We'd like to use analytics cookies to understand which symbols people look up, so we know what to research next. They are not used for advertising and we set none of them unless you say yes. What we store.
The notation of a results table, grouped because the symbols are meaningless apart: a p-value means nothing without the hypothesis it was computed under, a chi-squared statistic nothing without its degrees of freedom, a correlation nothing without the interval around it. Three of these records also carry a correction, because the p-value, the confidence interval and the conditional probability are each routinely taught in a form that is wrong.
A letter x with a horizontal bar above it, standing for the average of a sample.
A lower-case sigma, or a letter s, standing for the spread of a distribution.
A Greek letter chi raised to the power two, naming a distribution and its test.
A lower-case p followed by a probability, reporting how surprising a result would be.
A capital H with a subscript zero, naming the hypothesis being tested.
A capital H with a subscript one, naming the hypothesis accepted if the null is rejected.
Two numbers stacked inside tall brackets, counting selections without order.
A lower-case r, reporting how closely two variables move together.
A pair of bounds, giving a range of plausible values for an estimate.
A capital E with a variable in brackets, meaning the long-run average.
A capital P with two events separated by a vertical bar.
A capital N with a mean and variance in brackets, naming a bell-shaped distribution.
The letters df, or a Greek nu, counting the values free to vary.