Self-Assessment Questions

This section provides a review. Mark each statement on the Answer Sheet according to the factual materials contained in this issue and the opinions of the authors.

Article One (pp. 180–191)

1. Confrontation naming tests correlate highly with one another, as well as with measures of aphasia severity and discourse informativeness. This kind of evidence supports which of the following psychometric properties?

   • Test–retest reliability.

   • Inter-rater reliability.

   • Construct validity.

   • Capital validity.

   • Item difficulty.

2. Which of the following statements is true?

   • Tests developed under classical test theory are validated at the level of the item, while tests developed using item response theory are validated at the level of the test.

   • Tests developed under classical test theory are validated at the level of the test, while tests developed using item response theory are validated at the level of the item.

   • Tests developed under both classical test theory and item response theory are validated at the level of the item only.

   • Tests developed under both classical test theory and item response theory are validated at the level of the test only.

3. Within an item response theory framework, item difficulty is:

   • The number of off-target items contained within a given test.

   • The same thing as percent-correct in classical test theory.

   • The degree to which a response refines a patient's estimated ability.

   • The relative ease or difficulty a patient might experience in producing a correct response.

   • The degree to which a person possesses a given skill or attribute.

4. Standard error of measurement is:

   • Assumed to be constant or unchanged in classical test theory, regardless of how a person performs on a test.

   • Assumed to be constant or unchanged in item response theory, regardless of how a person performs on a test.

   • Conditional on a person's ability in classical test theory.

   • Conditional on a person's ability in item response theory.

   • Both A and D are true.

5. One potential clinical application of the study by Hula et al, which found that both of the computer-adaptive forms of the PNT yielded quick and precise estimations of a patient's naming impairment, is:

   • The computer-adaptive forms of the PNT could free up resources, giving clinicians more time to perform other relevant tasks.

   • The computer-adaptive forms of the PNT could also be useful for assessing auditory comprehension.

   • The computer-adaptive forms of the PNT should only be used for research studies.

   • The computer-adaptive forms of the PNT are boring for clinicians to administer.

   • That clinicians should avoid assessing anomia in acute care settings.

   Article Two (pp. 192–210)

6. One way to interpret a change in PROM score is to evaluate whether the change exceeded the amount of change that could be expected due to measurement error alone (i.e., “statistically important change”). This article recommends that clinicians estimate that threshold with the _______

   • Minimal clinically important difference (MCID).

   • Minimally important difference (MID).

   • Minimal detectable change (MDC).

   • Standard error of measurement (SEM).

   • Confidence interval (CI).

7. Another way to interpret a change in PROM score is to evaluate whether the change was clinically meaningful. The main way that this has been done for PROMIS and Neuro-QoL measures is by estimating the ________

   • Response hierarchy.

   • Minimally important difference (MID).

   • Minimal detectable change (MDC).

   • Standard error of measurement (SEM).

   • Confidence interval (CI).

8. A single PROM score could be interpreted by _____

   • Comparing that score to the mean of a general population normative sample with a T score.

   • Comparing that score to the mean of a clinical population normative sample with a T score.

   • Looking at published cutpoints (e.g., the result of “PRO-“bookmarking”).

   • Looking at published patientacceptable symptom state (PASS) values.

   • All of the above.

9. Which of the following is not an acceptable form of communication support to be offered during PROM administration?

   • The administrator defines concepts like “quality of life” that is the health construct being assessed.

   • The administrator defines words like “appointment.”

   • Formatting the short form in a way that is optimally readable.

   • The administrator repeats the item and response options.

   • The administrator allows the client to respond nonverbally, e.g., by pointing.

10. The normative sample for most PROMIS measures is:

    • Adults with one of five acquired neurological conditions: stroke, multiple sclerosis, amyolateral sclerosis, epilepsy, or Parkinson's disease.

    • Adults in the general population (i.e., with demographics that mirror the recent U.S. census).

    • Adults with traumatic brain injury (TBI).

    • Adults with spinal cord injury.

    • PROMIS measures do not have a normative sample.

    Article Three (pp. 211–224)

11. Which of the following ideas was not a motivation for the development of the ACOM?

    • Previously developed patientreported measures for aphasia were insufficiently specific and comprehensive in assessing communicative functioning.

