Interactive Statistical Tools
Experience statistics through hands-on simulation and visualization — from foundational concepts to advanced methods, these interactive tools transform abstract theory into tangible understanding.
← Return to HomepageFoundations and Intuition Builders
These tools make the most difficult conceptual ideas in introductory statistics visible: sampling distributions, confidence intervals, p-values, and the Central Limit Theorem. Ideal for in-class demonstrations and self-study.
🎓 Hypothesis Testing Logic
NewUnderstand the logical foundation of hypothesis testing through visual demonstrations of null and alternative hypotheses, p-values, and Type I and Type II errors.
- Visual explanation of null and alternative hypotheses
- p-value interpretation made clear
- Type I and Type II errors visualized
- Interactive decision-making scenarios
📖 Best used with: Chapter 10
🎯 CLT for Sample Means (x̄)
LiveObserve how sampling distributions of sample means emerge through simulation. This interactive tool demonstrates the Central Limit Theorem for quantitative data through repeated sampling.
- Sample from real-world datasets (income, test scores, heights)
- Observe normalization as sample size increases
- Adjust sample size and observe convergence to normal distribution
- Verify CLT: μx̄ = μ and σx̄ = σ/√n
📖 Best used with: Chapters 8, 9
🎯 CLT for Sample Proportions (p̂)
NewUnderstand how sampling distributions of proportions behave through simulation. This tool applies the Central Limit Theorem to categorical data, relevant for polls, quality control, and medical studies.
- Real-world scenarios: voter turnout, college degrees, health insurance
- Observe p̂ sampling distribution construction
- Verify normality conditions: np ≥ 10 and n(1−p) ≥ 10
- Confirm CLT: μp̂ = p and σp̂ = √(p(1−p)/n)
- Based on Census Bureau and Pew Research data
📖 Best used with: Chapters 8, 9
📊 Sampling Distributions
NewDistinguish between population distribution, sample distribution, and sampling distribution of the mean through simultaneous visualization of all three concepts.
- Three distributions displayed simultaneously
- Interactive parameter adjustment
- Observe standard error behavior
📖 Best used with: Chapter 8, Appendix B
📈 Confidence Intervals
LiveSee what “95% confidence” REALLY means through dynamic simulation. Generate hundreds of confidence intervals and observe the proportion that capture the true parameter.
- Visual demonstration of confidence level
- See both successful and failed intervals
- Adjust confidence level and sample size
- Understand the frequentist interpretation
📖 Best used with: Chapter 9
⚡ p-Value Simulator
LiveUnderstand what p-values really mean through interactive simulation. Generate data under the null hypothesis and see how often you get results as extreme as yours.
- Simulate null distribution in real-time
- Visualize your test statistic's extremeness
- Adjust effect size and sample size
- See Type I error rate empirically
📖 Best used with: Chapter 10
🎭 The Complete Dance: P-Values vs Confidence Intervals
NewA response to Geoff Cumming's famous "Dance of the p-Values" argument. This tool reveals the mathematical truth: the "dance" is not a defect — it's the very mechanism by which Type I error is controlled! Watch p-values AND confidence intervals dance side by side, and see that BOTH are perfectly calibrated.
- Side-by-side visualization: p-values vs confidence intervals dancing together
- Real-time calibration proof: rejection rate converges to α, coverage to 1−α
- Histogram showing p-values follow Uniform(0,1) under H₀
- Power analysis: see why "replication failure" reflects power, not p-value pathology
- Mathematical comparison table exposing the logical inconsistency in Cumming's argument
- Effect size scenarios: observe how power affects "significance instability"
📖 Best used with: Chapters 9, 10, 19 (New Statistics discussion)
📊 Distribution Explorer
LiveVisualize and manipulate probability distributions to build intuition. Explore Normal, Binomial, Poisson, Geometric, Exponential, Weibull, and more with interactive parameter controls.
- 12 probability distributions (3 discrete, 9 continuous)
- Real-time PDF/PMF and CDF visualization
- Parameter sliders with instant updates
- Mathematical formulas with real-world examples
- Calculate probabilities for any region
📖 Best used with: Chapters 6, 7, Appendix A
📊 Anscombe's Quartet & Datasaurus Dozen
NewOne of the most interesting demonstration in statistics! See how IDENTICAL summary statistics can hide COMPLETELY different data patterns. This single tool will change how you think about data analysis.
- Anscombe's Quartet (1973): four datasets with identical means, SDs, correlations
- Datasaurus Dozen (2017): 13 shapes with nearly identical statistics
- Illustrates why graphical exploration should precede numerical summary
- Interactive explorer to compare datasets side-by-side
- Real-world examples of why this matters in research, business, and medicine
📖 Best used with: Chapters 2, 14, 15
Data Analysis Tools
These tools resemble statistical software and calculators. They bridge textbook examples to practical data analysis, helping students learn how real analyses are done in practice.
