Note for PSY 290A with Professor Jacobs at UA
Note for PSY 290A with Professor Jacobs at UA
Popular in Course
Popular in Department
This 3 page Class Notes was uploaded by an elite notetaker on Friday February 6, 2015. The Class Notes belongs to a course at University of Arizona taught by a professor in Fall. Since its upload, it has received 17 views.
Reviews for Note for PSY 290A with Professor Jacobs at UA
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 02/06/15
4711 Lecture 22 0 Purpose of experimental design The whole purpose is to eliminate or minimize sources of error 0 To eliminate or minimize sources of bias 0 First step to cure problems from last few lectures o If we can control these biases we can uncover real relationships in the undifferentiated real world 0 Notion of error 0 Bizarre noncontuitive concept 0 Error an unspecified factor that affects one condition of the experiment in one way and other factors in other ways 0 Anything other than the one thing we are manipulating 0 Relative definition to the experiment 0 We always perform experiments to answer specific questions about the way the real world works 0 Any error or bias will make your conclusions and inferences about the real world weird and wrong and you won t be able to make logical inferences about it o Disallows us to receive control over the real world 0 Error 0 One error in another experiment is not relevant to an error in another I Error will occur in your study if you overlook relevant variables I It will occur in a study if you do an inadequate analysis of your data I Error will enter your study if you use inadequate sampling methods I Error will occur if you have any expectations of the experiment 0 Error may occur for an infinite number of reasons we have not found control for I Overlooking relevant variables will cause error 0 All relevant variables must be taken into account 0 Best way is to conduct an experiment is to learn about all of the relevant variables we already know so that you can discover something new 0 Biochemistry on schizophrenia o Bivalent quasiexperiment Used a sample of schizophrenics in a large hospital Analyzed the hormones in the pee of the subjects Those tested had less of the hormones than the control group Every researcher s nightmare happened I A grad student was reading their experiment but he noticed that the test was actually a quasiexperiment and he knew that diet affected the hormones and the schizoids had a poor unspiced diet and he reran the experiment with similar diets and there was no change in the hormone level so the other researchers wasted millions of dollars for that research experiment 0 O O O 0 Main point of controlling ERROR IS TO GET US DATA THAT SERVES AS THE BASE THAT SERVES AS STRONG INFERENCE 0 Fact in and of itself the data you get out of an experiment or study is perfect in the real world 0 0 Error has to do with relevant variables Anything less than a l to l correspondence and you have skrewed up somewhere More you minimize error the better the experiment 2quotd problem more sources of error Experimenters do not know how to analyze their data Example Rorschach tests were not very good people didn t see cause and effect when there was vice versa BE CAREFUL WHEN YOU READ THESE 5 WORDS SHOULD BE IN YOUR FINAL I If you are not a statistical studies expert have an expert check it out I Always ask if the statistics are important to make a conclusion I Do not believe that just because you have a PhD or an MD that you know it all You don t I If you have no statistical knowledge you have no place trying to make conclusions about the study 0 Inadequate sampling 0 If we are going to make general statements the subjects must represent the real world in some way Many experiments are horrible meaning they do not represent the real world Even well trained professionals make honest errors they know they are being watched Some investigators cheat Example one teacher gave kids an impossible experiment and they brought data back on 3 weeks IN FINALgt ALWAYS LOOK AT THE FOOTNOTES WHEN READING LITERATURE I If it is for a masters or PhD requirement do not trust the data because there is pressure on the students to receive particular results I Be aware of famous people publishing stuff from a famous lab 0 The primary researcher does not oversee all of the things many experimenters are done by his grad students etc List of things you might do if and when you enter this game I Positive results 0 At the masters level it does not matter if you have positive results 0 Place less emphasis on positive results I Closely monitor your experimenters 0 Nothing should come out of lab without it being checked by other people in the lab 0 Use doubleblind procedures 0 Keep both participants and experimenter in the dark about the hypothesis Love your enemies embrace your enemies 0 Make sure your study is checked over by your competitor because he doesn t want yours to be correct and if he can t disprove it that is good for you IN FINALgt SO MANY PITFALLS IN ANY ONE EXPERIMENT STUDY THAT WE CANNOT TAKE ANY ONE EXPERIMENT TOO SERIOUSLY DON T TRUST ONE EXPERIMENT EVER EVER EVER DO NOT TRUST THE RESULTS OF ONE STUDY WAIT TO SEE THIS STUDY DONE OVER AND OVER
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'