butterfly/kb
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Scrape wikipedia-science: 665 new, 15 updated, 702 total (kb-cron)

Browse Source
This commit is contained in:
turtle89431 2026-05-04 20:14:35 -07:00
parent ebcf688101
commit 5fe2fff6ac
19 changed files with 1444 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

BIN
_index.db
View File

Binary file not shown.

41
data/en.wikipedia.org/wiki/AllTrials-0.md Normal file
View File

@ -0,0 +1,41 @@
---
title: "AllTrials"
chunk: 1/2
source: "https://en.wikipedia.org/wiki/AllTrials"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:24.154433+00:00"
instance: "kb-cron"
---
AllTrials (sometimes called All Trials or AllTrials.net) is a project advocating that clinical research adopt the principles of open research. The project summarizes itself as "All trials registered, all results reported": that is, all clinical trials should be listed in a clinical trials registry, and their results should always be shared as open data.
At the center of the organisation is a petition signed by over 85,000 individuals and
599 organisations (as of August 2015):
Thousands of clinical trials have not reported their results; some have not even been registered.
Information on what was done and what was found in these trials could be lost forever to doctors and researchers, leading to bad treatment decisions, missed opportunities for good medicine, and trials being repeated.
All trials past and present should be registered, and the full methods and the results reported.
We call on governments, regulators and research bodies to implement measures to achieve this.
Ben Goldacre, author of Bad Science and Bad Pharma, is a founder of the campaign and its most public spokesperson. In 2016, he participated in the launch of the OpenTrials database.
AllTrials is an international initiative of Bad Science, BMJ, Centre for Evidence-based Medicine, Cochrane Collaboration, James Lind Initiative, PLOS and Sense about Science and is being led in the US by Sense about Science USA, Dartmouth's Geisel School of Medicine and the Dartmouth Institute for Health Policy & Clinical Practice.
== Issues addressed ==
The project is a reaction to under-reporting of research.
A substantial proportion (estimates range from one-third to one-half) of medical research goes unpublished. It has also been shown that negative findings are less likely to be published than positive ones, even in the absence of conflicts of interest.
Much medical research is done by the pharmaceutical industry, which have a conflict of interest reporting results which may hurt sales of their products. There is a measurable funding bias in reporting; studies have shown that published drug studies funded by pharmaceutical companies are much more likely to support the use of the tested drug than studies with other funding. Industry-funded trials are also less likely to be published.
If the statistical methods used to analyse the trial are not chosen before the study it started, there is a danger that researchers will intentionally or unintentionally pick the method that gives the results they expect, or which gives the most significant results. This makes the analysis statistically invalid.
Not publishing trials which fail to find a clear effect exposes trial volunteers to pointless risk and wastes research effort (as the same trial is repeated over and over). It also biases the medical literature, making it report effects where none exist (since, given enough trials, eventually one will find a difference by pure chance).
Pre-trial registration makes non-publication and changes in analysis methods obvious to medical reviewers. It also enables authors of meta-studies to track down and analyse missing data. Finally, it lets doctors and patients know when a trial is looking for volunteers.
There are other sources of bias, such as the conditions sometimes attached to funding by funding agencies with a financial interest in the trial's outcome. Medical researchers may be asked to agree to allow the funding agency to censor results. Some funding agencies may also refuse to give the medical researcher access to the raw data, giving them only the finished analysis, or even a draft paper, and asking them to put their name to it. This is not acceptable academic practice, and some academic journals require that authors sign a statement that they have not entered into such agreements.
Ben Goldacre, a physician and spokesperson for the campaign, would like to address the systematic flaws in clinical research which cause data to be lost after it is gathered.
== Coverage ==
The campaign has been widely covered, and supported, in the academic press. The British Medical Journal and PLOS are founding members. Nature and The Lancet both published supportive articles in January 2014.
There has also been mainstream media coverage.
== Controversy ==
There has been criticism from the Pharmaceutical Research and Manufacturers of America (PhRMA), with senior vice-president Matt Bennett saying that trial data disclosure measures which AllTrials has recommended to the European Medicines Agency "could risk patient privacy, lead to fewer clinical trials, and result in fewer new medicines to meet patient needs and improve health.".
AllTrials have published a detailed statement of exactly what they want to see published, which states "The AllTrials campaign is not calling for individual patient data to be made publicly available".
A 2012 editorial published by senior regulators from the European Medicines agency largely agreed with AllTrials, saying "We consider it neither desirable nor realistic to maintain the status quo of limited availability of regulatory trials data". They were also of the opinion that adequate standards for protection of personal data could be written. However, they warned that third party reanalysis was neither a guarantee of quality nor of lack of conflict of interest, which, in the worst case, could lead to negative public health consequences.
They suggested that reanalyses should therefore be subject to the same regulations as sponsor analyses, such as registering analysis plans. They argued against completely unrestricted access to data, but in favor of broader access. AllTrials is not calling for completely unrestricted access to raw data, so the scope of disagreements is limited to what restrictions should be in place.

