Calculation of Effect Size in Single-Subject Experimental Studies: Examination of Non-Regression-Based Methods

It is observed that meta-analysis studies have not been included in single-subject experimental studies as much as the experimental studies. In order to overcome this deficiency in the literature, regression-based and non-regression-based indexes that can be used as the effect size in single subject experimental studies have been recently developed. Since most of the regression-based indexes are affected by the serial dependency of single-subject experimental data, non-regression-based indexes that were less affected by this dependency and were preferred more than regression-based indexes were the main subject of this study. Although there are many indexes that are not based on regression in single-subject experimental studies, it is observed that most of the researchers prefer the percentage of non-overlapping data (PND) and percentage of zero data (PZD). There are many controversies in the literature especially on the use of the PND index. In the absence of a study describing the alternative indexes of PND and PZD in Turkey literature, the examination of non-regression-based indexes makes this study important. The aim of this study is to examine the non-regression methods used to calculate the effect size in single-subject experimental studies and to show how these methods will be applied in single-subject experimental research. In this aim, how to prepare the data, how to analyze it, how to synthesize it for more than one study and how to interpret the results are discussed. In this study, suggestions were made for the researchers based on 10 different indexes