A Tale of 2 Tails
This kind of test has two critical values.
What is two-tailed?
Compare a sample mean to a population mean when you don't have the population SD.
What is a one-sample t-test or single-sample t-test?
For any z-test or t-test, the measurement scale needed for the DV.
What is interval or ratio?
This hypothesis states that there is no difference between conditions.
What is the Null Hypothesis?
Increasing the sample size does this to the confidence interval.
What is make it narrower?
Do this kind of test to test for a difference but not a direction.
What is two-tailed?
Estimate a range of values for a mean when you know the population SD.
What is a confidence interval with z scores?
Statisticians consider this a "large" N.
What is 30 or more?
What you do to the Ho when your result is significant.
What is reject?
Cohen's d and r2 tell you this.
What is effect size?
Must hypothesize this to do a one-tailed test.
What is direction of the difference?
Compare a sample mean to a population mean when you know the population SD.
What is a one-sample z-test?
Each score is measured separately from the other scores.
What is independent observations?
When you fail to reject the Ho, your result is this.
What is not significant?
This distribution doesn't really exist but we use it anyway.
What is a sampling distribution?
An advantage of one-tailed tests.
What is more powerful/more likely to be significant?
Compare the means of two conditions in a between-subjects design.
What is an independent samples t-test?
What is normal?
Alpha is the probability of this outcome.
What is Type I error/rejecting a true Ho?
A t-distribution gets flatter if you have fewer of these.
What are degrees of freedom?
An advantage of two-tailed tests.
What is safer/can be significant in either direction?
Compare the means of two conditions in a within-subjects design.
What is a paired t-test or related samples/dependent samples t-test?
Equal variance in the different conditions.
What is homogeneity of variance?
When your sample statistic is close to the middle of the sampling distribution, do this.
What is fail to reject the Ho?
If this test is significant, you violated homogeneity of variance.
What is Levene's test?