Popular in Course
verified elite notetaker
Popular in Department
This 5 page Reader was uploaded by Haylee Smith on Tuesday October 6, 2015. The Reader belongs to a course at a university taught by a professor in Fall. Since its upload, it has received 11 views.
Reviews for ExperimentalPsychology1.pdf
I'm really struggling in class and this study guide was freaking crucial. Really needed help, and Haylee delivered. Shoutout Haylee, I won't forget!
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 10/06/15
Advanced Research Methods How do we know what we know Exploration into the unknown Methods you need to know Naturalistic observation Survey Case studycorrelational Experimentation Field vs laboratory Naturalistic observation Go watch Pros 0 You actually observe phenomena in the natural environment 0 Can be structured to allow for ne data collection 0 Can eliminate hawthorne effect 0 Little to no cost 0 Good for watching interations Cons No manipulation No control Cant code age Cant code emotions O O O O O O O 0 Survey Why don39t you just ask the guy Pros 0 Easy from a time and expense concern 0 For personality and attitude measures who knows you better then yourself Cons 0 Do you actually know yourself Lying vs lack of awareness o wording issues 0 Does it predict behavior Sometimesnot so much explicit attitudes and are they the only type of attitudes our theory should be concerned with Case Study Tell me everything Pros 0 Can generate enormous amounts of information regarding an individual or small group of people 0 Has the power to help understand and explain an individual with all this information 0 Very good for rare cases Cons o No 2 people in the entirety of our historyuniverse have lived the exact same life So what works for person A may not work for person B o Captures the uniqueness of individuals but does little to nothing to capture the similarity of individuals 0 N is too small to make conclusions across people not included in the study Consider the wild heavy drinking heavy smoking 91 year old uncle Jimbo Correlation How much does this relate to that Pros 0 An easy approach that allows you to quantitatively examine the relationship between 2 variables Cons o No manipulation so no causality can be determined 0 Despite evidence of a relationship between 2 variables correlation the relationship could be completely nonsensical or spurious 0 Possible 3rCI variable problems Experimentation The Gold Standard Two types to know 0 Field experiment Pros Operating in the quotreal worldquot 0 High external validity Lack of control cant measure everything 0 Low internal validity 0 Lab experiment Pros Can control and measure whatever you want 0 Temperature 0 High internal validity your manipulation did affect the variable and not something else that you failed to measure 0 Can be quite arti cial consider paper people 0 Low external validity generalizability What method would be best suited for investigating 0 Does sugar make children hyperactive lab 0 Are happy people more likely to donate to a charity than less happy people both o Is there a relationship between employee satisfaction and performance Field Generalizability Broad term encompassing the usefulness and applicability of nding to the broader population That is does what you found matter Two concepts to know 0 Reliability produces same resultsconsistency o Validity testing what you say its testing high external validity accompanies high generalizability Reliability 0 Concerns consistency o What measurements should be concerned with consistency 0 Many ways to assess reliability Behavioral measures interjudge agreement Psychological tests testretest internal consistency internal consistency all tests have this measurement error anything you observe but is not what you39re trying to capture vision problems amount of sleep distracting environment every measurement tool that exists has measurement error 0 Reliability is a necessary but not determining o The goal of assessing reliability and understanding the threats to reliability is to attempt to control and eliminate sources of error that limit the strength of our measures Validity o Are you measuring what you purport to be measuring Can the measure be used to make accurate decision 0 Concerns accuracy 0 Reliability is NECESSARY but NOT suf cient for validity Examples when could something be reliable but not valid A broken scale Concepts of validity quotTypesquot vs facesaspects Landy 1986 Trinitarian view 0 Content validity 0 Construct validity o Criterionrelated validity Predicative validity Concurrent validity not predicting behavior but the behavior is already shown 1St two quottypesquot investigate whether a test measures what it is supposed to measure Criterionrelated validity investigates whether a test can be used to make correctaccurate decisions 0 Used to make a bad decision Readings Landy Trinitarian 0 Construct validity are you measuring what you think you39re m measure 0 Content validity results match what your trying to measure 0 Criterion validity is it predictive Predictive validity Conclusive validity Unitarian landy o Validity 0 Ex big ve Point There are consequences for all tests and you can39t retake and change these tests to make the score different Scientists began picking and choosing what type of validity they wanted to support If I scientist wants to measure a construct they use construct validity If a scientist wanted to measure how fast a person can read and type they would use content validity MMPI is good at predicting what a psychologist will prescribe but it asks questions that are not relevant to types of mental illnesses ex Do you we the bed this is criterion validity Has to predict and measure something He argues that scientist39s theories are and have been supporting one type of validity but now all of them when it should and this damages science 0 Not only what it is but how it predicts 0 Collecting stamps only supporting a few types of validity not all Readings Cambell and Fitch Asking how do they know if they have something construct validity Multi method measuring the traits with multiple tests Multi trait measuring a test with two different constructs Convergent validity multiple methods of testing validity should work Within the same test the measure should correlate MMTM Used to gure out how items relate and are discriminant Note on Latent Variables Psychology is often concerned with latent or unobserved variables Latent variables psychological construct two properties 0 Abstract summaries of some phenomena in nature 0 Related to r connected with concrete observable entities or events Consider gravity as a construct How do we increase the strength of our methods and procedures 0 Triangulation Construct validation and MTMM o Aggregation simply averaging observation which means you need plenty of observation Ruler example 0 Standardization Why need for protocol Makes results reliable operational de nition How do we inject standardization into our research 0 Observation operational de nition training 0 Survey ask everyone the same question and rate them via the same standard Psychological constructs 1 Identify behaviors that relate to the theoretical construct 2 Identify other constructs that should be related OR unrelated to the theoretical construct being measured 3 Identify behaviors related to all of these other constructs and based on the relations among constructs determine whether each behavior is related to the construct being measured Prosocial behaviors kind and helpful Domain Sampling taking constructs and deciding which ones best explain the construct of your theory Ethics and Other concerns in social research Laud Humphreys Tearoom sex study Stanley Milgrim shock study Zimbardo Stanford Prison Experiment Hawthorne studies Institutional review board 1974 MTMM take multiple measures of the same things and see if they relate converge to what they are supposed to and don39t to what they are not supposed to Take measures that are similar and see if the converge on the same construct and then test to see if two different methods are discriminate on the same construct Discriminate similar methods of different constructs do not correlate and convergent they agree and go together validity used to validate constructs by relating them Dotted lines in article stand for convergent data Increase organization standardization and aggregation Test Validity reliability methods why a lab study here r eld study here internal and external validity random assignment and sampling aggregation and standardization triangulationhow you prove a construct construct is triangulation but MTMM different aspects of validity landy measurement error pros and cons methods