EXP PSY - M7

• Independent variable (IV)


  The dimension that the experimenter intentionally manipulates
• Dependent variable (DV)


  The particular behavior we expect to change because of our experimental treatment
• Operational definition


  Specifies the precise meaning of a variable within an experiment by defining it in terms of observable operations, procedures, and measurements
• Experimental operational definition


  • Explains the precise meaning of the independent variables by describing exactly what was done to create the various treatment conditions
• Measured operational definition


  • Describes what procedures are followed to assess the impact of different treatment conditions on the dependent variable
• Hypothetical constructs


  Concepts that cannot be observed directly, but whose existence is inferred from observable behaviors
• Operational definitions are important for both hypothetical constructs and directly observable variables
• Scales of measurement


  The degree of precision used to measure variables, ranging from comparative labels to precise quantitative measurements
• Dependent variables can be measured using different scales, which provide varying amounts of information
• Precise


  More precise than dividing subjects simply on the relative basis of low versus high doses
• Luxes


  Unit of measurement for room illumination that allows other researchers to evaluate and replicate an experiment more thoroughly than knowing only that a room was bright or dim
• Dependent variable


  Can be measured in more than one way, and the amount of information provided will depend on how the variable is measured
• Experiments testing the effects of "anxiety" on affiliation

  • Schachter (1959) experiment - subjects stated whether they wanted to wait with others, alone, or had no preference (nominal scale)
  • Sarnoff and Zimbardo (1961) experiment - subjects rated the intensity of their preference on a scale from 0 (very weak preference) to 100 (very strong preference) (interval scale)
• Researchers generally choose the highest level of measurement possible because it provides more information about a variable
• Levels/Scales of Measurement

  • Nominal
  • Ordinal
  • Interval
  • Ratio
• Nominal


  The simplest level; classifies items into two or more distinct categories that can be named
• Ordinal


  The next level; the magnitude of each value is measured in the form of ranks
• Interval


  Higher level; measures magnitude or quantitative size using measures with equal intervals between the values
• Ratio


  Highest level; measures magnitude or quantitative size using measures with equal intervals between all values and a true zero point
• Definition


  The same variable can have many definitions, so how can we know which definition is best? This question has no hard-and-fast answers. As with many other aspects of experimentation, what works well in one experiment may simply not be appropriate in another. Our definition must be objective and precise so that others can evaluate and replicate the procedures.
• Reliability


  Refers to the consistency of experimental operational definitions and measured operational definitions
• Reliable


  Good operational definitions are reliable: If we apply them in more than one experiment, they ought to work in similar ways each time
• Interrater Reliability


  The degree to which observers agree in their measurement of the behavior
• Test-Retest Reliability


  The degree to which a person's scores are consistent across two or more administrations of a measurement procedure
• Interitem Reliability


  The extent to which different parts of a questionnaire, test, or other instruments designed to assess the same variable attain consistent results
• Validity


  Means the operational definition accurately manipulates the independent variable or measures the dependent variable
• Face Validity


  The degree to which the validity of a manipulation or measurement technique is self-evident
• Content Validity


  How accurately a measurement procedure samples the content of the dependent variable
• Predictive Validity


  How accurately a measurement procedure predicts future performance
• Concurrent Validity


  Compares scores on the measuring instrument with another known standard for the variable
• Construct Validity


  How accurately an operational definition represents a construct
• Construct validity


  Have I succeeded in creating a measuring device that measures the construct I want to test? Are my operational definitions tapping only the construct I want to test?
• Researchers ask whether the data make sense in the context of the overall theoretical framework in which they operate
• One way of evaluating construct validity is to see whether the data follow expected patterns in relation to other concepts
• Internal validity


  The degree to which changes in the dependent variable across treatment conditions were due to the independent variable
• Internal validity establishes a cause-and-effect relationship between the independent and dependent variables
• Extraneous variables


  Factors that are not the focus of experiment but can influence the findings
• Confounding


  When the value of an extraneous variable systematically changes along with the independent variable
• Threats to internal validity


  • History
  • Maturation
  • Testing
  • Instrumentation
  • Statistical regression
  • Selection
  • Subject mortality
  • Selection interactions
• Standardization of the conditions, obtaining information about participants and procedural details, and choosing an appropriate research design can help control threats to internal validity