Research, week 11
Research, week 11 PSY 3392
Popular in Research Design and Analysis
Popular in Psychlogy
This 60 page Class Notes was uploaded by Yesenia Notetaker on Monday March 14, 2016. The Class Notes belongs to PSY 3392 at University of Texas at Dallas taught by Meredith Grant in Spring 2016. Since its upload, it has received 20 views. For similar materials see Research Design and Analysis in Psychlogy at University of Texas at Dallas.
Reviews for Research, week 11
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: 03/14/16
EXPERIMENTAL METHOD A research study that allows you to infer causality tlhrougln quot Manipulation Control Does an IV cause changes in a DV 2 or more levels of W often treatment and control REQUIREMENTS 39 r of the IV Experimenter manipuiates what participants experience 2 or more ievels in W Often quottreatmentquot or controi Cailied a control because it r 39 tor the in uence of ali other variables A conditions helldl constant except IV Usuain balance group39s 39 True experiments use random assignment 0i UWWUYS P OSSi b le 39 39cst Smrrcs VINCE l1ici39quottl1llllil Regular Cluxrumnmi li nrsi5lhgsl HQ INTERNAL VALIDITY To infer causa ity esfrcbilish infernal validity Covariu on TimeOrder rela rionship Elimination of possible alternu rives COMPARISON Descriptive Experimental Measure variables Manipulate variables lack con rrol Willi conirol Correla rians be rween variables Establish causal rela anslhips Research Questionquot Daes noise level in uence study e eeriveness TYPES OF EXPERIWMENTS ndependem Group39s Between subiec rs Repeated Measures Within subie c rs Complex Designs Mixed Measures QuasiaExpverimen rol INDEPENDENT GROUPS DESIGN GROUPS Mcn rched Groups Natural Groups I RANDOM GROUPS EXAMPLE Hypothesis Drinking liquid wiii reduce appetite Condi ons IV Expeirimemull No water i 2 02 glass of water Random assignment individual differences distributed across groups MATCHElD GROUPS EXAMPLE Hyporl resls Visual material will be recalled barter rl39lcrr l verlbcil material Solution Assess each porrlcipunf39s memory oibllr ty prior to the exp erlmenr and match them Visual Verbal 5 5 3 3 NMURAL GROUPS EXAMPLE Hypothesis adopted intents have less secure attachments than bioiogicoili intonts Hypothesis pregnant teens who have on abortion w illli be more ilikely to experience depression than pregnant teens who have the baby CAUSAILITY WITH NATURAL GROUPS introduce onother IV that IS manipulated by the experimenter Hypothesis Severely depressed patients will respond better to treotment A while moderater depressed potients will respond better to tredtment B Treatment A Treatment lB Severe depression Modemte depression INDEPENDENT ROUPS DESIIGN INDEPENDENT SAMPLES TTEST After receiving the medication the experimental group reported Signi cantly more energy M e 3406 SD 5amp6 compared to the Qti control group M E 2077 SD 493 1158 603 p 002 DJ C o mrol Experimental FACTORIAL ANIOVA A factorial ANOVA was conducted across experimental conditions to examine potential differences in terms of energy level There was a main effect of medication Planned followup comparisons with a Bonferroni correction revealed participants that received medication at any dose reported signi cantly higher energylevels compared to the control group F 3 56 567p 065 if 2 1278 50m 7 250m n 3 30 rt 30 n 30 In 30 EXAMPLES AND PRACTICE Ami am rm In Plastic Surgery for Barbara Games Hanna Network Inc 1 Details Ratings and Reviews Rc lmted Screenshuts M9 app I5 U e l w DID CH lPhO e IP33 D n I333 Hasiw 3952 IT 61mm Rama Hm THE END Repeated Measures Designs RMD b Each participant completes all levelsiconditions of the ill I Participants serve as own controls h Still ii I Old 9 roblern with independent design between groups r balanced l gr b Hard to overcome intact groups Advantages to Repeated Designs 1 Require iewevr participants 2 increases sensitivity more power 3 Can study change Disadvantages to Repeated Design I Practice effects changes in performance overtime because of 1 learning the task 2 boredomfatigue b To control counterbaiance balance order of conditions to average out practice effects b Complete ABBA block randomization b lncomplete all possible orders random starting order with rotation Latin Squares Counterbalance Example b Hypothesis People will have greater recall for emotion words paired with rnoocl congruent pictures than emotion words paired with miood incongruth pictures I IV Mood congruence 1 Congruent happy happy sad sad 2 Incongruent happy sad sad happy i DV Recognition of words after 10 minute distractor task What might we see if there was a practice effect 100 g 393 50 U 8 quotE H G B 0 Primacy imerme ate Recency Position in sequence J 39 Complete Counterbalancing b Each participant completes every condition several times in different orders each time D Eliniinetes praotioe effects for each participant 5 ABBA I Btock Randomization Complete Counterbalancing ABBA h Conditions in one sequence and then the reverse eg ABBA ABCCBA ABCDDBCA in g A B c D p c u A HH SS HS SH SH HS 55 HHI b Average position of condition equal thus participants should not perform differently on a condition because of timing Complete Counterbalancing ABBA Not Always Appropriate Ir Practice effEcts are nonlinear