Psych 325 1st Edition Lecture 17Review of Tuesday:-Studying One PersonoCase study-Single person or group-Detailed, descriptive study-Extraordinary circumstancesoWhy do them?-Extraordinary people are rare-Can't study a process in the laboWhat do they accomplish?-Uncover general principles-Insight into unstudied areaoWho uses them?-Clinical psychologists-Behavioral neuropsychologistsoConsciousness is complicated to study, so it is good to know examples: Peter Tripp--wake-athon, was out of his mind after 3 days. Gardener--stayed up for around 7 nightsand felt super refreshed. Stephen D.--thought he was a dog, extreme paranoia. Feral children--abandoned, never developed as fully-functioning people-Is this a useful scientific approach?oCase studies can be scientificoIt's a science if…-Developing or refining theory-Explanations use theory-Research is falsifiableoUseful because…-Provide ideas for future work-Can test existing theories-Inspire better questions-Provide findings to complement experiments or larger studies-Problems with case studiesoNo control over variablesoCan't do statisticsoOperational definitions are difficultoExternal validity-Just one person's behavior-Do results generalize?-ONE SIZE DOES NOT FIT WELLPart 2: Single Variable Research-Studying One VariableThese notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.-Single variable studiesoDescribe a sampleoValue generalizability mostoDon't directly address causes-Sometimes theoretically usefuloStriking behavior in a sampleoExpect one thing, find another-Milgram obedience/ Asch conformity studies-Body image-Judgment and decision makingoOK to use convenience sample-Prevalence of… PTSD among war vets-Proportion of…-Probability of… abortion in Midwest -Other examples: -Types of Single Variable Research-CensusoStudy every personoExpensive and difficult-Population surveyoStudy random sample of populationoRandom (hopefully) means generalizable-Cluster samplingoCreate list of locations where population is foundoRandomly sample locationsoRandomly sample within locationsoEasier to do-Error in Single Variable Research-Error is ALWAYS presentoMissing people in consensusoSelect presentative sampleoRespond incorrectly-Sampling erroroMargin of error oPotential range of variable-Sampling works welloNeed only small subsetoVery precise measurements-Random sampling more accurate than census-Ex: survey of satisfaction with educational resources among WI college students-Think about how the concept of sampling error applies to exit polls in elections-These polls are based samples of the larger population of voters-Good news: sampling works well-In any study you can NEVER EVER NEVER EVER study every single person-Forms of Population Studies-EpidemiologyoUsed in clinical psychologyoPrevalence of mental disorders-Public opinionoAttitudes and preferences of a particular populationoVoting, public policy-Marketing researchoWhat you want to buyoImportant for sales-Problems with population studies-Not easy to do!oMany approaches to sampling oTraining interviewers is difficultoPeople may respond badly-Americans are skeptical of surveys-Range of participant backgrounds can make writing questions difficultoLanguage issuesoInterpretation issuesoKeep questions simple!-Loss of experimental controlPart 3: Correlational Methods-What is correlational research?-GoaloUnderstand causal relationships between at least two variables-How?oMeasure or categorize variablesoAre they related? Vary together?-Correlation coefficients-Group comparisonsoIf so, may be causally related-Ex post facto datao"after the fact"oNo manipulationoMust acknowledge internal validity problems-See pp.