PsyTeachR DSRR Ch. 10

confidence interval

A type of interval estimate used to summarise a given statistic or measurement where a proportion of intervals calculated from the sample(s) will contain the true value of the statistic.

type I error

A false positive; When a test concludes there is an effect when there is really is no effect

binomial distribution

The distribution of data where each observation can have one of two outcomes, like success/failure, yes/no or head/tails.

power

The probability of rejecting the null hypothesis when it is false.

effect size

The difference between the effect in your data and the null effect (usually a chance value)

false negative

When a test concludes there is no effect when there really is an effect

bivariate normal

Two normally distributed vectors that have a specified correlation with each other.

univariate

Relating to a single variable.

NHST

Null Hypothesis Signficance Testing

Poisson distribution

A distribution that models independent events happening over a unit of time

effect

Some measure of your data, such as the mean value, or the number of standard deviations the mean differs from a chance value.

true positive

When a test concludes there is an effect when there is really is an effect

probability

A number between 0 and 1 where 0 indicates impossibility of the event and 1 indicates certainty

function

A named section of code that can be reused.

type II error

A false negative; When a test concludes there is no effect when there is really is an effect

null effect

An outcome that does not show an otherwise expected effect.

parameter

A quantity characterizing a population.

correlation

The relationship two vectors have to each other.

p value

The probability of seeing an effect at least as extreme as what you have, if the real effect was the value you are testing against (e.g., a null effect)

covariance matrix

Parameters showing how a set of vectors vary and are correlated.

normal distribution

A symmetric distribution of data where values near the centre are most probable.

alpha

(stats) The cutoff value for making a decision to reject the null hypothesis; (graphics) A value between 0 and 1 used to control the levels of transparency in a plot

false positive

When a test concludes there is an effect when there really is no effect

uniform distribution

A distribution where all numbers in the range have an equal probability of being sampled

simulation

Generating data from summary parameters

significant

The conclusion when the p-value is less than the critical alpha.

SESOI

Smallest Effect Size of Interest: the smallest effect that is theoretically or practically meaningful

beta

The false negative rate we accept for a statistical test.

discrete

Data that can only take certain values, such as integers.

The end!