CHAPTER 9- LECTURE ONLY September 24, 2019 9:29 AM PRACTICAL CONSIDERATIONS OF EXPERIMENTAL DESIGNS HOW DO WE DESIGN AN EXPERIMENT TO MAXIMIZE THE CHANCES THAT WE'LL BE ABE TO - Conclude that IV caused change across conditions - Detect an effect if it's there Experimental control - Ensuring only IV changes across conditions - Avoiding alternative explanations: experimenter effects, demand characteristics Optimizing our operational definitions of the IV and the DV - Avoiding ceiling and floor effects - Having an appropriate pairing of strong IV and sensitive DV Key Terms: Key Terms (Cozby Chapter 9) Operational definition of an IV manipulation Experimenter Expectancy Effects Straightforward Staged Demand Characteristics Floor & Ceiling Effects Self-report Operational definition of a measured variable (e.g., DV or in correlation design) Behavioural Physiological Experimenter Expectancy Effect: - Cognitive bias that occurs when a researcher knows what condition participants are in and therefore unconsciously manipulates an experiment in order to find the expected effect - Experimenter may treat the participants or data differently because they expected outcome of the study - Can be a threat to internal validity - e.g. o Clever Hans e.g. easy math, horse stomped his foot, horse had mathematical ability and language. Other people asked the questions, he could still answer. If the horse could not see the humans he could not answer. He did not have math ability but was really good at sensing human responses. Humans are giving off signs helping the horse guess the right answer. Therefore impacting the results o Maze Bright/ Maze Dull Rats: research assistants were told to train rats. Some are bright some are dull. They had to be trained PSYC 217 Page 1
how to run through the maze. Done for two weeks. After two weeks, bright rats ran faster than dull rats. There were not bright or dull rats, just randomly assigned. Actually bright rat group did better because research assistants were treating the groups differently. - Solution: double blind study. Participants are in blind condition, they do not which condition they are in. also, the researcher, the person directly dealing with participants should be in blind condition so they cannot unconsciously behave differently with the two groups. Demand Characteristics - Any feature of the study that informs participants of the purpose of the study - Or they just know the purpose of the study - Threat to internal validity - Can lead to participants help/ hurt you ? Be a good participant ? Mess up the results ? Social desirability- how people will respond when they know research is being conducted on them. When being asked things like what should be done for world hunger or saving the environment. Even if they feel the poor should be left hungry or saving the environment is not their responsibility, they might say so to be accepted by the society since they know someone else is going to read the response. - Thus, sometimes deception is needed. Operation Definition of IV Manipulation: - Straightforward manipulation o