Glossary

Predictions made prior to analysing the data, usually based on prior research and theory

Statement of the expected relationship between variables, usually that there is a relationship or effect

Measuring the participant variable (which is an extraneous variable) and ensuring an even distribution of participants with different values on that variable across conditions, while using random assignment

Different entities (in Psychology, usually people) are in two or more experimental conditions/groups

A variable that varies systematically with the independent variable. It can changes in the dependent variable across conditions, so that one does not know if the different scores on the dependent variable across conditions are due to the manipulation of the independent variable or due to the confound (or both).

The extent to which a measure reflects the variable of interest (and not some other construct)

Allow for fractions (at least in theory)

Variables are measured, not manipulated

Change the order of conditions across participants in a balanced way, in a within-subjects design (e.g., half the participants are exposed to condition A followed by condition B; half the participants are exposed to condition B followed by condition A)

Features of the experimental situation provide cues to the participant as to how they are expected to behave. The participant may conform to these expectations, leading to loss of internal validity. Thus, demand characteristics may be a confound.

In an experiment, the variable that is measured and that we expect to change as a result of the different values of the independent variable

Used to summarize, organize, and describe data

Can only be measured in whole number amounts

Spread of scores in a dataset

Extent to which research results can be generalized to common real-world behaviours and natural situations (i.e., a special type of external validity)

One or more variables (independent variable) is systematically manipulated to see its (their) effect on another variable (the dependent variable)

Extent to which the results of a study generalize to other populations and settings

Any variables that vary at random in the experiment (they could be participant variables or aspects of the experiment itself)

A design in which there are at least two independent variables (e.g., a 3 x 2 factorial design has one independent variable with three levels and another independent variable with two levels)

The type I error rate across a family of tests

The expectation (written as a statement) of what will happen in the context of a particular study

A variable that is manipulated by the experimenter in an experiment

Used on sample data to infer things about the population from which the sample was drawn

In a multifactorial ANOVA, the interaction refers to how the effect of one independent variable on the dependent variable changes according to the level of the other independent variable (it is not the effect of one independent variable on the other independent variable)

Each score indicates an actual amount, and there are equal units separating any two adjacent scores. Zero scores is possible, but does not necessarily indicate a zero amount

The weight of the tails relative to a normal distribution.

Leptokurtic: light tails; values are more concentrated around the mean
Platykurtic: heavy tails; values are less concentrated around the mean

Do not include in the study participants who have particular scores on an extraneous variable of concern

In a multifactorial ANOVA, a main effect is the effect of a single independent variable on the dependent variable

Checks whether the intended manipulation of the independent variable actually occurred (this is not the same as measuring the effects of the independent variable on the dependent variable)

The means of the levels of each variable while collapsing across the levels of the other variable

Used in an experiment, to hold a variable(s) constant across groups, where pairs (if there are two groups) of participants scoring the same or similarly on a particular variable are randomly assigned to the different conditions

Sum of all scores divided by the number of scores

Middle score in a dataset (when scores ordered from lowest to highest)

Most frequent score

A label is used to describe levels of a variable - numbers do not mean anything in a mathematical sense (they just represent a category)

Statement of the expected relationship between variables if there is no actual relationship between the variables; this is usually, but not always a statement that there will be no relationship or effect

A test that tests for an overall difference between group means, but does not tell us which groups differ significantly from each other

Data that are rank ordered (e.g., 1st, 2nd, 3rd, etc.)

In non-experimental contexts, the variable that we think might be changing as a result of changes in the predictor

The probability or the likelihood of getting that value for the test statistic or more extreme, in the long run, when the null hypothesis is true

Comparisons between pairs of means that are based on a priori predictions (vs. post hoc)

Comparisons between pairs of means, conducted when there were no a priori predictions

The probability that the statistical test will detect an effect, given that there is an effect in the population

In non-experimental contexts, the variable that we think might be causing change

Scores measure an actual amount, there is a true zero, and ratio statements can be made

Ability of a measure to produce the same results under the same conditions (e.g., specifically, test-retest reliability refers to whether the measurements are stable across repeated testings)

Distance between each datapoint's Y-score and the Y-value that would be predicted based on the regression model

In a correlational design, when difference between the lowest and highest scores on one of the measured variables is small

The distribution of means that we would get if we randomly sampled an infinite number of times from the population, samples of the same size (the size of the sample in our study), when the null hypothesis is true, and plotted those means on a frequency distribution

The extent to which a measure allows for precision in measurement (e.g., a rating scale from 1 to 3 is going to be less sensitive, and hence precise, than a rating scale from 1 to 7; in the former, two people might both respond with a low score of 1 but feel quite differently about whatever the item is asking them)

In a factorial design, the effects of one independent variable on the dependent variable at each level of the other independent variable

In a non-normal distribution, it is when one tail of the distribution is longer than another

Negative skew: when the tail points to the negative end of the spectrum; in other words, most of the values are on the right side of the distribution
Positive skew: when the tail points to the positive end of the spectrum; in other words, most of the values are on the left side of the distribution

Levels of the independent variable are substantially different from one another

systematic variation / unsystematic variation

a general principle or set of principles that explains known findings about a particular topic

In correlational research, the difficulty in knowing whether an unmeasured or uncontrolled, third variable, led to an apparent association between the two measured variables

Concluding that the alternate hypothesis is correct when it is in fact false (false positive, alpha)

Concluding that the alternate hypothesis is false when it is in fact correct (false negative, beta)

The same entitites (in Psychology, usually people) take part in all the different conditions