Assessment and testing, the psychometrics area

A counselor has the client complete the PHQ-9 at every session and reviews the trend with the client. What is this practice called?

1. criterion-referenced testing
   Criterion-referenced testing compares a score with a fixed standard of mastery; it does not name the practice of repeated measurement over time.
2. routine outcome monitoring
   Routine outcome monitoring, or measurement-based care, gives a brief standardized measure at regular intervals and uses the trend to guide treatment with the client.
3. summative evaluation
   Summative evaluation judges a program's overall results at its end; tracking one client's scores session by session is clinical monitoring.

What usually happens when the cutoff score on a screening test is lowered so that more clients screen positive?

1. reliability increases while validity stays the same
   Moving a cutoff changes how scores are classified, not the consistency of the scores themselves, so reliability is not what shifts.
2. specificity increases and sensitivity decreases
   Higher specificity with lower sensitivity is what happens when the cutoff is raised: fewer false positives but more missed cases.
3. sensitivity increases and specificity decreases
   A lower cutoff flags more people, so fewer true cases are missed (higher sensitivity) but more people without the condition are flagged (lower specificity).

A counselor asks a parent to record each time a child hits a sibling, noting what happened just before and just after. What type of assessment is this?

1. A projective personality assessment
   Projective techniques use ambiguous stimuli to infer inner conflicts; ABC recording is direct observation of behavior.
2. A structured diagnostic interview
   A structured diagnostic interview asks set questions to reach a diagnosis; it does not track the triggers and payoffs of one behavior at home.
3. A mental status examination
   A mental status examination is the counselor's snapshot of appearance, mood, thought, and cognition during an interview, not a parent's log of incidents.
4. A functional behavioral assessment
   Recording antecedents, behavior, and consequences (ABC data) is the core of a functional behavioral assessment, which identifies what triggers and maintains a behavior.

A new anxiety scale correlates highly with an established anxiety measure but only weakly with a measure of verbal intelligence. What does this pattern provide evidence of?

1. Test-retest and split-half reliability
   Reliability concerns the consistency of the scale's own scores, not how it relates to other measures.
2. Concurrent and predictive validity
   Criterion-related validity compares scores with a practical outcome measured now or later; correlating with other tests of related and unrelated constructs is construct evidence.
3. Content and face validity
   Content validity rests on expert judgment of how well items sample the domain, and face validity on appearance; neither involves correlations with other measures.
4. Convergent and discriminant validity
   High correlation with a measure of the same construct (convergent) and low correlation with an unrelated one (discriminant) together support construct validity.

A client scores 31 on the BDI-II at intake. How should the counselor use this result?

1. Combine it with interview data to plan care
   A BDI-II score of 31 is in the severe range of symptom severity; it is one source of information, combined with the interview, history, and observation to plan care.
2. Rely on it alone to choose the interventions
   Relying on one score alone to choose interventions ignores context such as risk, history, and the client's goals.
3. Treat it as a diagnosis of major depression
   A self-report inventory measures symptom severity; a diagnosis requires a clinical assessment against DSM-5-TR criteria.
4. Withhold it from the client to prevent distress
   Clients have a right to understand their results, and discussing them can support collaboration in treatment.

A counselor records six clients' session counts: 4, 7, 9, 12, 15, and 18. What is the median?

1. 12
   12 is the fourth score, one of the two middle values; with six scores the median falls halfway between the third and fourth.
2. 10.5
   With an even number of ordered scores, the median is the average of the two middle values: (9 + 12) / 2 = 10.5.
3. 10.8
   10.8 is the mean (65 / 6 ≈ 10.8), which uses every score rather than the middle position.

A counselor must retest clients after two weeks but wants to minimize their memory of specific items. Which reliability estimate is most relevant?

1. Inter-rater reliability
   Inter-rater reliability concerns agreement between scorers, not repeated testing of the same clients.
2. Test-retest (stability) reliability
   Retesting with the same form is exactly what lets clients remember specific items, which inflates the coefficient.
3. Parallel-forms reliability
   Parallel (alternate) forms use an equivalent version at retest, which reduces memory and practice effects while still estimating consistency.

A counselor refers a 10-year-old for a comprehensive individual intelligence evaluation. Which Wechsler scale is designed for this age?

