Informal Aggregation Technique for Software Engineering Experiments

One of the bases for development and standardization in any discipline is continuous empirical verification of knowledge. Thus, empirical replication is required to validate or establish experimental results. When these replications occur, there is also a need to synthesize the different results for a common result. Software Engineering experiments usually fall short of the requirements of the meta-analysis techniques currently in use for this purpose either in number, report or even design. Therefore, there is a need for a less rigorous technique that could serve even as a temporary tool but suitable to software engineering studies and also usable. This study presents an alternative less rigorous aggregation techniques that systematically synthesize the statistic ANOVA results by grouping treatments with seemingly equal level of evidence together. Each group is then ranked on an ordinal scale, grouped and interpreted.