J Korean Acad Prosthodont.  2012 Oct;50(4):258-270. 10.4047/jkap.2012.50.4.258.

Assessment of statistical errors of articles published in the Journal of the Korean Academy of Prosthodontics: 2006 - 2010

• Kang DW
• Seo Y
• Oh NS
• Lim HJ

• KMID: 2195467
• DOI: http://doi.org/10.4047/jkap.2012.50.4.258

Abstract

PURPOSE
Use of inappropriate statistical methods may lead to incorrect conclusions and a waste of valuable resources. The goal of this study was to assess the frequency and the types of several common statistical errors in the published articles of the Journal of the Korean Academy of Prosthodontics (JKAP) for a 5-year period.
MATERIALS AND METHODS
Of 336 articles in the JKAP published from 2006 to 2010, 255 articles using statistics were reviewed and classified by statistical method and year. The frequency and types of the statistical methods were examined, and the statistical errors were evaluated by the appropriateness of the experimental design, assumption check, independent outcomes, proper sample size and suitable use of statistical method. Statistical guidelines were completed based on the appropriateness.
RESULTS
Of the 255 articles using statistics, 193 articles (75.9%) used inferential statistics and 153 articles used SPSS statistical software (60.0%). Of the articles using inferential statistics, the three most frequently used statistical methods were ANOVA (41.5%), t-test (20.0%), and the nonparametric method (16.9%). The average rate of statistical errors was 61.2 percent, similar to the rate reported by several studies completed for the medical journal.
CONCLUSION
After the whole analysis of the difference among the groups, post-hoc tests for the pairwise comparisons are required. The optimal sample size calculation is an essential part of this study protocol. To minimize the occurrence of statistical errors, statistical guidelines were developed according to each statistical test procedure and will contribute to the academic improvement in the JKAP.

Keyword

Statistical error; Statistical method; Journal of the Korean Academy of Prosthodontics

MeSH Terms

Prosthodontics
Research Design
Sample Size

Figure

• 

  Fig. 1 A: 2 × 4 experiment design, B: 2 × 4 dependent experiment design, C: 2 × 2 × 3 experiment design.

Reference

1. Park DG, Choi YG, Kim YS, Shin SW. An assessment of statistical errors of articles in the Journal of Korean Academy of Prosthodontics: Comparison between Korean version and English version. J Korean Acad Prosthodont. 2009. 47:273–285.

2. Kim YJ, Jang HS, Kim BO. Evaluation of statistical methods in the Journal of Korean Academy of Periodontology published from 2000 to 2006. J Korean Acad Periodontol. 2009. 39:399–405.

3. Lim HJ. A Review of Statistical Methods in the Korean Journal of Orthodontics and the American Journal of Orthodontics and Dentofacial Orthopedics. Korean J Orthod. 2004. 34:371–379.

4. Kim JS, Kim DK, Hong SJ. Assessment of errors and misused statistics in dental research. Int Dent J. 2011. 61:163–167.

5. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986. 1:307–310.

6. Kuo YH. Extrapolation of correlation between 2 variables in 4 general medical journals. JAMA. 2002. 287:2815–2817.