Psych 7 Week 7
Popular in Intro to Experimental Psych
Popular in Department
This 5 page Class Notes was uploaded by Debbie Chen on Monday November 24, 2014. The Class Notes belongs to 7 at University of California Santa Barbara taught by PROTZKO in Fall2014. Since its upload, it has received 41 views.
Reviews for Psych 7 Week 7
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: 11/24/14
Psych 7 Lecture Notes Week 7 Correlation amp Regression Nonexperimental Designs 0 Longitudinal Design 0 A cohort is studied repeated over time same people Notation O O O O No manipulation Get temporality good OOO Individual differences are controlled for 39 Adulthood Cognitive Changes 0 Verbal intelligence scores hold steady with age while nonverbal intelligence scores decline 0 Advantages 39 Fewer participants 39 Good sensitively to change within a person 39 Equal conditions 0 Disadvantages 39 Order effects 0 No control group 0 Fatigue 0 Contrast 0 Practice 39 Dropout 0 Random 0 How many observations 39 More observations better grain to detect changestability 39 Must balance with order effectsfatiguepractice 0 Predicting O A variable at one time predicts something a second time Children who are spanked when they are 5 have more behavior problems later in life Notation Ospanking Obehavior Most common developmental design OOO What s wrong with this 39 Technically nothing if you re only interesting in predicting 39 But if you move into talking about causality it is a problem 0 Confounds 3rd variable 0 Controlling For 39 Take a measure of behavior problems at age 5 39 Os5 and Ob5 both go to Ob10 controlling for behavior problems at age5 39 Many other possible confounds 39 What controlling for is 39 Equations ROXO R0 0 Ymxb turns into yb0b1x O Posttest scores intercept some value for group O X is 0 or 1 control or experimental and b1 tells us if that make a difference 0 Group 0 some score 0 Group 1 some score more Yb0 b1x1 b2x2 0 Adding in b2x2 is controlling for pretest model Using slope because you re essentially putting a straight line onto your slope 39 Multiple regression Posttest scores 113 104 Pretest scores 108 99 When you control for YadJ39 Y BPlXiX9l Assume Bp 997 assuming that pretest scores are both the same 0 New posttest score 0 Get new posttest scores 0 Posttest score Bp Pretest score average score When you control for the pretest differences the posttest scores end up evening out What looked like an effect actually didn t exist 39 Controlling Is not holding fixed It is conditioning changing observed scores Taking the differences at age 5 into account what would our age 10 look like 39 Controlling amp Predictions CrossSectional Design 0 O 0 How many potential confounds are there in any study Infinite Some people believe MR can get you causality but it can t ltjust controls for the big ones but it can t control all Useful but not absolute Ice cream amp relationship success no matter what you control for the relationship doesn t go away highest causality 0996 but there is not causality spurious correlations Compare Ss of different ages studied at one point in time Notation 0 Advantages 39 Cheaper and faster than longitudinal Disadvantages 39 Infer developmental changes 39 A lot more comes with being a different age 39 May not line up with longitudinal 0 Adulthood cognitive changes 0 Crosssectional suggests decline longitudinal suggests more stability 0 Multiple regressions without holding anything constant 0 Ignore prediction is okay Week 7 Lecture 2 Case Studies and Observational Methods read independently Case Studies 0 Singlesubject extremely small numbers 0 Phineas Gage spike through brain which changed his personality 0 Prefrontal cortex responsible for emotion 0 Not responsible for language moving etc 0 Now we know this to be true 0 At rescue she was nonverbal 0 After 5 years 39 Nonverbal abilities were excellent 39 Never able to master grammar O Tells us things about language like 39 Critical period for grammar 39 Need input 39 Nonverbal and some vocabulary remain unhindered 0 Frere Jacques Brother John O is language necessary for thought Falsifiable 0 Had seizures long amp short 0 Nonverbal for periods of time 0 But he can do things like think amp travel remember operate things 0 This may show us language is not necessary but language has already affected the brain 0 Splitbrain patients 0 Cut Corpus Callosum to stop seizures 0 L hemisphere is the interpreter makes up story Case Studies 0 Advantages O Dramatic new knowledge 0 Extreme situations we cannot ever duplicate in a lab 0 Disadvantages 0 Single cases can be misinterpreted 39 Phineas Gage 13 years later he recovered O No counterfactual 39 Genie maybe she would have been impaired anyway 0 Manwho fallacy 39 Frere Jacques I know a man who couldn t speak but could think even if we trust his own judgment 0 Generalizability 39 Law of Large Numbers 39 Split brain patients individual differences in brain behavior 0 Noise amp errors every behavior gets interpretation How do we make case studies better 0 Replicate 0 Gather More Participants 0 Ignore Individual Differences Single Case Research Designs 0 Manipulation so not case studies 0 2 Main Designs 0 Multiple Baseline Design 39 ABAB Design 0 Stable Baseline Design 39 Ignore it 0 Multiple Baseline Design 0 Introduce treatment at different times for different subjectsbehaviorssituations 0 History effects Add more people 0 Suppose you want to see how the way you dress affects the way people treat you 39 Get a baseline measure 39 Change 39 Recover 39 Change 39 Change 2 O Marie s project 39 5 different outfits 39 Her conclusion we are shallow creatures and that we put ourselves into boxes 39 Problems 0 Biased reporting 0 Blinding 0 Monday Friday OKCupid People might be different 39 Should have been O 0 21 week project Limitations of Reversal Designs 39 Some effects of treatments are difficult to reverse 39 Most useful for seeing shortterm effects of treatments 39 Order effects are a problem 39 Can be very difficult to interpret Observational Research 0 NaturalisticObservations O O O 0 Complete and accurate picture of behaviors rather than to test hypothesis Analyze observations in the field and then form hypothesis Participation Concealment Robert Cialdini how to make people buy things 39 Low ball technique customer given price much lower than salesman intends to sell car in order to induce customer to agree to purchase 39 Students coming in for a research study change time more people come in if they re committed 0 Systematic Observations O Careful observation of one or more behaviors to a particular settings 39 Interested in only specific behaviors 39 Quantifiable 39 Hypothesisdriven 0 Archival Research
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'