Statistics

Statistics introduces data literacy through descriptive statistics, probability, discrete and continuous distributions, the central limit theorem, correlation, linear regression, and a conceptual introduction to confidence intervals and hypothesis testing — at a practical, accessible level without AP-depth derivations.

Who Should Take This

Ideal for high school juniors and seniors, college freshmen, and working professionals who need to read and communicate about data confidently. No calculus is required; basic algebra is sufficient. Participants finish ready to interpret statistical claims in news, research articles, and business reports.

What's Covered

1Descriptive Statistics

2Data Visualization and Interpretation

3Sampling and Study Design

4Probability

5Probability Distributions

6Introduction to Inference

7Correlation and Linear Regression

8Categorical Data Analysis

What's Included in AccelaStudy® AI

Adaptive Knowledge Graph

Practice Questions

Lesson Modules

Console Simulator Labs

Exam Tips & Strategy

13 Activity Formats

Course Outline

1Descriptive Statistics

8 topics

Describe measures of central tendency including arithmetic mean (sum divided by n), median (middle value after sorting), and mode (most frequent value), and identify which measure best summarizes different distribution shapes

Apply calculations of range, variance, and standard deviation by computing deviations from the mean, squaring them, averaging (for variance), and taking the square root (for standard deviation) of a data set

Apply the five-number summary (minimum, Q1, median, Q3, maximum) to construct box-and-whisker plots and identify outliers using the 1.5 × IQR fence rule

Apply z-score formula z = (x - μ)/σ to standardize individual data values, compare values from different distributions, and interpret z-scores as the number of standard deviations a value is from the mean

Describe distribution shapes as symmetric, left-skewed, right-skewed, uniform, or bimodal, and explain how skewness affects the relative positions of mean, median, and mode

Analyze the effect of outliers and data transformations (adding a constant or multiplying by a constant) on measures of center and spread, predicting which statistics change and by how much

Apply weighted mean calculations for grouped data and frequency-weighted means, and explain why the weighted mean equals the simple mean only when all group sizes are equal

Describe percentiles and their interpretation — the p-th percentile is the value below which p% of data fall — compute quartiles as the 25th, 50th, and 75th percentiles, and interpret an individual score's percentile rank

2Data Visualization and Interpretation

7 topics

Apply construction and interpretation of frequency tables, relative frequency tables, and cumulative frequency tables to summarize categorical and quantitative data sets

Apply interpretation of histograms by describing the shape, center, spread, and any unusual features, and distinguish histograms from bar charts by noting that histogram bars represent continuous intervals

Apply dot plots and stem-and-leaf plots to display small data sets, read individual values directly, and compare two distributions using back-to-back stem-and-leaf displays

Analyze two-way frequency tables to compute joint, marginal, and conditional frequencies, and interpret them to describe the association between two categorical variables

Analyze misleading data visualizations by identifying manipulated axis scales, cherry-picked time ranges, inappropriate chart types for data, and explain what presentation would more accurately convey the data

Apply cumulative frequency graph (ogive) construction and reading to determine percentiles and quartiles graphically, and explain how the ogive's slope at a point indicates local data density

Describe time series plots by identifying trends, seasonal patterns, and irregular fluctuations, and explain why seasonal adjustment is necessary before comparing monthly or quarterly data across time periods

3Sampling and Study Design

6 topics

Describe sampling methods including simple random sampling, systematic sampling, stratified sampling, cluster sampling, and convenience sampling, and identify potential sources of bias in each

Describe the difference between observational studies and controlled experiments, explain confounding variables, and identify why randomized controlled experiments provide stronger evidence for causation

Apply sample design evaluation to identify population, sample, parameter, and statistic in a study and explain how undercoverage, nonresponse, voluntary response bias, and question wording can produce unrepresentative samples

Analyze the effect of sample size on the variability of sample statistics, explain why larger samples produce more reliable estimates, and describe the trade-off between cost and precision in study design

Apply experimental design principles including control groups, placebo use, blinding (single- and double-blind), blocking, and randomization, and explain how each design element prevents specific confounding mechanisms

Analyze ethical issues in data collection including informed consent, privacy protection, and the risk of data misuse, and evaluate whether a described study design meets ethical standards for human and animal research

4Probability

8 topics

Describe the sample space, events, and basic probability axioms including P(S) = 1, P(A) ≥ 0, and the complement rule P(Aᶜ) = 1 - P(A), and compute probabilities for equally likely outcomes

Apply the addition rule P(A∪B) = P(A) + P(B) - P(A∩B) for general events and the simplified form for mutually exclusive events, and compute probabilities from Venn diagrams

