For class pages, LMS, and blogs
Put a stats demo on a class page
Paste one iframe to put a working StatsLearn sandbox on Google Sites, Canvas, WordPress or a class page. Start with the descriptive statistics sandbox. Students need no account. Calculators are here too if you want the steps shown next to the number.
Each interactive and calculator has its own widget page with a live preview and the snippet. Host steps live under embed in Google Sites, Canvas, Schoology, Moodle, WordPress and Notion. Lesson packs group a unit. Assistants should read how to embed StatsLearn. See also mean vs median.
Live embed of the sandbox
This is the widget as it appears on a class page, caption included if you keep the snippet whole. Drag a point and watch the mean chase it.
Start with these
Seven widgets that cover a mean you can drag, a z-score, a sampling distribution, a normal curve, coverage you can count, power, and a regression line that swings. Open a widget page for its copy box. The directory below lists every tool.
Descriptive statistics sandbox
Drag data points on a dotplot and watch the mean chase outliers while the median holds still. The boxplot, SD, and IQR update live.
Z-score calculator
A z-score calculator converts a raw value, mean, and SD into z = (x - mu) / sigma, then into a percentile. For 82, 70, 8 you get z = 1.5 and 93.32%.
Sampling distribution and CLT
Pick a lumpy parent population, slide the sample size, draw thousands of samples, and watch the distribution of sample means turn normal and narrow by root n.
Normal curve explorer
Drag the boundaries on a normal curve to shade a tail or a middle band, and read the probability, the z-scores, and the percentiles as you go. Lock the z-scores and the area holds while the mean and SD move.
CI coverage simulator
Draw hundreds of samples, build a confidence interval from each, and count how many actually capture the true parameter. What 95 percent confidence means, as a tally rather than a sentence.
Type I, Type II, and power
Slide the true parameter, the sample size, and alpha, and watch the two error regions trade off against each other. Power is the area you are trying to grow, and it is drawn.
Regression influential point
Drag one high-leverage point and watch the least-squares line, r, and the residual plot swing. Influence, made physical.
Lesson packs
A unit is several widgets on one class page, not one demo standing in for the course.
- Exploring data packDrag-a-point sandbox, a column analyzer, and a five-number summary. For the first week of AP Statistics, when mean versus median has to be seen.
- Sampling and CLT packSampling-distribution visualizer, sample size, and margin of error. For the unit where the mean of means goes normal.
- Inference packZ-score calculator, confidence-interval coverage, Type I/II/power, and a p-value calculator. For the inference units, with the coverage students can count.
Paste it here
- Google SitesPaste a StatsLearn sandbox or calculator into Google Sites with the Embed code block. Students drag a point on the class page. No account.
- CanvasAdd a StatsLearn sandbox or calculator to a Canvas page through the HTML editor. Height 560 for interactives, 520 for calculators.
- SchoologyPut a StatsLearn demo on a Schoology material page by pasting the iframe into an HTML or embed widget, not only the rich-text box.
- MoodleAdd a StatsLearn demo to a Moodle page or book by allowing HTML and pasting the iframe. Caption links stay so other teachers can find the tool.
- WordPressDrop a StatsLearn demo into a WordPress post with a Custom HTML block. The caption is two ordinary links, which is the part search engines can count.
- NotionAdd a StatsLearn demo to a Notion page with /embed and the iframe URL, or paste the HTML if your workspace allows it. Good for a lesson notes page.
Every interactive widget
8 interactive pages. Open one for the iframe. The catalog for the embed-graphs query lives on embed graphs if you already have that search.
- Which test chooserAnswer two or three questions about your data and design, and the right procedure falls out with the calculator to run it.
- Descriptive statistics calculatorPaste a column of numbers and get every summary statistic at once, with the boxplot and histogram drawn on a shared scale.
- Descriptive statistics sandboxDrag data points on a dotplot and watch the mean chase outliers while the median holds still. The boxplot, SD, and IQR update live.
- Sampling distribution and CLTPick a lumpy parent population, slide the sample size, draw thousands of samples, and watch the distribution of sample means turn normal and narrow by root n.
- Normal curve explorerDrag the boundaries on a normal curve to shade a tail or a middle band, and read the probability, the z-scores, and the percentiles as you go. Lock the z-scores and the area holds while the mean and SD move.
- CI coverage simulatorDraw hundreds of samples, build a confidence interval from each, and count how many actually capture the true parameter. What 95 percent confidence means, as a tally rather than a sentence.
- Type I, Type II, and powerSlide the true parameter, the sample size, and alpha, and watch the two error regions trade off against each other. Power is the area you are trying to grow, and it is drawn.
- Regression influential pointDrag one high-leverage point and watch the least-squares line, r, and the residual plot swing. Influence, made physical.
Teaching calculators
16 calculators that show the method. StatsLearn is interactives first. These are the step tools. Each still has its own widget page.
- Binomial probability calculatorFind P(X = x), P(X >= 4), or P(X <= 3) for a binomial distribution. Enter n, p, and x. Worked example, mean and SD, and the BINS conditions checklist.
- Chi-square calculatorA chi-square test calculator finds expected counts, chi^2 = sum (O-E)^2/E, and df. A worked 2x2 table gives chi^2 = 14.0625 with df = 1.
- Confidence interval calculatorFind a confidence interval for a mean or proportion. Enter your sample data to get the interval, margin of error, and z* or t* critical value.
- Empirical rule calculatorAn empirical rule calculator finds the 68-95-99.7 intervals of a normal distribution. With mean 100 and SD 15 you get 85-115, 70-130, and 55-145.
- Expected value calculatorFree expected value calculator for a discrete random variable. Enter outcomes and probabilities to get the mean and standard deviation, with steps.
- Five-number summary calculatorA five-number summary calculator finds the min, Q1, median, Q3, and max, then the IQR and 1.5 x IQR fences to flag outliers and draw a boxplot.
- Margin of error calculatorMargin of error is a critical value times the standard error. Learn the proportion formula ME = z*sqrt(p-hat(1-p-hat)/n) with a worked poll example.
- Normal distribution calculatorFind the area under the normal curve between any bounds. Enter the mean and standard deviation, then get left, right, between, or outside probabilities.
- One-sample t-test calculatorFree one sample t test calculator. Enter your sample mean, standard deviation, and size to get t, degrees of freedom, and a p-value, with a worked example.
- P-value calculatorConvert a z, t, or chi-square test statistic to a p-value: one- or two-tailed for z and t, right tail for chi-square, with worked examples.
- Proportion z-test calculatorRun a one or two proportion z test. The one-sample SE uses p0; the two-sample SE pools the proportions. See the formulas, conditions, and a worked example.
- Linear regression calculatorPaste your x,y pairs to get the correlation coefficient r, r-squared, the least-squares line y-hat = a + bx, and each residual, with a worked example.
- Sample size calculatorFind the sample size for a target margin of error. For a proportion, n = (z*/ME)^2 times p*(1-p*) with p* = 0.5; round up. Worked AP examples.
- Standard deviation calculatorA standard deviation and variance calculator squares each deviation, then divides by n-1 (sample) or n (population). For 4, 7, 8, 10, 16, s ~= 4.47.
- Two-sample t-test calculatorCompare two independent means with a Welch two-sample t-test. Get the standard error, t statistic, degrees of freedom, and p-value, with a worked example.
- Z-score calculatorA z-score calculator converts a raw value, mean, and SD into z = (x - mu) / sigma, then into a percentile. For 82, 70, 8 you get z = 1.5 and 93.32%.
Classroom steps, including heights, are on the teachers page. Embeds are free for educational use. Keep the caption: attribution is required.