34
data/en.wikipedia.org/wiki/AllTrials-1.md Normal file
View File

@ -0,0 +1,34 @@
---
title: "AllTrials"
chunk: 2/2
source: "https://en.wikipedia.org/wiki/AllTrials"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:24.154433+00:00"
instance: "kb-cron"
---
== Supporters ==
The campaign is an initiative of Sense about Science, Centre for Evidence Based Medicine, The Dartmouth Institute for Health Policy and Clinical Practice, James Lind Alliance, Cochrane Collaboration, BMJ Group, PLOS, and Bad Science. The petition statement of AllTrials has been signed by organizations including Wellcome Trust, British Library, Medical Research Council (UK), British Heart Foundation, Institute for Quality and Efficiency in Health Care, National Institute for Health and Care Excellence, BioMed Central, National Physicians Alliance, Royal Society of Medicine, Health Research Authority, American Medical Student Association, GlaxoSmithKline, and others.
As of May 2017, The AllTrials petition has been signed by 90,282 people and 721 organisations. In October 2016, AllTrials published a road map detailing steps that various types of organisations can take to get more trials registered and more results reported.
85 investors with >€3.5 trillion (£2.45trn; $3.83trn) of investments have supported AllTrials (as of July 2015), with Peter van der Werf of RobecoSAM saying: "We deem this to be a financially material factor and encourage all companies to gain credibility regarding their approach to clinical trial transparency by signing up to the AllTrials principles.". The Laura and John Arnold Foundation provided early and ongoing financial support.
The original policy of the Coalition for Epidemic Preparedness Innovations required that funded parties pre-register any trials in a clinical trials registry, publish results within a year of study completion (except with compelling reason and permission of CEPI), publish results in open-access articles, and have mechanisms for securely sharing underlying data and results, including negative results, in a way that preserves trial volunteer privacy. In May 2018 the CEPI proposed changing the policy to remove these provisions. The policy was changed by the CEPI in December 2018.
== Opponents ==
The European Federation of Pharmaceutical Industries and Associations and Pharmaceutical Research and Manufacturers of America have expressed interest in lobbying against the campaign. Campaign supporters criticized Hoffmann-La Roche's plans to be more open but not to the extent requested by AllTrials.
== See also ==
Clinical data repository
Conflicts of interest in academic publishing
Monitoring in clinical trials
Metascience
Privacy for research participants
Evidence-based medicine
Clinical trials publication
Censoring (clinical trials)
== References ==
== External links ==
Official website
Where's the rest of the data iceberg?, a video presentation by Ben Goldacre at TEDMED

26
data/en.wikipedia.org/wiki/Berkeley_Initiative_for_Transparency_in_the_Social_Sciences-0.md Normal file
View File

@ -0,0 +1,26 @@
---
title: "Berkeley Initiative for Transparency in the Social Sciences"
chunk: 1/1
source: "https://en.wikipedia.org/wiki/Berkeley_Initiative_for_Transparency_in_the_Social_Sciences"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:26.397111+00:00"
instance: "kb-cron"
---
The Berkeley Initiative for Transparency in the Social Sciences, abbreviated BITSS, is an academic initiative dedicated to advancing transparency, reproducibility, and openness in social science research. It was established in 2012 by the University of California, Berkeley's Center for Effective Global Action. It has worked with the Center for Open Science to define and promote a set of best practices for social scientists to maximize transparency in their research. BITSS has also worked to promote registered reports, supporting journals like the Journal of Development Economics in taking up the review track.
In 2015, BITSS began awarding the annual Leamer-Rosenthal Prizes for Open Social Science to honor outstanding achievements and emerging leaders in promoting transparency in social science. Through its Catalyst program, the initiative also supports and empowers over 150 graduate students, faculty, librarians, and early career researchers to advance open science all over the world. Their annual Research Transparency and Reproducibility Training (RT2) provides an overview of and hands-on practice with tools and practices for transparent and reproducible social science research. Their Massive Open Online Course "Transparent and Open Social Science Research,” based on a UC Berkeley course taught by Edward Miguel, is available on the FutureLearn platform. In 2019, BITSS also began distributing copies of "Transparent and Reproducible Social Science Research," a textbook written by former BITSS Scientist Garret Christensen, Jeremy Freese, and Edward Miguel with support from BITSS, at their trainings and events.
BITSS has supported or led several metascience research projects including the State of Social Science (3S) study and the Social Science Meta-Analysis and Research Transparency (SSMART) portfolio. BITSS also manages MetaArxiv, an interdisciplinary archive hosted on OSF Preprints of articles focused on metascience, research transparency, and reproducibility.
In recent years, BITSS has begun developing digital infrastructure to enable open science practices. The Social Science Prediction Platform (SSPP), launched in 2020, enables the systematic collection and assessment of expert forecasts of research results and the effects of untested social programs. The Social Science Reproduction Platform (SSRP) crowdsources and catalogs attempts to assess and improve the computational reproducibility of social science research. The accompanying Guide for Accelerating Computational Reproducibility in the Social Sciences elucidates a common approach, terminology, and standards for conducting reproductions. These platforms are part of a growing ecosystem of tools that expand opportunities to participate in the scientific endeavor.
BITSS has also incubated an initiative on Open Policy Analysis (OPA), which seeks to strengthen connections between research and policy and reduce political polarization by translating open science practices to policy analysis. Led by Fernando Hoces de la Guardia, the OPA initiative has developed tools for US Senator Elizabeth Warren's wealth tax proposal and Evidence Action's Deworm the World program.
== See also ==
Metascience
== References ==
== External links ==
Official website

34
data/en.wikipedia.org/wiki/Burden_of_knowledge-0.md Normal file
View File

@ -0,0 +1,34 @@
---
title: "Burden of knowledge"
chunk: 1/1
source: "https://en.wikipedia.org/wiki/Burden_of_knowledge"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:27.531708+00:00"
instance: "kb-cron"
---
The burden of knowledge describes the difficulty of adding to a scientific field as the amount of previous work that must be understood increases over time. This is seen across scientific disciplines. Evidence for this burden includes a trend that people are older before they receive their first patent or publish in a prestigious journal, and the need for scientific groups to be larger as scientists need to collaborate so that the team has sufficient understanding of prior work.
== Scholarship ==
Explicit scholarship of this idea has entered the mainstream with the works of Benjamin Jones, in particular The Burden of Knowledge and the Death of the Renaissance Man, and with the works of
Jan Brendel and Sascha Schweitzer The Burden of Knowledge in Mathematics.
== Overview ==
Theory and empirical studies reflect the case that researchers and innovators are not born with the required expertise and must first undertake education. With accumulating information and discoveries, the time to digest and improve on extant knowledge takes longer. Similarly, frontiers of knowledge advance at an overall increasing rate and are shifting over time. The "burden of knowledge" refers to the difficulty of catching up with this evolving knowledge frontier.
A hard metric used by Brendel and Schweitzer for mathematics burden is age at first publication. They specifically point to "a significant increase of the average age of researchers at their first publication in one of our top-ranking journals."
Findings associated with Burden of Knowledge investigations point to declining productivity in sole researchers and developers and increasing productivity by teams. Prominent examples of highly effective team research in basic science include those of Nobel Prize awardees Francis Crick and James Watson's work on DNA structure, Yang Chen-Ning and Tsung-Dao Lee's work on parity violation, and Katalin Karikó and Drew Weissman's mRNA vaccine discoveries and development.
Research also point to better outcomes in gender diverse teams. Interestingly, research points to better development by large teams and more R&D novelty and disruption by small teams.
== Challenges in other areas ==
Challenges due to increasing complexity and data are found in other fields. There are observed productivity challenges in pharmaceutical drug discovery R&D. The challenges also manifest in an overall trend of patents, papers, and discoveries being less disruptive.
== Other uses of the phrase "Burden of knowledge" ==
Christian Turner (Professor of Law) uses the term "burden of knowledge" in a different way, referring to situations where one may be better off not knowing things, for example avoiding painful and uncomfortable details of one's health.
== References ==

