Null hypothesis
A null hypothesis is a statistical claim about a population or data-generating process that serves as a reference assumption and is evaluated by a hypothesis test, often asserting no effect or difference.
Statistics Every Programmer Needs
A null hypothesis is a specified default claim about a population or data-generating process that a hypothesis test evaluates evidence against.
- How it is used
- Provides the default claim against which p-values guide regression and distribution-testing decisions.
- How deep it goes
- Defines test-specific claims and interprets p-values against a chosen threshold, without developing a general theory of hypothesis testing.
- Examples
- Uses running-time regression, raisin classification, and Benford leading-digit counts to give distinct null claims.
- What you can do after
- Apply it via regression or chi-square p-values to decide whether to reject a stated claim at a predetermined threshold.
Read how Statistics Every Programmer Needs explains null hypothesis →
Statistics Slam Dunk
A null hypothesis is a statistical proposition describing a baseline condition such as no effect, difference, or association.
- How it is used
- Frames significance tests as decisions about whether observed differences are due to chance.
- How deep it goes
- Explains baseline claims, p-value thresholds, rejection decisions, and t-test inputs without formal proofs.
- Examples
- Uses NBA home-road fouls and free throws, salary Gini coefficients, collections treatments, drugs, exam scores, and branch-service ratings.
- What you can do after
- Apply it via R's t.test to decide whether two-group differences meet a predefined significance threshold.
Timeless Algorithms, The Foundational Ideas
A null hypothesis is a baseline statistical claim, commonly asserting that an effect or difference is absent.
- How it is used
- Uses the null hypothesis as the reference claim for decisions under uncertainty in the Neyman–Pearson framework.
- How deep it goes
- Develops hypothesis specification, significance thresholds, decisions, and error risks through worked examples.
- Examples
- Illustrates the null with beverage fill volumes, a clinical trial recovery rate, and a manufacturing defect rate.
- What you can do after
- Evaluate and understand test decisions without treating rejection or failure to reject as proof.
Read how Timeless Algorithms, The Foundational Ideas explains null hypothesis →
Grokking Statistics
A hypothesis stating that a claimed population value or relationship represents the status quo to be evaluated.
- How it is used
- Frames claims about population means, directional differences, and linear correlation for hypothesis testing.
- How deep it goes
- Defines H0, contrasts it with Ha, and applies rejection rules without deriving the underlying test proofs.
- Examples
- Uses soda volumes, tea-order guessing, 142-gram chip bags, and sports-drink sales versus temperature.
- What you can do after
- Apply it via SciPy to state H0, calculate a test result, and reject or fail to reject it at a chosen threshold.
Causal AI
A null hypothesis is a statistical claim, typically asserting the absence of an effect or association, that is evaluated by a hypothesis test.
- How it is used
- Frames statistical decisions about conditional independence and causal-model refutation through competing claims and p-value thresholds.
- How deep it goes
- Explains hypothesis direction, p-values, significance levels, and causal-testing limitations, but stops short of deriving test statistics.
- Examples
- Uses a transportation DAG, chi-squared tests, d-separation assertions, stock-price patterns, and DoWhy refuters.
- What you can do after
- Apply it via pgmpy or DoWhy to test conditional independences and assess whether causal assumptions are refuted.
Grokking Bayes
A null hypothesis is the hypothesis assigned a designated reference role in a classical statistical test, typically representing no effect or no difference.
- How it is used
- Frames experiments around a baseline claim before measuring whether observed results provide evidence of change.
- How deep it goes
- Defines the null model, test statistic, tail probability, and rejection rule without developing their mathematical derivations.
- Examples
- Uses a 500-visitor website experiment with 63 conversions against a 10% baseline.
- What you can do after
- Evaluate whether observed data are unusually extreme under a specified baseline using a classical hypothesis test.
Also covered in
What to know first
- varianceVariance is the expected squared deviation of a random variable from its mean and measures the dispersion of its values.
- false positiveA false positive is a positive result that incorrectly indicates a target condition is present when it is absent.
- meanThe mean is a measure of central location equal to the arithmetic average of numerical observations, or the probability-weighted average of a random variable’s possible values.
- p-valueA p-value is the probability, assuming a specified null hypothesis, of obtaining a test statistic or result at least as extreme as the observed one.
Related concepts
- alternative hypothesiscontrasts with · An alternative hypothesis is a statistical proposition that contrasts with the null hypothesis by asserting an effect, difference, relationship, or other departure.
- false positiveused by · A false positive is a positive result that incorrectly indicates a target condition is present when it is absent.
- p-valuemeasured by · A p-value is the probability, assuming a specified null hypothesis, of obtaining a test statistic or result at least as extreme as the observed one.
- hypothesiskind of · A hypothesis is a proposed proposition, explanation, prediction, or model whose validity or plausibility can be assessed against evidence.
- hypothesis testingused by · Hypothesis testing is a formal statistical procedure for evaluating evidence from observed data about a specified hypothesis, often a null hypothesis.
- binomial distributionused by · A binomial distribution is the probability distribution of the number of successes in a fixed number of independent trials, each with the same probability of success.
Questions about null hypothesis
What is null hypothesis?
A null hypothesis is a statistical claim about a population or data-generating process that serves as a reference assumption and is evaluated by a hypothesis test, often asserting no effect or difference.
Which Manning books explain null hypothesis?
Statistics Every Programmer Needs, Statistics Slam Dunk, Timeless Algorithms, The Foundational Ideas, Grokking Statistics, Causal AI and Grokking Bayes.
What should I know before learning null hypothesis?
Manning books that teach null hypothesis usually introduce variance, false positive, mean and p-value first.
What is the difference between null hypothesis and alternative hypothesis?
A null hypothesis is a statistical claim about a population or data-generating process that serves as a reference assumption and is evaluated by a hypothesis test, often asserting no effect or difference. An alternative hypothesis is a statistical proposition that contrasts with the null hypothesis by asserting an effect, difference, relationship, or other departure.
What is null hypothesis a kind of?
Null hypothesis is a kind of hypothesis. A hypothesis is a proposed proposition, explanation, prediction, or model whose validity or plausibility can be assessed against evidence.
What is null hypothesis also called?
Null hypothesis is also called 𝐻0.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.