    • Newer, increasingly accessible psychometric approaches offered advantages over the classical test theory model.

    • Existing impairment-level measures of language ability were not responsive to aphasia treatment.

    • Social participation was increasingly recognized as an important goal of aphasia rehabilitation.

    • There was consensus among government agencies, the healthcare industry, and clinical researchers that health outcomes should be described from patients' perspectives.

12. Which of the following statements about the internal structure of the ACOM is true?

    • The current version of the ACOM is based on a unidimensional measurement model that showed adequate fit to the data.

    • The ACOM provides score estimates with excellent internal consistency reliability for the subdomains of talking, comprehension, writing, and naming, in addition to a score estimate for overall communicative functioning.

    • Variability in ACOM scores is attributable primarily to a general factor that influences all items, regardless of their subdomain.

    • The ACOM treats item content regarding both language and nonlanguage cognitive functions as a single, undifferentiated construct, but excludes content related to writing to achieve adequate model fit.

    • None of the above.

13. Members of which of the following potential stakeholder groups reviewed the initial candidate item pool for the ACOM?

    • Persons with aphasia.

    • Communication partners of persons with aphasia.

    • Speech-language pathologists who aphasia rehabilitation services.

    • A and C.

    • A, B, and C.

14. Standard errors of measurement derived from IRT models and from internal consistency reliability estimates under the classical test theory model fail to capture which of the following sources of error:

    • Random response error due to momentary distractions, fluctuations of attention, and similar variation that occur within a testing session.

    • Random error attributable to variation in item content.

    • Longitudinal variability attributable to instability of the trait being measured or other factors contributing to the response that occur between testing sessions.

    • A and B.

    • None of the above.

15. Which of the following pieces of evidence most clearly support the validity of the ACOM as a measure of change in communicative functioning due to aphasia treatment?

    • Winans-Mitrik et al's finding of robust positive ACOM change scores that correlated positively with change on the comprehensive aphasia test.

    • Hula et al's finding that the ACOM showed adequate fit to an item-level bifactor model.

    • Doyle et al's finding that mean score estimates for talking, writing, and comprehension were not different between self and surrogate reports on the ACOM.

    • Hula et al's finding that a 12-item computer-adaptive version of the ACOM agreed well with the full 50-item version.

    • Hula et al's finding that ACOM scores correlated 0.72 with ratings on the BDAE Aphasia Severity Rating Scale.

    Article Four (pp. 225–239)

16. Why are cognitive interviews important in the construction and development of patient-reported outcomes?

    • They contribute to the validity and reliability of the instruments.

    • They help ensure the content and format of the items are acceptable to key stakeholders.

    • They help ensure that the content and format of the instrument is understandable to people who represent the population(s) for whom the questionnaire will be used.

    • They help ensure the instrument contains relevant information and has not omitted important information.

    • All of the above.

17. Please evaluate the following statement: “Raw scores are recommended for use in interpreting the results of IRT-based instrument administration, including comparisons across people or time points.” This statement is:

    • True. Raw scores allow for valid and reliable comparisons across people and time.

    • False. There is no valid way to compare across people or time points using any type of scoring for IRT-based instruments because the adaptive testing results in different items being used at different testing time points.

    • False. Raw scores are not valid for use in comparing scores across time or people, but once scores are converted to a standard scale such as the T-scale, these comparisons can be made.

    • Partially false. Raw scores can be used to interpret computerized adaptive test administrations, but not static short forms.

18. Which of the following variables seems to be associated with CPIB scores most strongly and consistently across populations included in analyses described in this article?

    • Self-reported speech/voice/language symptom severity.

    • Clinician-rated speech/voice/language symptom severity.

    • Instrumental measures/objective measures of speech/voice/language symptom severity.

    • Depression.

    • Age.

19. Which statement best describes the appropriateness of the CPIB, like other patient-reported outcomes, for use with people with aphasia?

    • The CPIB is appropriate for people with mild-moderate aphasia as long as the clinician provides appropriate communication support.

    • Use of the CPIB and other patient-reported outcomes is important to ensure the viewpoints of people with aphasia are included in research and clinical activities.

    • IRT-based instruments such as the CPIB may facilitate greater ease of patient-reported outcomes for people with aphasia because adaptive testing often results in shorter item sets.

