Chapter 2 Research Methods Research methods in psychology Empirical science use observable data to help explain and understand the world around us Inductive Reasoning drawing inferences based on observation Good scientific theories are falsifiable make predictions that can be tested prediction at a time Science progresses by testing and disconfirming theories one Hypotheses predictions very specific testable falsifiable Theories higher order knowledge structures organize and explain facts often used to generate hypotheses Theories should make testable predictions test the specific prediction independent variable what we manipulate dependent variable what we measure Scientific method make observations construct hypotheses test with experiment draw conclusions report results Measurement operational definition How must be very specific someone could replicate in another experiment memory recall of words studied performance on exam ability to remember names of people one just met Operational definition self report participants written or oral accounts of thoughts feelings actions interviews questionnaires social desirability bias lack of awareness Behavior objective observation of actions in lab or neutral setting RT reaction time accuracy trained coders time intensive Biological physiology bodily responses heart rate skin conductance EEG expensive equipment Multiple measures Measurement happiness ask someone self report how happy are you right now how happy are you generally compared to your peers are you typically more or less happy please rate how frequently you experience laughing a light heart warmth towards others ask their friends family other report observe facial expressions behavior brain activity biological Other considerations measurement Reliability tendency to produce the same result whenever it is used to measure the same things how consistently a test measures the variable of interest Validity are we measuring the variable we are trying to measure how accurately a measure captures the construct of interest low validity inaccurate conclusions and predictions high validity accurate conclusions and predictions Testing a hypothesis choose research design descriptive research correlational research experimental research research design draw a sample from the population of interest random selection Descriptive research goal describe behavior and mental processes not trying to make predictions Naturalistic observations Case study Survey Naturalistic observation observations of behavior in the environment in which it typically occurs Jane Goodall Strengths realistic setting Limits can only describe no private or rare events self influence and demand characteristics observer bias poor control Case study intensive examination of a specific person or situation Phineas gage Strengths detailed rare or private phenomena Limits may not be representative Surveys questionnaire or interviews given to many people Strengths lots of data fast inexpensive Limits question wording response bias Correlational Research examines the relationship between two variables is one variable related to another How strong is that relationship sleep and positive mood ask students in hallway rate your mood 1 negative to 7 positive Positive correlation variables increase or decrease in the same direction hours spent studying and grade on exam Negative correlation as one variable 2 increases the other decreases Zero correlation no predictive value number of classes missed and number of smoothies consumed Strengths of correlational research can test predictions evaluate theories suggest new hypotheses useful when cannot manipulate the variables Limits correlation is not causation correlation doesn t equal causation you can never definitively assume causation from a correlational relationship reverse causation married people are more likely to be happy reciprocal causation helping people increases happiness third variable problem ice cream sales are associated with shark attacks Experimental research goal explain behaviors and mental processes by systematically manipulating variables make predictions and infer causal relationships independent variable IV variable manipulated by the experimenter expected to cause a change in another variable Dependent variable DV outcome variable measure by experimenter expected to be influenced by the independent variable unlike descriptive research observers natural phenomenon without intervening unlike correlational research make correlation between two variables experimental research causes some change i e intervenes measure outcome controls the environment Strengths isolate cause and effect relationships Limits confounds other variables that may have influenced the DV controlled often unnatural setting Internal vs external validity internal validity internal control ability to isolate cause effect relationships was the research done properly external validity realistic findings generalize to other settings and samples often a tradeoff between the two 3