Research Methods in Psychology: Correlational Designs and Scientific Method - Prof. Z. A. , Study notes of Psychology

Download Research Methods in Psychology: Correlational Designs and Scientific Method - Prof. Z. A. and more Study notes Psychology in PDF only on Docsity! 3. Research Methods Getting Started - Make an observation of behavior, and come up with a theory to explain that behavior - Generate a hypothesis based on that theory - Theory: Alcohol intoxication impairs driving skills (general observation about behavior) - Data: Measure seconds a person can stand on one leg after drinking alcohol - Hypothesis: People who are intoxicated will show less motor coordination -“Good Theory” = generate testable hypotheses ex. Not The Interpretation of Dreams -Serendipitous finding= unexpected finding ex. Torsten Wiesel and David Hubel- cats respond to lines and edges, instead of at dots –from projector jamming -Operations definitions= identify and quantify variables so that they can be measured ex. Rating fun 3 Research Designs 1. Descriptive Designs: Observe behaviors 2. Experimental Designs: Manipulate behaviors 3. Correlational Design: Compare (co-related) Behaviors 1. Descriptive (observational) study- Research method that involves observing, noting, and classifying behavior *Types 1. Naturalistic: Where the observer is apart from and makes no attempt to alter or change the situation - Good: studying without effecting the study 2. Participant: Where the researcher is actively involved in the situation - Good: have control over experiment - Konrad Lorenz’s imprinting: after a short time after birth, geese imprint who their mother is. He wanted to see if it was innate or learned. *3 Decisions 1. Done in the laboratory or in the natural environment? - more control in laboratory - more like reality if in natural environment? 2. How collecting the data?- case studies, surveys, questionnaires 3. Visible Observer? - Reactivity is an effect that occurs when the knowledge that one is being observed alters the behavior being observed - Observer bias: systematic errors in observation that occur due to an observer’s expectations ∴Experimenter Expectancy Effect: actual change in the behavior due to observer bias Ex. Rosenthal’s Study half-told that rats are good at running mazes; learned the task more quickly half-no expectation; slow Ex. - How many people wash hands after using the restroom? - Who washes more dishes, girls or guys? *Longitudinal Studies: developmental design ex. How intellectual abilities change over the adult years *Cross-Sectional studies: comparing different groups to make inferences Ex. Intellectual abilities of young adults vs old adults 2. Correlational Designs (examine how variables are naturally related in the real world without manipulation) - Compare 2 factors or measures, but there is no manipulation of factors Ex. Violent TV vs. aggression, GPA vs. Study time, learning ability vs. brain size, depression vs suicide - Correlation does not equal causation!-directionality problem (you’re not manipulating things so there is no way to know) - Remember, correlation is not an experiment - Often show this kind of data with a scatterplot *Degrees of Correlations - strength of correlations is measured by the r statistic (0 = no correlation, +/-1.0 perfect correlation) - perfect correlation: Weight in kilograms is the same as weight in pounds * Issues - Correlation does not equal causation! (A -> B or B -> A) - The third-variable problem: a third variable could the cause of a third variable (C -> A and C -> B) Ex.Someone wanted to investigate if there was a correlation between shoe size and intelligence. Went to elementary school and measured IQ and shoe size at recess and it showed a strong correlation. why? (maybe 3rd grader would be smarter than 1st grader and have bigger shoes) Ex. Having high energy cause working long hours and wanting to have fun after work : Ex. people’s favorite color is red tend to be more aggressive Ex. cuter babies get more attention from their parents and develop faster 3. Experimental Design (test causal hypotheses by measuring and manipulating variables) -manipulate independent variables (ex. # of hours working) dependent (how much fun women want) *Scientific Method (systemic procedure of observing and measuring phenomena to answer questions about what happens, when it happens, what causes it, and why) - Develop theory( a model of interconnected ideas and concepts that explains what’s observed and makes predictions) (i.e., general principles about behavior) - Generate hypothesis (prediction) - Design a method that manipulates conditions w/ an experiment - Collect data - Analyze findings using statistics to describe and evaluate data -Replication (repetition of an experiment to confirm the results) *Example! Theory: the stresses associated with school influence rate of depression among high school students Hypothesis: Students placed in stressful situations will report lower rates of happiness Participants - Random assignment: setting up groups of participants so that each individual has an equal chance of being assigned to any level of independence variable - typically what you want. because the groups won’t be equal - A sample must be representative of a population so you can generalize the results -Population= the group you want to study -Sample = subset of population; the group of people that you have - Random sampling= each participant has an equal chance -Experience Sampling=take several samples of participants’ experiences over time ex. Writing how he feels at each class Ex. Emotional change throughout the day - Convenient sampling: easier and more realistic ex. Students at our school - Truman and Dewey -Random Assignment: participants assigned randomly to the control or experimental group =the participants are on average, equal -Meta-analysis= “study of studies” that combines the findings of multiple studies to arrive at a conclusion ex. Whether self-esteem is related to school achievement Data: Groups of HS students participating in two types of computerized tests: Moderate (control) and Difficult (experimental group) - Follow- up questionnaire taken after the computerized test that measures how one is feeling that day (Daily Happiness Index: you want to come up with an operational definition) *Collecting Data 1. in the laboratory or in a natural environment?