7.1 Research Domain Map and Exam Logic

Why Research Reasoning Matters on Part 1

The Examination for Professional Practice in Psychology (EPPP) is a two-part exam in 2026. EPPP Part 1-Knowledge assesses general psychology knowledge from graduate training across eight content domains. Research Methods and Statistics is weighted at 7%, the smallest of the eight, but it is fully scored and it props up reasoning in nearly every other domain.

The other domains are Assessment and Diagnosis (16%), Ethical/Legal/Professional Issues (16%), Treatment/Intervention/Prevention/Supervision (15%), Cognitive-Affective Bases (13%), Growth and Lifespan Development (12%), Social and Multicultural Bases (11%), and Biological Bases (10%).

The January 2026 ASPPB Candidate Handbook fixes the logistics. Part 1 has 225 total items: 175 scored and 50 unscored pretest items, with a working time of 4 hours and 15 minutes, delivered at Pearson VUE test centers across the U.S. and Canada. Pretest items are seeded for future development and are not identified to the candidate. The recommended passing score is a scaled score of 500 on a 200-800 scale, which corresponds to roughly 65-70% correct on a typical form, though individual jurisdictions may set their own cutoff.

Do the arithmetic: 7% of 175 scored items is about 12 questions. That is enough to swing a borderline result, but it is not where most of the exam lives. The practical implication is to master high-yield concepts (validity threats, statistic selection, reliability vs. validity) rather than chasing obscure formulas. Because pretest items are invisible, answer every research stem as if it counts.

How Research Items Are Built

Research knowledge on the EPPP is rarely a bare calculation. A vignette may describe a study and ask which design was used (experimental, quasi-experimental, correlational, longitudinal, cross-sectional, single-case, or program evaluation), what conclusion is justified, which threat to validity is present, or which statistic fits the variables. Distractors are usually plausible-sounding but commit one of three errors: overstating causation, ignoring a sampling limit, or selecting a statistic that does not match the measurement scale.

A reliable exam habit is to ask four questions before hunting for a formula:

1. What is the research question? Description, association, prediction, or causation?
2. What variables were measured or manipulated, and at what scale (nominal, ordinal, interval, ratio)?
3. How were participants selected and assigned? Random selection affects generalization; random assignment affects internal validity.
4. What conclusion is the author trying to make, and does the design support it?

Those four questions usually reveal whether a causal claim is justified, whether a sampling bias weakens generalization, or whether a measure is too unreliable to anchor a strong interpretation.

Study Priorities

• Distinguish description, association, prediction, and causation.
• Connect each design feature to the validity threat it controls.
• Read tables and statistical results as evidence, not decoration.
• Separate clinical significance from a merely statistically detectable result.
• Treat evidence-based practice as the integration of research evidence, clinical expertise, and client characteristics.

Common trap: choosing the option with the most technical wording. A statistically dense distractor ("the hierarchical multiple regression confirmed mediation") is often wrong because the described design (a one-time correlational survey) cannot support mediation claims. The EPPP rewards the candidate who reads like a scientifically trained psychologist: choose answers that match the design, respect measurement limits, and interpret uncertainty accurately. The research domain also protects against overconfident practice.

A finding from a narrow sample may not generalize to a client with different cultural, developmental, medical, or service-setting characteristics; a result can be statistically detectable yet too small to change care.

A Worked Example of Exam Reasoning

Consider a representative item. "A community agency offers a new six-week parenting group. Forty parents who self-referred attend, complete a stress measure before and after, and show a statistically significant drop in stress. Which conclusion is best supported?" Walk the four-question routine. The research question is whether the group reduces parenting stress. The variable is parenting stress, measured twice. Participants self-selected, so there is no random selection and no random assignment. The design is a single-group pretest-posttest, a weak quasi-experimental structure.

The justified conclusion is that stress decreased among these self-referred attendees, not that the group caused the decrease, because history, maturation, regression to the mean, and self-selection are uncontrolled. The credited answer hedges causation; the trap answer claims the group works.

This pattern recurs across the domain. EPPP research stems reward the candidate who states the most defensible conclusion and rejects the most overreaching one. A short checklist for the test itself:

• Read the conclusion first, then test it against the design rather than reading the whole stem cold.
• Underline the design verbs: manipulated, assigned, randomly selected, measured, observed. These verbs distinguish experimental from correlational claims.
• Watch causal words: caused, produced, led to, increased. Reserve them for adequately controlled experiments.
• Prefer the cautious, qualified option when designs are weak, and the precise statistic when designs are sound.
• Do not be seduced by jargon; a technically worded distractor is still wrong if the design cannot support it.

Because the domain is only about 12 scored items, efficient candidates learn the high-yield concepts cold (validity threats, reliability versus validity, statistic selection, effect size versus significance, and evidence-based practice) and avoid sinking study time into rare procedures. The payoff is twofold: the dozen research items become near-automatic, and the same reasoning sharpens performance on Assessment, Treatment, and Ethics vignettes, where evaluating evidence quality is also rewarded. In short, treat this chapter as the analytic backbone of the whole exam, not an isolated 7% silo.