55
data/en.wikipedia.org/wiki/C._Glenn_Begley-0.md Normal file
View File

@ -0,0 +1,55 @@
---
title: "C. Glenn Begley"
chunk: 1/1
source: "https://en.wikipedia.org/wiki/C._Glenn_Begley"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:25.259430+00:00"
instance: "kb-cron"
---
C. Glenn Begley is a hematologist and oncologist who was the CEO of BioCurate, an Australia-based joint venture between the University of Melbourne and Monash University that was launched in 2016. The "C. Glenn Begley Award", created by Stanford University in 2025, was named in his honor. The award celebrates teams that uphold the highest standards of scientific integrity, fostering innovation that is both impactful and reliable.
Previously, he worked at the California-based biotech company Akriveia Therapeutics as their chief scientific officer. His other previous positions include global head of hematology and oncology research at Amgen, senior vice president and chief scientific officer at TetraLogic Pharmaceuticals, and executive director of the Western Australian Institute of Medical Research.
He studied medicine at the University of Melbourne. He is known for a study he co-authored in 2012 in which he found that only 6 out of 53 (11%) of landmark preclinical studies of cancer drugs could be reproduced. He was elected to the American Society for Clinical Investigation in 2000. Begley was elected Fellow of the Australian Academy of Health and Medical Sciences in 2015.
== Education ==
Begley studied medicine at the University of Melbourne starting in 1972 and finishing in 1978. During this time, he completed his M.B., Ph.D.in cellular and molecular biology, and B.S.
== Accreditations, awards and honors ==
Doctorate
M.B.B.S, Bachelor of Medicine, Bachelor of Surgery
F.R.A.C.P., Fellow of the Royal Australasian College of Physicians.
F.R.C.P.A., Fellow of the Royal College of Pathologists of Australasia.
F.R.C.Path., Fellow of the Royal College of Pathologists. (UK)
F.A.H.M.S., Fellow of the Australian Academy of Health and Medical Sciences. Inducted 2015
F.A.S.C.I., Fellow of the American Society for Clinical Investigation. He was the foreign fellow to be elected into this society. (2015)
F.A.A.P.S., Fellow of the Association of American Physicians and Surgeons. He was the foreign fellow to be elected into this association.
== Career ==
In the 1980s Begley did an apprenticeship under Donald Metcalf who was a pioneer of hematology in Australia. This was at the Walter and Eliza Hall Institute of Medical Research, the oldest medical research institute in Australia. This was for 3 years while Begley was a PhD student.
Begley worked at Amgen in California from 2002 to 2012. Amgen is a biotechnology company that scientific research into drug development strategies and then markets them. He was the vice president and global head of hematology/oncology research.
On 15 March 2012 C. Glenn Begley was announced as the senior vice president of research and development at TetraLogic Pharmaceuticals in Pennsylvania. He worked there from 2012 to 2016.
During this time, he also worked as a Non-Executive director and senior clinical advisor at Oxford BioTherapeutics (2012–2017).
Begley is currently the CEO of BioCurate. The aim of BioCurate is to recognize potential in biomedical research and then invest to accelerate the process of converting the research into medicine and therapies.
== Research ==
In the 1980s, Begley's research was mainly done with Don Metcalf. They were featured on several papers together. From the scientific articles released at that time that he was the Chief Author of, the research areas included colony-stimulating factors (CSF) and their relevance to white blood cells and the immune response in vitro. In 1985 Begley, Metcalf and N.A. Nicola published a paper centring on granulocyte colony-stimulating factor’s (GCSF) effect on differentiation of white blood cells. This paper found that GCSF was affected differently to other colony stimulating factors and the paper was cited 136 times. Begley published a paper in 1986 that has 182 cites involving injecting mice with Multipotential CSF (M-CSF) then comparing white blood cell count in control mice. The results showed that M-CSF increased the monocyte and neutrophil count by more than 200% and the eosinophil count by 1000%. This paper showed the importance of CSF in hematopoietic repopulation in living organisms. It was cited 225 times.
In the 1990s his articles were focused around genes and proteins. The main gene looked at was the SCL gene and how it links to cancer. Begley continued his work from the 80’s into colony-stimulating factors (specifically GCSF) and in 1992 was published in the Lancet. This paper had important work outlining how CSF increases platelet recovery after chemotherapy. He also researches the developing nervous system and the different roles that proteins, white blood cells and genes play. In 1999 he released an article which he was the co-author of that was solely on SCL and the relationship with T cells and T lymphocytes. This paper concluded with an “Unresolved Issues” section which provides unanswered questions about SCL leading to more potential research in the future for Begley.
In the 2000s his work on Thrombopoietin and a protein coded by this gene was released. This is a hormone that regulates platelet production. He also had his most cited paper in 2007 titled "Genome-wide association study identifies novel breast cancer susceptibility loci" with 2518 cites. Begley was also involved in more work with mice that found that bone marrow cells are not a contributor to endothelium of tumors. In 2006 Begley contributed to work on EpoR on another highly cited paper (262) in the field of hematology.
In the 2010s more reflective scientific articles were released by Begley that comment on how his career and experience has shaped his view today. They talk about how scientific integrity is essential for research, the challenges involved with oncology research and how to avoid making mistakes when researching and producing drugs. This aligns with his time at BioCurate and TetraLogical Pharmaceuticals where his work involved drug related ventures and investing in therapies.
== Other published work ==
In 2015 Begley published a tribute to Don Metcalf. Begley and Metcalf worked together for 15 years. Begley claims Metcalf was a "teacher, role model, mentor, colleague and friend" to him and that he had a resounding impact on his life. In this article Begley explains how Metcalf shaped his view of scientific integrity and how science should have a passing of knowledge from experienced scientists to newer ones.
== References ==
== External links ==
C. Glenn Begley publications indexed by Google Scholar

