RESEARCH METHODS AP Psychology Ms Brown Myers Ch

RESEARCH METHODS AP Psychology Ms. Brown Myers – Ch. 1

LIMITS OF INTUITION & COMMON SENSE “The naked intellect is an extraordinarily inaccurate instrument. ” - Madeleine L’Engle (1977)

Why Science? • How can we best understand why people think, feel, and act as they do? • Hunches? • Common sense? • Intuition? • Psychology needs a scientific approach to separate “common sense” and “hunches” from actual credible evidence. • Differ from person to person • Not always 100% correct

Overconfidence • The tendency to overestimate our abilities, including knowledge (the illusion of knowledge) • Faster • Smarter • Stronger • More aware • More observant • Even after being presented with evidence that contradicts previous assumptions about ourselves, most claim, “Well I was close. ”

Draw a bike. A B C D

A I bet you were pretty confident you knew how to draw a bike… Our brains allow us to be overconfident to shield us from everything we DON’T know – “the illusion of knowledge” effect. How many of you are claiming, “I knew it all along!” (hindsight bias!)

Hindsight Bias • The tendency to believe, after learning an outcome, that one would have foreseen it. • “I knew it all along phenomenon” • Hindsight is 20/20 – after the fact, it is easy to see why an event occurred, someone acted in a certain way, etc • Psychological findings often seem already known or like common sense because we constantly observe psychological phenomena or “knew it all along. ”

The Scientific Method • A method in which scientists form hypotheses from theories, conduct Theory - an explanation that organizes data and predicts observations (NOT fact) 1) Theories ad Le … to G or ene re rat fin e e… research, and then refine theories in the light of research observations. 3) Research and Observations 2) Hypothesis Lead to… Hypothesis - a testable prediction, often implied by a theory Replication – repeating the study to see if the basic finding extends to other participants in other circumstances

DESCRIPTIVE METHODS Description is the starting point of any science. These types of research methods describe behaviors/attitudes, not explain them.

Case Study • An observation technique in which one person is studied in depth with the hope of revealing universal principles. • Pros: • Insight into specific cases that possibly could not be studied due to ethics • Cons: • Individual studies are hard to generalize to large populations • Some events/circumstances cannot be replicated (ethics) • Ex: brain lesion studies, instances of socially isolated (“feral”) children

Survey • A technique for ascertaining the self-reported attitudes or behaviors of people, usually by questioning a representative, random sample of them. • Must have a representative and random sampling from the population for generalization to be possible • Population – all of the cases in a group from which samples may be drawn for study • Representative – reflective of the population • Random sample – a sample in which all individuals have an equal chance of inclusion in the study • Generalization – the ability to reflect results from the random sample on the entire population

Survey • Pros: • Can study large populations from a representative sample • Cons: • Self-reporting is not always reliable • Social-desirability effect – the tendency of respondents to answer questions in a manner that will be viewed favorably by others. • Samples are not always representative • Results are largely based on how questions are worded (“aid to the needy” instead of “welfare”) • Ex: Gallup polls, Kinsey Report on sexuality

Naturalistic Observation • Observing and recording behavior in naturally occurring situations without trying to manipulate and control the situation. • The observer must not manipulate or stage the situation. • Pros: • Observe people/animals in real, not artificial, environments • Cons: • No control over events or variables • Ex: videotaping mothers and children together in different cultures, recording students’ self-seating patterns in the lunchroom

CORRELATIONAL METHODS After describing phenomena, researchers want to see if certain variables relate to each other. When observed variables seem to relate to each other, it is said they correlate.

