AP Statistics Last-Minute Review (Night Before)

title: "AP Statistics Last-Minute Review (Night Before)" description: "The night-before AP Statistics checklist: must-know formulas, conditions, calculator commands, common traps, and morning-of advice. Skim in 45 minutes." date: "2026-01-15" examDate: "May AP Exam" topics:

• Formula Sheet
• Common Traps
• Calculator Tips

The exam is tomorrow. This is not the time to learn new content — it's the time to skim, reset, and sleep. Spend 30-45 minutes on this page, then put your notes away.

Must-know formulas

| Quantity | Formula | |---|---| | $z$-score | $z = \\frac{x - \\mu}{\\sigma}$ | | Sample mean SE | $\\frac{\\sigma}{\\sqrt{n}}$ (or $\\frac{s}{\\sqrt{n}}$ for $t$) | | Sample proportion SE | $\\sqrt{\\frac{p(1-p)}{n}}$ (or $\\sqrt{\\frac{\\hat p (1 - \\hat p)}{n}}$) | | 1-prop $z$ test stat | $z = \\frac{\\hat p - p_0}{\\sqrt{p_0(1-p_0)/n}}$ | | 1-sample $t$ test stat | $t = \\frac{\\bar x - \\mu_0}{s/\\sqrt{n}}$, $df = n - 1$ | | 2-sample $t$ test stat | $t = \\frac{\\bar x_1 - \\bar x_2}{\\sqrt{s_1^2/n_1 + s_2^2/n_2}}$ | | Chi-square stat | $\\chi^2 = \\sum \\frac{(O - E)^2}{E}$ | | Binomial mean / sd | $\\mu = np$, $\\sigma = \\sqrt{np(1-p)}$ | | Geometric mean | $\\mu = 1/p$ | | LSRL slope | $b = r \\cdot \\frac{s_y}{s_x}$ | | Linear combinations | $E(aX + bY) = aE(X) + bE(Y)$; if independent, $\\text{Var}(aX + bY) = a^2 \\text{Var}(X) + b^2 \\text{Var}(Y)$ |

These (and more) are on the formula sheet provided with the exam — but knowing them cold saves time.

Conditions checklist

| Procedure | Conditions | |---|---| | 1-prop $z$ | Random; $10\\%$; Large Counts ($n\\hat p, n(1-\\hat p) \\geq 10$) | | 2-prop $z$ | Random (both); $10\\%$ (both); Large Counts (all 4 expected $\\geq 10$) | | 1-sample $t$ | Random; $10\\%$; Normal ($n \\geq 30$, OR pop stated normal, OR no skew/outliers in graph) | | 2-sample $t$ | Same as 1-sample, applied to both groups | | Paired $t$ | Random; $10\\%$; Normal applied to the differences | | Chi-square | Random; $10\\%$; All expected counts $\\geq 5$ | | LinReg $t$ | LINE: Linear, Independent residuals, Normal residuals, Equal SD |

The 4-step framework (every CI and test)

1. State — parameter in context + hypotheses (for tests).
2. Plan — name the procedure + check all conditions in context.
3. Do — test statistic + $p$-value (or CI bounds).
4. Conclude — compare $p$ to $\\alpha$, restate decision in context. Never say "accept $H_0$."

Top 10 traps that cost real points

1. "$p$-value is the probability $H_0$ is true." WRONG. It's conditional on $H_0$ being true.
2. "95% probability the true mean is in this interval." WRONG. The 95% refers to the procedure.
3. "Accept $H_0$" — say "fail to reject $H_0$" instead.
4. Skipping conditions — even if obvious, write them out for full points.
5. Random selection vs random assignment — random selection $\\to$ generalizable; random assignment $\\to$ causal.
6. Mixing up $\\sigma$ and $\\sigma_{\\bar x}$ — sampling distribution SD shrinks by $\\sqrt{n}$.
7. Forgetting $r^2$ interpretation in context — "$r^2 = 0.83$ means 83% of the variation in $y$ is explained by the linear model with $x$."
8. Wrong test choice — paired $t$ when data are matched, not 2-sample $t$.
9. Conditions on chi-square — must check ALL expected counts $\\geq 5$, not just observed.
10. Conclusions out of context — always restate in the language of the problem.

Calculator quick reference

• normalcdf(lower, upper, μ, σ): probability for normal distribution.
• invNorm(area, μ, σ): find the $x$ value with given left-tail area.
• binomcdf(n, p, k): $P(X \\leq k)$ for binomial.
• binompdf(n, p, k): $P(X = k)$ exactly.
• geomcdf(p, k) / geompdf(p, k): geometric.
• Tests: 1-PropZTest, 2-PropZTest, T-Test, 2-SampTTest, χ²-Test, χ²GOF-Test, LinRegTTest.
• CIs: 1-PropZInt, 2-PropZInt, TInterval, 2-SampTInt, LinRegTInt.
• Stat → Edit: enter data into L1, L2; 1-Var Stats L1 for summary stats.

Battery check tonight. Bring spares.

Score boundaries (recent years)

Approximate ranges out of 100 total points:

• 5: ~70+
• 4: ~58-69
• 3: ~45-57
• 2: ~33-44
• 1: below ~33

You only need to answer ~70% correctly to score a 5. You can get several MCQs wrong and miss parts of FRQs and still earn the top score. Don't panic if a problem stumps you.

Morning-of checklist

• ☐ 8 hours of sleep.
• ☐ Real breakfast (protein + slow carbs, not just sugar).
• ☐ 2 sharpened #2 pencils, blue/black pens.
• ☐ Approved calculator + spare batteries.
• ☐ Photo ID + AP ID label sheet.
• ☐ Water bottle, snack for the break.
• ☐ Watch (without alarm) if your room doesn't have a clock.
• ☐ Arrive 30 minutes early.

During the exam

1. MCQ: Mark and skip anything taking >90 seconds. Come back.
2. FRQ: Read all 6 first. Start with the one you're most confident on.
3. FRQ structure: Always State, Plan, Do, Conclude. Even if you're rushed, hitting these 4 sections earns partial credit on every problem.
4. Investigative task: Read it twice. Don't panic. The novel scenario combines familiar concepts.
5. Don't leave anything blank on FRQs. Even partial setups (defining the parameter, writing hypotheses, identifying the test) earn points.

One last thing

You've prepared. The work is done. Trust it. Show up rested, breathe between sections, and remember that the rubric wants to give you points — your job is to write clearly enough, in context, that it can.

Good luck. You've got this. 🎯

Need more focused review? Browse the AP Statistics topic library → or revisit the FRQ practice guide →.