In order to do the analysis, the results had been observed from 75 different reports. Some of the reports which had been represented had a completely individual research group. Among the different reports, around 68 did the assessment of the outcomes at post, 8 also collected some of the follow up information. These post effects data had been based on some of the end points. Also the 8 also collected some of the follow up information. Only 7 of these included the follow up data. The final study sample consisted of the examples from 21st CLCC and some other kind of local initiatives which had been supported by different organizations. Each of the effect had been corrected for any specific sample bias.
In order to study the model, a random effects model had been analyzed. In order to implement the consistent strategy was to do the calculation of the effect size for each analysis. In the first analysis from the different programs which were there at post, the average of the effect sizes has been studied to get one effect. A random effect model had been used for the purpose of the analysis. This random effect model does the assumption that there may be variation in the SMD’s which are there across the studies. Throughout the analysis, around 0.5 % probability was used. The mean impact for doing the study of the groupings had been reported with the help of 0.5 confidence intervals. Other than this, the homogeneity analysis had also been conducted in order to do the analysis if the SMD’s estimated the same population effect. This analysis had been based on the Q statistic which had been distributed as a chi-square which has around k-1 degrees of freedom. Here the valye of k is the number of studies. Other than this the I2 statistic has also been obtained. After the analysis, two primary methodological features had been coded as present or absent. In addition to this the use of the randomized data had been given.