I’m new to statistics
Start with OpenStax’s introductory text, which introduces statistics through collecting, analyzing, interpreting, and presenting data.
Start with: Introductory Statistics, Section 1.1
RedLectern topic
Introductory statistical reasoning: data, probability, inference, and models.
For: Learners at any starting point, from first-time students to people refreshing a specific skill.
After this page you can: Choose an appropriate introductory route and explain what a statistical result does and does not show.
Page reviewed
Part of: Mathematics
Statistics is the science of collecting, analyzing, interpreting, and presenting data. It helps summarize observed data and use probability to draw conclusions from samples about larger populations. A statistic estimates a population parameter, which is a numerical characteristic of the whole population.
Study it by moving from how data are gathered and summarized to probability, inference, and models. Learn to describe data with graphs and numerical summaries, then ask how sample-based estimates carry uncertainty and how relationships can be represented with regression. Practice interpreting results in context and checking how data were collected and what assumptions a method uses. You can follow these areas in order or choose the one that fits your current question.
Start with OpenStax’s introductory text, which introduces statistics through collecting, analyzing, interpreting, and presenting data.
Start with: Introductory Statistics, Section 1.1
Use MIT’s open course materials to revisit probability and statistics topics with readings, assignments, and exams.
Start with: Introduction to Probability and Statistics
Use OpenIntro’s chapter resources to go directly to topics such as data summaries, probability, inference, or regression.
Start with: OpenIntro Statistics
Descriptive statistics uses graphs and numerical summaries to organize data and show features such as center, spread, and shape.
Its chapter resources include numerical summaries and displays for quantitative and categorical data.
Median
The value that divides ordered data into two halves.
For the values 2, 4, and 6, find the mean and median.
The mean is 4 and the median is 4.
Probability assigns measures to possible outcomes and provides a mathematical way to reason about randomness.
This section defines probability, outcomes, events, sample spaces, and conditional probability.
Probability
A measure of how likely an outcome or event is, on a scale from zero to one.
Conditional probability
The probability of A calculated from the reduced sample space B.
A fair coin has two equally likely outcomes. What is the probability of heads on one toss?
The probability is one half, or 0.5.
Statistical inference uses sample data and probability to estimate population quantities and express uncertainty about those estimates.
The section explains how a sample statistic can estimate a population parameter and introduces inferential statistics.
Sample
A portion or subset of a population selected for study.
Confidence interval
A range of plausible values for a regression function, specified with a probability.
A survey uses a sample to estimate the average for a larger population. Name one reason to report an interval as well as the estimate.
An interval communicates uncertainty around the sample-based estimate of the population value.
Regression models describe relationships between variables and can be used to estimate responses, with uncertainty considered when interpreting predictions.
The course includes linear regression and offers lecture notes, problem sets, and exam materials.
The handbook explains why model predictions need uncertainty estimates and how confidence intervals are used.
Regression
A method for representing relationships between variables and predicting a response from explanatory variables.
A fitted line predicts 10 when x is 2; the observed response at x = 2 is 13. What is the residual, using observed minus predicted?
The residual is 3.
No recent updates are available.
David M. Diez, Mine Çetinkaya-Rundel, and Christopher D. Barr · 2019
The Open Textbook Library review notes its coverage from data collection and visualization through inference and logistic regression. A University of Texas syllabus names it as the primary textbook.
David Freedman, Robert Pisani, and Roger Purves · 2007
A Times Higher Education review describes it as an undergraduate introductory text. A University of Chicago course listed its fourth edition as required reading.
George Casella and Roger L. Berger · 2002
The Mathematical Association of America recommends it for undergraduate mathematics libraries. Stony Brook lists the second edition as required text for a mathematical statistics course.
John W. Tukey · 2019
NIST identifies Tukey’s original book as a seminal work in exploratory data analysis.
John A. Rice · 2006
MIT OpenCourseWare lists it as a textbook for a course covering statistical techniques used in science and industry.
Listed as Statistics courses: