Why we need to report more than 'Data were Analyzed by t-tests or ANOVA'

Weissgerber, Tracey L.; Garcia‐Valencia, Oscar; Garovic, Vesna D.; Milić, Nataša; Winham, Stacey J.

doi:10.7554/elife.36163

Cited by 55 publications

(52 citation statements)

References 30 publications

(41 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two reviewers applied the assessment criteria (see Appendix 2) independently to appraise the statistics reporting quality of included studies. This checklist was established based on the SAMPL guidelines (6) and other previously published studies (10-12), and the items were modified to be listed in a simple and readable manner. All of the logistic regression assessment items droved from Zhang’s research (13), and Cox regression items were from Zhu’s research (14).…”

Section: Methodsmentioning

confidence: 99%

Assessing the statistical reporting quality in high-impact factor urology journals

Dai

et al. 2020

Preprint

View full text Add to dashboard Cite

11Backgrounds Observational studies plays an important role in urology studies, But few studies have 12 paid attention to the statistical reporting quality of observational studies. The purpose of this study 13 was to investigate the frequency and evaluate the reporting quality of statistical methods of the 14 published observational studies in urology. 15 Methods The five urology journals were selected according to the 5-year impact factor. A systematic 16 literature search was performed in PubMed for relevant articles. The quality of statistical reporting was 17 assessed according to assessment criteria. 18 Results A total of 193 articles were included in this study. The mean statistical reporting score of 19 included articles was 0.42 (SD=0.15), accounting for 42% of total score. The items that must be 20 reported with a reporting rate more than 50% were: alpha level (n=122, 65.2%), confidence intervals 21 (n=134, 69.4%), name of statistical package (n=158, 84.5%) and exact P-values (n=161, 86.1%). The 22 items with a reporting rate less than 50% were: outliers (n=2, 1.0%) and sample size (n=13, 6.7%). For 23 multivariable regression models (liner, logistic and Cox), variables coding (n=27, 40.7%), validation 24 analysis of assumptions (n=58, 40.3%), interaction test (n=43, 30.0%), collinearity diagnostics (n=5, 25 3.5%) and goodness of fit test (n=6, 5.9%) were reported. Number of authors more than 7(OR=2.06, 26 95%CI=1.04-4.08) and participation of statistician or epidemiologist (OR=1.73, 95%CI=1.18-3.39) 27 were associated with the superior reporting quality. 28 Conclusion The statistical reporting quality of published observational studies in 5 high-impact factor 29 urological journals was alarming. We encourage researchers to collaborate with statistician or 30 epidemiologist. The authors, reviewers and editors should increase their knowledge of statistical 31 methods, especially new and complex methods. 32

show abstract

Section: Methodsmentioning

confidence: 99%

Assessing the statistical reporting quality in high-impact factor urology journals

Dai

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Health was monitored by weight (twice weekly), food and water intake, and general assessment of animal activity, panting, and fur condition…. The maximum size the tumors allowed to grow in the mice before euthanasia was 2000 mm 3 ." [205] Item 17.…”

Section: Examplementioning

confidence: 99%

“…However, evidence shows that the majority of publications fail to include key information and there is significant scope to improve the reporting of studies involving animal research [1][2][3][4]. To that end, the NC3Rs published the ARRIVE guidelines in 2010.…”

Section: Introductionmentioning

confidence: 99%

Reporting animal research: Explanation and Elaboration for the ARRIVE guidelines 2019