Apply conditional probability P(A|B) = P(A∩B)/P(B) to compute updated probabilities given partial information and distinguish conditional probability from joint probability in problem setups

Apply Bayes' theorem P(A|B) = P(B|A)P(A)/P(B) to update prior probabilities with new evidence, using tree diagrams or two-way tables to organize medical screening and quality control calculations

Apply expected value E(X) = ΣxP(x) to evaluate fair games, insurance premiums, and decision-making under uncertainty, and compute expected value and variance of a discrete random variable

Analyze independence between two events by verifying P(A∩B) = P(A)·P(B) and contrast with the multiplication rule for dependent events, explaining why independence and mutual exclusivity are different concepts

Apply tree diagram construction to enumerate outcomes of multi-stage probability experiments and calculate probabilities by multiplying along branches, verifying that all branch probabilities at each node sum to 1

Describe the law of large numbers stating that the sample proportion of successes converges to the true probability as trial count grows, and explain why this does not imply that a sequence of losses makes a win more likely (gambler's fallacy)

5Probability Distributions

9 topics

Describe the binomial distribution conditions (fixed n, two outcomes, constant p, independent trials), compute P(X = k) = C(n,k)pᵏ(1-p)ⁿ⁻ᵏ, and calculate mean np and standard deviation √(np(1-p))

Apply binomial distribution calculations to find probabilities of exactly k, at least k, and at most k successes in quality control, survey response, and clinical trial contexts

Describe the normal distribution's bell curve shape, the empirical rule (68-95-99.7%), and how μ controls center while σ controls spread, identifying that area under the curve equals probability

Apply the standard normal table to find probabilities P(X < x), P(X > x), and P(a < X < b) for normal distributions by converting to z-scores and using cumulative area tables

Describe the Central Limit Theorem stating that sample means from any population with mean μ and std σ are approximately normally distributed with standard error σ/√n for large samples

Apply the Central Limit Theorem to compute probabilities for sample means by standardizing x̄ to z = (x̄ - μ)/(σ/√n) and using normal probability calculations for sampling distribution problems

Apply geometric distribution P(X = k) = (1-p)^(k-1)p for the number of trials until the first success, compute its mean 1/p, and use it to model quality inspection, customer wait times, and repeated independent trials

Apply the normal approximation to the binomial distribution for large n by checking np ≥ 10 and n(1-p) ≥ 10, using μ = np and σ = √(np(1-p)) to standardize and compute approximate binomial probabilities

Apply uniform distribution calculations on [a, b] by recognizing that the PDF is constant at 1/(b-a), compute P(c < X < d) = (d-c)/(b-a), and identify the mean as (a+b)/2 and standard deviation as (b-a)/√12

6Introduction to Inference

7 topics

Describe confidence intervals as a range of plausible values for a population parameter, interpret a 95% confidence interval correctly (not as probability the parameter is in the interval), and explain the margin of error

Apply confidence interval construction for a population proportion p̂ ± z√(p̂(1-p̂)/n) using z = 1.96 for 95%, identify conditions for validity (np̂ ≥ 10, independence), and interpret the result in context

Describe hypothesis testing by defining null and alternative hypotheses, explain the p-value as the probability of observing results as extreme as the sample if H₀ is true, and state the decision rule using significance level α

Describe Type I error (rejecting a true null hypothesis at rate α) and Type II error (failing to reject a false null hypothesis at rate β), and explain the trade-off between the two error types when setting α

Analyze p-values by correctly interpreting p = 0.03 vs p = 0.45 in a given context, distinguish statistical significance from practical significance, and explain why a small p-value does not prove the alternative hypothesis with certainty

Apply one-proportion z-test setup by writing H₀: p = p₀ and Hₐ: p ≠ p₀ (two-tailed), computing the test statistic z = (p̂ - p₀)/√(p₀(1-p₀)/n), and comparing to critical value ±1.96 for α = 0.05

Apply confidence interval for a population mean x̄ ± t*(s/√n) when σ is unknown, recognize that the t-distribution has heavier tails than the normal, and identify the correct degrees of freedom df = n - 1

7Correlation and Linear Regression

8 topics

Describe the Pearson correlation coefficient r as a measure of linear association strength and direction, interpret values of r near ±1 as strong and near 0 as weak, and explain that r measures only linear relationships

Apply the least-squares regression line ŷ = b₀ + b₁x by identifying slope b₁ = r(sy/sx) as predicted change in y per unit x, and y-intercept b₀ as the predicted y when x = 0