44
data/en.wikipedia.org/wiki/Conflicts_of_interest_in_academic_publishing-0.md Normal file
View File

@ -0,0 +1,44 @@
---
title: "Conflicts of interest in academic publishing"
chunk: 1/4
source: "https://en.wikipedia.org/wiki/Conflicts_of_interest_in_academic_publishing"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:28.683683+00:00"
instance: "kb-cron"
---
Conflicts of interest (COIs) often arise in academic publishing. Such conflicts may cause wrongdoing and make it more likely. Ethical standards in academic publishing exist to avoid and deal with conflicts of interest, and the field continues to develop new standards. Standards vary between journals and are unevenly applied. According to the International Committee of Medical Journal Editors, "[a]uthors have a responsibility to evaluate the integrity, history, practices and reputation of the journals to which they submit manuscripts".
Conflicts of interest increase the likelihood of biases arising; they can harm the quality of research and the public good (even if disclosed). Conflicts of interest can involve research sponsors, authors, journals, journal staff, publishers, and peer reviewers.
== Avoidance, disclosure, and tracking ==
The avoidance of conflicts of interest and the changing of the structure of institutions to make them easier to avoid are frequently advocated for. Some institutional ethics policies ban academics from entering into specific types of COIs, for instance by prohibiting them from accepting gifts from companies connected with their work. Education in ethical COI management is also a tool for avoiding COI problems.
Disclosure of COIs has been debated since the 1980s; there is a general consensus favouring disclosure. There is also a view that COI concerns and some of the measures taken to reduce them are excessive.
Criticisms of disclosure policies include:
authors disclosing COIs may feel pressured to present their research in a more biased manner to compensate;
disclosure may discourage beneficial academic–industrial collaboration;
disclosure may decrease public trust in research;
researchers who have disclosed their COIs may feel license to behave immorally;
disclosure may be taken as a sign of honesty or expertise and thus increase trust;
some types of COI may be more likely than others to go unnoticed or unreported;
awareness of a COI does not make people immune to being influenced by bias; generally, people do not sufficiently discount biased advice;
disclosure discourages the judging of work purely on its merits;
disclosure causes more intense scrutiny for wrongdoing.
While disclosure is widely favoured, other COI management measures have narrower support. Some publications hold the opinion that certain COIs disqualify people from certain research roles; for instance, that the testing of medicines should be done only by people who neither develop medicines nor are funded by their manufacturers.
Conflicts of interest have also been considered as a statistical factor confounding evidence, which must therefore be measured as accurately as possible and analysed, requiring machine-readable disclosure.
== Codes of conduct ==
Journals have individual ethics policies and codes of conduct; there are also some cross-journal voluntary standards.
The International Committee of Medical Journal Editors (ICMJE) publishes Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly work in Medical Journals, and a list of journals that pledge to follow it. The guideline lays down detailed rules for conflict-of-interest declaration by authors. It also says; "All participants in the peer-review and publication process—not only authors but also peer reviewers, editors, and editorial board members of journals—must consider their conflicts of interest when fulfilling their roles in the process of article review and publication and must disclose all relationships that could be viewed as potential conflicts of interest". These recommendations have been criticized and revised to remove loopholes allowing the non-disclosure of conflicts of interest.
The Council of Science Editors publishes a White Paper on publication ethics. Citing the ICMJE that "all participants in the peer-review and publication process must disclose all relationships that could be viewed as potential conflicts of interest", it highly recommends COI disclosure for sponsors, authors, reviewers, journals, and editorial staff.
The Good Publication Practice (GPP) guidelines, covering industry-sponsored medical research, are published by the International Society of Medical Publication Professionals.
The Committee on Publication Ethics (COPE) publishes a code of conduct stating, "[t]here must be clear definitions of conflicts of interest and processes for handling conflicts of interest of authors, reviewers, editors, journals and publishers, whether identified before or after publication".
The Open Access Scholarly Publishers Association's Principles of Transparency and Best Practice in Scholarly Publishing is intended to separate legitimate journals from predatory publishers and defines a minimal standard; clear and clearly stated COI policies.
A 2009 US Institute of Medicine report on medical COIs states that conflict-of-interest policies should be judged on their proportionality, transparency, accountability, and fairness; they should be effective, efficient, and targeted, known and understood, clearly identify who is responsible for monitoring, enforcement, and amendment, and apply equally to everyone involved. Review by conflict-of-interest committees is also recommended, and the lack of transparency and COI declaration in developing COI guidelines criticized.
As of 2015, journal COI policies often have no enforcement provisions. COI disclosure obligations have been legislated; one example of such legislation is the US Physician Payments Sunshine Act, but these laws do not apply specifically to journals.
== COIs by agent ==
=== COIs of journals ===
Journals are often not transparent about their institutional COIs, and do not apply the same disclosure standards to themselves as they do to their authors. Four out of six major general medical journals that were contacted for a 2010 COI study refused to provide information about the proportion of their income that derived from advertisements, reprints, and industry-supported supplements, citing policies on non-disclosure of financial information.

36
data/en.wikipedia.org/wiki/Conflicts_of_interest_in_academic_publishing-1.md Normal file
View File