eg primacyx recency I B B NE NE f Ir Anticipation effects present egu anticipation of time passed UIIHIiII A BCCBAABCCBA til 24 36 36 24 12 i2 24 36 36 24 12 Complete Counterbalancing Block Randomization gt Black a all conditions in random order Block 1 H5 1 SS HH SH Block 39 5H I HH f Block 3 Squot SH And so an V Complete Counterbalancing lBlock Randomization cont b How many blocks do you need I Have enough blocks when the average position for each condition is approximately equal Blocks HH SS HS SH 4 150 275 350 225 8 200 275 288 238 16 244 256 244 2 f Counterbalancing Complete b All conditions are administered to each participant several times using different orders each time p F b Controls for practice effects quotfor every participant k Egg ABBA block randomization b Incomplete All conditions are administered to a participant only once Data for a single participant IS confounded by practice effects Eg All possible orders Latin Square Random Starting Order With Rotation Incomplete Counterbaiancing All Possibie Orders Ir Each participant randomly assigned to one of all possible orders I Best option for incomplete design D However N possibie orders Participant Order Conditions Orders P1 HSA i 2 1 P2 HAS 3 6 P3 1 SuiIA i 4 24 P4 SAH a 120 P5 AHS a P6 ASH 10 3628800 1quot Incomplete Counterbalancing Random Starting Order with rotation gt Start with a random order then for each row rotate one to the left to N rows P Every condition in every ordinal position exactly once 15quot Order C D B A 2quot l rder EDquot 3rd Order 8 A C D Order A C B Incomplete Counterioaliancing LatniSquare II IEach condition appears in each ordinal position 7 Each condition precedes and foiiows each other condition exactly once I Not to worry Not too common 9 COW PEOUU OED ETJJ PUO oppom pone memme omewp WHEEQMIzg iiii 5 Incomplete Counterbalancing Steps ll 2 3i 0 0 9 Generate Orders Randomly assign individuals to orders Make sure that at least one participant is assigned to each order Decide technique and be prepared to randomize BEFORE you start the experiment Incomplete Counterbalancing Advantages Disadvantages b Often more feasible I Individual participants data cannot be h Many canditions interpreted gt May not make sense 7 99 hm Wm Wm h I Need at least one w participant per order Iv Don t have stimuli Ir Need to be fast eggs infants Again Counterbalancing Complete Incomipiete b All conditions are I All Conditions are administered to each admI iStEFEd W a 7 participant several partrcrpont oniy once times using different b Data for a single orders each time participant IS confounded by practice b Controls for practice effects effects quotfor every p E All pbl d 43 i possr e or ers partmpant Latin Square Random h Egg ABBA biock Starting Order With randomization Rotati 27 x Repeated Measures Design gt Differentiai Transfer performance in one condition is dependent on the condition that precedes it D Solution independent design Put it together Independent Design Repeated Design b Between Subjects b Within Subjects Ir Confound differences Ir Confoundi Practice between groups effects p Central randomization k Control counterbalancing b Compiete ABBA Block Randomization b incomplete All possible orders Latin Square Random Starting Order With Rotation Goal Maximize internai Validityf y xquot The Analysis Independent Independent sample39s Factorial ANOVA Design ttest Repeated Design Paired Samplles ttest Repeated Measures ANDVA Wilson Brewer Et iRosenbluth 2014 b Voter ID laws citizens show proof of identificationquot before voting Can way from stating name to it require various types of photo IiD I Supporters prevents fraud h Opponents requirements disproportionately impact certain groups preventing voting Wilson Brewer El Rosenluth 2014 b The sample 1435 participants from 2012 Imperative Congressional Election study demographically representative data collected online gt The study h Participants indicate their support for voter ll laws after viewing one of three images Ir Control no image b Treatment A white voter P Treatment B black voter in Also measured political ideology and racial resentment b The results white respondents showed greater support for voter liD laws in the AA iage condition compared to other conditions gt What type os study was this I Independet groups between subjects design with random assignment Is this an independent sample design h Hypothesis Job applications submitted with traditional l American ames will be reviews more favorably than i those submitted with less common ethnic names 5 Using repeated measures will give more power and will be i able to Process more quotjob applications ll I Hypothesis individuals will recommend more lenient sentencing for criminals who express more remorse than for criminals who do not express remorse I Could be independent or repeated measurements b Hypothesis Individuals will consume more calories when they feel bad about themselves than when they feel good about themselves I There could be priming it will depend on operational definition Eg if quotfeeling badquot is clinically depressed we need independent groups because it is a natural group I Hypothesis mood will be affected by the emotional Review I Method
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'