Free EPPP Assessment and Diagnosis Practice Questions
The Assessment and Diagnosis content area represents 16% of the EPPP (Part 1-Knowledge) and covers psychometrics, assessment methods and instruments, differential diagnosis, base rates, and outcome measurement. Review each answer and explanation as you practice.
The EPPP format is 225 questions in 255 minutes; this page is a focused content-area sample.
What Assessment and Diagnosis Covers on the EPPP
ASPPB's test specifications for this content area include:
- Psychometrics: test construction, standardization, reliability, validity, sensitivity and specificity, and fairness and bias
- Assessment models and methods, including interviews, self-report, multi-informant reports, and direct observation
- Commonly used instruments and their appropriate use with different populations
- Differential diagnosis, classification systems, and dimensional versus categorical approaches
- Base rates, epidemiology, heuristics, and other influences on interpreting assessment data
- Theories of psychopathology, outcome measurement, and the use of technology in assessment
How to Study Assessment and Diagnosis
- Practice psychometric calculations such as the standard error of measurement, confidence intervals, and predictive values until they are quick.
- Study diagnoses in pairs that are easy to confuse, focusing on duration, onset, and the feature that separates them.
- Remember that a low base rate lowers positive predictive value even when a test is accurate.
EPPP Assessment and Diagnosis Sample Questions with Answers
Sample Question 1 — Assessment and Diagnosis
A screening test for a disorder has a sensitivity of 90% and a specificity of 90%. It is given to 1,000 people in a population where the base rate of the disorder is 5%. Of the people who screen positive, approximately what percentage actually have the disorder?
- A. 10%
- B. 32% (Correct answer)
- C. 50%
- D. 90%
Correct answer: B
Explanation: B is correct because 50 people have the disorder and 45 of them screen positive; 950 do not and 10% of them, 95 people, also screen positive, so 45 of 140 positives (about 32%) are true positives. A is incorrect because 10% is the false-positive rate among people without the disorder, not the proportion of positives who have it. C is incorrect because predictive value depends on the base rate, and with a 5% base rate false positives outnumber true positives. D is incorrect because 90% is the test's sensitivity, the proportion of people with the disorder who test positive, which is not the same as positive predictive value.
Sample Question 2 — Assessment and Diagnosis
On the MMPI-2, a marked elevation on which validity scale MOST suggests that a test taker is overreporting psychological problems?
- A. L (Lie)
- B. F (Infrequency) (Correct answer)
- C. VRIN (Variable Response Inconsistency)
- D. K (Correction)
Correct answer: B
Explanation: B is correct because the F scale consists of items rarely endorsed by most people, so a marked elevation suggests overreporting of symptoms, severe distress, or random responding. A is incorrect because elevations on L suggest an unsophisticated attempt to present oneself in an unrealistically favorable light. C is incorrect because VRIN detects inconsistent or random responding to items with similar or opposite content, not a consistent pattern of overreporting. D is incorrect because elevations on K suggest defensiveness or underreporting of problems.
Sample Question 3 — Assessment and Diagnosis
An 82-year-old nursing home resident who was cognitively intact last week becomes disoriented and has trouble keeping attention on a conversation. The confusion worsens in the evening and improves the next morning. The resident was recently diagnosed with a urinary tract infection. Which diagnosis is MOST likely?
- A. Delirium due to another medical condition (Correct answer)
- B. Brief psychotic disorder
- C. Major neurocognitive disorder due to Alzheimer's disease
- D. Major depressive disorder with psychotic features
Correct answer: A
Explanation: A is correct because acute onset over days, impaired attention, symptoms that fluctuate during the day, and evidence of a medical cause are the defining features of delirium. B is incorrect because the main picture is disturbed attention and awareness rather than delusions or hallucinations, and a medical cause is present. C is incorrect because Alzheimer's disease has a gradual onset over months to years and does not produce an abrupt decline in a previously intact person. D is incorrect because depression does not typically cause acute fluctuating disorientation, and the timing points to the infection.
Sample Question 4 — Assessment and Diagnosis
Cronbach's coefficient alpha is an estimate of which type of reliability?
- A. Test-retest stability
- B. Internal consistency (Correct answer)
- C. Alternate-forms equivalence
- D. Interrater agreement
Correct answer: B
Explanation: B is correct because coefficient alpha reflects how strongly the items of a single administration intercorrelate, which is internal consistency reliability. A is incorrect because test-retest reliability requires giving the same test twice and correlating the two sets of scores. C is incorrect because alternate-forms reliability correlates scores on two different but equivalent versions of a test. D is incorrect because interrater reliability concerns consistency between scorers or observers, not among items.
