands, dynamic assessment, and when standard scores should not be reported.

By Elena R.Published Reviewed by PrepSolution Editorial Team12 min read
Inside this articleSpeech-Language Pathology
Section preview 01

Where does assessment sit on the SLP Praxis?

The ETS Praxis Study Companion for test 5331 lists 132 selected-response questions in 150 minutes, split across three content categories of about 44 questions each. Category II is "Screening, Assessment, Evaluation, and Diagnosis". Within it, "Approaches to Assessment and Evaluation" includes "Selecting appropriate assessment instruments, procedures, and materials", and the assessment procedures run across all nine practice areas.

Read this section
On this page

Assessment questions on the SLP exam test two skills at once. You have to read the numbers correctly, and you have to know when the numbers should not be used at all. A standard score of 78 means something precise; the same score from a test normed on a different population may mean nothing.

Work the 11 questions below and commit to an answer before you read each explanation. For the full question bank across all nine practice areas, see the SLP Praxis exam prep course.

PrepSolution wrote every question on this page. None is an ETS or ASHA exam item. Each answer rests on ASHA's Practice Portal or the OpenStax Introductory Statistics textbook, as named in the explanation and listed under References, and every calculation was checked in Python. This is exam practice, not clinical guidance. The PrepSolution Editorial Team checked each key against the source; a certified speech-language pathologist has not reviewed it.

Where does assessment sit on the SLP Praxis?

The ETS Praxis Study Companion for test 5331 lists 132 selected-response questions in 150 minutes, split across three content categories of about 44 questions each. Category II is "Screening, Assessment, Evaluation, and Diagnosis". Within it, "Approaches to Assessment and Evaluation" includes "Selecting appropriate assessment instruments, procedures, and materials", and the assessment procedures run across all nine practice areas.

ASHA states that when the exam moves to PSI on January 11, 2027 as the National Speech-Language Pathology Examination, the content, format and passing standards remain unchanged, so this material applies on either side of the switch.

Scores at a glance

Many standardized language tests report standard scores with a mean of 100 and a standard deviation of 15. Each test manual gives its own scale, so always check it. On that scale, the conversions below follow from the z-score formula, z = (score minus mean) divided by the standard deviation, and the normal distribution. Percentiles were computed in Python and rounded.

Standard scorez-scorePercentile (about)Empirical rule check
115+1.0084thAbout 68% fall between 85 and 115
1000.0050thThe mean
85-1.0016th(100 - 68) / 2 = 16% fall below
78-1.477thBetween one and two standard deviations below
70-2.002nd (2.3%)(100 - 95) / 2 = about 2.5% fall below
Standard score to z-score to percentile, for a test with mean 100 and standard deviation 15. Percentiles are from the normal distribution, computed in Python.

Questions on test types and scores

Question 1: Choosing the kind of test

A school team needs to decide whether a 7-year-old's language skills fall well below those of same-age peers for an eligibility decision. Which type of assessment is designed to answer that question?

  • A. A criterion-referenced checklist of classroom vocabulary
  • B. A norm-referenced standardized test
  • C. An informal probe of wh- questions
  • D. A parent interview

Best answer: B. ASHA describes norm-referenced tests as standardized tests designed to compare and rank test takers in relation to one another, against a large group of statistically selected individuals usually of the same age or grade. Comparing a child with same-age peers is that purpose.

  • A is not the best answer. ASHA says criterion-referenced tests measure performance against predetermined criteria, not against a group of peers.
  • C is not the best answer. ASHA gives an informal probe of wh- questions as an example of a criterion-referenced procedure, which answers a narrower question.
  • D is not the best answer. A parent interview is valuable case-history information, but it does not rank the child against a norming group.

Question 2: What a percentile rank says

A child scores at the 25th percentile on a norm-referenced vocabulary test. Which statement is correct?

  • A. The child answered 25% of the items correctly
  • B. The child is 25% below the average score
  • C. The child scored better than 75% of the norming group
  • D. The child performed as well as or better than 25% of the same-age norming group

Best answer: D. ASHA explains that a norm-referenced score is typically reported as a percentile ranking, and that an individual who scores in the 90th percentile has performed as well as or better than 90% of others of the same age or grade in the norming group. The same reading applies at the 25th percentile.

  • A is not the best answer. A percentile rank describes standing in the norming group, not the proportion of items answered correctly.
  • B is not the best answer. A percentile is not a percentage distance from the average.
  • C is not the best answer. This reverses the ranking. At the 25th percentile, about 75% of the norming group scored higher.

Question 3: Converting a standard score

A test reports standard scores with a mean of 100 and a standard deviation of 15. A child earns a standard score of 78. What is the z-score, to two decimal places?

  • A. -1.47
  • B. -0.22
  • C. -2.20
  • D. +1.47

Best answer: A. OpenStax defines the z-score as the number of standard deviations a value lies above or below the mean, with values below the mean negative. (78 - 100) / 15 = -22 / 15 = -1.47. That is about the 7th percentile on a normal distribution.

