Research Methods Pgs 63 85 Research Methods Overview Why can t we rely on our judgment for conducting research Types of Research Naturalistic Observation Case Studies Correlational Designs Experiments Types of measurements Ethical guidelines Statistics Why can t we rely on our judgment for conducting research Reasons why We make We are not impartial We make errors in When relying on human judgment goes wrong Prefrontal Lobotomy Heuristics Heuristics mental or rules of thumb that help us to streamline our thinking and make sense of the world Heuristics Friends or foes Pros Cons Intend to save time and effort Work well Oversimplify complex concepts Can lead Representative Heuristic Judging a book by its cover We judge the probability of something based on how Base rate fallacy Availability Heuristic We estimate the likelihood of things based on how easily they come to mind Hindsight Bias I knew that was going to happen Overestimating how well we could have forecasted outcomes Overconfidence Overestimating our ability to make correct predictions Types of Research Scientific Method Skills that allows us to examine research questions Avoids heuristics biases Allows us to test hypotheses Different Types of Research Naturalistic Observation Case Studies Correlational Designs Experiments Naturalistic Observation Studies people in their environment Example Naturalistic Observation Pros and Cons Pro High external validity how much we can Con Low internal validity extent to which we can draw Case Studies Examine one person or a small number of people over an extended period of time Case Studies Pros and Cons Pros Existence proofs Rare Useful insights Cons Hypotheses Correlational Designs Examine the extent to which two variables are associated or related Ex What are correlations A measure of how related two variables are The stronger the correlation the more Correlation r values Correlations can be positive negative or zero Correlations are measured by r r can range from to Examples of r values r r r e r o c S m a x E y g o o h c y s P l of Beers the Night Before Shoe Size Classes Attended Positive Correlations As one the other increase As one the other decreases Negative Correlations As one the other As one the other decreases increases Zero Correlations relationship between the two variables Your Turn Correlation Research 1 Find a partner and develop a list of two numeric variables that you think would be associated i e amount of sleep and GPA and survey observe 5 10 people in class 2 Roughly graph your findings in a scatterplot 3 Discuss the findings with your partner and share your conclusions How do you tell which correlation is strongest The higher the of r the stronger the relationship Sign doesn t matter Quick Quiz Which Correlation is strongest A B C D Correlation DOES NOT Equal Causation When two things are related it DOES NOT mean one the other You can infer causation from a correlation alone Example There s a positive correlation between hours of TV watched and blood pressure in kids Media Errors Low Self Esteem Shrinks Brain A S u r p r i s i n g S e c r e t t o a i n S c h o o l L o n g L i f e S t a y Housework Cuts Breast Cancer Risk F e a r o f h e l l m a k e s u s r i c h e r F e d s a y s Wearing a helmet puts cyclists At risk suggests research Winning World Cup lowers heart attack deaths Eating fish prevents crime Low Self Esteem Shrinks Brain A Surprising Secret to a Long Life Stay in SchoolHousework Cuts Breast Cancer RiskFear of hell makes us richer Fed saysWearing a helmet puts cyclistsAt risk suggests researchWinning World Cup lowers heart attack deathsEating fish prevents crime Illusory Correlations Judgment errors Example we tend to remember things when they confirm our beliefs Fallacy of positive instances Nonevents are forgotten Ex Great Fourfold Table of Life How can we determine causes Experiments Experiments of participants to conditions Manipulation of an variable Random Assignment Randomly sorts participants into groups Confounding variable a difference between groups other than the Variables Independent variable IV Variable the experimenter to see if it causes change in the DV Dependent variable DV Variable the experimenter to see if the IV had an effect on it Example of Study and Variables A researcher wants to know whether what you eat for breakfast effects school performance She has the subjects in her first period class eat a donut for breakfast and has subjects in her second period class eat frosted mini wheats for breakfast She then measures subject s performance on a quiz later that day Is this an experiment Why or why not What is the independent variable What is the dependent variable Pros and Cons Pros infer cause and effect Good internal validity Controls for other variables Cons External validity Placebo Effect Improvements resulting from expectation alone Example Controlling for the Placebo Effect Include a placebo group participants receive a fake treatment i e a sugar pill Blind participants if they re in the experimental or placebo group Experimenter Expectancy Effects Researchers unintentionally bias the outcomes of their study Example Controlling for Experimenter Expectancy Effects Double Blind Designs researchers nor subjects know is in the experimental or control group Hawthorne Effect Participants knowledge that they are being studied can influence their behavior Example Controlling for the Hawthorne Effect Covert observation Participant observation Be aware of it Demand Characteristics Cues that participants pick up from a study that allow them to generate guesses regarding the researcher s May alter their behavior