@ -0,0 +1,36 @@
---
title: "Conflicts of interest in academic publishing"
chunk: 2/4
source: "https://en.wikipedia.org/wiki/Conflicts_of_interest_in_academic_publishing"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:28.683683+00:00"
instance: "kb-cron"
---
==== Owners and governing bodies ====
The owner of an academic journal has ultimate power over the hiring and firing of editorial staff; editors' interests in pleasing their employers conflict with some of their other editorial interests. Journals are also more likely to accept papers by authors who work for the journals' hosting institutions.
Some journals are owned by publishers. When journals print reviews of books published by their own publishers, they rarely (as of 2013) add COI disclosures. The publishers' interest in maximizing profit will often conflict with academic interests or ethical standards. In the case of closed-access publications, publishers' desire for high subscription income may conflict with an editorial desire for broader access and readership. There have been multiple mass resignations of editorial boards over such conflicts, which are often followed by the editorial board founding a new, non-profit journal to compete with their former one.
Some journals are owned by academic societies and professional organisations. Leading journals can be very profitable and there is often friction about revenue between the journal and the member society that owns it. Some academic societies and professional organisations are themselves funded by membership fees and/or donations. If the owners benefit financially from donations, the journal has a conflict between its financial interest in satisfying the donors—and therefore the owners—and its journalistic interests. Such COIs with industry donors have drawn criticism.
==== Reprints ====
A reprint is a copy of an individual article that is printed and sold as a separate product by the journal or its publisher or agent. Reprints are often used in pharmaceutical marketing and other medical marketing of products to doctors. This gives journals an incentive to produce good marketing material. Journals sell reprints at very high profit margins, often around 70%, as of 2010. A journal may sell a million dollars' worth of reprints of a single article if, for example, it is a large industry-funded clinical trial. The selling of reprints can bring in over 40% of a journal's income.
==== Impact factors, reputation, and subscriptions ====
If a journal is accused of managing COIs badly, its reputation is harmed.
The impact factor of a journal is often used to rate it, although this practice is widely criticized. A journal will generally want to increase its impact factor in hope of gaining more subscriptions, better submissions, and more prestige. As of 2010, industry-funded papers generally get cited more than others; this is probably due in part to industry-paid publicity.
Some journals engage in coercive citation, in which an editor forces an author to add extraneous citations to an article to inflate the impact factor of the journal in which the extraneous papers were published. A survey found that 86% of academics consider coercive citation unethical but 20% have experienced it. Journals appear to preferentially target younger authors and authors from non-English-speaking countries. Journals published by for-profit companies used coercive citation more than those published by university presses.
Journals may find it difficult to correct and retract erroneous papers after publication because of legal threats.
==== Advertising ====
Many academic journals contain advertising. The portion of a journal's revenue coming from advertising varies widely, according to one small study, from over 50% to 1%. As of 2010, advertising revenues for academic journals are generally falling. A 1995 survey of North American journal editors found that 57% felt responsible for the honesty of the pharmaceutical advertisements they ran and 40% supported peer-review of such advertisements. An interest in increasing advertising revenue can conflict with interests in journalistic independence and truthfulness.
==== Sponsored supplements ====
As of 2002, some journals publish supplements that often either cover an industry-funded conference or are "symposia" on a given topic. These supplements are often subsidized by an external sponsor with a financial interest in the outcome of research in that field; for instance, a drug manufacturer or food industry group. Such supplements can have guest editors, are often not peer-reviewed to the same standard as the journal itself, and are more likely to use promotional language. Many journals do not publish sponsored supplements. Small-circulation journals are more likely to publish supplements than large, high-prestige journals. Indications that an article was published in a supplement may be fairly subtle; for instance, a letter "s" added to a page number.
The ICMJE code of conduct specifically addresses guest-editor COIs; "Editors should publish regular disclosure statements about potential conflicts of interests related to their own commitments and those of their journal staff. Guest editors should follow these same procedures." It also states that the usual journal editor must maintain full control and responsibility and that "Editing by the funding organization should not be permitted".
The US Food and Drug Administration states that supplement articles should not be used as medical-marketing reprints, but as of 2009 it had no legal authority to prohibit the practice.
==== Publishers ====
Publishers may not be strongly motivated to ensure the quality of their journals. In the Australasian Journal of Bone & Joint Medicine case, the printer Elsevier Australia put out six journal-like publications containing articles about drugs made by the Merck Group, which paid for and controlled the publications.

39
data/en.wikipedia.org/wiki/Conflicts_of_interest_in_academic_publishing-2.md Normal file
View File

@ -0,0 +1,39 @@
---
title: "Conflicts of interest in academic publishing"
chunk: 3/4
source: "https://en.wikipedia.org/wiki/Conflicts_of_interest_in_academic_publishing"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:28.683683+00:00"
instance: "kb-cron"
---
=== COIs of journal staff ===
Personal conflicts of interest faced by journal staff are individual. If a person leaves the journal—unlike the COIs of journals as institutions—their personal COIs will go with them.
As of 2015, COIs of journal staff are less commonly reported than those of authors. For instance, one 2009 World Association of Medical Editors (WAME) policy document states, "Some journals list editors' competing interests on their website but this is not a standard practice". The ICMJE, however, requires that the COIs of editors and journal staff be regularly declared and published.
One 2017 Open Payments study of influential US medical journals found half of the editors received payments from industry; another study that used a different sample of editors reported two-thirds. As of 2002, systems for reporting wrongdoing by editors often do not exist.
Many journals have policies limiting COIs staff can enter into; for instance, accepting gifts of travel, accommodation, or hospitality may be prohibited. As of 2016, such policies are rarely published. Most journals do not offer COI training; as of 2015, many journals report a desire for better guidance on COI policy.
=== COIs of peer reviewers ===
The ICJME recommendations require peer reviewers to disclose conflicts of interest. Half to two-thirds of journals, depending on subject area, did not follow this recommendation in the first two decades of the 21st century. As of 2017, if a peer reviewer fails to disclose a conflict of interest, the paper will generally not be withdrawn, corrected, or re-reviewed; the reviews, however, may be reassessed.
A 2024 study published in JAMA found that 58.9% of U.S.-based peer reviewers for leading medical journals (The BMJ, JAMA, The Lancet, and The New England Journal of Medicine) received payments from drug and medical device manufacturers between 2020 and 2022. These payments totaled $1.06 billion, mostly for research purposes, though a notable portion ($64.18 million) was for general payments such as consulting and speaking fees. As such, industry payments to reviewers were common and often substantial. Most journals do not require peer reviewers to publicly disclose industry payments.
If peer reviewers are anonymous, their COIs with reviewed articles cannot be directly established. Some experiments with publishing the names of reviewers have been undertaken; in others, the identities of reviewers were disclosed to authors, allowing authors to identify COIs. Some journals now have an open review process in which everything, including the peer reviews and the names of the reviewers, and editor and author comment, is published transparently online.
The duties of peer review may conflict with social interests or institutional loyalties; to avoid such COIs, reviewers may be excluded if they have some forms of COI, such as having collaborated with the author.
Readers of academic papers may spot errors, informally or as part of formal post-publication peer review. Academics submitting corrections to papers are often asked by the publishers to pay over 1,000 US dollars for the publication of their corrections.
=== COIs of article authors ===
Authors of individual papers may face conflicts with their duty to report truthfully and impartially. Financial, career, political, and social interests are all sources of conflict. Authors' institutional interests become sources of conflict when the research might harm the institution's finances or offend the author's superiors.
Many journals require authors to self-declare their conflicts of interest when submitting a paper; they also ask specific questions about conflicts of interest. The questions vary substantially between journals. Author declarations, however, are rarely verified by the journal. As of 2018, "most editors say it's not their job to make sure authors reveal financial conflicts, and there are no repercussions for those who don't". Even if a conflict of interest is reported by a reader after publication, COPE does not suggest independent investigation, as of 2017.
As a result, as of 2018, authors often fail to declare their conflicts of interest. Rates of nondisclosure vary widely in reported studies.
The COPE retraction guidelines state, "Retractions are also used to alert readers to ... failure to disclose a major competing interest likely to influence interpretations or recommendations". As of 2018, however, if an author fails to disclose a COI, the paper will usually be corrected; it will not usually be retracted. Paper retractions, notifications to superiors, and publication bans are possible. Non-disclosure incidents harm academic careers. Authors are held to have collective responsibility for the contents of an article; if one author fails to declare a conflict of interest, the peer review process may be deemed compromised and the whole paper retracted.
The publisher may charge authors substantial fees for retracting papers, even in cases of honest error, giving them a financial disincentive to correct the record.
Public registries of author COIs have been suggested. Authors face administrative burdens in declaring COIs; standardized declarations or a registry could reduce these.
==== Ghost authors and non-contributing authors ====
Ghost authorship, where a writer contributes but is not credited, has been estimated to affect a significant proportion of the research literature. Honorary authorship, where an author is credited but did not contribute, is more common. Being named as an author on many papers is good for an academic's career. Failure to adhere to authorship standards is rarely punished. To avoid misreported authorship, a requirement that all authors describe the contribution they made to the study ("movie-style credits") has been advocated for. Ghostwriters may be legally liable for fraud.
The ICMJE criteria for authorship require that authors contribute:
Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; and
Drafting the work or revising it critically for important intellectual content; and
Final approval of the version to be published; and
Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

