Psychology 120 Exam #3 Flashcards: ANOVAs

What are the assumptions of the one-way ANOVA?

• 1.) Scores are measured on an interval or ratio scale. 

• 2.) Scores are sampled from a normally distributed population of scores.

• 3.) The variance of scores between groups is about equal or homogeneous across all populations sampled. (homogeneity of variance).

What are confidence Intervals?

• A process that produces a range of scores within which the true population parameter would fall in 95% of samples.  

What are degrees of freedom (df)?

The denominator of variance (MS), which is used to relate the number of participants contributing to our estimates and the things we are estimating.

What are error bars?

• A visual indicator of variability on a bar graph. They can represent the standard deviation, standard error of the mean, and or the 95% confidence interval. 

What are the mean squares (MS)?

• The average distance of scores from their mean in squared units (it is the calculation of variance).

What is k?

K is a variable used to represent the number of levels of the IV in an ANOVA.

What is statistical power?

• The statistical ability to reject the null hypothesis when it is false. It can be affected by the sample size, variability of the DV, alpha (criterion), and effect size. The more statistical power, the better, preferably, you want to have statistical power greater than 0.80. 

What is the Choen's d?

The measure of effect size for differences between two means.

What is the F statistic?

The F statistic is a family of distributions that we can use to make informed decisions about the probability that we found our obtained F as the result of a real effect as opposed to sampling error.

What is the one-way ANOVA?

• A statistical test used when you have one independent variable with more than two levels. 

What is the Sum of Squares (SS)?

The numerator of the variance (MS), that represents the sum of squared differences from a mean.

Why are Post-Hoc Tests Used?

• Because ANOVA is an omnibus test (meaning that it looks for all possible differences at once), we use post hoc testing to further break down a statistically significant analysis to see which pairwise analyses are actually significant.

What is treatment variance?

• Variance is introduced to the data by the independent variable.

What is error variance?

• Variance is introduced to the data by other (unknown) reasons. 

What is variance?

The variety of scores that different people get on any dependent variable that we measure.

Within Groups Variance

Treatment Variance and Error Variance

Between Groups Variance

Error Variance

What is a significant interaction?

A significant interaction means that the effects of one independent variable (on the dependent variable) are different across the levels of the other independent variable.

What is a Two-Way ANOVA?

A statistical test used for when you have two independent variables, each with multiple levels of the IV and one dependent variable. This kind of statistical test can be useful because it can detect the effects of two independent variables alone (main effects), as well as interactions between the independent variables.

What is a Main Effect?

The effect of one IV alone ignoring the IV.

What is an interaction?

A significant interaction means that the effect of one independent variable (on the dependent variable) are different across the levels of the other independent variable.

What effects does a Two-Way ANOVA look at?

1.) The effect of the first IV by itself.

2.) The effect of the second IV on itself.

3.) The effect of any interaction between the first and second IVs.

What is collapsing?

Collapsing is the process of calculating the M, sum of values, and n for each independent variable alone. We complete collapsing so that we can measure the effects of the 2 IVs independently.