psych 7 week 9
Popular in Intro to Experimental Psych
Popular in Department
This 4 page Class Notes was uploaded by Debbie Chen on Friday December 5, 2014. The Class Notes belongs to 7 at University of California Santa Barbara taught by PROTZKO in Fall2014. Since its upload, it has received 40 views.
Reviews for psych 7 week 9
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: 12/05/14
0 Causes random researcher error phacking O Phacking 39 Don t put a lot of tests 20 because it s easy to reject should run them all on the same days 39 20 DV and one of them is significant 0 Replication 39 Existence proof 39 Random 0 Type II beta 0 Failed to reject when you should have 39 Randomly researcher error not enough people 39 Power ability to not have a type two error 39 CHOOSING A SAMPLE SIZE POWER ANALYSIS 39 Power is a statistical test that determines optimal sample size based on probability of correctly rejecting the null hypothesis 39 Power 1 p Type II error 39 Effect sizes range and desired power 0 Smaller effect sizes require larger samples to be significant 0 Higher desired power demands a greater sample size 0 Researchers usually use power around 8 0 Replication 0 Second study unable to replicate is it really a null effect 0 Type III NHT 0 Effect is in the wrong direction 0 Parametric 39 Makes assumptions about the population that your sample comes from 39 Assumes at least an interval scale 39 Central tendency mean 0 Nonparametric 39 No assumptions about the population 39 Assumes either nominal or ordinal scale 39 Frequency based mode exact probability Week 9 Interactions Interactions 0 When the effect ofX on Y depends on something else M 0 Main Effect The effect ofX on Y 3 Examples 0 Main effects are the effects of each iv considered separately O O O Instrument training and instruction and IQ 39 We manipulate a new factor think or rigorous Heritage and insult and anger 39 We use an already existing factor Height and age and income 39 Ourfactor is continuous not categorical 0 Interpreting O O O O 0 When you have an interaction you lose the ability to talk about the main effects Does insulting somebody cause an increase in stress hormones 39 Can t say because there is an interaction 39 Moderated effectconditional effect But what if the main effect is still signi cant 39 Pooling data there is still an effect 39 Too bad 39 You still lose the ability Confound v Moderator 39 Confound is an alternate explanation is there a cause at all 39 Moderator is not saying that the cause doesn t work it39s saying that it39s a conditional cause ModeratedConditional Statements 39 Unconditional response Yes 39 Conditional response Not if my ex is there 39 ls Daniel going to the party 0 Depends 39 ls insults going to increase stress 0 Depends 0 Where are they from 39 Interactions cause conditional statements 39 We can t interpret main effects 39 Creates conditionals and subgroups 39 Insulting a southerner increase their stress hormones NEVER DICHOTOMIZE A CONTINUOUS VARIABLE It is NOT exploratory 39 It increases the likelihood of type I and II errors depending on data 39 Meaning you can t trust or interpret what it says Advantages 39 Test more than one hypothesis 39 Potentially confounding variables built into design 39 Enables interaction effects to be tested
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'