Research Methods Enduring Understanding: It is actually way more exciting than it sounds!!!! Psychologists use various methods to conduct their scientific research, each having benefits and drawbacks. Research Methods Problems Research Methods Problems: 1.) Hindsight Bias 2.) Overconfidence 3.) The Barnum Effect Hindsight Bias The tendency to believe, learning the outcome, that you knew it all along. Cubs FINALLY win World Series in 2016. AFTER they win, you said you knew they d win all along. Really, did you? Or are you just telling a fib? Or is it hindsight bias? Monday Morning Quarterbacking!!! Overconfidence 82% of U.S. drivers consider themselves to be in the top 30% of their group in terms of safety. 81% of new business owners felt they had an excellent chance of their businesses succeeding. When asked about the success of their peers, the answer was only 39%. (Now that's overconfidence!!!) Overconfident driving skills 1
The Barnum Effect It is the tendency for people to accept very characterizations of themselves and take them to be accurate. Another fatal flaw Perceiving order in events. Flip a coin 20 times. What type of pattern do you think you ll see? Random things (like a number coming up on the roulette wheel) are completely random. There is no real pattern. Ever. In random sequences, patterns and streaks (like repeating digits) occur more often than people expect.. Critical Thinking Smart thinking Elements Examines assumptions Assesses the source Discerns hidden values Confirms evidence Assesses conclusions The Amazing James Randi! TesT Yourself 1.) Mrs. Yoder notices that the last four times she has been to Giant she has scored a parking spot right up front! She knows she is on a lucky streak! 2.) Brian is often called a Monday Morning Quarterback by his friends for saying he knew Nelson Agholor would be a bust. 3.) Janelle, a senior in high school with a 3.0 GPA, is filling out her college applications. When asked by her friends and family what schools she is apply to & which ones she ll get in, she lists Princeton, Yale, Stanford, & Harvard and says she thinks she will get into all of them, except maybe Harvard, which is her reach school. TesT Yourself 4.) Shrek & Fiona break up. Their classmate, Donkey, tells his mother that he knew all along the two of them were not going to make it. 5.) Janelle, a student in your class, is certain that Mrs. Yoder does not like her. For the last three class sessions, Yoder has not called on her to answer a question, even though her hand was raised. 6.) Carolina just took the latest Buzzfeed test online, the one about which drink you are most like, she feels like the computer read her mind because the results describe her perfectly! 2
Applied vs. Basic Research Applied Research has: YOU CAN USE IT!!! Basic Research explores questions that you may be curious about, but not intended to be used. Studying how kissing changes when you get older is interesting but that s about it. Research on therapies for drug addicts has a clear purpose. Research Methods Terms A theory is an using an integrated set of principles that organizes observations and predicts behaviors or events, often highly researched A hypothesis is a, often implied by a theory. In psych, it is a statement between or among variables. Ie participating in class leads to better grades Research Methods Terms Replication importance: If a result is true, it should show itself over and over. If a study shows that gratitude increases positive well-being, then replication of that study should show the same results. Replicated studies with the same results can lead to advancing our knowledge. Generalizability: The PSYCH-nTific Method Develop a question Develop a hypothesis Design study and collect data Analyze the data Publish the findings In Psych We Test Hypotheses & Refine Theories Using: Descriptive methods: describe behaviors with case studies, surveys or naturalistic observations Correlational methods: Experimental methods: manipulate variables to discover effects and determine 3
How does a researcher choose which method to use? It depends on the research question. Each research method has advantages and disadvantages. Descriptive Methods Naturalistic Observation Case Studies Surveys Naturalistic Observation Observing and recording behavior in natural environment as it naturally occurs. No control - just an observer. ADVANTAGES Allows description of behavior as it occurs in the environment. DISADVANTAGES Allows researcher little or no control of the situation. Observations may be. Ethical boundary spying? Beware of the Hawthorne Effect The alteration in behavior of the subjects in a study due to: Whether the lights were brighter or dimmer, production went up in the Hawthorne electric plant. Laboratory Observation Case Studies investigation of an individual, small group, or situation (typically over an extended period of time). ADVANTAGES Allows use of sophisticated equipment. DISADVANTAGES Allows researcher only limited control of the situation. Observations may be biased. ADVANTAGES Provides in-depth information on individuals. DISADVANTAGES Vital information may be missing, making the example difficult to interpret AND/OR the person s memories may be selective or inaccurate. Does not allow conclusions about cause and effect. Behavior may differ from behavior in the environment. Unusual examples can shed light on situations / problems that are unethical or impractical to study in other ways. The individual may not be. Limits generalizability. Time consuming & costly 4