Apply residual analysis by computing residuals eᵢ = yᵢ - ŷᵢ, plotting them against x or ŷ, and evaluating whether patterns in a residual plot suggest the linear model is appropriate or a curved model is needed

Analyze the correlation vs causation distinction by describing lurking variables, Simpson's paradox examples, and why observational association between two variables does not establish that one causes the other

Apply the coefficient of determination r² to quantify the proportion of variability in y explained by the linear regression on x, and compare model fit across different scatter plots using r and r²

Analyze influential observations and leverage points by identifying how a single high-leverage point near the boundary of x values can dramatically change the slope of a regression line

Apply extrapolation vs interpolation caution by explaining that predictions within the observed range of x (interpolation) are more reliable than those outside (extrapolation), and identifying how far a prediction extends beyond the data range

Describe non-linear regression models (exponential and power) by identifying data patterns that suggest each model type, and apply logarithmic linearization to transform exponential data into linear form for regression analysis

8Categorical Data Analysis

7 topics

Apply relative frequency (proportion) calculations to summarize categorical data in bar charts and pie charts, comparing groups on the same response variable using side-by-side or segmented bar graphs

Apply marginal and conditional distribution extraction from two-way tables to compare the distribution of a response variable across levels of an explanatory variable and identify apparent associations

Analyze Simpson's paradox examples where a trend present in aggregated data reverses when data are broken into subgroups, explaining how lurking variables embedded in subgroup composition drive the reversal

Apply confidence interval calculations for a single proportion to determine sample size requirements for a desired margin of error, using the conservative estimate p̂ = 0.5 when no prior estimate is available

Analyze matched-pairs study design by explaining that pairing on confounding variables (age, weight, pre-existing condition) and computing differences within pairs removes between-subject variability, producing a more powerful comparison

Apply two-way table probability extraction to compute joint probability P(A ∩ B) = cell count / total, marginal probability P(A) = row total / grand total, and conditional probability P(A|B) = cell count / column total directly from a table

Describe observational vs experimental data using examples, and explain why drawing causal conclusions from observational studies requires ruling out confounders through design controls or sensitivity analysis

Scope

Included Topics

Descriptive statistics (mean, median, mode, range, standard deviation, variance, quartiles, IQR), data distributions and shapes (symmetric, skewed, uniform, bimodal), five-number summary and box plots, sampling methods (random, systematic, stratified, cluster, convenience) and bias, experimental vs observational study design, basic probability rules (complement, union, intersection, conditional), Bayes' theorem and updating probability, expected value for discrete distributions, discrete distributions (binomial, geometric), continuous distributions (uniform, normal — z-scores and standard normal table), central limit theorem for sample means, confidence intervals for proportions and means (conceptual), hypothesis testing introduction (null vs alternative hypothesis, p-value interpretation, Type I and II errors), correlation coefficients and scatter plots, linear regression (least-squares line, interpreting slope and intercept), residuals and model fit, reading and interpreting common data visualizations (histograms, dot plots, stemand-leaf, pie charts)

Not Covered

Chi-square tests, ANOVA, and multi-factor regression (AP Statistics level)
Bayesian inference beyond Bayes' theorem introduction
Non-parametric tests
Time series analysis and ARIMA models
Formal mathematical proofs of distribution theorems

Start Statistics — free

Adaptive learning that maps what you already know and closes the gaps. No credit card, no trial clock, no paywall.

Enroll Free

More free Foundational Academics courses

Statistics

Who Should Take This

What's Covered

What's Included in AccelaStudy® AI

Course Outline

Describe measures of central tendency including arithmetic mean (sum divided by n), median (middle value after sorting), and mode (most frequent value), and identify which measure best summarizes different distribution shapes

Apply calculations of range, variance, and standard deviation by computing deviations from the mean, squaring them, averaging (for variance), and taking the square root (for standard deviation) of a data set

Apply the five-number summary (minimum, Q1, median, Q3, maximum) to construct box-and-whisker plots and identify outliers using the 1.5 × IQR fence rule

Apply z-score formula z = (x - μ)/σ to standardize individual data values, compare values from different distributions, and interpret z-scores as the number of standard deviations a value is from the mean

Describe distribution shapes as symmetric, left-skewed, right-skewed, uniform, or bimodal, and explain how skewness affects the relative positions of mean, median, and mode

Analyze the effect of outliers and data transformations (adding a constant or multiplying by a constant) on measures of center and spread, predicting which statistics change and by how much

Apply weighted mean calculations for grouped data and frequency-weighted means, and explain why the weighted mean equals the simple mean only when all group sizes are equal

