Chapter 5 of 816% of exam

Assessment and Diagnosis

This domain covers psychometrics, test construction, assessment methods and instruments, classification systems, and the biases that affect clinical judgment. Several items require a short calculation, so practice the core formulas.

Reliability and measurement error

Reliability questions test the core formulas and what each reliability estimate captures. Measurement error is expressed as the standard error of measurement and used to build confidence bands. Standard scores and percentiles must be converted quickly using the normal curve.

Standard error of measurement

SEM equals the SD times the square root of (1 minus reliability); a 95% band is about the obtained score plus or minus 1.96 SEM.

AERA/APA/NCME Standards (2014)

Spearman-Brown

Corrects a split-half correlation to full length: 2r / (1 + r).

Spearman (1910); Brown (1910)

Standard scores

T-scores have mean 50 and SD 10; one SD above the mean is about the 84th percentile.

Standard normal distribution

Validity and decision accuracy

Validity is accumulated evidence that scores mean what they are claimed to mean. Items distinguish sources of validity evidence and ask about prediction error. Base rates change what a positive screening result means, a point that is easy to miss.

Construct validity

Convergent and discriminant correlations together support construct validity.

Campbell & Fiske (1959); Cronbach & Meehl (1955)

Criterion-related validity

Predictive validity uses a criterion measured later; concurrent validity uses one measured at the same time.

AERA/APA/NCME Standards (2014)

Standard error of estimate

SD of the criterion times the square root of (1 minus r squared).

Linear regression

Base rates

Positive predictive value falls as the base rate falls, even with good sensitivity and specificity.

Meehl & Rosen (1955)

Items, fairness and instruments

Test construction covers item analysis, item response theory and fairness. The outline also asks for commonly used instruments and their appropriate use. Fairness questions separate group mean differences from bias in how a test predicts.

Item difficulty

Item variance p(1 - p) is largest at p = .50, which allows maximum discrimination when guessing is not a factor.

Classical test theory

Item response theory

The a, b and c parameters describe discrimination, difficulty and the lower asymptote (pseudo-guessing).

Lord (1980)

Differential prediction

A test-criterion relationship that differs across groups is predictive bias, separate from mean differences or item-level DIF.

AERA/APA/NCME Standards (2014), ch. 3; Cleary (1968)

MMPI-2 validity scales

F elevations suggest overreporting or random responding; L reflects naive virtuous self-presentation; K reflects defensiveness.

MMPI-2 Manual (Butcher et al., 2001)

Classification and clinical judgment

The outline asks about classification systems, dimensional versus categorical approaches, epidemiology and base rates. It also names the factors that distort interpretation, such as heuristics and cultural bias. Research on clinical judgment shows where unaided judgment tends to go wrong.

RDoC

NIMH's Research Domain Criteria is a dimensional research framework and is not meant as a diagnostic guide.

NIMH About RDoC

Incidence vs prevalence

Incidence counts new cases in a period among those at risk; prevalence counts all existing cases.

CDC Principles of Epidemiology

Clinical vs statistical prediction

Mechanical prediction has generally equaled or outperformed clinical judgment.

Meehl (1954); Grove et al. (2000)

Judgment errors

Availability and illusory correlation lead clinicians to see frequencies and relationships the data do not support.

Tversky & Kahneman (1974); Chapman & Chapman (1967)

Keep going: the full EPPP Part 1 — Examination for Professional Practice in Psychology guide covers every section of the exam. EPPP Part 1 Study Guide — 2026 Edition — PDF + EPUB, $19.99 · 14-day refund →

Practice this topic →← All chapters

Studying in order?

Practice stays free. The full EPPP Part 1 — Examination for Professional Practice in Psychology study guide is the material itself, taught start to finish — a downloadable PDF + EPUB you keep.

Get the book — $19.99