Examples of boilerplate text from PLOS ONE papers based on targeted n-gram searches (sentence level).

Statistical analysis is error prone. A best practice for researchers using statistics would therefore be to share data among co-authors, allowing double-checking of executed tasks just as co-pilots do in aviation. To document the extent to which this ‘co-piloting’ currently occurs in psychology, we surveyed the authors of 697 articles published in six top psychology journals and asked them whether they had collaborated on four aspects of analyzing data and reporting results, and whether the described data had been shared between the authors. We acquired responses for 49.6% of the articles and found that co-piloting on statistical analysis and reporting results is quite uncommon among psychologists, while data sharing among co-authors seems reasonably but not completely standard. We then used an automated procedure to study the prevalence of statistical reporting errors in the articles in our sample and examined the relationship between reporting errors and co-piloting. Overall, 63% of the articles contained at least one p-value that was inconsistent with the reported test statistic and the accompanying degrees of freedom, and 20% of the articles contained at least one p-value that was inconsistent to such a degree that it may have affected decisions about statistical significance. Overall, the probability that a given p-value was inconsistent was over 10%. Co-piloting was not found to be associated with reporting errors.

Showing statistical analysis of study data.

Dataset for the statistical analysis of the article "Empowerment through Participatory Game Creation: A Case Study with Adults with Intellectual Disability".

BackgroundIn medical practice, clinically unexpected measurements might be quite properly handled by the remeasurement, removal, or reclassification of patients. If these habits are not prevented during clinical research, how much of each is needed to sway an entire study?Methods and ResultsBelieving there is a difference between groups, a well-intentioned clinician researcher addresses unexpected values. We tested how much removal, remeasurement, or reclassification of patients would be needed in most cases to turn an otherwise-neutral study positive. Remeasurement of 19 patients out of 200 per group was required to make most studies positive. Removal was more powerful: just 9 out of 200 was enough. Reclassification was most powerful, with 5 out of 200 enough. The larger the study, the smaller the proportion of patients needing to be manipulated to make the study positive: the percentages needed to be remeasured, removed, or reclassified fell from 45%, 20%, and 10% respectively for a 20 patient-per-group study, to 4%, 2%, and 1% for an 800 patient-per-group study. Dot-plots, but not bar-charts, make the perhaps-inadvertent manipulations visible. Detection is possible using statistical methods such as the Tadpole test.ConclusionsBehaviours necessary for clinical practice are destructive to clinical research. Even small amounts of selective remeasurement, removal, or reclassification can produce false positive results. Size matters: larger studies are proportionately more vulnerable. If observational studies permit selective unblinded enrolment, malleable classification, or selective remeasurement, then results are not credible. Clinical research is very vulnerable to “remeasurement, removal, and reclassification”, the 3 evil R's.

Results of statistical analysis.

Statistical analysis.

Variables Used for Statistical Analysis.

BackgroundThere is widespread evidence that statistical methods play an important role in original research articles, especially in medical research. The evaluation of statistical methods and reporting in journals suffers from a lack of standardized methods for assessing the use of statistics. The objective of this study was to develop and evaluate an instrument to assess the statistical intensity in research articles in a standardized way.MethodsA checklist-type measure scale was developed by selecting and refining items from previous reports about the statistical contents of medical journal articles and from published guidelines for statistical reporting. A total of 840 original medical research articles that were published between 2007–2015 in 16 journals were evaluated to test the scoring instrument. The total sum of all items was used to assess the intensity between sub-fields and journals. Inter-rater agreement was examined using a random sample of 40 articles. Four raters read and evaluated the selected articles using the developed instrument.ResultsThe scale consisted of 66 items. The total summary score adequately discriminated between research articles according to their study design characteristics. The new instrument could also discriminate between journals according to their statistical intensity. The inter-observer agreement measured by the ICC was 0.88 between all four raters. Individual item analysis showed very high agreement between the rater pairs, the percentage agreement ranged from 91.7% to 95.2%.ConclusionsA reliable and applicable instrument for evaluating the statistical intensity in research papers was developed. It is a helpful tool for comparing the statistical intensity between sub-fields and journals. The novel instrument may be applied in manuscript peer review to identify papers in need of additional statistical review.

Statistical analysis of the similarity between topics on LexisNexis and Web of Science data.

R scripts in this fileset are those used in the PLOS ONE publication "A snapshot of translational research funded by the National Institutes of Health (NIH): A case study using behavioral and social science research awards and Clinical and Translational Science Awards funded publications." The article can be accessed here: http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0196545This consists of all R scripts used for data cleaning, data manipulation, and statistical analysis used in the publication.There are eleven files in total:1. "Step1a.bBSSR.format.grants.and.publications.data.R" combines all bBSSR 2008-2014 grant award data and associated publications downloaded from NIH Reporter. 2. "Step1b.BSSR.format.grants.and.publications.data.R" combines all BSSR-only 2008-2014 grant award data and associated publications downloaded from NIH Reporter. 3. "Step2a.bBSSR.get.pubdates.transl.and.all.grants.R" queries PubMed and downloads associated bBSSR publication data.4. "Step2b.BSSR.get.pubdates.transl.and.all.grants.R" queries PubMed and downloads associated BSSR-only publication data.5. "Step3.summary.stats.R" performs summary statistics6. "Step4.time.to.first.publication.R" performs time to first publication analysis.7. "Step5.time.to.citation.analysis.R" performs time to first citation and time to overall citation analyses.8. "Step6.combine.NIH.iCite.data.R" combines NIH iCite citation data.9. "Step7.iCite.data.analysis.R" performs citation analysis on combined iCite data.10. "Step8.MeSH.descriptors.R" queries PubMed and pulls down all MeSH descriptors for all publications11. "Step9.CTSA.publications.R" compares the percent of translational publications among bBSSR, BSSR-only, and CTSA publications.