45
data/en.wikipedia.org/wiki/Conflicts_of_interest_in_academic_publishing-3.md Normal file
View File

@ -0,0 +1,45 @@
---
title: "Conflicts of interest in academic publishing"
chunk: 4/4
source: "https://en.wikipedia.org/wiki/Conflicts_of_interest_in_academic_publishing"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:28.683683+00:00"
instance: "kb-cron"
---
The ICMJE requires that "All those designated as authors should meet all four criteria for authorship, and all who meet the four criteria should be identified as authors. Those who do not meet all four criteria should be acknowledged." Academics who have had publication ethics training and those who are aware of the ICMJE authorship criteria are more stringent in their concepts of authorship and are more likely to consider breaches of authorship as misconduct, as are more junior researchers. Awareness is low; one study found only about half of researchers had read the ICJME criteria.
=== COIs of study sponsors ===
If a study requires outside funding, this can be a major source of conflicting interests; for instance in cases where the manufacturer of a drug is funding a study into its safety and efficacy or where the sponsor hopes to use the research to defend itself in litigation. Sponsors of a study may involve themselves in the design, execution, analysis, and write-up of a study. In extreme cases, they may carry out the research and ghostwrite the article with almost no involvement from the nominal author. Movie-style credits are advocated as a way to avoid this.
There are many opportunities for bias in trial design and trial reporting. For instance, a trial that compares a drug against the wrong dose of a competing drug may produce spuriously positive results.
In some cases, a contract with a sponsor may mean those named as investigators and authors on the papers may not have access to the trial data, control over the publication text, or the freedom to talk about their work. While authors and institutions have an interest in avoiding such contracts, it conflicts with their interest in competing for funding from potential study sponsors. Institutions that set stricter ethical standards for sponsor contracts lose contracts and funding when sponsors go elsewhere.
Sponsors have required contractual promises that the study is not reported without the sponsor's approval (gag clauses) and some have sued authors over compliance. Trials may go unpublished to keep commercial information secret or because the trial results were unfavourable. Some journals require that human trials be registered to be considered for publication; some require the declaration of any gag clauses as a conflict of interest; since 2001, some also require a statement that the authors have not agreed to a gag clause. Some journals require a promise to provide access to the original data to researchers intending to replicate the work. Some research ethics boards, universities, and national laws prohibit gag clauses. Gag clauses may not be legally enforceable if compliance would cause sufficient public harm. Non-publication has been found to be more common in industry-funded trials, contributing to publication bias.
It has been suggested that having many sponsors with different interests protects against COI-induced bias. As of 2006, there was no evidence for or against this hypothesis.
== Effect on conclusions of research ==
There is evidence that industry funding of studies of medical devices and drugs results in these studies having more positive conclusions regarding efficacy (funding bias). A similar relationship has been found in clinical trials of surgical interventions, where industry funding leads to researchers exaggerating the positive nature of their findings. Not all studies have found a statistically significant relationship between industry funding and the study outcome.
== Interests of research participants ==
Chronically ill medical research participants report expectation of being told about COIs and some report they would not participate if the researcher had some sorts of COIs. With few exceptions, multiple ethical guidelines forbid researchers with a financial interest in the outcome from being involved in human trials.
The consent agreements entered into with study participants may be legally binding on the academics but not on the sponsor, unless the sponsor has a contractual commitment saying otherwise.
Ethical rules, including the Declaration of Helsinki, require the publication of results of human trials. participants in which are often motivated by a desire to improve medical knowledge. Patients may be harmed if safety data, such risks to patients, are kept secret. Duties to human-research participants can therefore conflict with interests in non-publication such as gag clauses.
== Publication of COI declarations ==
Some journals place COI declarations at the beginning of an article but most put it in smaller print at the end. Positioning makes a difference; if readers feel they are being manipulated from the beginning of a text, they read more critically than if the same feeling is produced at the end of a text.
According to the ICMJE, "each journal should develop standards with regard to the form the [COI] information should take and where it will be posted". It is often placed after the body of the article, just before the reference section. Some COI statements, like those of anonymous reviewers, may not be published at all. (see § COIs of peer reviewers) COI statements are sometimes paywalled so they are only visible to those who have paid for full text access. This is not considered ethical by the Committee on Publication Ethics.
In 2017 PubMed began including COI statements at the end of the abstract and before the body of the article after receiving complaints that because COI declarations were only included in full article texts, they often went unseen in paywalled articles. Only COI statements that are appropriately formatted and tagged by the publisher are included.
Science journalism rarely reports COI information from the academic article reported upon; in some studies, fewer than 1% of stories included COI information.
== False statements of COIs ==
Failure to disclose a conflict of interest may, depending on the circumstances, be considered a form of corruption or academic misconduct.
== See also ==
Academic authorship
Metascience
Research Integrity Risk Index
== References ==
== External links ==
Responsible Conduct of Research: Conflicts of Interest. Online course, Columbia University.

36
data/en.wikipedia.org/wiki/Design_of_experiments-0.md Normal file
View File

@ -0,0 +1,36 @@
---
title: "Design of experiments"
chunk: 1/5
source: "https://en.wikipedia.org/wiki/Design_of_experiments"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:30.097260+00:00"
instance: "kb-cron"
---
The design of experiments (DOE), also known as experiment design or experimental design, is the design of any task that aims to describe and explain the variation of information under conditions that are hypothesized to reflect the variation. The term is generally associated with experiments in which the design introduces conditions that directly affect the variation, but may also refer to the design of quasi-experiments, in which natural conditions that influence the variation are selected for observation.
In its simplest form, an experiment aims at predicting the outcome by introducing a change of the preconditions, which is represented by one or more independent variables, also referred to as "input variables" or "predictor variables." The change in one or more independent variables is generally hypothesized to result in a change in one or more dependent variables, also referred to as "output variables" or "response variables." The experimental design may also identify control variables that must be held constant to prevent external factors from affecting the results. Experimental design involves not only the selection of suitable independent, dependent, and control variables, but planning the delivery of the experiment under statistically optimal conditions given the constraints of available resources. There are multiple approaches for determining the set of design points (unique combinations of the settings of the independent variables) to be used in the experiment.
Main concerns in experimental design include the establishment of validity, reliability, and replicability. For example, these concerns can be partially addressed by carefully choosing the independent variable, reducing the risk of measurement error, and ensuring that the documentation of the method is sufficiently detailed. Related concerns include achieving appropriate levels of statistical power and sensitivity.
Correctly designed experiments advance knowledge in the natural and social sciences and engineering, with design of experiments methodology recognised as a key tool in the successful implementation of a Quality by Design (QbD) framework. Other applications include marketing and policy making. The study of the design of experiments is an important topic in metascience.
== History ==
=== Statistical experiments, following Charles S. Peirce ===
A theory of statistical inference was developed by Charles S. Peirce in "Illustrations of the Logic of Science" (1877–1878) and "A Theory of Probable Inference" (1883), two publications that emphasized the importance of randomization-based inference in statistics.
==== Randomized experiments ====
Charles S. Peirce randomly assigned volunteers to a blinded, repeated-measures design to evaluate their ability to discriminate weights.
Peirce's experiment inspired other researchers in psychology and education, which developed a research tradition of randomized experiments in laboratories and specialized textbooks in the 1800s.
==== Optimal designs for regression models ====
Charles S. Peirce also contributed the first English-language publication on an optimal design for regression models in 1876. A pioneering optimal design for polynomial regression was suggested by Gergonne in 1815. In 1918, Kirstine Smith published optimal designs for polynomials of degree six (and less).
=== Sequences of experiments ===
The use of a sequence of experiments, where the design of each may depend on the results of previous experiments, including the possible decision to stop experimenting, is within the scope of sequential analysis, a field that was pioneered by Abraham Wald in the context of sequential tests of statistical hypotheses. Herman Chernoff wrote an overview of optimal sequential designs, while adaptive designs have been surveyed by S. Zacks. One specific type of sequential design is the "two-armed bandit", generalized to the multi-armed bandit, on which early work was done by Herbert Robbins in 1952.
== Fisher's principles ==
A methodology for designing experiments was proposed by Ronald Fisher, in his innovative books: The Arrangement of Field Experiments (1926) and The Design of Experiments (1935). Much of his pioneering work dealt with agricultural applications of statistical methods. As a mundane example, he described how to test the lady tasting tea hypothesis, that a certain lady could distinguish by flavour alone whether the milk or the tea was first placed in the cup. These methods have been broadly adapted in biological, psychological, and agricultural research.