Correlational Research • A measure of the extent to which two factors vary together, and thus of how well either factor predicts the other. • How are two things related? • How strong is this relationship? • Can the relationship shape predictions? • Scatterplots – a graphed cluster of dots, each of which represents the values of two variables. The amount of scatter suggests the strength of the correlation (little scatter = more correlation)

Positive Correlation • Two variables rise or fall together • The taller you are, the more you weigh • The more you smoke, your risk of cancer increases • As temperature rises, crime rate increases • As the ocean level decreases, the fish population decreases

Negative Correlation • Two variables relate inversely to each other – as one rises, the other falls. • The more you brush your teeth, your risk of cavities decreases • The more years spent in jail, the lower the education level • The more you hold a baby, the less it cries • The more hours spent watching TV, the less time spent doing HW

No Correlation • Two variables do not seem to be related • People born later in the year and intelligence level

Correlation coefficient • The mathematical expression of the relationship, from 0 to 1, +/- indicating correlation type • Measures how well either one predicts the other and how strong that relationship/prediction is (0 = no relationship) • r = +0. 37 • • • r correlation coefficient (relationship) + indicates direction of relationship (positive or negative) 0. 37 Indicates strength of relationship (0. 00 weak to 1. 00 strong) r = +1. 00 r = -1. 00 r=0

Practice Positive or Negative Correlation? Strong , Moderate, or Weak Correlation? Those with higher rates of depression tend to have higher risks of suicide. • +0. 71 • The louder the music while studying, the lower the exam performance. • -0. 13 • The more you observe aggression, the more aggression you display. • • Fairly strong relationship • Fairly weak relationship • +0. 46 • Moderate relationship

Correlation IS NOT Causation • Correlations cannot fully predict future behaviors/attitudes, regardless of how strong the correlation coefficient is. • Ex: low self esteem is correlated with depression, however this does not mean that low self-esteem directly causes depression. Could cause (1) Low self esteem Depression OR (2) Depression Could cause Low self esteem OR (3) Distressing events or biological predisposition Could cause Depression Low self esteem

Correlational Research • Pros: • Can measure the extent of a relationship • Cons: • Correlation is not causation (just because two things are related does not mean one causes the other)

EXPERIMENTAL METHODS Because many factors influence everyday behaviors/attitudes, psychologists need to isolate and control variables to establish cause and effect relationships. They do this using experiments.

Experiments • A research method in which an investigator manipulates one or more factors (independent variables) to observe the effect on some behavior or mental process (the dependent variable). By random assignment of participants, the experimenter aims to control other relevant factors. • Manipulate the factors of influence • Hold other variables constant • Unlike correlational studies that uncover naturally occurring relationships, an experiment manipulates a factor to determine its effect.

Variables – anything that can vary • Independent variable • The variable manipulated by the experimenter • Dependent variable • The outcome being studied as a result of the ind. variable • Ex: Hypothesis – Pill X can reduce the symptoms of schizophrenia. • Ind. - Drug • Dep. – symptoms of schizophrenia • Mnemonic (memory aid): make a if-then statement “IF the independent variable THEN the dependent variable. ”

Operational Definition • Specific definition of the independent and dependent variables • Very specific especially when variables are vague • Ex: Hypothesis – Pill X can reduce the symptoms of schizophrenia. • Ind – Drug • Operationalize: Pill X, placebo • Dep – symptoms of schizophrenia • Operationalize: the amount of dopamine activity in the brain, the prevalence of disorganized speech

Research groups • Experimental group • Participants that receive the independent variable • Control group • Participants NOT exposed to the independent variable • Serves as a comparison for evaluating the effect of the ind. variable (sets a base-line) • Ex: Hypothesis – Pill X can reduce symptoms of schizophrenia. Group A receives Pill X, while Group B receives a placebo (fake drug). • Exp. – Group A receiving Pill X (ind. variable) • Control – Group B receiving placebo

Placebos • Placebo – a substance or treatment that has no effect apart from a person’s belief in it. • Placebo effect – a person receiving the placebo may report to positive effects due to a belief in the drug/treatment • Single blind study – participants do not know if they are in the experimental or control group • Double blind study – participants nor researchers know who is in the experimental or control group • Reduces researcher bias - process where the scientists performing the research unconsciously influence the results, in order to portray a certain outcome.

