Indian Journal of Dental Research

EDITORIAL
Year
: 2019  |  Volume : 30  |  Issue : 2  |  Page : 163-

Data interpretation and statistical significance


SM Balaji 
 Executive Editor, Indian Journal of Dental Research, Director and Consultant, Balaji Dental and Craniofacial Hospital, Chennai, Tamil Nadu, India

Correspondence Address:
S M Balaji
Executive Editor, Indian Journal of Dental Research, Director and Consultant, Balaji Dental and Craniofacial Hospital, Chennai, Tamil Nadu
India




How to cite this article:
Balaji S M. Data interpretation and statistical significance.Indian J Dent Res 2019;30:163-163


How to cite this URL:
Balaji S M. Data interpretation and statistical significance. Indian J Dent Res [serial online] 2019 [cited 2019 Aug 23 ];30:163-163
Available from: http://www.ijdr.in/text.asp?2019/30/2/163/259222


Full Text

Internationally, there has been a recent uproar in the biostatistical field. It started with a comment in nature to end the “P ≤ 0.05.”[1] It started with three statisticians who called for the abandonment that immediately garnered support from researchers across the world.[1] A series of articles and commentaries begin to appear, the most prominently being in the American Statistical Association Journal's Supplementary Issue 1 of 2019 as “Statistical Inference in the 21st Century: A World.”[1],[2] The reliance on the P ≤ 0.05 has transgressed to monster proportion to obscure the findings of the research enterprise. In pursuit of statistical significance, the clinical significance and translational effects are lost midway.[2]

The researchers should be starting to live and look beyond the arbitrary statistical significance as well as looking to see through the maze of the clinical implications. They would need to garner basic ideas of misconceptions and 360° viewpoint of studies. Application of logic, analytical skill, design, ethical principles, and bias also should be taken into consideration. It is also important to note that the current status of science and statistics, using advanced techniques, permits the study of small effects.[3] There is a real need to employ this aspect in dental research as many clinically important conditions result from comparatively small changes of various biological parameters, such as pH or oxygenation. In this process, we must hold in mind that all “large effects” could have been reported earlier as many might “have already picked the low hanging fruit.”[4]

It is in this view that in the research enterprise, the reproducibility and misconduct have emerged as a hard challenge.[5] It poses a threat to the current integrity of science. The existing approaches suffer from several crises. Most of the presently published results are unreliable due to growing problems with research and publication practices. There is a need for epochal changes in approach to changes in research outlook, empowerment of scientists and the need for more accurate, inspiring, and compelling researches.[4],[5],[6]

I hope that Indian Dental Researchers would also take a cue from the international discussion centered on “P ≤ 0.05” “reproducibility” as well as stay parallel in research design, concepts, hypothesis generation and sample sizes to yield research outputs that benefit humanity.

References

1It's time to talk about ditching statistical significance. Nature 2019;567:283.
2Wasserstein RL, Schirm AL, Lazar NA. Moving to a world beyond “P < 0.05”. Am Stat 2019;73:S1-19.
3Rao UK. Concepts in sample size determination. Indian J Dent Res 2012;23:660-4.
4Fanelli D. Opinion: Is science really facing a reproducibility crisis, and do we need it to? Proc Natl Acad Sci U S A 2018;115:2628-31.
5Eisner DA. Reproducibility of science: Fraud, impact factors and carelessness. J Mol Cell Cardiol 2018;114:364-8.
6Balaji SM. Self-reporting of improper data interpretation. Indian J Dent Res 2015;26:339.