A researcher is interested in bullying but learns it is difficult to differentiate between actual bullying and normal playful interactions between third graders during recess. After developing her definition of bullying she asks her three graduate students to observe behavior at an elementary school during recess and identify bullying incidents. At the end of the day the researcher compares the independent results of the three researchers. What type of reliability is this researcher using?

Question

Quizplus · Accepted Answer

Inter-rater reliability is being used in this scenario because the researcher is comparing the results of multiple observers to see if they are all identifying the same instances of bullying. Test-retest reliability involves repeating the same test over time to see if results are consistent. Split-half reliability involves splitting a test into two halves and comparing the results of each half. Concurrent/benchmark reliability involves comparing the results of a new test to an established, reliable test.

A Researcher Is Interested in Bullying but Learns It Is