@article{ajams20241232,
author={Ntumi, Simon},
title={Estimating Effect Sizes, Heterogeneity Parameters and Weighted Standard Deviation (WSD) of Postgraduate Theses using Meta-Analytic and Systematic Review Methods},
journal={American Journal of Applied Mathematics and Statistics},
volume={12},
number={3},
pages={41--54},
year={2024},
url={https://pubs.sciepub.com/ajams/12/3/2},
issn={2328-7292},
abstract={The purpose of the study was to conduct a meta-analytic from an institutional repository by exploring the statistical analysis and reporting practices of postgraduate students¡¯ theses in the University of Cape Coast. To achieve this, the study was nested into the quantitative approach where archival data were retrieved from University of Cape Coast institutional repository (UCCIR). The 2020 PRISMA chart flow was used to extract 778 researched studies from the UCCIR. The study found overall medium resultant significant effect size indicating the extracted studies may have some limited practical applications (Min. ES=.378; p=.000; Hg=1.72, z=12.20; Std Err=.238; n=752; Max. ES=.430; p=.012; Hg=.812, z=14.12, Std Err=.623; Overall=.591, p=.000**). Again, the results from the heterogeneity analysis also showed that there was some degree of probability sampling errors in the extracted studies. Also, the study found that most of the extracted studies were likely to produce similar results and conclusions as a result of similar statistical methods employed by the students (T<SUP>2</SUP>=.627, p=.023**, df=6, Z=16.12; CI=95%, n=752; Pq=.934, p=.002**, df=6, Z=12.01; CI=95%; Tau=.723, p=.004**, df=6, Z=30.01; CI=95%; n=752; WSD=.075, p=.013**, df=6, Z=17.23; CI=95%; n=752). From the ensued findings, it was concluded that there were some statistical misapplications and misinterpretations of studies results leading to different conclusions and recommendations from the studies. The study recommended that to improve upon the robustness and rigorousness of postgraduate studies in Ghana, research mechanisms such as series and timely training workshops should be put in place by the university to re-orient and expose postgraduate students to modern statistical methods.},
doi={10.12691/ajams-12-3-2}
publisher={Science and Education Publishing}
}