CLASSIC CASE STUDIES IN PSYCHOLOGY The Survey Says. Provides a amount of information on a lot of people. Cheap / Easy Anonymous Questionnaires and interviews that ask people directly about their experiences, attitudes, or opinions. ADVANTAGES DISADVANTAGES Can ask about things that aren t easy to observe. Most type of study used in psych. If sample is non-representative or biased (Volunteer bias), it may be impossible to generalize from the results. : people respond in a way they know the researcher wants them to The of surveys may be biased and participants are not always honest response rate Subjective self-report Population & Samples Population: Psychologists would like to generalize what they learn to all people but due to time and money, they have to pick a sample of the total population Sample: of a population who is actually IN the study Random sample (random selection): every person has an equal chance of participating (names in a hat) Representative sample: characteristics that are similar to those in the population (CB East kids; 10 kids from each grade) A random sample typically generates a representative sample Sampling bias: An in the sampling process that allows some members of a population to be likely than others to be included in a study. N = population size n = size Abbreviations Example: Of the students enrolled in Yoder s AP Psych Course (N = 109), the most frequent complaint was too much reading (n = 58). Cross-Sectional and Longitudinal Studies Which descriptive technique would be best utilized in each case: a case study (CS), naturalistic observation (NO) or survey (S). 1. Mr. Lucabaugh wants to determine whether he should use the building funds for a renovated student courtyard or a faculty exercise room. 2. An army doctor wants to see how soldiers are handling the transition back to civilian life. 3. A parent is curious to know how their child behaves when away from home at school. 4. An animal researcher wants to prove that squirrels run in packs as wolves do. Cohort Effects: 5. The Central Bucks School Board is trying to determine if teachers and students feel the building facilities are adequate and safe. 5
Correlational Research *Important to note that surveys can also be used to determine correlations. Correlational Research Determines if a relationship exists between two things, how the relationship is, & if it s. Can NOT say that one variable CAUSES another. There is a positive correlation between ice cream and murder rates. Does that mean that ice cream causes murder? Correlational Research Shows how closely related two sets of scores are to each other How well does A predict B? How well is A related to B? No manipulation of variables in correlational studies so RRRrrrrr. Netflix & Not so Chill A positive or correlation, shows variables that either rise or fall together # of HW assignment turned in increases, a person s grade in AP Psyche also increases A negative or INVERSE correlation, A measurement called a correlation coeffiecient (represented as r) describes the strength and direction of the relationship. You will not have to calculate! # of hours watching Netflix increases, a person s GPA decreases 6
How to Read a Correlation Coefficient Correlations Correlation coefficients range from The number (-.38) represents the of the correlation and the sign represents the DIRECTION A correlation coefficient of -.70 indicates a stronger relationship than a coefficient of +.65 A correlation of 0 shows that the two sets are not related Correlation Does NOT Equal Causation Illusory Correlations Illusory correlation: I saw a few very short students getting A s on my quizzes in class, therefore, short kids must be smarter! Correlations are useful for predictions and avoidance of unethical experiments where it is impossible to manipulate the independent variable Death of a family member and levels of depression the following year 7
Third Variables Third variable problem: Positive correlation between murder rate and the sale of ice-cream. Third variable which includes a variable Weather is cold, fewer people are out interacting with others and less likely to purchase ice-cream. Hot outside, more social interaction and more ice-cream being purchased Weather is the variable that confounds the relationship between ice-cream sales and murder Predict whether the correlation would likely be positive (P) or negative (N): 1. The number of fast food restaurants: the obesity rate in the U.S. 2. The average U.S. household income: annual gross profit reported by U.S. retailers. 3. The illiteracy rate: the presence of Head Start or early intervention education programs. 4. The number of hours spent commuting to and from work; the amount of dinners cooked at home from scratch. 5. Hours spent learning a skill; proficiency in the skill. Experimental Research Experimental Research Explores cause and effect relationships. Drinking too much Redbull causes -like behavior. EXPERIMENTS allow the investigator to CONTROL a situation & decrease the possibility that unnoticed, outside factors will influence the results. In an experiment, some event, treatment, or condition, is changed, controlled, or manipulated by the psychologist so they can observe the effects. PRO - ONLY research method to: When psychologists set up an experiment, they think, If I do X, the people in my study will do Y. X Independent variable Y Dependent variable CON - possibility of confounding variables & other problems with methodology 8