  • B is not the best answer. -0.22 divides the 22-point gap by the mean of 100 instead of by the standard deviation.
  • C is not the best answer. -2.20 divides the 22-point gap by 10, not by the standard deviation of 15.
  • D is not the best answer. The score is below the mean, so the z-score must be negative. OpenStax states that values smaller than the mean have negative z-scores.

Question 4: Two standard deviations below

On the same test (mean 100, standard deviation 15), a child scores 70. Using the empirical rule, about what percentage of the norming group would score lower than 70?

  • A. About 16%
  • B. About 5%
  • C. About 2.5%
  • D. About 0.15%

Best answer: C. A score of 70 is (70 - 100) / 15 = -2, two standard deviations below the mean. OpenStax's empirical rule says about 95% of values lie within two standard deviations of the mean. The remaining 5% is split between the two tails, so about 2.5% fall below 70. The exact normal value is 2.3%.

  • A is not the best answer. About 16% is the share below one standard deviation (a score of 85), not two.
  • B is not the best answer. 5% is both tails together. Only the lower tail is below 70.
  • D is not the best answer. About 0.15% is the share below three standard deviations, a score of 55.

Question 5: One standard deviation below

A test has a mean of 100 and a standard deviation of 15. Roughly what percentile does a standard score of 85 represent?

  • A. About the 5th percentile
  • B. About the 16th percentile
  • C. About the 34th percentile
  • D. About the 85th percentile

Best answer: B. A score of 85 is (85 - 100) / 15 = -1, one standard deviation below the mean. By OpenStax's empirical rule, about 68% of scores lie within one standard deviation, leaving 32% in the two tails and about 16% below 85. The normal distribution gives 15.9%.

  • A is not the best answer. The 5th percentile is about 1.64 standard deviations below the mean, lower than 85.
  • C is not the best answer. 34% is the share between the mean and one standard deviation below it, not the share below 85.
  • D is not the best answer. A standard score is not a percentile. A score of 85 is below the mean, so its percentile is well under 50.

Question 6: Reading a score band

A score report gives a child's standard score of 82 with a 90% confidence band and a 95% confidence band. Which band will be wider?

  • A. The 95% band
  • B. The 90% band
  • C. Both are the same width
  • D. It depends on whether the score is above or below the mean

Best answer: A. OpenStax shows that for the same data the 95% confidence interval is wider than the 90% interval, because being more confident that the interval contains the true value requires a wider interval.

  • B is not the best answer. The 90% band is the narrower one. A lower confidence level gives a smaller interval.
  • C is not the best answer. The two bands differ because they use different confidence levels.
  • D is not the best answer. The width depends on the confidence level and the measurement error, not on which side of the mean the score falls.

Questions on when scores are valid

Question 7: A child outside the norming sample

A clinician evaluates a 5-year-old who speaks mainly Vietnamese at home. The only standardized test available was normed on monolingual English speakers. How should the standard scores be handled?

  • A. Report them, then add 10 points to adjust for the second language
  • B. Report them as usual, since the test is standardized
  • C. Report only the percentile rank, which is not affected by the norming sample
  • D. Do not determine or report standard scores, because the norming sample does not represent the child

Best answer: D. ASHA states that standard scores should not be determined if the norming sample does not adequately represent the individual being assessed, and that standardized test scores are not valid and are not reported for someone who is not reflected in the normative sample. The test may still give descriptive information.

  • A is not the best answer. ASHA describes no point adjustment. Adding points does not make a mismatched norming sample valid.
  • B is not the best answer. Being standardized does not make a test valid for a person outside its norming sample.
  • C is not the best answer. A percentile rank is also a norm-referenced score from the same mismatched sample.

Question 8: Translating a test

A colleague offers to translate an English standardized language test into Spanish on the spot so that a standard score can be reported for a Spanish-speaking child. What does ASHA say?

  • A. Translation is acceptable if the translator is fluent
  • B. Translation is acceptable if a bilingual SLP gives the test
  • C. Translation invalidates the results, and standard scores cannot be reported
  • D. Translation is acceptable for receptive subtests only

Best answer: C. ASHA states that translation of a standardized assessment invalidates the results and that standard scores cannot be reported when the assessment has been translated. It explains that translating without validation studies may weaken psychometrics such as content validity.

  • A is not the best answer. The translator's fluency does not restore validity. The problem is the test's psychometrics, not the language skill of the person translating.
  • B is not the best answer. A bilingual examiner does not change the fact that the translated test was not normed in that form.
  • D is not the best answer. ASHA makes no exception for receptive subtests.

Question 9: Difference or disorder?

A 6-year-old who is learning English as a second language performs poorly on a vocabulary probe. The clinician teaches a small set of new words in a short session, then retests, and the child learns and uses them quickly. What does this result suggest?

  • A. A language difference rather than a language disorder
  • B. A language disorder, because the first probe was low
  • C. An invalid assessment, because teaching is not allowed during testing
  • D. A hearing loss that needs referral

Best answer: A. This is dynamic assessment: test, teach and retest. ASHA states that people able to change their language by learning a skill are likely to have language differences, while those unable to make changes are more likely to have a language disorder. Quick learning points to a difference.

