Abstract
Studies in applied behavior analysis have used percentage agreement as an index of interobserver reliability. Because percentage agreement is inflated by chance agreements, kappa has been recommended as a preferred alternative. In this article, we present a procedure to estimate maximum and minimum kappa from percentage agreement with limited information. The procedure will facilitate comparisons between studies and the reassessment of the reliabilities of previous studies.
Original language | English (US) |
---|---|
Pages (from-to) | 375-378 |
Number of pages | 4 |
Journal | Behavioral Assessment |
Volume | 6 |
Issue number | 4 |
State | Published - Sep 1984 |
All Science Journal Classification (ASJC) codes
- Psychiatry and Mental health