Experiments • Pros: • Variables can be controlled and manipulated • Can determine cause-and-effect • Can be replicated • Cons: • Labs can not always duplicate real-life environments • Can be expensive • Ethics can prohibit certain experiments

Research Method COMPARING RESEARCH METHODS Basic Purpose How Conducted What is Manipulated? Weakness Descriptive To observe and record behavior Case studies, surveys, or naturalistic observations Nothing No control of variables; single cases may be misleading Correlational To detect naturally occurring relationships; to assess how well one variable predicts another Compute statistical association, sometimes among survey responses Nothing Does not specify cause and effect Experimental To explore cause and effect Manipulate one or Independent more factors; variable(s) random assignment to groups Crash Course – Research Methods Sometimes not feasible; results may not generalize to other contexts; not ethical to manipulate certain variables

Ethics - Animals • Why do we experiment with animals? • Can be similar to humans • Can perform actions that cannot be performed on humans • Problems • Results from animal testing cannot be FULLY generalized to humans because we are not exactly alike • At what point is an action unethical? Starvation? • Social/emotion deprivation? • Termination? •

Ethics - Humans • American Psychological Association (APA) provides a set of guidelines for ethical psychological studies. • The American Psychological Association commits to its vision through a mission based upon the following values: • Continual pursuit of excellence • Knowledge and its application based upon methods of science • Outstanding service to its members and to society • Social justice, diversity, and inclusion • Ethical action in all that we do

To Sum Up • Get proper informed consent • No informed consent is needed in public places. • Informed consent is needed where there is an expectation of privacy. • Informed consent must describe the nature of the experiment, their right to withdraw, and how they can obtain their results. • Avoid or limit deception • Researcher considers a cost-benefit analysis – if the scientific outcomes outweigh the costs then deception can be considered. • Deception cannot be considered if it is expected to cause severe emotional or physical distress. • In all cases, minimize harm. • Always debrief participants • Psychologists must always debrief their research participants as soon as possible after the study.

STATISTICAL REASONING

Isn’t this AP Psych, not AP Stats? • After gathering dating with research methods, it needs to be organized and summarized using statistics so we can make inferences about the data. • We make sense out of data using statistics.

Measures of Central Tendency • a single score that represents a whole set of scores/data • Mean – the average of the scores/values • Median – the middle score /value in a distribution • Mode – the most frequent score/value • Eg: Test scores – 60, 70, 80, 80, 80, 90, 100 • Mean: 80 • Median: 80 • Mode: 80

Measures in Variation • Allows us to see the variation or difference in a set of scores/values. • Range - the difference between the highest and lowest scores/values. • Standard deviation – a measure of how much scores/values vary around the mean. σ= Sum of (deviations from the mean)2 Number of scores

Range and Standard Deviation • Eg: Test scores- 60, 70, 80, 80, 80, 90, 100 • Mean: 80 • Median: 80 • Mode: 80 • Range: 100 -60 = 40 • The measure of dispersion of data • Standard Deviation: 11 • The degree to which scores differ from each other and vary around the mean value for the set (will always fall between 0 and ½ of range) As the standard deviation approaches 0, the closer the scores are to each other. • As the standard deviation approaches ½ of range, the more dispersed the score are. •

Mean, Median, Mode Distribution of Quiz Scores 6 Frequency 5 4 3 2 1 0 0 20 40 Quiz Scores 60 80 100 120 Standard Deviation = 11 points (how much scores vary from each and around the mean)

Normal Distributions • In a typical distribution of data: • 68% of scores will be within 1 standard deviation above or below the mean • 95% of scores are within 2 standard deviations above or below the mean. • 99. 7% of scores are within 3 standard deviations above or below the mean.

Mean, Median, Mode Distribution of Quiz Scores 6 Frequency 5 4 3 2 1 0 0 20 40 Quiz Scores Not a perfect distribution, therefore the quiz is not standardized. 60 80 100 -1 to +1 St. Dev. = 77% -2 to +2 St. Dev. = 100% 120

Statistical Significance • a statistical statement of how likely it is that an obtained result occurred by chance. • Uses a fancy formula that we do not need to know… HOWEVER… When a researcher says… It means… “When the results from the experimental group and the control group were compared, they were found to be statistically significant. ” The results could not have statistically occurred by chance…. There is a direct relationship between the variables. “When the results from the experimental group and the control group were compared, they were NOT found to be statistically significant. ” There is not enough statistical evidence to say that the variables are in fact directly related and the results could have just occurred by chance.