PSYC 217 Lecture 2: Week 3 Lecture Notes - 2
Document Summary
Multiple levels/conditions; only expected cause differs; minimizes cofounds. > a variable that co-varies along with the iv. > could explain all or part of the result. Hypothesis: people who sleep longer the night before a test will perform better than those who sleep for less time - variables. Internal validity: the ability to infer the iv causes changes in the dv: covariation between two variable, temporal precedence, eliminate plausible alternative explanations. Experimental control: only the iv changes across conditions - no cofounds! Allows researchers to balance out random variables across different conditions on average. * you cannot have a cofounding variable in a correlational design but you can have third variables. Experimental control: only the iv (which we manipulate) changes across conditions. Random assignment of people to condition: balances in uence of random effects across the conditions, especially as sample size increases.