  • B is not the best answer. A low first score is exactly what dynamic assessment looks past. The child's response to teaching is the deciding information.
  • C is not the best answer. ASHA describes dynamic assessment, which builds teaching into the procedure, as a legitimate method alongside standardized tests and language samples.
  • D is not the best answer. Nothing in this result points to hearing. ASHA does recommend hearing screening in other contexts, but fast word learning is not a sign of hearing loss.

Question 10: Dialect features on a test

A child who speaks African American English is scored on a test normed on General American English. Several "errors" are features of the child's dialect. What is the best interpretation?

  • A. Count the features as errors, since the test manual defines correct answers
  • B. Consider the dialect; standard score interpretations are not valid if the norming sample does not represent the child, and a dialect is not a disorder
  • C. Refer for a hearing evaluation, since the child is not producing standard forms
  • D. Repeat the same test until the child produces the standard forms

Best answer: B. ASHA says clinicians consider the influence of dialect patterns on assessment measures, which are typically based on General American English, and that standard score interpretations are not valid if the norming sample is not representative of the child. It also states that using a dialect unfamiliar to the listener is not an indication of a spoken language disorder.

  • A is not the best answer. Scoring dialect features as errors treats a difference as a disorder, which ASHA warns against.
  • C is not the best answer. Dialect use is not a sign of hearing loss.
  • D is not the best answer. Repeating the test does not fix a norming mismatch, and ASHA does not treat dialect forms as something to correct by retesting.

Question 11: Measuring progress in therapy

A school SLP wants to check every two weeks whether a second grader is learning the specific grade-level words targeted in intervention. Which approach does ASHA describe for this kind of question?

  • A. Readministering the full norm-referenced test every two weeks
  • B. Comparing the child's percentile rank each month
  • C. An informal criterion-referenced procedure built for the target skill
  • D. A hearing rescreen every two weeks

Best answer: C. ASHA says criterion-referenced procedures can be developed informally to address specific questions, such as automatic recognition of grade-appropriate words in print, and to assess response to intervention. That fits a short, repeated check on targeted words.

  • A is not the best answer. A norm-referenced test is built to rank the child against peers, not to track a narrow set of targets every two weeks.
  • B is not the best answer. A percentile rank describes standing in a norming group and is not designed to show whether specific targets were learned.
  • D is not the best answer. A hearing rescreen does not measure vocabulary learning.

How should you study assessment for this exam?

Learn the score scale until conversions are automatic: 85 and 115 are one standard deviation from 100, 70 and 130 are two. Then practice the second skill, which carries as many marks: deciding whether a score may be used at all. Every time you see a bilingual child, a translated test or a dialect speaker in a question, check the norming sample before you read the number.

Research design, sensitivity and specificity are covered in the research, EBP and ethics question set, and child language milestones and dialect cases are in the child language and speech sound cases. The SLP Praxis study guide shows how the three content categories fit together.

Elena R.

PrepSolution Content Editor, Speech-Language Pathology

About PrepSolution

References

  1. [1] ETS (2021). The Praxis Study Companion: Speech-Language Pathology (5331). ETS Praxis. ETS Praxis
  2. [2] American Speech-Language-Hearing Association (2026). Assessment Tools, Techniques, and Data Sources (Practice Portal). ASHA. ASHA
  3. [3] American Speech-Language-Hearing Association (2026). Spoken Language Disorders (Practice Portal). ASHA. ASHA
  4. [4] American Speech-Language-Hearing Association (2026). Multilingual Service Delivery in Audiology and Speech-Language Pathology (Practice Portal). ASHA. ASHA
  5. [5] Illowsky, B., Dean, S., et al. (2023). Introductory Statistics 2e, section 6.1: The Standard Normal Distribution. OpenStax. OpenStax
  6. [6] Illowsky, B., Dean, S., et al. (2023). Introductory Statistics 2e, section 8.1: A Single Population Mean Using the Normal Distribution. OpenStax. OpenStax
  7. [7] American Speech-Language-Hearing Association (2026). National Speech-Language Pathology Examination. ASHA. ASHA

Help us improve this guide

How helpful was this article?

One tap helps us understand what’s useful.

Not helpful → Very helpful

Frequently asked questions

The ETS Praxis Study Companion for test 5331 lists Screening, Assessment, Evaluation, and Diagnosis as one of three content categories, about 44 of 132 questions, or 33%.

On a test with a mean of 100 and a standard deviation of 15, a score of 85 is one standard deviation below the mean, which is about the 16th percentile.

ASHA describes norm-referenced tests as comparing and ranking test takers against a norming group, usually of the same age or grade, and criterion-referenced tests as measuring performance against predetermined criteria or standards.

Only if the child is represented in the test's norming sample. ASHA says standard scores should not be determined when the norming sample does not represent the individual, and cannot be reported when a test has been translated.

A test, teach and retest method. ASHA says people who can change their language after learning a skill are likely to have a language difference, and those who cannot are more likely to have a language disorder.

Keep practicing for the SLP Praxis.

Review the study tools, access options and price, or start with free practice.

Related Articles