A Review of Published Studies Involving Parametric Manipulations of Treatment Integrity

摘要

We reviewed parametric analyses of treatment integrity levels published in 10 behavior analytic journals through 2017. A total of 19 experiments (13 articles) met criteria for review. The objectives were to: (1) determine the number of experiments in which levels of treatment integrity were systematically manipulated; (2) identify the types of errors that were investigated; (3) determine which parts of the intervention procedure were manipulated; and (4) assess the degree to which these errors affected consumer behavior. We also summarized other relevant study features. Although results documented that desired behavior change may occur when treatments feature errors, outcomes are less predictable than treatment outcomes during high-integrity implementation. Our findings also revealed infrequent use of descriptive assessment and that nearly half of the studies introduced both errors of omission and commission. The consequence was the most common treatment component during which errors were introduced. We discuss the general findings from the identified literature as well as directions for future research.