🎓 FISHER-25 Statistical Calculator
LiveHonoring Sir Ronald A. Fisher's 100-year anniversary (1925–2025) of Statistical Methods for Research Workers — the book that launched modern statistics! A comprehensive, professional-grade calculator with ALL essential statistical procedures, beautiful visualizations, and Fisher's complete legacy documented within.
- Complete statistical suite: descriptive stats, hypothesis tests, confidence intervals
- Correlation & regression analysis with visual scatter plots
- Histogram, box plot, QQ plot, and essential visualizations
- Six data columns (C1–C6) with predefined datasets including Anscombe's Quartet
- Reference material on Fisher's contributions to modern statistics
📖 Best used with: Chapters 2–4, 9–17
🌟 Golden Statistical Decision Tree
CompleteRevolutionary intelligent guidance system that asks YOU about your data, then selects and performs the optimal statistical procedure. No more guessing which test to use — answer simple questions about your samples and research design, and let the Golden Tree guide you to the correct method every time.
- 16 statistical methods: comprehensive coverage
- Intelligent question-based navigation system
- Automatic method selection based on your data structure
- Performs the complete analysis with proper interpretation
- Teaches correct statistical thinking through the process
📖 Best used with: Chapters 10–13
🔄 Bootstrap Resampling: The Ultimate Tool
NewA magnificent tribute to Bradley Efron's revolutionary 1979 bootstrap innovation! Experience resampling with replacement, watch bootstrap distributions emerge in real time, and see why this method transformed statistical inference. Hover over Efron's portrait to discover his legendary contributions.
- Interactive portrait: hover to see Efron's contributions to statistical science
- Four CI methods: percentile, basic, standard error, and BCa
- Enter your own data or use examples from textbook Chapter 18
- Bootstrap any statistic: mean, median, SD, quartiles
- Visual demonstration of “with replacement” sampling
- Observe bootstrap distribution construction with green-shaded confidence intervals
📖 Best used with: Chapter 18
📉 Regression Playground
LiveFit regression models and see how assumptions affect results. Add outliers, adjust heteroscedasticity, and understand diagnostic plots interactively.
- Linear and polynomial regression
- Interactive residual plots
- Assumption violation scenarios
- Model comparison and validation
📖 Best used with: Chapters 15, 16
🎯 Power Analysis Pro – Enhanced
NewUnderstand α, β, and statistical power through interactive visualization. See null and alternative distributions overlap in real time, calculate sample sizes, and compare scenarios side by side.
- Power visualizer: observe α, β, and power regions change as you adjust parameters
- Effect size tool: see what small, medium, large effects actually look like
- Full calculator: multiple tests (t-tests, proportions, ANOVA, correlation)
- Scenario comparison: compare several study designs side by side
📖 Best used with: Chapters 10, 11
Bayesian–Frequentist Comparison Tools
These tools show both paradigms side by side: confidence intervals vs. credible intervals, p-values vs. Bayes factors. They support Chapter 19’s goal of understanding connections rather than promoting “statistics wars”.
🎯 Bayesian–Frequentist Comparison: Mean Estimation
NewRevolutionary tool honoring Thomas Bayes! Compare frequentist confidence intervals with Bayesian credible intervals (HPD) side by side. See both statistical paradigms in action — the future of statistical software is harmonizing these two approaches. Works with σ known or unknown.
- Two paradigms, one interface — frequentist CI vs Bayesian HPD interval
- Thomas Bayes portrait — hover for biographical information about Thomas Bayes
- Enter raw data or summary statistics
- Choose σ known (normal) or σ unknown (t distribution)
- Adjust Bayesian prior strength interactively
- Posterior distribution visualization with intervals overlaid
📖 Best used with: Chapters 9, 19
🎯 Bayesian–Frequentist Comparison: Proportion Estimation
NewComplete the paradigm comparison! Estimate population proportions using BOTH frequentist and Bayesian approaches. Perfect companion to the mean estimation tool — see how confidence intervals and credible intervals compare for categorical data. Essential for understanding polls, quality control, and medical studies.
- Two paradigms for proportions — frequentist CI vs Bayesian credible interval
- Beta–Binomial Bayesian analysis with adjustable prior
- Automatic Wilson score interval on the frequentist side
- Posterior Beta distribution visualization
- See the paradigms harmonize for large samples
📖 Best used with: Chapters 9, 19
🎯 Bayesian–Frequentist Comparison: One-Sample t Test
NewCompare the frequentist one-sample t test with a Bayesian t test side by side. Featuring portraits of William Gosset (Student's t) and Thomas Bayes. Demonstrates how p-values and Bayes factors provide different summary measures of statistical evidence.
- Dual hypothesis testing — frequentist t test vs Bayesian test in one interface
- Historical portraits honoring Gosset and Bayes
- Enter raw data or summary statistics
- Set null hypothesis value (μ₀) interactively
- Get p-value and Bayes factor simultaneously
- Visual comparison of test results and interpretations
📖 Best used with: Chapters 10, 19