Measurement and Statistics_Dasch_Date__021710

# Measurement and - ← ← one tailed test when you have a directional hypothesis • you specify a positive or negative direction • i.e

This preview shows page 1. Sign up to view the full content.

Statistics Basics, continued - if confidence intervals DO NOT overlap then there is a statistically significant difference between the two groups even if it overlaps slightly (lower boundary of one group with upper boundary of another) then there is not a significant difference - two tailed test: when you do not have a directional hypothesis there is an effect, but you do not specify if it is positive or negative i.e. music has an effect on people’s moods o 2.5% confidence that you’re wrong in EACH direction o room for error for each direction is cut in half i.e. eating healthy food effects mood
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ← ←- one tailed test: when you have a directional hypothesis • you specify a positive or negative direction • i.e. “Don’t Stop Believing” by Journey makes people happy • i.e. going to the gym helps you lose weight ← ←- Type I errors: when we believe there is an effect in the population but there actually isn’t • alpha level- probability of Type I error = .05 ← ←- Type II errors: when we believe there isn’t an effect in the population but there actually is • beta level- probability of Type II error = .20 ←...
View Full Document

## This note was uploaded on 08/30/2010 for the course PSYC 209 taught by Professor Hoffman during the Spring '08 term at University of Delaware.

Ask a homework question - tutors are online