Psych Stats Chapter 8 Statistical Significance, Confidence Intervals, Effect Size, and Power

Statistically Significant

Results that are highly unlikely to have happened if in reality the null hypothesis is true

Type 1 Error

rejecting the null hypothesis when you should not (saying there is an effect when there is not

Meta-analysis

A type of research study that combines results from all relevant research

Nullification Fallacy

describes the mistaken assumptions that nonsignificant findings that are caused only by the absence of an effect of the independent variable

Significance Theory

describes the mistaken assumptions that statistically significant findings are meaningful or useful in the practical context

Effect Size

A measure of the strength of an association between two variables; or practical rather than the statistical importance of a research finding

Power Analysis

An analysis conducted prior to the start of the study that has the goal of determining the sample size needed to achieve a specified level of power

Replication

(also called reproducibility) the notion in science that all findings should survive repeated testing before they are accepted

Replication crisis

a crisis that exists in psychology and other sciences as a result of the failure to replicate findings that were previously accepted

TRUE OR FALSE: confidence intervals are always two tailed

TRUE: we want to determine the upper and lower limits of the confidence interval