160 research outputs found
Talks Hartgerink
No full textA project to track and save my talks, slides, etc. for sharing. These will be indexed by academic year, starting 2015-2016
688,112 statistical results: Content mining psychology articles for statistical test results
No full textA dataset of 688,112 statistical test results reported according to the standards prescribed by the American Psychological Association (APA), mined from 50,845 articles out of 167,318 published by the APA, Springer, Sage, and Taylor & Francis. Mining from Wiley and Elsevier was actively blocked. Metadata for each article are included. All journals included, scripts, etc. are available at https://github.com/chartgerink/2016statcheck_data and preserved at http://dx.doi.org/10.5281/zenodo.5981
Contributions towards understanding and building sustainable science
No full textThis dissertation focuses on either understanding and detecting threats to the epistemology of science (chapters 1-6) or making practical advances to remedy epistemological threats (chapters 7-9). Chapter 1 reviews the literature on responsible conduct of research, questionable research practices, and research misconduct. Chapter 2 reanalyzes Head et al (2015) their claims about widespread p-hacking for robustness. Chapter 3 examines 258,050 test results across 30,710 articles from eight high impact journals to investigate the existence of a peculiar prevalence of -values just below .05 (i.e., a bump) in the psychological literature, and a potential increase thereof over time. Chapter 4 examines evidence for false negatives in nonsignificant results throughout psychology, gender effects, and the Reproducibility Project: Psychology. Chapter 5 describes a dataset that is the result of content mining 167,318 published articles for statistical test results reported according to the standards prescribed by the American Psychological Association (APA). In Chapter 6, I test the validity of statistical methods to detect fabricated data in two studies.Chapter 7 tackles the issue of data extraction from figures in scholarly publications. In Chapter 8 I argue that "after-the-fact" research papers do not help alleviate issues of access, selective publication, and reproducibility, but actually cause some of these threats because the chronology of the research cycle is lost in a research paper. I propose to give up the academic paper and propose a digitally native "as-you-go" alternative. In Chapter 9 I propose a technical design for this
688,112 statistical results: Content mining psychology articles for statistical test results
No full textIn this data deposit, I describe a dataset that is the result of content mining 167,318 published articles for statistical test results reported according to the standards prescribed by the American Psychological Association (APA). Articles published by the APA, Springer, Sage, and Taylor & Francis were included (mining from Wiley and Elsevier was actively blocked). As a result of this content mining, 688,112 results from 50,845 articles were extracted. In order to provide a comprehensive set of data, the statistical results are supplemented with metadata from the article they originate from. The dataset is provided in a comma separated file (CSV) in long-format. For each of the 688,112 results, 20 variables are included, of which seven are article metadata and 13 pertain to the individual statistical results (e.g., reported and recalculated p-value). A five-pronged approach was taken to generate the dataset: (i) collect journal lists; (ii) spider journal pages for articles; (iii) download articles; (iv) add article metadata; and (v) mine articles for statistical results. All materials, scripts, etc. are available at https://github.com/chartgerink/2016statcheck_data and preserved at http://dx.doi.org/10.5281/zenodo.59818. View Full-Text Keywords: statistics; p-values; psychology; content mining; mining; error
Analyzing DECREASE trials for extent of data fabrication
No full textThe DECREASE set of trials investigated the effectiveness of beta-blockers to prevent perioperative mortality in non-cardiac surgery. Perioperative stands for the period of the surgical procedure, typically including admission, anaesthesia, surgery, and recovery. Upon being discovered as unreliable (Commissie Vervolgonderzoek Wetenschap...; Commissie Vervolgonderzoek 2012 2012; Onderzoekscommissie Wetenschappelijke...), the DECREASE studies were excluded in a 2014 meta-analysis (Bouri et al. 2014), which found that conclusions with respect to the effectiveness of beta-blockers were reversed and increase perioperative mortality instead of decreasing it.
The reports into the DECREASE studies by the Erasmus University indicate that data fabrication was likely, but the extent of the data fabrication was not clearly indicated or deemed estimable. In this project, we aim to investigate the extent of data fabrication in the DECREASE studies by statistical analysis. We expect to confirm the conclusions of the reports that data fabrication is likely
- …
