10.2 Last-Week Review Map
Key Takeaways
- The final week consolidates the eight CACREP domains; it is not the time to start new textbooks.
- Attack your weakest equally weighted domain first because every area is worth the same 20 items.
- Lock down the small set of formulas and definitions the CPCE actually tests (mean/median/mode, SD, reliability/validity, DSM basics).
- The day before: review logistics, high-yield error-log cues, and rest — no marathon cramming.
10.2 Last-Week Review Map
The final seven days are for consolidation, not discovery. Because the CPCE weights all eight CACREP core areas equally (20 items, 12.5% each), the highest-yield move is to raise your floor in your weakest domains rather than polish your strongest. Use three inputs to decide what to review: your eight domain subscores from practice tests, your error log, and the official CPCE Content Outline published by CCE.
A concrete seven-day map
| Day | Focus | Activity |
|---|---|---|
| Day 7–6 | Two weakest domains | Re-read core concepts, then a 40-item targeted drill; write one rule per miss |
| Day 5–4 | Next two weakest | Mixed 60–80 item sets; rotate domains to rebuild switching |
| Day 3 | Formulas + theorists | Active recall of the limited testable set (below) |
| Day 2 | Error-log cues | Re-test the exact items you missed; confirm the cue fires |
| Day 1 | Logistics + rest | Confirm ID, location/time, do light review only |
Do not add a new prep book in the last week. New resources at this stage scatter attention and inflate anxiety without raising your floor.
The high-yield facts the CPCE actually tests
The Assessment & Testing and Research & Program Evaluation domains reward a small, fixed set of definitions. Memorize these cold so you spend exam time reasoning, not deriving:
- Measures of central tendency: mean (average, pulled by outliers), median (middle, resistant to outliers), mode (most frequent). In a positively skewed distribution, mean > median > mode.
- Variability: range, variance, and standard deviation. About 68% of scores fall within ±1 SD, 95% within ±2 SD, 99.7% within ±3 SD of the mean (the normal curve).
- Standard scores: z-scores (mean 0, SD 1), T-scores (mean 50, SD 10), IQ/deviation scores (mean 100, SD 15), percentiles.
- Reliability vs validity: reliability = consistency (test-retest, internal consistency/Cronbach's alpha, inter-rater); validity = accuracy (content, criterion, construct). A test can be reliable but not valid; it cannot be valid without being reliable.
- Correlation: ranges −1.0 to +1.0; magnitude shows strength, sign shows direction; correlation does not establish causation.
Theorist and model crosswalk
The Counseling & Helping Relationships and Human Growth & Development domains hinge on matching models to people:
- Rogers — person-centered; unconditional positive regard, empathy, congruence.
- Beck — cognitive therapy; automatic thoughts and cognitive distortions.
- Ellis — REBT; A-B-C model and irrational beliefs.
- Glasser — reality/choice theory; WDEP.
- Erikson — eight psychosocial stages.
- Piaget — four cognitive stages (sensorimotor through formal operations).
- Kohlberg — moral development (preconventional, conventional, postconventional).
- Maslow — hierarchy of needs.
Ethics is open-and-shut points — if you know the source
Professional Counseling Orientation & Ethical Practice rewards knowing the ACA Code of Ethics structure. High-frequency anchors: maintain confidentiality except for duty-to-warn (Tarasoff), suspected abuse, or imminent harm; avoid harmful dual relationships; obtain informed consent; practice within your scope of competence; document accurately. When two answers seem ethical, choose the one that protects client welfare and leaves the cleanest documentation trail — the CPCE rewards the conservative, code-aligned action over the convenient one.
Taper and protect your floor
Plot your eight subscores on a simple grid each day; you are looking for the lowest bar to rise, not the tallest to grow. By Day 2, stop introducing material and start re-testing yourself on prior misses so the error-log cues become automatic. Sleep matters: a rested candidate recovers two to three borderline items that a cramming candidate misreads. The final week's goal is a calm, even profile across all eight CACREP areas — not a spike in one and a hole in another, because the equal weighting makes the hole cost just as much.
Finish Day 1 by confirming logistics with your program: the CPCE is administered through your university or testing site, your school sets the passing cut score (commonly a raw score around the institution's chosen threshold; there is no single national pass mark), and you should know your specific number before you walk in.
A one-page "brain dump" sheet to rehearse
Many candidates carry a single mental sheet they can mentally reconstruct in the first two minutes of the exam (on permitted scratch paper if allowed). Keep it to facts that are easy to forget under pressure and high-frequency on the CPCE:
- Normal curve: 68 / 95 / 99.7 within ±1, ±2, ±3 SD.
- Standard scores: z (0, 1); T (50, 10); IQ (100, 15); percentile rank.
- Skew rule: positive skew → mean > median > mode; negative skew reverses it.
- Reliability ≠ validity; reliability is necessary but not sufficient for validity.
- Erikson's eight stages in order; Piaget's four in order; Kohlberg's three levels.
- Group stages (Tuckman): forming, storming, norming, performing, adjourning.
- ACA confidentiality exceptions: duty to warn, suspected abuse, imminent harm, court order.
Rehearse reproducing this list from memory each of the last three days until it is automatic. Writing it down at the start of the exam frees working memory: instead of re-deriving a z-score formula mid-question, you glance at your sheet and spend the 84 seconds on reasoning. Stop the final-week review on a confident note — a calm, rested candidate who knows these anchors cold outperforms an exhausted one who studied an extra two hours the night before.
On a normal distribution, approximately what percentage of scores fall within one standard deviation of the mean?
A test consistently yields the same results across administrations but does not actually measure the construct it claims to. This test is best described as: