Instrument Selection - Eppp Skills

1. Which validity evidence is supported by convergent and discriminant patterns of correlations?

   Construct validity evidence. It confirms the test measures the intended theoretical construct through expected correlation patterns.

2. Which concept describes the score range likely to contain a client true score given measurement error?

   Confidence interval around an obtained score. It accounts for measurement error to estimate the range where the true score likely falls.

3. What type of reliability is evaluated by correlating scores from two administrations over time?

   Test-retest reliability. This approach measures temporal stability by comparing scores from the same test given at different times.

4. Which norm group feature is most important to evaluate for fairness in interpretation?

   Representativeness of the normative sample to the client. Similarity to the client's demographics ensures unbiased and applicable score interpretations.

5. Which property is threatened when test scores show systematic group differences due to bias?

   Validity, especially measurement invariance and fairness. Bias undermines equitable measurement across groups, compromising interpretive accuracy.

6. Which norm type is most appropriate when comparing a student score to same-grade classmates?

   Grade-based norms. These norms facilitate evaluation against peers at the same educational level.

7. Which validity evidence is supported when a test predicts a future outcome (e.g., relapse, GPA)?

   Predictive criterion-related validity. It indicates the test can forecast future behaviors or outcomes based on current scores.

8. Which psychometric property is most critical when selecting a test for diagnostic decisions?

   High reliability and strong validity for the intended diagnostic use. Reliability ensures consistent results, while validity confirms the test accurately identifies the diagnosis in question.

9. Which score scale has mean $100$ and standard deviation $15$ in many cognitive tests?

   Standard score (e.g., IQ metric: $M=100$, $SD=15$). This metric standardizes scores for easy comparison across tests and populations.

10. Which score expresses rank position from $1$ to $99$ relative to the norm group?

    Percentile rank. It indicates the percentage of the norm group scoring below the individual.

11. Choose the best action when a client is outside the test normative age range.

    Select an instrument normed for that age or interpret with extreme caution. Norms must match the client's age to ensure valid comparisons; otherwise, results may mislead.

12. What is the standard error of measurement (SEM) formula in terms of $SD$ and $r_{xx}$?

    $SEM = SD\sqrt{1-r_{xx}}$. This formula quantifies measurement error using the standard deviation and reliability coefficient.

13. Which selection rule best matches a test to a referral question: broad screening or focused diagnosis?

    Match instrument purpose and scope to the referral question. Alignment ensures the test's breadth and focus address the specific clinical or evaluative need.

14. What is the approximate $95\%$ confidence interval formula using $SEM$ (normal assumption)?

    Obtained score $\pm 1.96\times SEM$. This uses the SEM to bound the obtained score within approximately 95% probability under normality.

15. Which option is preferred for screening when missing true cases is the primary concern?

    High sensitivity (accepting more false positives). Prioritizing detection of true cases reduces misses, even if it increases unnecessary follow-ups.

16. What type of reliability is evaluated by agreement between two independent raters?

    Interrater reliability. This method quantifies consistency in scoring or observation across different evaluators.

17. Which score expresses deviation from the mean in $SD$ units using $z=\frac{X-M}{SD}$?

    z score. This transformation standardizes raw scores for statistical analysis and comparison.

18. Identify the best instrument choice when low base rates require detecting rare conditions accurately.

    A test with strong likelihood ratios and high specificity. These properties minimize false positives and enhance diagnostic accuracy in low-prevalence scenarios.

19. What type of reliability is estimated by correlating items within a single test administration?

    Internal consistency reliability (e.g., coefficient alpha). This method assesses how well items measure the same construct without needing multiple administrations.

20. Which reliability approach is best for measures vulnerable to practice effects across retesting?

    Alternate (parallel) forms reliability. Equivalent versions minimize memory or learning biases that could inflate scores on repeated testing.

21. Which option is preferred when false positives have high cost (e.g., labeling, exclusion decisions)?

    High specificity (accepting more false negatives). This minimizes erroneous classifications that could lead to harmful consequences.

22. Identify the better choice for monitoring treatment change over weeks: high internal consistency or high test-retest?

    High test-retest reliability (and good sensitivity to change). Temporal stability reduces error in repeated measures, while sensitivity detects meaningful shifts.

23. Which norm type is most appropriate when interpreting a child score relative to same-age peers?

    Age-based norms. These norms allow comparison to developmental peers, accounting for age-related changes.

24. Which validity evidence is supported when a test correlates strongly with an established measure now?

    Concurrent criterion-related validity. It shows the test aligns with current benchmarks, supporting its use for immediate applications.

25. Which validity evidence is supported when test content adequately samples the target domain?

    Content-related validity evidence. It demonstrates that items represent the full breadth and depth of the construct being assessed.

Study these flashcards