- artificial or not 2.written description of what was seen or as a running tally of prespecified categories of behavior -Do data address your question? ex. Not advertisements 2. Reliability (stability or consistency)= measure is stable and consistent over time in similar conditions 3. Accuracy= the extent to which an experimental measure is free from error a. random error- the value of error is different each time b. systematic error= the value of error is consistent •Ethics Issues Institutional review board (IRB) : -consists of both scholars, administrators, legal advisors; review research to ensure that it meets the accepted standards. -for reviewing proposed research to ensure that it meets the accepted standards of science and provides for the physical and emotional well-being of research participants 1. Privacy ex. Monitoring people remotely Ex. Too personal questions 2. Confidentiality- personal information is not released to the public 3. Anonymity- no personal information is collected, never be traced to any individual 1. Informed Consent- participants make a knowledgeable decision to participate A. Or Deception- involve either misleading participant about the study’s goals or not fully revealing what will take place----Require DEBRIEFING after experiment 2. Relative risks of participation-should be very specific mental or physical health -the risk/benefit ration= analysis of whether the research is important enough to warrant placing participants at risk Ex. Tuskeegee Experiments on Black Men -lying to ppl –“ it’s a cure” -denied the treatment -consent was iffy, money offer is hard to be rejected -racial discrimination= only experimenting on African Americans. Ex. Milgram Obedience Study- right after Holocaust , 유유유유유 유 유유유유유 유유? -position of authority made the experiment continue -informatitive study -unethical: traumatize, making to think im capable of killing Ex. The Stanford Prison Experiments •Measuring the Body Directly 1. EEG- Electroencephalogram: measures electrical activity of the brain 2. PET- Positron Emission Tomography: inject radioactive substance, can see metabolic activity in the brain, i.e. brain activity 3. fMRI- Functional Magnetic Resonance Imaging: measure blood’s o2 level to detect brain activity 4. TMS- Transcranial Magnetic Stimulation: emit magnetic pulse to temporarily knock out brain function -hold a wand, knock out a specific region of the brain, what is responsible for what -measure hormones Advantage Disadvantage Descriptive Studies -valuable in early stages of research -real-world setting -observer bias; observer’s presence can change reactivity Longitudinal Studies -see developmental changes -expensive, long time, may lose participants over time Cross-sectional Studies -faster, less expensive -third variable problem Correlational Studies -rely on naturally occurring relationship -real-world setting -cannot show the direction of cause/effect -third variable problem Experiments --causal relationships; no directionality problem -in an artificial setting Cross- cultural Studies -make psychology more applicable to the world -the effect of culture on some variable of interest -translations don’t convey same meaning -misunderstanding during the research process Case Studies -extensive data about one or a few individuals or organizations -very comprehensive Ex. Seun_Hui cho- cannot figure out the reason behind his shooting -can be very subjective -not possible to generalize the results Interactive Methods -(self-report: questionnaire)gather data from a large # of people -easy to administer, cost-efficient, fast way -explore new lines of questioning -determine how responses vary over time -people can introduce biases into their answers or may not recall information accurately (socially desirable responding, faking good, describe in positive ways <better-than-average effect>) Response performance methods -simple way to study cognition and perception -less affected by observer bias or subject reactivity -can be costly and time consuming -less likely real-world setting Brain activity methods -map the brain to show brain regions involved in different tasks -show only the brain regions active -correlations: don’t know whether a particular brain region is necessary for a particular task -third variable/ directionality problem * Response Performance: researchers quantify perceptual or cognitive processes in response to a specific stimulus 1. Reaction time Ex/ a figure flashed on a screen is red or blue, whether that figure is red or blue AJD round or square 2. Response Accuracy- the response is right or wrong 3. Stimulus Judgment- response to different stimuli *Brain/ Body Activity Directly Measured Ex. Frightened: tensed muscles, faster heart beat=Psychophysiological assessment: blood pressure, blood temperature, perspiration rate, etc Polygraphs (“lie detectors’)= lying causes more arousal 1. Electrophysiology= measure electrical activity in the brain -put electrodes on scalp- measured by Electrocephalogram (EEG) ex. Sleeping but hard on cognition which reflect too many recordings of brain activity 2. Brain Imaging= make brain structures and brain activity visible; measure changes in the rate of blood flow (carrying nutrients to active brain regions A. Positron Emission Tomography (PET)- assess metabolic activity by using a radioactive substance injected into the bloodstream; release glucose; measure the density B. Magnetic Resonance Imaging (MRI)- high quality, magnetic-field, image; find location of damage–energy is released from brain tissue in a form that can be measured by detectors A. Functional MRI (fMRI)= create images of working human brain- measure blood flow C. Transcranial Magnetic Stimulation (TMS)= use strong magnets to briefly interrupt normal brain activity as a way to study brain regions. -see whether a particular brain is necessary for a particular task *Animals -John Watsons’ Behaviorism- humans are subject to the same laws of nature as other animals -Animal models provide advantages when the research involved is not feasible in humans;