Analyzing Results: Statistical Tests and ANOVA

Flashcards covering key statistical tests, ANOVA, and experimental design concepts.

Chi-Square Test

A nonparametric test that compares frequencies to determine if they represent those in the population, often using a 2 x 2 contingency table, requiring subjects to be tested individually.

T-Test for Independent Groups

A parametric test used to evaluate interval or ratio data from a two-group experiment, relating differences between treatment sets of data drawn from the same population.

T-Distribution and Sample Size

The exact shape changes depending on sample size; small samples vary more from the population mean, while larger samples resemble a normal curve.

Matched-Pair t-test (Paired t-test)

Used for dependent samples where measurements from the same subjects are compared, common in before-and-after comparisons.

ANOVA (Analysis of Variance)

A statistical procedure used to evaluate differences among three or more treatment means by dividing all variance into component parts.

Within-Subjects Design

The same participants test all conditions corresponding to a variable.

Between-Subjects Design

Different participants are assigned to different conditions corresponding to a variable.

F-Ratio

The ratio of variability between groups to variability within groups; a large F-ratio indicates statistically significant differences.

MSw (Mean Square Within)

Represents variability from error sources in ANOVA.

MSb (Mean Square Between)

Represents variability from error + treatment effects in ANOVA.

Post Hoc Tests (a posteriori)

Follow-up tests conducted after finding a significant F-ratio in ANOVA.

Pre-Planned Comparisons (a priori)

Hypotheses formed before data collection to be tested after ANOVA.

ANCOVA (Analysis of Covariance)

Controls for potential moderating variables and adjusts treatment effects for pre-existing differences between groups.

One-Way Repeated Measures ANOVA

Used for within-subjects design to analyze the effect of one IV across multiple groups tested on the same participants, focusing on main effects.