Chapter 13 Objectives

After studying this chapter and completing the structured exercises in the module, you should be able to:

• Differentiate among methods for obtaining, analyzing, and reporting evidence to support the validity of particular score interpretations.
  
  • Convergent and Discriminant Validity Evidence
    
  • Evidence of Predictive Validity
    
  • Evidence of Concurrent Validity
    
  • Evidence of Internal Structure
    
  • Evidence of Group Differences
    
  • Evidence of Lack of Bias in Items/Test Scores
    
• Interpret the needed forms of validity evidence in a unified manner in given contexts of assessment use.
  
• Describe key concepts from classical reliability theory: observed scores, true scores, and standard error of measurement.
  
• Describe the standard error of measurement and reliability coefficient as two alternate methods for estimation of score reliability.
  
• Interpret different types of reliability coefficients, errors estimated, and data collection designs for estimating errors of measurement (Split-half, K-R 20, Cronbach’s alpha, test-retest, inter-rater and parallel forms reliability coefficients).

The purpose of the Chapter Highlights is to help you remember key concepts. As you review the summary, click on underlined terms (terms with activated links) to retrieve glossary definitions. If too many terms seem unfamiliar or you miss too many items on the structured comprehension exercises and tests provided for each chapter in this computer module, you may want to re-read the chapter or see your instructor.