Inter-Rater Reliability
Reliability is very important to consider in research studies. There are many different types of reliability and one type is called inter-rater reliability. Inter-rater reliability studies are often conducted when you have more than one person who will be evaluating and scoring data. If inter-rater reliability is high, then different people are recording approximately the same values for the behavior they are evaluating.
Many of you are familiar with Albert Bandura’s Bobo doll experiment in 1961. Bandura was a proponent of the social learning theory. In his experiment, he attempted to determine if children would model aggressive behavior.
Let’s say you are planning to conduct a modern-day social learning study and plan to record and rate the aggressive interactions after children have watched acts of violence on television. The potential acts of aggression will be rated from video tapes of the children as they play with a life-sized, blow-up toy. Before you conduct your study, you must train your four research assistants to make sure that they are identifying the same behaviors and end up with the same scores.
Create a report in a 3- to 4-page Microsoft Word document with a detailed description explaining how to train the assistants so that they have high inter-rater reliability. Search Online Library to find some inter-rater reliability studies and see if you can get any ideas. Make sure you operationalize aggression before you start to explain your training plan. Be specific about what your research assistants should do to increase the similarity of their rating scores. Submit the reliability training plan for your research assistants.
