Statistical Significance and Meaningfulness

The soundness of a scientific conclusion of a research paper should be founded on more than statistical analysis. Other aspects such as the correct interpretation of the statistical results are critical in arriving at sound conclusions. Commonly used is the P value, which is applied in the summarization of increased complexity and quantity of data used in scientific studies. This way, the P value has gained prominence as a means to derive conclusions out of studies due to the analytical capabilities of the P value. Many scholars and readers have held the gold standard value of the P value at 0.05 or a 95% confidence level, a position that is disputed by Nahm (2017). The P value indicates the probability of a given statistical model or better, it indicates how incompatible data is with the null hypothesis. The P value, therefore, focuses more on the null hypothesis of a study as described by Nahm (2017).

Drawing upon the reasoning provided in the above paragraph, it is appropriate to express that the footnote was not consequential. The basis of rejecting the null hypothesis ought to have been expanded beyond the P value. Adjusting the P value should not be relied upon as a means to arrive at a conclusion. The interpretation of the conclusion of a scientific study should not be solely based on a single value, which has further been compromised to fit the wishes of the researcher through manipulation. As explained by Wasserstein and Lazar (2016), scientific conclusions, business, or policy decisions should not be arrived at only based on whether a p-value goes beyond a specific threshold. Wasserstein and Lazar (2016) further explain that a p-value is not a good measure of evidence when it comes to models or hypothesis. The researchers should provide more evidence about the footnote.

References

Nahm, F. S. (2017). What the P values really tell us. The Korean journal of pain, 30(4), 241-242.

Wasserstein, R. L., & Lazar, N. A. (2016). The ASA’s statement on p-values: context, process, and purpose. The American Statistician, 70(2), 129-133.