PSY 350 Lecture Notes - Lecture 32: Variance, Null Hypothesis, Test Statistic
Document Summary
Identify situations where an independent-samples t-test is appropriate. Calculate the pooled variance for cases where the sample sizes are unequal. You are comparing two samples with different people in each sample. They are independent because there is no connection between any person in sample. The null hypothesis is that the two samples come from the same underlying population there"s no real difference between them. The alternative hypothesis is that the two samples come from different populations. Even if they started out in the same population (before the study), at the end they are in different populations. The basic conceptual formula for any t-test is the same: For a one sample t-test, we calculate this by: In an independent-samples t-test, instead of considering a single mean, we look at the difference between means. There is enough error in each sample, so we need to account for all of it.