All data from dataset used in statistical analysis, in the format of testing.

Summary of the statistical analysis applied in the two experiments.

Statistical measures and methods in JAMA articles published in 1990, 2000, and 2010*.

BackgroundIndividual participant data (IPD) meta-analyses that obtain “raw” data from studies rather than summary data typically adopt a “two-stage” approach to analysis whereby IPD within trials generate summary measures, which are combined using standard meta-analytical methods. Recently, a range of “one-stage” approaches which combine all individual participant data in a single meta-analysis have been suggested as providing a more powerful and flexible approach. However, they are more complex to implement and require statistical support. This study uses a dataset to compare “two-stage” and “one-stage” models of varying complexity, to ascertain whether results obtained from the approaches differ in a clinically meaningful way. Methods and FindingsWe included data from 24 randomised controlled trials, evaluating antiplatelet agents, for the prevention of pre-eclampsia in pregnancy. We performed two-stage and one-stage IPD meta-analyses to estimate overall treatment effect and to explore potential treatment interactions whereby particular types of women and their babies might benefit differentially from receiving antiplatelets. Two-stage and one-stage approaches gave similar results, showing a benefit of using anti-platelets (Relative risk 0.90, 95% CI 0.84 to 0.97). Neither approach suggested that any particular type of women benefited more or less from antiplatelets. There were no material differences in results between different types of one-stage model. ConclusionsFor these data, two-stage and one-stage approaches to analysis produce similar results. Although one-stage models offer a flexible environment for exploring model structure and are useful where across study patterns relating to types of participant, intervention and outcome mask similar relationships within trials, the additional insights provided by their usage may not outweigh the costs of statistical support for routine application in syntheses of randomised controlled trials. Researchers considering undertaking an IPD meta-analysis should not necessarily be deterred by a perceived need for sophisticated statistical methods when combining information from large randomised trials.

Chi-square test indicating the relationship between processing parameters and consumption patterns of Munkoyo. Processing parameters and consumption patterns with P-values in the same row with different letters are significantly different (p-value<0.05; α after Bonferroni correction to correct for multiple testing is 0.05/14 = 0.0035).

Detailed methods, statistical analysis, figures, and references.

Data that was used to train the SVM. As the train-test data were assigned randomly for every training iteration, the individual data used for generating the subfigures b–e are not separately listed, as these cannot be manually recreated but depend on the train-test assignment by the algorithm. (ZIP)

Example boilerplate text from ANZCTR studies with the highest number of matches per topic (sentence level).

This study compared the effects of cold water immersion (CWI) and hot water immersion (HWI) on muscle recovery following a muscle-damaging exercise protocol in women. Thirty healthy women (23.3 ± 2.9 years) were randomly assigned to either the CWI, HWI, or control (CON) groups. Participants completed a standardised exercise protocol (5 x 20 drop-jumps), followed by a 10 min recovery intervention (CWI, HWI, or CON) immediately and 120 min post-exercise. Physiological responses, including muscle oxygen saturation (SmO2), core and skin temperature, and heart rate, were assessed at baseline, immediately post-exercise, after the first recovery intervention (postInt), and during 30 min follow-up. Recovery was evaluated through maximal voluntary isometric contraction, muscle swelling, muscle soreness ratings, and serum creatine kinase at baseline, 24, 48, and 72 h post-exercise. A mixed-effects model was used to account for repeated measures over time. Results showed lower SmO2 values in the CWI compared to the HWI group at 20 min (Δ-6.76%, CI: −0.27 to −13.25, p = 0.038) and 30 min (Δ-9.86%, CI: −3.37 to −16.35, p = 0.001), and compared to CON at 30 min (Δ-7.28%, CI: −13.77 to −0.79, p = 0.022). Core temperature was significantly higher in the HWI than the CWI group (postInt and 30 min), while it was significantly lower in the CWI group than CON (30 min). CWI caused a substantial decrease in skin temperature compared to HWI and CON between postInt and 30 min follow-up (all p < 0.001). Skin temperature was higher in the HWI group compared to CON at postInt and throughout 30 min follow-up (all p < 0.001). No significant differences in recovery markers were observed between CWI and HWI groups, although HWI led to slightly higher creatine kinase levels (24 h and 72 h) and greater muscle swelling (24 h) compared to CON. Despite distinct acute physiological responses to CWI and HWI, neither improved subjective or objective recovery outcomes during the 72 h follow-up compared to CON in women following a muscle-damaging exercise protocol.Trial registration numberNCT04902924 (ClinicalTrials.gov), SNCTP000004468 (Swiss National Clinical Trial Portal).