    • The use of patient-reported outcomes for people with aphasia must be limited to those people who can demonstrate accurate, independent reading comprehension of the items.

    • All of the above items except D are true.

20. What is the one population for whom the original version of the CPIB (with the item stem referring to “condition”) should not be used because cognitive interviews have found that the wording is unacceptable to stakeholders?

    • People with cognitive-communication disorders.

    • People with aphasia.

    • People who have had a laryngectomy.

    • People who are transgender.

    • People with facial paralysis.

    Article Five (pp. 240–255)

21. “Drift rate” in the diffusion model refers to:

    • Starting point in the decision-making process (i.e., response bias).

    • Nondecision time (e.g., early perceptual and late motor processes).

    • The rate of evidence accumulation (i.e., processing efficiency).

    • The average rate of the ex-Gaussian distribution.

    • Level of response caution (i.e., speed–accuracy tradeoff ).

22. One limitation of the standard diffusion model is that:

    • It requires effortful hand-coding of responses.

    • It is designed to characterize simple two-choice tasks that may not be directly clinically relevant.

    • It is important from a historical perspective, but out of date.

    • It accounts for speed, but not accuracy.

    • It is novel and has not been extensively studied.

23. If an individual sets an overly cautious speed–accuracy tradeoff, they…

    • Will present with lower than necessary accuracy.

    • Will present with faster than ideal processing times.

    • Will set a narrow boundary separation parameter in the diffusion model.

    • Will perform efficiently during treatment and functional communication.

    • Will present with slower than necessary processing times.

24. If an individual sets an overly impulsive speed–accuracy tradeoff, they…

    • Will present with lower than necessary accuracy.

    • Will frequently provide responses after their optimal response time cutoff in picture naming.

    • Will set a wide boundary separation parameter in the diffusion model.

    • Will perform efficiently during treatment and functional communication.

    • Will present with slower than necessary processing times.

25. The point of adaptive returns is calculated using the diffusion model and is defined as:

    • The optimal “sweet spot” starting point to initiate a response.

    • The balanced “sweet spot” in response caution where an individual performs close to their accuracy asymptote, but in as little time as possible.

    • The point when people with aphasia decide they are ready to return to the clinic to focus on adaptation strategies.

    • The reward rate optimal boundary.

    • The optimal “sweet spot” calculated by varying drift rate.

    Article Six (pp. 256–274)

26. Which of the following questions can the MPT-Naming model address?

    • Which of two items is more likely to cause a specific type of error during a naming attempt?

    • How has a client's speech fluency been affected by Alzheimer's disease?

    • What level of education does a client have?

    • Does the client have problems with reading or writing?

    • Does the client have problems with speech perception?

27. Which of the following questions can the MPT-Naming model not address?

    • How long does it take to describe a picture?

    • What is the probability of a specific type of naming response for a client naming a particular item?

    • How much does an item challenge a subcomponent of the naming process?

    • How impaired is a client on a subcomponent of the naming process?

    • How many similar sounding words are there in the language for a particular item?

28. Which of the following is a situation that can be accurately described by a compensatory model?

    • There are two possible strategies to achieve a goal, and one is better than the other, but they both achieve the same goal to some extent.

    • To achieve a goal, a particular sequence of steps must be successfully completed in order.

    • A client's test scores are being affected by something that the investigator did not account for in the testing procedure.

    • A client is asking for help navigating insurance claims.

    • A client is seeking psychological counseling for depression.

29. Which of the following is a situation that can be described by a noncompensatory model?

    • There are two possible strategies to achieve a goal, and one is better than the other, but they both achieve the same goal to some extent.

    • To achieve a goal, a particular sequence of steps must be successfully completed in order.

    • A client's test scores are being affected by something that the investigator did not account for in the testing procedure.

    • A client is asking for help navigating insurance claims.

    • A client is seeking psychological counseling for depression.

30. Which of the following is a subcomponent of the picture-naming process?

    • Visual object recognition.

    • Retrieving an abstract word form.

    • Constructing the metrical frame and phonological form of the word.

    • Executing and monitoring speech articulation.

    • All of the above.

Publication History

Article published online:
14 July 2021