Meta-analysis traditionally relies on parametric methods to synthesize quantitative data across studies, with effect sizes and their associated variabilities being central to this process. However, in many cases, articles reviewed for meta-analysis do not provide complete quantitative data. Often, only qualitative or descriptive statements are available, such as the direction (increase or decrease) of observed effects, without reference to the magnitude or variability. These studies pose a challenge for traditional parametric meta-analysis. This article proposes a non-parametric approach to overcome these limitations, offering an innovative solution for synthesizing data when traditional effect sizes are unavailable. We discuss several non-parametric techniques, including rank-based methods and bootstrapping, to analyze and combine results from studies that report only directional information. The article demonstrates how these methods can be applied in practice and introduces new statistical tools for meta-analysis that accommodate studies with limited quantitative data. By extending meta-analysis methodology to non-parametric approaches, this work is expected to provide researchers with a wider toolkit for synthesizing findings from a broader range of studies, ultimately enhancing the generalizability and robustness of meta-analytic results in general. The non-parametric approach presented is particularly relevant for fields where studies frequently report qualitative data or when effect size information is basically missing.