Page references indicate where the entry is discussed in greater detail and words in bold indicate terms defined elsewhere in the glossary.
Aliased effects Occur when some main or interaction effects cannot be estimated because they are confounded. Aliasing is done on purpose to simplify the design of the experiment. [p. 91]
Alpha (α) See Critical value.
Alternative hypothesis (H1) Is the hypothesis that an association between variables is present or that an effect exists. It can be a general statement that the null hypothesis is not true, or it can be more specific and state the direction of the association or effect.
Analysis of covariance (ANCOVA) Is a statistical model with at least one categorical and one continuous predictor variable and a continuous outcome.
Analysis of variance (ANOVA) Is a statistical model with one or more categorical predictor variables and a continuous outcome.
Ancillary variable An outcome variable that is unrelated to the hypothesis or research question but is measured to provide more information about the subjects or the experimental system. Such variables are useful for quality control checks.
Argument Is an input into an R function. For example, if a function takes a number and returns the square root, the number is an argument to the square-root function.
Autocorrelation Is the ‘self-correlation’ of a vector across time or space.
Bias Is the difference between a measured or calculated value and the true value.
Biological effects Are variations that arise from intrinsic differences between biological samples or sample material. Examples include the sex or age of the subjects. [p. 52]
Biological unit Is the entity about which we would like to make an inference, test a hypothesis, estimate some property, or draw a conclusion in an experiment. This may or may not be the same as the experimental unit. [p. 95]
Blinding Occurs when either the experimenter, the subject, or both are unaware of the treatment conditions while the experiment is being conducted. Blinding can remove some sources of bias.