STAT-2126EL Lecture 7: T-test for two related samples, Correlation, and Logic
Document Summary
When a single group of individuals is measured more than once on the same dependent variable. Difference scores: difference score = d = x2 - x1. You cannot say if there is an effect size you do not get a significance. Less subjects needed for the study (money, rarity of subjects, time) Helps reduce or eliminate variance due to individual difference (especially when there is variability of your dependent measure in the population, or many factors affect the dependent measure). When you are examining the differences between two populations (male/female) We measure the strength of the relationship between two or more variables. We can determine the degree of correlation and the direction of this correlation. Any consistent relationship can be used to predict future events. Can support but not prove a theory. There may be no understanding as to why the relationship exists. Cannot show a causal relationship between the two variables.