Describe percentiles and their interpretation — the p-th percentile is the value below which p% of data fall — compute quartiles as the 25th, 50th, and 75th percentiles, and interpret an individual score's percentile rank

Apply construction and interpretation of frequency tables, relative frequency tables, and cumulative frequency tables to summarize categorical and quantitative data sets

Apply interpretation of histograms by describing the shape, center, spread, and any unusual features, and distinguish histograms from bar charts by noting that histogram bars represent continuous intervals

Apply dot plots and stem-and-leaf plots to display small data sets, read individual values directly, and compare two distributions using back-to-back stem-and-leaf displays

Analyze two-way frequency tables to compute joint, marginal, and conditional frequencies, and interpret them to describe the association between two categorical variables

Analyze misleading data visualizations by identifying manipulated axis scales, cherry-picked time ranges, inappropriate chart types for data, and explain what presentation would more accurately convey the data

Apply cumulative frequency graph (ogive) construction and reading to determine percentiles and quartiles graphically, and explain how the ogive's slope at a point indicates local data density

Describe time series plots by identifying trends, seasonal patterns, and irregular fluctuations, and explain why seasonal adjustment is necessary before comparing monthly or quarterly data across time periods

Describe sampling methods including simple random sampling, systematic sampling, stratified sampling, cluster sampling, and convenience sampling, and identify potential sources of bias in each

Describe the difference between observational studies and controlled experiments, explain confounding variables, and identify why randomized controlled experiments provide stronger evidence for causation

Apply sample design evaluation to identify population, sample, parameter, and statistic in a study and explain how undercoverage, nonresponse, voluntary response bias, and question wording can produce unrepresentative samples

Analyze the effect of sample size on the variability of sample statistics, explain why larger samples produce more reliable estimates, and describe the trade-off between cost and precision in study design

Apply experimental design principles including control groups, placebo use, blinding (single- and double-blind), blocking, and randomization, and explain how each design element prevents specific confounding mechanisms

Analyze ethical issues in data collection including informed consent, privacy protection, and the risk of data misuse, and evaluate whether a described study design meets ethical standards for human and animal research

Describe the sample space, events, and basic probability axioms including P(S) = 1, P(A) ≥ 0, and the complement rule P(Aᶜ) = 1 - P(A), and compute probabilities for equally likely outcomes

Apply the addition rule P(A∪B) = P(A) + P(B) - P(A∩B) for general events and the simplified form for mutually exclusive events, and compute probabilities from Venn diagrams

Apply conditional probability P(A|B) = P(A∩B)/P(B) to compute updated probabilities given partial information and distinguish conditional probability from joint probability in problem setups

Apply Bayes' theorem P(A|B) = P(B|A)P(A)/P(B) to update prior probabilities with new evidence, using tree diagrams or two-way tables to organize medical screening and quality control calculations

Apply expected value E(X) = ΣxP(x) to evaluate fair games, insurance premiums, and decision-making under uncertainty, and compute expected value and variance of a discrete random variable

Analyze independence between two events by verifying P(A∩B) = P(A)·P(B) and contrast with the multiplication rule for dependent events, explaining why independence and mutual exclusivity are different concepts

Apply tree diagram construction to enumerate outcomes of multi-stage probability experiments and calculate probabilities by multiplying along branches, verifying that all branch probabilities at each node sum to 1

Describe the law of large numbers stating that the sample proportion of successes converges to the true probability as trial count grows, and explain why this does not imply that a sequence of losses makes a win more likely (gambler's fallacy)

Describe the binomial distribution conditions (fixed n, two outcomes, constant p, independent trials), compute P(X = k) = C(n,k)pᵏ(1-p)ⁿ⁻ᵏ, and calculate mean np and standard deviation √(np(1-p))

Apply binomial distribution calculations to find probabilities of exactly k, at least k, and at most k successes in quality control, survey response, and clinical trial contexts

Describe the normal distribution's bell curve shape, the empirical rule (68-95-99.7%), and how μ controls center while σ controls spread, identifying that area under the curve equals probability

Apply the standard normal table to find probabilities P(X < x), P(X > x), and P(a < X < b) for normal distributions by converting to z-scores and using cumulative area tables

Describe the Central Limit Theorem stating that sample means from any population with mean μ and std σ are approximately normally distributed with standard error σ/√n for large samples

Apply the Central Limit Theorem to compute probabilities for sample means by standardizing x̄ to z = (x̄ - μ)/(σ/√n) and using normal probability calculations for sampling distribution problems

Apply geometric distribution P(X = k) = (1-p)^(k-1)p for the number of trials until the first success, compute its mean 1/p, and use it to model quality inspection, customer wait times, and repeated independent trials

