Testing and Individual Differences: Types of Reliability and Validity

Testing and Individual Differences: Types of Reliability and Validity

The following examples all involve a particular type of reliability or validity. Use your reasoning skills to label each of them.

  • Inter-rater reliability
  • Content validity
  • Intra-rater reliability
  • Predictive validity
  • Test-retest reliability
  • Construct  validity
  • Equivalent-form reliability
  • Face validity
  • Split-half reliability
  • Criterion

 \n

  1. Predictive validity: A researcher presents data that indicates that students who earn high scores on the SAT perform better in college than those who earned low scores.

 \n

  1. Intra rater reliability: After a long day of evaluating, an art show judge gives the same score to a photograph as she had given it earlier in the day.

 \n

  1. Content validity: A student complains to her teacher that the test on “ The American Civil War”  has several items on it about post-war “Reconstruction”, a topic not yet covered.

 \n

  1. Construct validity: A graduate student builds a definition of “common sense” and then develops an assessment he claims accurately measures the most critical elements of  that elusive concept

 \n

  1. Face validity: An AP Psychology teacher accidentally gives his class a test from his course AP Furniture Design.  His students immediately realize this isn’t what they have studied.

 \n

  1. Criterion validity: A college admission officer argues the only way to fairly evaluate whether ACT scores actually forecast success in college is to compare those scores to other measures of the high school student’s ability

 \n

  1. Test retest reliability: In a longitudinal study of personality, 10 participants receive scores on a measure of conscientiousness that almost exactly matches their scores on the same measure 20 years earlier.

 \n

  1. Inter rater reliability: Four judges of a drama competition event independently rank each of the plays they viewed in the same order

 \n

  1. Split half reliability: A class collectively scores 87% on the first fifty items of an exam and 86% on the second 50 items.

 \n 10.Equivalent form reliability: A student earns two nearly identical scores on separate SAT tests space one month apart