Sert

Ahluwalia

Alam³

et al. 2019

Preprint

View full text Add to dashboard Cite

Improving the reproducibility of biomedical research is a major challenge. Transparent and accurate reporting are vital to this process; it allows readers to assess the reliability of the findings, and repeat or build upon the work of other researchers. The NC3Rs developed the ARRIVE guidelines in 2010 to help authors and journals identify the minimum information necessary to report in publications describing in vivo experiments.Despite widespread endorsement by the scientific community, the impact of the ARRIVE guidelines on the transparency of reporting in animal research publications has been limited. We have revised the ARRIVE guidelines to update them and facilitate their use in practice. The revised guidelines are published alongside this paper. This Explanation and Elaboration document was developed as part of the revision. It provides further information about each of the 21 items in ARRIVE 2019, including the rationale and supporting evidence for their inclusion in the guidelines, elaboration of details to report, and examples of good reporting from the published literature.As a new international working groupthe authors of this publication, we have revised the guidelines to update them and facilitate their uptake; the ARRIVE guidelines 2019 are published alongside this paper [13]. We have updated the recommendations in line with current best practice, reorganised the information and classified the items into two sets. The ARRIVE Essential 10 constitute the minimum reporting requirement and the Recommended Set provides further context to the study described. The two sets help authors, journal staff, editors and reviewers use the guidelines in practice, and allow a pragmatic implementation with an initial focus on the most critical issues. Once the Essential 10 are consistently reported in manuscripts, items from the Recommended Set can be added to journal requirements over time until all 21 items are routinely reported in all manuscripts. Full methodology for the revision and the allocation of items into sets is described in the accompanying publication [13].A key aspect of the revision was to develop this Explanation and Elaboration document to provide background and rationale for each of the 21 items of ARRIVE 2019. Here we present additional guidance for each item and subitem, explain the importance of reporting this information in manuscripts that describe animal research, elaborate on what to report, and provide supporting evidence. Each subitem is also illustrated with examples of good reporting from the published literature. Box 1: GlossaryBias: Introduction of a systematic error in the estimated effect of an intervention, caused by inadequacies in the design, conduct, or analysis of an experiment.Effect size: Quantitative measure that estimates the magnitude of differences between groups, or relationships between variables. Experimental unit:Biological entity subjected to an intervention independently of all other units, such that it is possible to assign any two experimental units to different t...

show abstract

“…Although fundamental research methodologies do not typically require sophisticated statistical analysis, studies with animals in most cases are akin to clinical trials, particularly those where multiple groups of animals or different genotypes or backgrounds undergoing different treatments are included. As such, they should be approached in a similar way as a clinical study, and similar guidelines for study design, power analysis, sample size, and statistical approach should apply . For statistical analysis, such studies (typically drawn from quantitative data) may require the application of appropriate general linear models (eg, analysis of variance, analysis of covariance, linear regression) and if necessary, post hoc multiple t tests can be used, where the threshold for significance is adjusted for multiple comparisons (see below for further details).…”

Section: Presentation and Analysis Of Preclinical Studiesmentioning

confidence: 99%

“…As such, they should be approached in a similar way as a clinical study, and similar guidelines for study design, power analysis, sample size, and statistical approach should apply. (11) For statistical analysis, such studies (typically drawn from quantitative data) may require the application of appropriate general linear models (eg, analysis of variance, analysis of covariance, linear regression) and if necessary, post hoc multiple t tests can be used, where the threshold for significance is adjusted for multiple comparisons (see below for further details). For repeated measures designs, linear mixed-effects models are preferable over repeated measures analysis of variance, although the latter is acceptable.…”

Section: Presentation and Analysis Of Preclinical Studiesmentioning

confidence: 99%

New Guidelines for Data Reporting and Statistical Analysis: Helping Authors With Transparency and Rigor in Research

Nguyen

Rivadeneira

Civitelli

2019

Journal of Bone and Mineral Research

View full text Add to dashboard Cite

Why we need to report more than 'Data were Analyzed by t-tests or ANOVA'

Cited by 55 publications

References 30 publications

Assessing the statistical reporting quality in high-impact factor urology journals

Assessing the statistical reporting quality in high-impact factor urology journals

Reporting animal research: Explanation and Elaboration for the ARRIVE guidelines 2019

New Guidelines for Data Reporting and Statistical Analysis: Helping Authors With Transparency and Rigor in Research

Contact Info

Product

Resources

About