A meta-analytic review was conducted to assess the current knowledge regarding caregiver training effectiveness for human-human and human-canine dyads. The results showed that most canine-related sources (66%; n=19) were case studies reporting a decrease of learner undesired behavior when using oral instruction/advice (21%; n=6). Most of the human-related research used singlecase designs (57%; n=26) reporting an increase in desired learner behavior (22%; n=10) when caregivers received multi-component training packages, including two or more approaches (17%, n=8). The meta-analysis of betweengroup-design studies (n=18) revealed that interventions had a large effect (Hedges' g=0.88, 95%CI [0.68-1.07]), with packages yielding a slightly larger moderate effect (Hedges' g=0.76, 95%CI [0.60-0.91]) than oral instruction/advice alone (Hedges' g=0.74, 95%CI [0.32-1,15]). Although the shown effectiveness of caregiver training is promising, the results should be interpreted cautiously. Due to the preponderance of case studies within caninerelated literature and the insufficient reporting of data across sources, only few studies could be included in the meta-analysis. Overall, more systematic and comparative research regarding the efficacy of caregivers in behavior change programs across species is needed. 174 words Keywords: systematic-review; human-dog relationship; caregiver-training; interventions qualitative nature. They often describe in great detail the assessment and/or treatment of one or more participants by integrating all information about the case into a unified and related idea or set of ideas (Sturmey, 2009; Virues-Ortega & Moreno-Rodriguez, 2008). Further, their interest is frequently related to reporting new findings that lack replication with group designs, hence, informing about new methods, novel applications of established techniques or unpredicted effects of assessments or treatments (Virues-Ortega & Moreno-Rodriguez, 2008). By contrast, SCDs are special adaptations of interrupted time-series designs and provide a strong basis for establishing causal, or functional, inference by (a) operationally defining the dependent variable; (b) conducting baseline measurements; and (c) replicating experimental conditions (e.g., A-B-A-B) with each subject (