Ch 2 The need for good research design o Facilitated communication 1990s autism treatment Based on the idea that autism was actually a motor disorder Douglas Biklen Facilitator sits next to child and guides the child s hand over the keyboard allowing the child to type o Children made great progress in communication Wrote poetry told parents I love you Some students made allegations of sexual abuse against their parents No physical evidence just the communicators via the facilitators o Results showed that the words came solely from the minds of the facilitators Some still practice facilitated communication Research Design Matters o Even well educated intelligent people can be fooled o Well planned study designs can help eliminate o Prefrontal lobotomy Developed by Moniz Patients appeared calmer and use spread Relied on subjective clinical reports not research Controlled studies found it to be ineffective How we can be fooled o Heuristics mental shortcuts that helps us streamline thinking Pro evolutionary value Con can oversimplify reality If you were to drive from Reno Nevada to San Diego California in what compass direction would you travel Southwest o Representativeness like goes with like We ignore how common behaviors actually are in a population and commit the base rate fallacy o Which is more dangerous traveling by plane or by car 40 000 in car crashes 1 000 in airplane crashes o Availability heuristic I knew it off the top of my head Estimating the likelihood of an occurrence based on the ease with which it comes to mind o Cognitive biases Systematic errors in thinking that can lead to confidence in false conclusions Hindsight bias I knew it all along Overconfidence the tendency to overestimate our abilities to make correct predictions The Scientific Method Toolbox o Allows us to test specific hypotheses derived from broader theories of how things work o Theories are never proven but hypotheses can be confirmed or disconfirmed o We can use a number of different types of scientific method tools to gain information Naturalistic observation o Watching behavior in real world settings without interfering o High degree of external validity extent to which findings generalize to real world o Low degree of internal validity extent to which we can draw cause and effect inferences Case studies o Studying one person or a small number of people for an extended period of time o Common with rare types of brain damage or mental illness o Help provide existence proofs demonstrations that a given psychological phenomenon can occur But can be misleading or anecdotal Self report measures and surveys o Psychologists often need to ask people about themselves or others Self report measure assesses characteristics such as personality or mental illness Surveys ask about a person s opinions or abilities Choosing participants o Population everybody o Population of interest who you re interested in studying o Sample portion of population of interest that is selected for a study o Representative sample a group of individuals selected from a population for study which matches the population as much as possible e g on important characteristics such as age sex etc o Random sample helps to ensure the representativeness of the sample Evaluating measures o To trust results measures must have Reliability consistency of measurement Validity extent to which a measure assesses what it claims to measure A test must be reliable to be valid but a reliable test can still be completely invalid Self report measures o Pros Easy to administer Direct self assessment of person s state ideas Inexpensive o Cons Accuracy is skewed for certain groups Potential for dishonesty Response sets tendency to distort responses Positive impression management Malingering Correlational designs o Correlation how strongly two variables are related o Correlational study looks for relationships between variables o o o o o o Correlation coefficient direction and strength of relationship between variables Ranges from 1 to 1 Positive one variable increases as another increases goes in one direction Negative one variable increases as another decreases goes in opposite directions Correlational designs Illusory correlation perception of a statistical association between two variables where none exists Correlation vs causation Correlation is NOT causation Three possible explanations A causes B B causes A C causes both A and B Determining causation Experimental design is necessary This is because in an experiment variables are purposefully manipulated rather than simply measuring already existing differences What makes a study an experiment Random assignment of participants to conditions Experimental group receives manipulation Control group does not receive manipulation Manipulation of an independent variable Dependent variable what experimenter measures to see whether manipulation had an effect Cause and effect possible to infer with random assignment and manipulation of the independent variable Experimental concerns Confounds any difference between the experimental control groups aside from the Independent variable Makes independent variable effects uninterpretable Placebo effect improvement resulting from the mere expectation of improvement To reduce participant must remain blind but still some effects Nocebo effect harm resulting from the expectation of harm Experimenter expectancy effect when researchers hypotheses lead them to unintentionally bias a study outcome Clever Hans horse that could calculate complex math problems Double blind designs can decrease this Demand characteristics cues that allow participants to generate guesses regarding the researcher s hypotheses Disguising the study s purpose or using filler items helps Preventing harm Institutional Review Board IRB Informed Consent Confidentiality Debriefing Deception Animal research o Animal research goes through the Institutional Animal Care and Use Committee IACUC o Only 7 8 of psychological research uses animals o Vast majority of animals are rodents and birds o Benefits Conduct medical research that is unethical in humans Development of new vaccines medical treatments Gain understanding of brain and body physiology o Costs 20 million animals are killed every year May experience some pain distress Animal Rights Activists destroy labs Not always applicable to human body Statistics The Language of Research o Descriptive statistics numerical characteristics of the nature of the