Practice probability, data analysis, distributions, hypothesis testing, and statistical inference with free online tests from easy to harder levels.
Statistics and probability connect random events with real data. The practice sequence moves from probability rules and descriptive analysis to distributions, hypothesis testing, and statistical inference — the steps used to decide what sample evidence can tell us about a larger population.
Each test focuses on a different point in the journey from uncertainty to a defensible statistical conclusion.
Begin with uncertainty itself. Practice sample spaces, basic probability rules, independent events, complements, and counting situations while learning to distinguish possible outcomes from likely outcomes.
Move from possible outcomes to observed information. Practice reading tables and charts, comparing data sets, and interpreting mean, median, mode, range, center, and spread rather than treating those quantities as isolated calculations.
Study how values are arranged rather than looking at one observation at a time. Practice normal, binomial, and Poisson-style ideas together with center, spread, z-scores, and the behavior of values across a distribution.
Ask whether the evidence is strong enough to challenge a claim. Practice null and alternative hypotheses, p-values, significance levels, Type I and Type II errors, and making statistical decisions without overstating what the test proves.
Use sample information to reason about a population. Practice estimation, confidence intervals, sampling distributions, standard error, and interpreting how much uncertainty remains when conclusions are based on sample data.
The arithmetic is only one part of the reasoning. A useful solution connects the claim, the sample evidence, the probability of observing that evidence, and the conclusion justified by the chosen significance level.
Identify the null hypothesis and the alternative hypothesis before using the sample results.
Evaluate the sample result under the assumption that the null hypothesis is true.
Use the p-value and the chosen threshold to determine whether the evidence is sufficiently unusual.
Describe what the evidence supports without claiming certainty beyond what the statistical procedure can establish.
Mean, median, spread, tables, and charts summarize the sample or data set in front of you. They organize what was observed without automatically extending the result to a larger population.
Confidence intervals, hypothesis tests, standard errors, and sampling distributions help quantify uncertainty when a sample is used to draw conclusions about a population.
Statistics can produce precise-looking calculations from a poorly interpreted problem. Check the reasoning as well as the arithmetic.
A probability, mean, z-score, standard error, and p-value answer very different questions.
Keep the observed sample statistic separate from the population quantity you may be trying to estimate.
Probability and sampling variability explain why different samples can produce different results.
Statistical evidence supports conclusions with uncertainty; it does not usually establish absolute certainty.
When two samples or groups look different, the important question is whether the difference is large enough to be meaningful compared with the amount of variation that could occur naturally.
Short explanations for ideas that often look similar in formulas but mean very different things statistically.
Probability starts with a model or assumptions and studies the outcomes that could occur. Statistics starts with observed data and uses those observations to describe patterns or draw conclusions about a larger process or population.
The mean is found by adding the observations and dividing by the number of observations. The median is the middle value after the data are ordered. An extreme value can affect the mean much more strongly than the median.
A z-score describes the location of a value relative to the mean in units of standard deviation. A positive z-score places the value above the mean, while a negative z-score places it below the mean.
A p-value measures how unusual the observed result, or a result at least as extreme, would be under the assumptions of the null hypothesis. A smaller p-value indicates stronger evidence against the null hypothesis.
No. A hypothesis test evaluates evidence under a statistical model. A small p-value can provide evidence against the null hypothesis, but it does not establish the alternative with absolute certainty.
A Type I error occurs when the null hypothesis is rejected even though it is actually true. The chosen significance level is connected to controlling the probability of this type of error.
A confidence interval uses sample information to give a range of plausible values for an unknown population parameter. Its interpretation depends on the statistical procedure and confidence level used to construct it.
Standard error describes the variability of a sample statistic across repeated samples. A smaller standard error generally means the statistic provides a more precise estimate of the corresponding population quantity.
A distribution describes how values or probabilities are arranged. Its center, spread, and shape provide information that individual observations cannot show by themselves.
Find the first incorrect decision rather than only checking the final arithmetic. Determine whether you misunderstood the probability model, interpreted the data incorrectly, chose the wrong distribution, confused a statistical quantity, or wrote a conclusion that was stronger than the evidence justified.