1. WAIS-5
   The WAIS-5 (2024) is the adult scale, starting at age 16, too old a range for this child.
2. WPPSI-IV
   The WPPSI-IV is for preschool and early school-age children, from about 2 years 6 months to 7 years 7 months.
3. WISC-V
   The WISC-V covers ages 6:0 to 16:11, so it is the Wechsler scale for a 10-year-old.

Before interpreting a personality inventory for a recently immigrated client, what should a counselor assess to judge whether the U.S. norms are appropriate?

1. The client's years of schooling in the country of origin
   Education can matter for reading level, but it does not tell whether the client resembles the U.S. norm group culturally and linguistically.
2. The client's level of acculturation and English proficiency
   Instruments normed on other groups call for caution; acculturation and language proficiency show how far U.S. norms may apply to this client.
3. The inventory's test-retest reliability over two weeks
   Stability over time is a property of the instrument; it says nothing about whether the norms fit this particular client.

A fourth grader earns a grade-equivalent score of 7.2 on a math achievement test. How should the counselor interpret this score?

1. The student scored like an average 7th grader on 4th-grade material
   A grade equivalent of 7.2 means the raw score matches that of a typical seventh grader in the second month, on the test the fourth grader took.
2. The student should be placed in 7th-grade math instruction next year
   The student was never tested on seventh-grade content, so the score cannot show readiness for seventh-grade instruction.
3. The student scored higher than 72% of 4th graders in the norm group
   That would be a percentile rank of 72; a grade equivalent is a different kind of score.

A client completes a forced-choice (ipsative) career inventory and scores highest on the Artistic scale. What is the MOST accurate interpretation?

1. She has stronger artistic ability than people in a national norm group
   An ipsative format compares no one with a norm group, and an interest inventory measures preferences, not ability.
2. She prefers artistic activities over the other options offered
   Ipsative scores are relative within the person: a high Artistic score shows what she chose over the other options, not how she compares with others.
3. She would succeed in an artistic field more than in others
   Interest scores do not predict success; they describe likes, and ipsative ones only relative to her other choices.

A client's score is exactly one standard deviation below the mean on a normally distributed test. What is the client's approximate percentile rank?

1. 16th
   About 50% of scores fall below the mean and about 34% lie between the mean and 1 SD below it, so about 16% fall below z = −1.
2. 32nd
   32% is roughly the share of scores more than 1 SD from the mean in both tails combined, not the share below this one score.
3. 34th
   34% is the area between the mean and 1 SD below it, not the area below the score.

On a stress scale, the 25th percentile score is 20 and the 75th percentile score is 35. What is the interquartile range?

1. 7.5
   7.5 is half the IQR, the semi-interquartile range (quartile deviation), not the IQR itself.
2. 27.5
   27.5 is the midpoint between the two quartiles, a measure of location rather than spread.
3. 15
   The interquartile range is Q3 − Q1: 35 − 20 = 15, the spread of the middle 50% of scores.

A counselor wants to correlate two supervisors' rank orderings of 12 trainees from best to worst. Which correlation coefficient fits these data?

1. point-biserial correlation
   The point-biserial correlation pairs a true dichotomy with a continuous variable; both variables here are rankings.
2. phi coefficient
   Phi correlates two dichotomous variables, such as pass/fail with yes/no, not rank orders.
3. Spearman's rho
   Spearman's rho is the correlation for ordinal data such as two sets of rankings.

A test question is answered correctly far less often by one group than another group of equal overall ability. This is evidence of what?

1. Restriction of range in scores
   Restriction of range is a narrow spread of scores that weakens correlations; it does not describe group differences on a single item.
2. Differential item functioning
   DIF means equally able test takers from different groups have different chances of answering an item correctly, which flags possible item bias.
3. Low test-retest reliability
   Test-retest reliability concerns score stability over time, not how one item behaves across groups of equal ability.

A test has a standard deviation of 15 and a reliability coefficient of .91. What is the standard error of measurement?

1. 4.50
   SEM = SD × √(1 − r) = 15 × √.09 = 15 × .30 = 4.50.
2. 13.65
   13.65 is 15 × .91, multiplying the SD by the reliability itself instead of by √(1 − r).
3. 14.31
   14.31 is 15 × √.91, taking the square root of the reliability rather than of 1 − r.
4. 1.35
   1.35 is 15 × .09: it uses 1 − r but forgets to take the square root.