Jalil Younesi
Abstract
Evaluating the effectiveness of behavioral interventions in various educational and therapeutic situations in experimental designs requires identifying the structure of collected data and appropriate methods of data analysis. Considering that researchers mainly use ANCOVA to analyze the data of the pretest-posttest ...
Read More
Evaluating the effectiveness of behavioral interventions in various educational and therapeutic situations in experimental designs requires identifying the structure of collected data and appropriate methods of data analysis. Considering that researchers mainly use ANCOVA to analyze the data of the pretest-posttest control group designs in dyadic samples, in this article with a research example of the data collected from a pretest-posttest control group design, the application and the accuracy of the ANCOVA results and the actor-partner interdependence model (APIM) approach have been compared. The results showed that the APIM model contains more information than ANCOVA; The APIM model makes it possible to estimate the effects of actors and partners and consider them in interpretation and conclusions. When the significance level of rejecting or accepting the hypothesis is close to the critical level of 0.05, there is a contradiction between the results of these two methods, and in these cases, one should act very cautiously for interpretation and final conclusions. In cases where the significance level is far from 0.05, there is no problem and the results of the two approaches are consistent. The shift in focus to the unit of analysis allows researchers to study the complex processes and underlying processes of social interactions with much greater validity.
