Increasing Physicians’ Awareness of the Impact of Statistics on Research Outcomes

Bridge, Patrick D.; Sawilowsky, Shlomo S.

doi:10.1016/s0895-4356(98)00168-1

Cited by 206 publications

(43 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Shapiro-Wilk (SW) test was performed to determine whether a data set of each group is well-modeled by the normal distribution, that is, with the p value greater than 0.05 [45]. To assess statistical significance of the differences between two groups, a two sided Student's t-test was used if both groups followed the normal distribution, otherwise, the Wilcoxon rank-sum test was used [46]. The p value is equal to or less than 0.05 regarded as statistically significant in this study.…”

Section: Methodsmentioning

confidence: 99%

Correlation analysis between 2D and quasi-3D gamma evaluations for both intensity-modulated radiation therapy and volumetric modulated arc therapy

Kim

Choi

et al. 2016

Oncotarget

View full text Add to dashboard Cite

The aim of this work was to investigate correlations between 2D and quasi-3D gamma passing rates. A total of 20 patients (10 prostate cases and 10 head and neck cases, H&N) were retrospectively selected. For each patient, both intensity-modulated radiation therapy (IMRT) and volumetric modulated arc therapy (VMAT) plans were generated. For each plan, 2D gamma evaluation with radiochromic films and quasi-3D gamma evaluation with fluence measurements were performed with both 2%/2 mm and 3%/3 mm criteria. Gamma passing rates were grouped together according to delivery techniques and treatment sites. Statistical analyses were performed to examine the correlation between 2D and quasi-3D gamma evaluations. Statistically significant difference was observed between delivery techniques only in the quasi-3D gamma passing rates with 2%/2 mm. Statistically significant differences were observed between treatment sites in the 2D gamma passing rates (differences of less than 8%). No statistically significant correlations were observed between 2D and quasi-3D gamma passing rates except the VMAT group and the group including both IMRT and VMAT with 3%/3 mm (r = 0.564 with p = 0.012 for theVMAT group and r = 0.372 with p = 0.020 for the group including both IMRT and VMAT), however, those were not strong. No strong correlations were observed between 2D and quasi-3D gamma evaluations.

show abstract

Section: Methodsmentioning

confidence: 99%

Correlation analysis between 2D and quasi-3D gamma evaluations for both intensity-modulated radiation therapy and volumetric modulated arc therapy

Kim

Choi

et al. 2016

Oncotarget

View full text Add to dashboard Cite

show abstract

“…The differences in these means were examined statistically. The datasets generated were not normally distributed (data not shown); skew in the data may affect Student’s t -test (Bridge and Sawilowsky, 1999). Thus, the non-parametric Mann–Whitney test was conducted on the datasets.…”

Section: Methodsmentioning

confidence: 99%

The Presence of the DNA Repair Genes mutM, mutY, mutL, and mutS is Related to Proteome Size in Bacterial Genomes

García-González¹,

Rivera-Rivera²,

Massey³

2012

Front. Gene.

View full text Add to dashboard Cite

DNA repair is expected to be a modulator of underlying mutation rates, however the major factors affecting the distribution of DNA repair pathways have not been determined. The Proteomic Constraint theory proposes that mutation rates are inversely proportional to the amount of heredity information contained in a genome, which is effectively the proteome. Thus, organisms with larger proteomes are expected to possess more efficient DNA repair. We show that an important factor influencing the presence or absence of four DNA repair genes mutM, mutY, mutL, and mutS is indeed the size of the bacterial proteome. This is true both of intracellular and other bacteria. In addition, the relationship of DNA repair to genome GC content was examined. In principle, if a DNA repair pathway is biased in the types of mutations it corrects, this may alter the genome GC content. The presence of the mismatch repair genes mutL and mutS was not correlated with genome GC content, consistent with their involvement in an unbiased DNA repair pathway. In contrast, the presence of the base excision repair genes mutM and mutY, whose products both correct GC → AT mutations, was positively correlated with genome GC content, consistent with their biased repair mechanism. Phylogenetic analysis however indicates that the relationship between the presence of mutM and mutY genes and genome GC content is not a simple one.

show abstract

“…before vs. after design) are more powerful to begin with (given the reduction in error variance associated with individual differences), thus non-parametric approaches offer little power advantage under these conditions. 3 Among the reviewed studies, the most relevant to our discussion are those with between-subject-design, but 75% of these used parametric analyses. Taken together, these results suggest that parametric analyses should not be the primary statistical method for analyzing CPT data as they currently are.…”

Section: Discussionmentioning

confidence: 99%

Experimental Comparison of Parametric Versus Nonparametric Analyses of Data From the Cold Pressor Test

Treister¹,

Nielsen²,

Stubhaug³

et al. 2015

The Journal of Pain

View full text Add to dashboard Cite

Parametric statistical methods are common in human pain research. They require normally distributed data, but this assumption is rarely tested. The current study analyzes the appropriateness of parametric testing for outcomes from the cold pressor test (CPT), a common human experimental-pain test. We systematically reviewed published CPT studies to quantify how often researchers test for normality and how often they use parametric vs. non-parametric tests. We then measured the normality of CPT data from 7 independent small-to-medium cohorts and one study of >10,000 subjects. We then examined the ability of two common mathematical transformations to normalize our skewed data-sets. Lastly, we performed Monte Carlo simulations on a representative dataset to compare the statistical power of the parametric t-test vs. the nonparametric Wilcoxon Mann Whitney (WMW) test. We found that only 39% of published CPT studies (47/122) mentioned checking data distribution, yet 72% (88/122) used parametric statistics. Furthermore, among our 8 data sets, CPT outcomes were virtually always non-normally distributed and mathematical transformations were largely ineffective in normalizing them. The simulations demonstrated that the non-parametric WMW test had greater statistical power than the parametric t-test for all scenarios tested– for small effect sizes, the WMW had up to 300% more power.

show abstract

Increasing Physicians’ Awareness of the Impact of Statistics on Research Outcomes

Cited by 206 publications

References 41 publications

Correlation analysis between 2D and quasi-3D gamma evaluations for both intensity-modulated radiation therapy and volumetric modulated arc therapy

Correlation analysis between 2D and quasi-3D gamma evaluations for both intensity-modulated radiation therapy and volumetric modulated arc therapy

The Presence of the DNA Repair Genes mutM, mutY, mutL, and mutS is Related to Proteome Size in Bacterial Genomes

Experimental Comparison of Parametric Versus Nonparametric Analyses of Data From the Cold Pressor Test

Contact Info

Product

Resources

About