Apply the normal approximation to the binomial distribution for large n by checking np ≥ 10 and n(1-p) ≥ 10, using μ = np and σ = √(np(1-p)) to standardize and compute approximate binomial probabilities

Apply uniform distribution calculations on [a, b] by recognizing that the PDF is constant at 1/(b-a), compute P(c < X < d) = (d-c)/(b-a), and identify the mean as (a+b)/2 and standard deviation as (b-a)/√12

Describe confidence intervals as a range of plausible values for a population parameter, interpret a 95% confidence interval correctly (not as probability the parameter is in the interval), and explain the margin of error

Apply confidence interval construction for a population proportion p̂ ± z*√(p̂(1-p̂)/n) using z* = 1.96 for 95%, identify conditions for validity (np̂ ≥ 10, independence), and interpret the result in context

Describe hypothesis testing by defining null and alternative hypotheses, explain the p-value as the probability of observing results as extreme as the sample if H₀ is true, and state the decision rule using significance level α

Describe Type I error (rejecting a true null hypothesis at rate α) and Type II error (failing to reject a false null hypothesis at rate β), and explain the trade-off between the two error types when setting α

Analyze p-values by correctly interpreting p = 0.03 vs p = 0.45 in a given context, distinguish statistical significance from practical significance, and explain why a small p-value does not prove the alternative hypothesis with certainty

Apply one-proportion z-test setup by writing H₀: p = p₀ and Hₐ: p ≠ p₀ (two-tailed), computing the test statistic z = (p̂ - p₀)/√(p₀(1-p₀)/n), and comparing to critical value ±1.96 for α = 0.05

Apply confidence interval for a population mean x̄ ± t*(s/√n) when σ is unknown, recognize that the t-distribution has heavier tails than the normal, and identify the correct degrees of freedom df = n - 1

Describe the Pearson correlation coefficient r as a measure of linear association strength and direction, interpret values of r near ±1 as strong and near 0 as weak, and explain that r measures only linear relationships

Apply the least-squares regression line ŷ = b₀ + b₁x by identifying slope b₁ = r(sy/sx) as predicted change in y per unit x, and y-intercept b₀ as the predicted y when x = 0

Apply residual analysis by computing residuals eᵢ = yᵢ - ŷᵢ, plotting them against x or ŷ, and evaluating whether patterns in a residual plot suggest the linear model is appropriate or a curved model is needed

Analyze the correlation vs causation distinction by describing lurking variables, Simpson's paradox examples, and why observational association between two variables does not establish that one causes the other

Apply the coefficient of determination r² to quantify the proportion of variability in y explained by the linear regression on x, and compare model fit across different scatter plots using r and r²

Analyze influential observations and leverage points by identifying how a single high-leverage point near the boundary of x values can dramatically change the slope of a regression line

Apply extrapolation vs interpolation caution by explaining that predictions within the observed range of x (interpolation) are more reliable than those outside (extrapolation), and identifying how far a prediction extends beyond the data range

Describe non-linear regression models (exponential and power) by identifying data patterns that suggest each model type, and apply logarithmic linearization to transform exponential data into linear form for regression analysis

Apply relative frequency (proportion) calculations to summarize categorical data in bar charts and pie charts, comparing groups on the same response variable using side-by-side or segmented bar graphs

Apply marginal and conditional distribution extraction from two-way tables to compare the distribution of a response variable across levels of an explanatory variable and identify apparent associations

Analyze Simpson's paradox examples where a trend present in aggregated data reverses when data are broken into subgroups, explaining how lurking variables embedded in subgroup composition drive the reversal

Apply confidence interval calculations for a single proportion to determine sample size requirements for a desired margin of error, using the conservative estimate p̂ = 0.5 when no prior estimate is available

Analyze matched-pairs study design by explaining that pairing on confounding variables (age, weight, pre-existing condition) and computing differences within pairs removes between-subject variability, producing a more powerful comparison

Apply two-way table probability extraction to compute joint probability P(A ∩ B) = cell count / total, marginal probability P(A) = row total / grand total, and conditional probability P(A|B) = cell count / column total directly from a table

Describe observational vs experimental data using examples, and explain why drawing causal conclusions from observational studies requires ruling out confounders through design controls or sensitivity analysis

Scope

Included Topics

Not Covered

Start Statistics — free

Apply confidence interval construction for a population proportion p̂ ± z√(p̂(1-p̂)/n) using z = 1.96 for 95%, identify conditions for validity (np̂ ≥ 10, independence), and interpret the result in context