If you have any feedback for the author, E-mail me
REQUIRED TEXT: Experimental Methodology eighth edition by Larry B. Christensen, Allyn and Bacon, 2001 ISBN #0-205-30832-5 ------------------------------------------------------------ Chapter 3 Experimental Research Approach Systematically changing one variable at a time under controlled conditions in order to describe cause and effect relationships. Causation: Rule out alternative explanations Identify causation by: Method of Agreement: Observe common variables surrounding a behavior, e.g. alcohol and water => drunk :. water => drunk Method of Difference: Observe different behaviors holding all variables constant except one. e.g. drug vs placebo on behavior. Joint Method of Agreement and Difference: Combined approach Method of Concomitant Variation: Variation in one variable results in parallel variation in another variable, e.g. different groups of participants receive different amounts of a variable (drug) resulting in ordered changes in behavior across the groups. More than correlations which must eliminate all possible alternative interpretations. Necessary Condition = variable must be present for the behavior to occur, (but even in its presence, the behavior may not occur, e.g. not all who consume alcohol become alcoholics, but you must consume alcohol before becoming alcoholic. Sufficient Condition = variable always produces the behavior, (but the behavior could occur in other ways), e.g. excessive bourbon drinking can produce an alcoholic, but beer, wine, vodka, etc. can also. Causation => both. Position of Falsification: Scientific experiments are designed to test hypotheses that are derived from theory. If a hypothesis is supported by an experiment, only means that the hypothesis has SURVIVED the test, it is not necessarily CONFIRMED. If a hypothesis fails to be supported by an experiment, the prediction is FALSIFIED. A theory is NEVER proved, but only 'not yet disconfirmed' = best status of theory in science. Any event (e.g. an experiment) is imbedded in a network of other related events and for any given effect (e.g. behavior) to occur a balanced series of these events must occur. Causation of human behavior is extremely complex with many antecedent variables being sufficient and/or necessary. Psychological Experiment = Objective observation of phenomena which are made to occur in a strictly controlled situation in which one or more factors are varied and the others are kept constant. As observers, researchers could perceive what they expect rather than what the participant's behavior. :. we must identify potential mistakes and take necessary steps to avoid the errors. p67 Behavior is externally observable and internal events (i.e. emotions) are measured by their externally, observable behavioral manifestations. Psychologist present stimuli which indirectly influence behavior ( S - O - R) in a controlled environment which eliminates extraneous variables (constancy) and allows only one variable to change (variation) at a time. p69 Artificiality is a potential disadvantage when one generalizes from the controlled, experimental environment to the real world. The generalization may never be assumed but demonstrated. (e.g. fetal alcohol syndrome generalization from mice to humans & Darley and Latane's work on safety when perceived to be in a group or be alone was impossible to control outside of the lab but could then be generalized from lab to uncontrolled environment.) But this is a difficult and time consuming process. ------------------------------------------------------------ My office is in room 116C of HGH (826-4571) and I invite you to consult with me on any matter during my office hours or any time that is agreeable to you and me.Go back to the beginning
Copyright © 2002, Dr. John M. Morgan, All rights
reserved -
This page last edited Nov 25, 2002
If you have any feedback for the author, E-mail
me