2Research DefinitionsAn experiment is a process in which an investigator devises two or more different experiences (treatments) for subjects or participants.Involves a control group and one or more treatment groups
3Research VariablesIndependent Variable (IV): the controlled variable in a study; hypothesized to have an effect on the dependent variableIn a true experiment, this is an experimental (manipulated) variable.In a quasi-experiment, this is a subject variable (not subject to manipulation).ExperimentalTypes of Independent Variables:Experimental: systematically controlled or manipulated by the experimenter; levels are set and participants are assigned to the different levelsSubjects (Organismic): a characteristic of the participant; not subject to manipulation e.g. intelligence, personality, etc.The type of variable affects the type of analyses performed and the way the results of statistical analyses are interpretedTo determine, ask: did the condition exist for the subject prior to the experiment?Some variables could go either way (usually environmental variables) e.g. lighting, noise, temperatureLevels of the Independent Variable:An independent variable can have two or more levelsLevels can be independent or dependent (before and after; matched subjects)Subject
4Research VariablesRegardless of type, there will always be two or more levels of the independent variable.Variable: a characteristic or phenomenon that may take on different values; variables must vary!The levels of the independent variable can be:Independent of one another: between-subjects designDependent on one another: within-subjects design
5Research VariablesDependent Variable (DV): an outcome of interest that is observed and measured by the researcher; hypothesized to be affected by the independent variable
6Defining Research Variables Operational Definition: a definition of a variable in terms of the operations used to manipulate it or measure itMust be precise enough that anyone reading a review of your research could replicate your experiment exactly.
7Problem VariablesExtraneous (Nuisance) Variables: uncontrolled variables which can affect the experimental outcomeExtraneous variables become confounding variables when their values change systematically along with the independent variable in an experiment.