PA #9-12

• Validity


  The degree to which a certain inference from a test is appropriate, useful or meaningful
• Validity


  The extent to which a test measures what it claims to measure. It addresses what the test is accurate about.
• Validity


  The extent to which a test does the job desired of it; the evidence may be either empirical or logical.
• If a test is reliable, then, necessarily, it is a valid test. But if a test is valid, it may not necessarily be a reliable test.
• Validity is the most important concept in test construction.
• External validity


  The extent to which the results of a research study can be generalized to different situations, different groups of people, different settings, different conditions etc.
• Internal validity


  The extent to which a study is free from flaws and that any differences in a measurement are due to an independent variable.
• Face validity


  The test looks/appears like the test is valid. The test items look appropriate. Not considered to be a true validity.
• Content-related validity


  The extent to which the content of the test represents a balanced and adequate sampling of the outcomes (domain) about which inferences are to be made.
• Content-related validity


  The extent to which the items on a test sample the skills or knowledge needed for achievement in a given field or task.
• Content-related validity


  The extent that it provides an adequate representation of the conceptual domain it is designed to cover.
• Content-related validity


  Based on a set of goals and objectives: table of specification (TOS). The teacher-made test which is based on a particular curriculum or course outline has this kind of validity.
• Content-related validity


  Appropriate for achievement tests or tests whose parameters are defined.
• Content-related validity


  • Outcome weight and number of items for different objectives
• Criterion-related validity


  Correlation between test scores and a criterion. A criterion is a standard against which to compare a trait or characteristic being measured.
• Characteristics of a criterion


  • Available
  • Reliable
  • Free from bias
  • Measurable
  • Relevant
• Predictive validity


  The accuracy with which a test is indicative of performance on a future criterion measure; the relationship between scores on a test and later performance on a criterion.
• Concurrent validity


  The relationship between test scores and current performance on some criterion.
• Factors influencing Criterion-related Validity


  • Test variables differ
  • Criteria differ
  • Groups differ
  • Variability differs
• Construct related validity


  The extent to which a test measures a theoretical construct or trait.
• Construct Explication


  The process of bringing the construct down to the test which involves setting up a nomological network (behaviors related and unrelated to the construct) and ending up with theorizing about the construct.
• Strategies for establishing construct validity


  • Group differences or Contrasting groups
  • Experimentation
  • Developmental changes
  • Standardized tests
  • Internal structure
  • Correlation matrices
  • Factor analysis
• Types of validity


  • Content
  • Criterion: Concurrent
  • Criterion: Predictive
  • Construct
• Types of Tests


  • Achievement tests
  • Aptitude tests
  • Ability tests
  • Personality tests
  • Intelligence tests