Independent Variable Independent Variable: Dependent Variable Dependent Variable: Hopefully the independent variable brings about change. If there is a drug in an experiment, the drug is almost always the independent variable. The independent variable (IV) is the CAUSE and the dependent variable (DV) is the EFFECT. If the dependent variable changes ONLY when the IV changes, we can conclude the changes in the IV caused the changes in the DV. CollegeBoard LOVES independent & dependent variables. Sugar, Sugar You are my Null Hypothesis Null hypotheses: IV has no impact on DV Researchers work to Sugar has no impact - Yeah, right! "Hyperactivity is unrelated to eating sugar" is an example of a null hypothesis. If the null hypothesis is tested and found to be false, using statistics, then a connection between hyperactivity and sugar ingestion may be. Operational Definitions The definition of behaviors or qualities Explain what you mean in your hypothesis. How will the variables be measured in real life terms. Let s say your hypothesis is that chocolate causes violent behavior. What do you mean by chocolate? What do you mean by violent behavior? Operationalize This! Create operational definitions for the following variables: 1.Happiness 2.Intelligence 3.Popularity 4.Good music 9
Do Cell Phone Use and Driving Mix? Experimental vs. Control Group Experimental group Members of an experiment who are to the treatment variable or manipulation by the researcher; represents the treatment group. Control group Participants in an experiment who are not exposed to the treatment variable; this is the group. PARTICIPANTS FOR THE CONTROL & EXPERIMENTAL GROUPS MUST BE RANDOMLY SELECTED & ASSIGNED!!!! Same, same but VERY diff! Random sampling is the first step in selecting a population to study and will result in a test group of that population. Random assignment (which only applies to EXPERIMENTS) is the manner in which participants in the experimental or control group and this should be representative as well with each participant having an change of being in either group. Random Assignment Controlling for confounding variables such as parental intelligence and environment Participants Potential problems with experiments 1 - confounding variables 2 placebo effect 3 - assignment problems 4 - bias a.) experimenter bias b.) demand characteristics (participant bias) Confounding Variable The interference by a third so as to distort the being studied between two other variables Could be the difference between the two groups other than the IV Example: Vitamin X vs. Vitamin Y. Group 1 run in the morning. Group 2 in the afternoon. Do you see a problem with this? 10
Beware of Confounding Variables If I wanted to prove that smoking causes heart issues, what are some confounding variables? The object of an experiment is to prove that. A confounding variable is anything that could cause change in B, that is not A. Another Problem Placebo - An action or substance given to members of the control group; the that has no benefit, but is administered as if it does PLACEBO EFFECT - the of performance due to the subject s about the results - For instance, the subject may feel better simply because they THINK they are going to. (the power of the MIND!!!) QuickTime and a TIFF (Uncompressed) decompressor are needed to s ee this picture. Lifestyle and family history may also effect the heart. ASSIGNMENT PROBLEMS Sometimes we CAN NOT randomly assign participants to a particular group. For instance, sometimes we will compare boys to girls or old people to young - these types of studies are called. There are potentially confounding variables with these. Types of Bias Experimenter bias - (experimenter expectancy effect) - a phenomenon that occurs when a about the outcome of a study influence the obtained (may not even be aware he s doing this) Demand characteristics - (participant bias) - the the participants gain about the study suggesting how they should respond Clever Hans Reading Analysis Did Hans know math? What was Hans doing? What problem do you think this may illustrate with conducting research? Ways to control for biases & improve experimentation Random assignment Single & double blind studies Standardization Replication Reliability & validity 11
Single & Double Blinds Single-blind study Experimenter knows who is in which group do not Double-blind study experimenter nor participants know who is in which group Eliminates experimenter bias Ways to Improve Experimentation STANDARDIZATION - the set of instructions & procedures are used so the experiment is exactly the same for everyone REPLICATION - MUST be able to reproduce experiment using different subjects & experimenters but procedures & manipulations - (similar) results in order to be considered reliable. VALIDITY vs. RELIABILITY How CONSISTENTLY individuals score on a test or research study. Validity - Does my Experiment test what it is supposed to test? Internal Validity: How well constructed was the experiment to control for variables? (think INSIDE the exp.) Content Validity: Learning History of Pysch & Research Methods, test questions on this content and NOT Abnormal Psych & Treatment IQ tests are strictly timed, may be a good measure of speed of processing. However, if researchers do not believe processing speed is an important component of intelligence, content validity would be questioned External Validity: How well do the results of the experiment other settings or participants? (think OUTSIDE) Types of Experiments BETWEEN SUBJECTS DESIGN - WITHIN SUBJECTS DESIGN - the behavior of an experimental participant BEFORE receiving treatment is compared to his behavior AFTER receiving treatment* If two treatments are used, the order of them could cause an effect - can eliminate this. Ex: 12