AP Statistics TI-84 and Desmos Cheat Sheet (2027)
Everything you need at the keyboard, on one printable page: which procedure to pick, where it lives on the TI-84, and what to type in the Bluebook Desmos calculator. Built for the revised 2026–27 course, so removed topics are left out. For the explanations behind each row, read the full calculator guide.
1. Pick the procedure
| You have… | Use | TI-84 (STAT ▸ TESTS) | Desmos |
|---|---|---|---|
| One proportion (x successes in n) | 1-proportion z | 1-PropZTest or 1-PropZInt | zproptest(x,n) |
| Two proportions | 2-proportion z | 2-PropZTest or 2-PropZInt | zproptest(x1,n1,x2,n2) |
| One mean, or paired differences | 1-sample t | T-Test or TInterval | ttest(L) or ttest(n,x̄,s); paired: ttest(L1-L2) |
| Two independent means | 2-sample t | 2-SampTTest or 2-SampTInt, Pooled: No | ttest(L1,L2) or ttest(n1,x̄1,s1,n2,x̄2,s2) |
| Counts in a two-way table | χ² homogeneity / independence | χ²-Test with matrix [A] | chisqtest(col1,col2,…) |
| Two quantitative variables | Linear regression | LinReg(a+bx) L1, L2 | Table, then y1 ~ m x1 + b |
Desmos opens Confidence Interval and Hypothesis Test panels under each of these. Type your null value in the Hypothesis Test panel and pick Left, Both or Right.
2. Distributions
| You need | TI-84 (DISTR = 2nd VARS) | Desmos |
|---|---|---|
| Normal area between a and b | normalcdf(a, b, μ, σ) | normaldist(μ,σ).cdf(a,b) |
| Normal value with area p to its left | invNorm(p, μ, σ) | normaldist(μ,σ).inversecdf(p) |
| t area between a and b | tcdf(a, b, df) | tdist(df).cdf(a,b) |
| t critical value (area p to the left) | invT(p, df) | tdist(df).inversecdf(p) |
| Binomial P(X = k) | binompdf(n, p, k) | binomialdist(n,p).pdf(k) |
| Binomial P(X ≤ k) | binomcdf(n, p, k) | binomialdist(n,p).cdf(k) |
| Chi-square area between a and b | χ²cdf(a, b, df) | chisqdist(df).cdf(a,b) |
On the TI-84, type 1E99 for ∞ and −1E99 for −∞ (the E comes from 2nd then the comma key). For a C% interval the t or z critical value has area (1 + C)/2 to its left.
3. Summary statistics and regression
| You need | TI-84 | Desmos |
|---|---|---|
| Everything at once | STAT ▸ CALC ▸ 1-Var Stats | stats(L) |
| Sample standard deviation | Sx | stdev(L) |
| Quartiles Q1, Q3 | Q1, Q3 in 1-Var Stats | quartile(L,1), quartile(L,3) |
| Least-squares line | LinReg(a+bx) | y1 ~ m x1 + b |
| Correlation r | Turn Stat Diagnostics on first | corr(L1,L2) |
| Residuals | 2nd STAT ▸ NAMES ▸ RESID | Created automatically as e1 |
4. Conditions to check before any inference
- Every procedure: random sample or random assignment, and independent observations (sampling without replacement: n is at most 10% of the population).
- Proportions: at least 10 expected successes and 10 expected failures (use p₀ for a test, p̂ for an interval; check each group for two proportions).
- Means (t): the population is roughly normal, or n is large (about 30 or more), or a small sample shows no strong skew or outliers. For paired data, check the differences.
- Chi-square: every expected count is at least 5.
- Regression: a roughly linear scatterplot and a residual plot with no pattern.
5. Traps that cost points
- Desmos
quartile()matches the AP method.quantile()does not. - Desmos
inversecdftakes the area to the left only. - The Desmos Hypothesis Test null box starts at 0. Type your own null value or the p-value is wrong.
- In a two-proportion test, the z statistic uses the pooled proportion but Desmos
.stderris the unpooled value used for the interval. - Both calculators give a decimal df for a two-sample t. That is fine. By hand, use min(n₁−1, n₂−1).
- For a very small p-value Desmos shows “< 0.0001”. Write p < 0.0001, not p = 0.
- For a chi-square test on the TI-84, the expected counts are stored in matrix [B]. You need them for the conditions.
6. The free-response frame
- Name the procedure and the parameter, in context.
- Check the conditions using the numbers from the problem.
- Give the test statistic, df if any, and p-value (or the interval). Label probability work: write the distribution with its parameters, as in
normaldist(mean = 109, standard deviation = 16)or binomcdf(20, 0.6, 14). - Conclude in context. Compare p to α, then say what it means about the population. For an interval: “We are C% confident that the interval from __ to __ captures the true __.”
As an Amazon Associate, MathInSite earns from qualifying purchases.
Check any answer step by step with the confidence interval & hypothesis test calculator or the regression calculator, find formulas on the statistics formulas cheat sheet, and see which calculator to bring.
Desmos syntax tested in the College Board practice calculator. Updated for the May 2027 exam. Last reviewed October 2026.
Frequently Asked Questions
-
Where are the hypothesis tests on a TI-84?
Press STAT, then arrow right to TESTS. The z-tests, t-tests, proportion tests, intervals and the chi-square test are all in that menu. Probability distributions such as normalcdf and binomcdf are in the DISTR menu, which you open with 2nd then VARS.
-
Can I bring a printed cheat sheet to the AP Statistics exam?
No. College Board supplies a formula and table reference during the exam, and personal notes are not allowed in the exam room. Use this sheet to study and to learn where each function lives on your calculator.
-
Which calculator functions were removed from AP Statistics for 2027?
The chi-square goodness-of-fit test, the t-test for a regression slope, and the geometric distribution were removed from the course. The calculators still have those tools, but they are no longer tested.