EPPP Part 2 Skills common mistakes
EPPP Part 2 Skills · Understand the traps and how to avoid them.
Choosing the answer that best matches what you would personally do in a clinical situation rather than the answer the item is testing (e.g., picking a favored therapy orientation when the stem asks what is empirically supported).
Part 2 items are written so that several options are clinically plausible; candidates default to personal training and theoretical loyalty under time pressure instead of identifying the construct being assessed.
What to do instead
Before reading the options, restate the stem in your own words and label the construct (e.g., 'this is asking about base rates', 'this is asking about informed consent'). Eliminate any option that reflects your preference but does not address that construct, and practice with mixed-orientation vignettes so you rehearse construct-first reasoning.
Ignoring base rates and prevalence when interpreting a positive test result, and treating sensitivity and specificity as if they directly give the probability the client has the disorder.
Candidates conflate P(result | disorder) with P(disorder | result), and vignettes deliberately omit or bury prevalence so the intuitive answer is wrong.
What to do instead
Memorize the distinction between sensitivity/specificity (test properties) and positive/negative predictive value (depend on base rate). For any screening vignette, explicitly ask 'what is the base rate here?' and reason that low-prevalence settings produce many false positives even with good tests. Work through at least ten 2x2 table problems with varying prevalence until the direction of the effect is automatic.
Selecting the most severe or restrictive diagnosis (e.g., major depressive disorder, schizophrenia) for a vignette that actually meets criteria for a milder or adjacent condition, or missing that substance use or a medical condition better explains the symptoms.
Vignettes include salient severe symptoms and candidates anchor on them, skipping the differential and the exclusion criteria that Part 2 items test.
What to do instead
Use a fixed differential routine: list the 3-4 plausible diagnoses, then check duration, functional impairment, and exclusion criteria (substance, medical, other disorder) for each. Practice writing one sentence per vignette explaining why the runner-up diagnosis fails, which forces attention to the discriminating criterion.
Applying the ethics code as a rigid rulebook and picking the most absolute option (e.g., 'must report', 'must refuse') instead of the option that reflects the code's hierarchy and the least intrusive effective step.
Candidates memorize standards as prohibitions and do not rehearse the decision-making sequence for competing obligations such as confidentiality versus harm to others, or multiple relationships in small communities.
What to do instead
Learn a concrete decision model (identify stakeholders, relevant standards and law, consult, document, act, re-evaluate) and apply it to every ethics vignette. Prefer the option that protects the client while using the least restrictive means, and remember that consultation and documentation are almost always part of the correct sequence. Drill 20 vignettes where the 'obvious' absolute answer is a distractor.
Treating statistical significance as evidence of practical importance, or misreading p-values, confidence intervals, and effect sizes in research vignettes.
Training emphasizes null hypothesis significance testing, so candidates equate p < .05 with a large or meaningful effect and overlook overlap in confidence intervals.
What to do instead
For every research item, extract three things separately: effect size, confidence interval, and p-value. Practice stating in one sentence what each does and does not tell you, and practice identifying when non-overlapping versus overlapping confidence intervals do or do not support a group difference.
Confusing related developmental milestones and norms (e.g., age of first words versus phrases, object permanence versus conservation, typical versus atypical regression) when answering growth and lifespan items.
Milestone facts are learned as lists without a timeline, and vignettes present a single behavior without asking candidates to place it on a developmental sequence.
What to do instead
Build a one-page timeline with ages on one axis and domains (motor, language, cognitive, social) on the other, and add red-flag thresholds for regression. For each vignette, first state the child's age and expected level, then judge whether the described behavior is delayed, typical, or advanced before looking at options.
Attributing a group difference found in a cross-cultural or social psychology vignette to an internal trait while ignoring situational, cultural, or measurement explanations the item is testing.
The fundamental attribution error affects test-takers too, and candidates underweight acculturation, stereotype threat, test bias, and norming issues when they are the intended construct.
What to do instead
For any group-difference item, run a checklist: could this reflect measurement bias or norming, situational pressure, stereotype threat, or acculturation level? Practice identifying the named social-cognitive phenomenon in each stem before evaluating options, and prefer interpretations that consider context over dispositional ones unless the item specifies an individual-level question.
Answering neuropsychology and psychopharmacology items from vague associations (e.g., 'dopamine means psychosis') rather than localizing function or matching drug class to mechanism and side effect.
Biological content is memorized as isolated facts, and vignettes require linking lesion site or neurotransmitter system to a specific behavioral profile or adverse effect.
What to do instead
Create paired flashcards that link structure or transmitter to function and to a classic lesion or drug example (e.g., frontal lobe to executive function and disinhibition; SSRIs to serotonin syndrome signs). For medication vignettes, identify drug class, mechanism, and the most characteristic adverse effect before reading options.
Misallocating time and answering in strict order, so that lengthy vignettes early in a section consume time and later items are rushed or left blank.
Part 2 items are longer and more applied than fact-recall items, and candidates do not rehearse pacing or flag-and-return strategies under timed conditions.
What to do instead
Take at least three full-length timed practice sets and record time per item. Adopt a rule: if an item is not resolved within roughly its average allotment, choose the best current answer, flag it, and move on. Review flagged items only after completing the section, and track whether your errors cluster at the end of sections.
Skipping the stem's qualifiers such as 'FIRST', 'MOST likely', 'BEST', 'EXCEPT', or 'initially', and choosing an option that is true but does not answer the question asked.
Under time pressure candidates read the stem once quickly and pattern-match to a familiar concept, and qualifiers are the easiest words to lose.
What to do instead
Circle or mentally highlight qualifiers before reading options, and after selecting an answer, re-read the stem and confirm the option satisfies the exact qualifier. Practice with items that contain a true-but-wrong distractor and a correct answer that is only the first step, so you learn to rank options rather than merely recognize them.