Statistics Key Terms flashcards
This deck defines the vocabulary of introductory and inferential statistics: population and sample, parameter and statistic, null and alternative hypotheses, significance level, p-value, Type I and Type II error, power, confidence intervals, standard error, the central limit theorem, the normal distribution and the empirical rule, correlation and the coefficient of determination, confounding, and the main sampling designs. Each card states one definition precisely, including the wording that separates it from the common misreading.
Several of these terms are graded on precision rather than gist. A p-value is a probability computed on the assumption that the null hypothesis is true, not the probability that the null hypothesis is true; a 95 percent confidence level describes the long-run behavior of the procedure, not the probability that one particular interval contains the parameter; and Type I and Type II errors are defined by what you did with the null hypothesis. Exam questions and lab reports both turn on exactly these distinctions.
The deck
How to study this deck
- For the p-value and confidence interval cards, write the definition out in full sentences rather than just recognizing it. Both are marked on their conditional wording — 'assuming the null hypothesis is true' and 'in repeated samples' — and recognition alone will not reproduce those clauses.
- Build the two-by-two error table from memory: null true or false down one side, reject or fail to reject across the top. Filling in the four cells yourself is faster and more reliable than memorizing 'false positive' and 'false negative' as bare labels.
- Pair every definition with the wrong version of it. Knowing that a p-value is not the probability the null is true, and that failing to reject the null does not prove it, is what the harder multiple-choice distractors are built from.