Sample Question 5 — Assessment and Diagnosis
For a test meant to spread examinees out across a broad range of ability, with items that cannot be answered correctly by guessing, the average item difficulty (p) that maximizes discrimination is closest to:
- A. .10
- B. .30
- C. .50 (Correct answer)
- D. .90
Correct answer: C
Explanation: C is correct because an item passed by half the examinees allows the greatest number of distinctions between those who pass and fail, so p near .50 maximizes variability and discrimination. A is incorrect because an item passed by only 10% of examinees is very hard and separates only the highest scorers from everyone else. B is incorrect because a p of .30 describes a fairly difficult item, which yields less score variance than an item of moderate difficulty. D is incorrect because an item passed by 90% of examinees is very easy and distinguishes only the lowest scorers from everyone else.
Sample Question 6 — Assessment and Diagnosis
In a three-parameter item response theory model, the lower asymptote of an item characteristic curve represents:
- A. The steepness of the curve, or how well the item separates ability levels
- B. The chance that a very low-ability examinee answers correctly by guessing (Correct answer)
- C. The proportion of total score variance due to true differences in ability
- D. The ability level at which half of examinees answer the item correctly
Correct answer: B
Explanation: B is correct because the c parameter is the lower asymptote of the curve, the probability of a correct response by examinees of very low ability, which reflects guessing. A is incorrect because the slope of the curve is the a parameter, which indexes item discrimination. C is incorrect because that describes a reliability coefficient from classical test theory, not a parameter of an item curve. D is incorrect because that location on the ability scale is the b parameter, which indexes item difficulty.
Sample Question 7 — Assessment and Diagnosis
Cohen's kappa is preferred over simple percent agreement for estimating interrater reliability because kappa:
- A. Corrects for the agreement expected by chance alone (Correct answer)
- B. Measures the stability of each rater's judgments over time
- C. Is unaffected by how common each rating category is
- D. Can be used with continuous ratings such as total test scores
Correct answer: A
Explanation: A is correct because kappa compares observed agreement with the agreement expected by chance given each rater's category frequencies, so it is not inflated by chance matches. B is incorrect because stability over time is test-retest or intrarater reliability, not agreement between raters. C is incorrect because kappa is affected by category base rates and can be low when one category is very rare. D is incorrect because kappa is designed for categorical judgments; continuous ratings are usually assessed with the intraclass correlation.
Sample Question 8 — Assessment and Diagnosis
An intelligence test has a standard deviation of 15 and a reliability coefficient of .96. A child obtains a score of 100. What is the approximate 95% confidence interval for the child's true score?
- A. 94 to 106 (Correct answer)
- B. 97 to 103
- C. 91 to 109
- D. 85 to 115
Correct answer: A
Explanation: A is correct because the SEM equals 15 times the square root of (1 - .96), or 15 x .20 = 3, and a 95% interval is about plus or minus 1.96 SEM, roughly 6 points. B is incorrect because plus or minus one SEM gives only about a 68% confidence interval. C is incorrect because plus or minus three SEM gives about a 99.7% interval, wider than needed for 95%. D is incorrect because this range uses the test's standard deviation rather than its standard error of measurement.
Sample Question 9 — Assessment and Diagnosis
A 20-item anxiety scale has a reliability coefficient of .60. The developer adds 20 more items that are comparable in content and quality to the originals. What is the MOST likely result?
- A. Reliability will stay the same, since item quality is unchanged
- B. Reliability will increase, as the Spearman-Brown formula predicts (Correct answer)
- C. Reliability will fall, because longer tests cause more fatigue
- D. Reliability will double, since the number of items has doubled
Correct answer: B
Explanation: B is correct because the Spearman-Brown prophecy formula estimates that lengthening a test with comparable items raises reliability; doubling this test predicts a reliability of about .75. A is incorrect because reliability depends on test length as well as item quality, and adding comparable items reduces the proportion of error variance. C is incorrect because fatigue can matter for very long tests, but adding comparable items generally increases reliability, as Spearman-Brown predicts. D is incorrect because reliability cannot exceed 1.0, and the Spearman-Brown formula predicts an increase to about .75, not a doubling.
Sample Question 10 — Assessment and Diagnosis
A personnel psychologist uses a regression equation to predict an applicant's job performance from a test score and wants to state a range within which the applicant's actual performance is likely to fall. Which statistic is needed?
- A. The standard error of estimate (Correct answer)
- B. The standard deviation of the test scores
- C. The standard error of the mean
- D. The standard error of measurement
Correct answer: A
Explanation: A is correct because the standard error of estimate describes the spread of actual criterion scores around predicted scores and is used to build a confidence interval around a predicted criterion score. B is incorrect because the spread of predictor scores says nothing about how accurately the criterion is predicted. C is incorrect because the standard error of the mean describes the variability of sample means and is used for inferences about a population mean. D is incorrect because the standard error of measurement builds a confidence interval around an obtained test score, not around a predicted criterion score.
Keep Practicing
Take the 10-question EPPP quick-start practice exam across all eight content areas, or return to the EPPP practice exam hub.