1/23
Vocabulary flashcards covering inferential statistical tests, including one-sample, independent, and dependent t-tests, one-way and mixed factorial ANOVAs, effect size benchmarks, error types, and Pearson correlation coefficients.
Omnibus Test
An overall hypothesis test (such as an ANOVA) used when evaluating >2 groups or time points to determine whether there is a difference across all conditions that is greater than would be expected due to chance.
Pairwise Comparisons
Post hoc t-tests performed between each combination of groups or time points to decompose a significant omnibus main effect or interaction.
Partial Eta Squared (ηp2)
An effect size measure in ANOVA that quantifies the proportion of variance explained by a factor, with conventional benchmarks defined as small (0.01), medium (0.06), and large (0.14).
Levene's Test
A statistical test evaluated prior to interpreting an ANOVA to verify the assumption of equality (homogeneity) of variances between groups.
Mixed ANOVA
An analysis of variance design that incorporates both a between-groups factor (such as experimental vs. control) and a within-subjects/repeated-measures factor (such as pre-test vs. post-test).
Interaction Effect
A condition in a factorial or mixed design where the effect of one factor differs across the levels of another factor (for example, when groups change differently over time).
Main Effect
The overall effect of a single independent factor (such as Time or Group) on a dependent variable, collapsing across all other factors in the model.
Pearson Correlation Coefficient (r)
A statistic measuring the direction and strength of the linear association between two continuous, quantitative (interval or ratio) variables, with values bounded between −1 and +1.
Degrees of Freedom for Pearson Correlation
The degrees of freedom used in testing the significance of a Pearson correlation, calculated as df=n−2.
Spurious Correlation
A strong mathematical association observed between two variables that is not causal, arising instead by pure coincidence or driven by an unmeasured third confounding variable.
One-Sample T-Test
An inferential statistical test used to determine whether the mean of a continuous variable from a single sample significantly differs from a specified known or hypothesized value of interest (μ).
Degrees of Freedom for a One-Sample T-Test
The parameter dictating the shape of the t-distribution in a one-sample test, calculated as df=n−1.
Dependent Sample T-Test
A test (also referred to as a paired-samples t-test) comparing two related sets of observations from the same individuals (such as pre- and post-intervention scores) by examining whether the mean change score differs from 0.
Independent Sample T-Test
A test evaluating whether there is a statistically significant difference between the means of two distinct, independent groups on a continuous dependent variable.
Degrees of Freedom for an Independent Sample T-Test
The degrees of freedom associated with comparing two independent groups, calculated as df=n1+n2−2.
Two-Tailed Test
A non-directional hypothesis test in which the critical rejection region is divided equally between both extremes of the sampling distribution (2α in each tail).
One-Tailed Test
A directional hypothesis test that concentrates the entire significance rejection region (α) into a single designated tail of the sampling distribution.
Type I Error (α)
The incorrect rejection of a true null hypothesis (a false positive), with the acceptable risk probability typically set by convention at α=0.05.
Type II Error (β)
The failure to reject a false null hypothesis when a true effect is present (a false negative), the probability of which is denoted by β.
Statistical Power
The probability of correctly rejecting a false null hypothesis, defined mathematically as Power=1−β and heavily influenced by sample size.
Familywise Error Rate
the cumulative inflation of Type I error probability that occurs when conducting multiple pairwise t-tests instead of a single omnibus ANOVA.
How do we interpret a Mixed ANOVA?
the interaction effect is evaluated first; main effects are only examined if the interaction fails to reach statistical significance.
What must the variables be to do a pearson’s correlation test?
The variables must be:
continuous
quantitative
measured on a interval/ration scale