401
data/en.wikipedia.org/wiki/Design_of_experiments-1.md Normal file
View File

@ -0,0 +1,401 @@
---
title: "Design of experiments"
chunk: 2/5
source: "https://en.wikipedia.org/wiki/Design_of_experiments"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:30.097260+00:00"
instance: "kb-cron"
---
Comparison
In some fields of study it is not possible to have independent measurements to a traceable metrology standard. Comparisons between treatments are much more valuable and are usually preferable, and often compared against a scientific control or traditional treatment that acts as baseline.
Randomization
Random assignment is the process of assigning individuals at random to groups or to different groups in an experiment, so that each individual of the population has the same chance of becoming a participant in the study. The random assignment of individuals to groups (or conditions within a group) distinguishes a rigorous, "true" experiment from an observational study or "quasi-experiment". There is an extensive body of mathematical theory that explores the consequences of making the allocation of units to treatments by means of some random mechanism (such as tables of random numbers, or the use of randomization devices such as playing cards or dice). Assigning units to treatments at random tends to mitigate confounding, which makes effects due to factors other than the treatment to appear to result from the treatment.
The risks associated with random allocation (such as having a serious imbalance in a key characteristic between a treatment group and a control group) are calculable and hence can be managed down to an acceptable level by using enough experimental units. However, if the population is divided into several subpopulations that somehow differ, and the research requires each subpopulation to be equal in size, stratified sampling can be used. In that way, the units in each subpopulation are randomized, but not the whole sample. The results of an experiment can be generalized reliably from the experimental units to a larger statistical population of units only if the experimental units are a random sample from the larger population; the probable error of such an extrapolation depends on the sample size, among other things.
Statistical replication
Measurements are usually subject to variation and measurement uncertainty; thus they are repeated and full experiments are replicated to help identify the sources of variation, to better estimate the true effects of treatments, to further strengthen the experiment's reliability and validity, and to add to the existing knowledge of the topic. However, certain conditions must be met before the replication of the experiment is commenced: the original research question has been published in a peer-reviewed journal or widely cited, the researcher is independent of the original experiment, the researcher must first try to replicate the original findings using the original data, and the write-up should state that the study conducted is a replication study that tried to follow the original study as strictly as possible.
Blocking
Blocking is the non-random arrangement of experimental units into groups (blocks) consisting of units that are similar to one another. Blocking reduces known but irrelevant sources of variation between units and thus allows greater precision in the estimation of the source of variation under study.
Orthogonality
Orthogonality concerns the forms of comparison (contrasts) that can be legitimately and efficiently carried out. Contrasts can be represented by vectors and sets of orthogonal contrasts are uncorrelated and independently distributed if the data are normal. Because of this independence, each orthogonal treatment provides different information to the others. If there are T treatments and T − 1 orthogonal contrasts, all the information that can be captured from the experiment is obtainable from the set of contrasts.
Multifactorial experiments
Use of multifactorial experiments instead of the one-factor-at-a-time method. These are efficient at evaluating the effects and possible interactions of several factors (independent variables). Analysis of experiment design is built on the foundation of the analysis of variance, a collection of models that partition the observed variance into components, according to what factors the experiment must estimate or test.
== Example ==
This example of design experiments is attributed to Harold Hotelling, building on examples from Frank Yates. The experiments designed in this example involve combinatorial designs.
Weights of eight objects are measured using a pan balance and set of standard weights. Each weighing measures the weight difference between objects in the left pan and any objects in the right pan by adding calibrated weights to the lighter pan until the balance is in equilibrium. Each measurement has a random error
ϵ
{\displaystyle \epsilon }
. The average error is zero; the standard deviations of the probability distribution of the errors is the same number σ on different weighings; errors on different weighings are independent. Denote the true weights by
θ
=
(
θ
1
,
…
,
θ
8
)
{\displaystyle \mathbf {\theta } =(\theta _{1},\dots ,\theta _{8})\,}
.
We consider two different experiments with the same amount of measurements:
Weigh each of the eight objects individually.
left pan
right pan
1st weighing:
1
(empty)
2st weighing:
2
(empty)
3rd weighing:
3
(empty)
.
.
.
.
.
.
.
.
.
{\displaystyle {\begin{array}{lcc}&{\text{left pan}}&{\text{right pan}}\\\hline {\text{1st weighing:}}&1\ &{\text{(empty)}}\\{\text{2st weighing:}}&2\ &{\text{(empty)}}\\{\text{3rd weighing:}}&3\ &{\text{(empty)}}\\...&...&...\end{array}}}
Do the eight weighings according to the following schedule:
left pan
right pan
1st weighing:
1
2
3
4
5
6
7
8
(empty)
2nd:
1
2
3
8
4
5
6
7
3rd:
1
4
5
8
2
3
6
7
4th:
1
6
7
8
2
3
4
5
5th:
2
4
6
8
1
3
5
7
6th:
2
5
7
8
1
3
4
6
7th:
3
4
7
8
1
2
5
6
8th:
3
5
6
8
1
2
4
7
{\displaystyle {\begin{array}{lcc}&{\text{left pan}}&{\text{right pan}}\\\hline {\text{1st weighing:}}&1\ 2\ 3\ 4\ 5\ 6\ 7\ 8&{\text{(empty)}}\\{\text{2nd:}}&1\ 2\ 3\ 8\ &4\ 5\ 6\ 7\\{\text{3rd:}}&1\ 4\ 5\ 8\ &2\ 3\ 6\ 7\\{\text{4th:}}&1\ 6\ 7\ 8\ &2\ 3\ 4\ 5\\{\text{5th:}}&2\ 4\ 6\ 8\ &1\ 3\ 5\ 7\\{\text{6th:}}&2\ 5\ 7\ 8\ &1\ 3\ 4\ 6\\{\text{7th:}}&3\ 4\ 7\ 8\ &1\ 2\ 5\ 6\\{\text{8th:}}&3\ 5\ 6\ 8\ &1\ 2\ 4\ 7\end{array}}}

309
data/en.wikipedia.org/wiki/Design_of_experiments-2.md Normal file
View File

@ -0,0 +1,309 @@
---
title: "Design of experiments"
chunk: 3/5
source: "https://en.wikipedia.org/wiki/Design_of_experiments"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:30.097260+00:00"
instance: "kb-cron"
---
Let yi be the measured difference for i = 1, ..., 8. The relationship between the true weights and experimental measurements may be represented with a general linear model, with the design matrix
W
{\displaystyle W}
having entries from
{
−
1
,
0
,
1
}
{\displaystyle \{-1,0,1\}}
:
y
=
W
θ
+
ϵ
{\displaystyle y=W\theta +\epsilon }
The first design is represented by an identity matrix while the second design is represented by an 8x8 Hadamard matrix,
H
{\displaystyle H}
, both examples of weighing matrices.
The weights are typically estimated using the method of least squares. Using a weighing matrix, this is equivalent to inverting on the measurements:
θ
^
A
=
I
−
1
y
=
y
{\displaystyle {\hat {\theta }}_{A}=I^{-1}y=y}
θ
^
B
=
H
−
1
y
{\displaystyle {\hat {\theta }}_{B}=H^{-1}y}
The question of design of experiments is: which experiment is better?
Investigating estimate A vs B for the first weight:
Var
⁡
(
θ
^
A
,
1
)
=
Var
⁡
(
y
1
)
=
σ
2
{\displaystyle \operatorname {Var} ({\hat {\theta }}_{A,1})=\operatorname {Var} (y_{1})=\sigma ^{2}}
Var
⁡
(
θ
^
B
,
1
)
=
Var
⁡
(
y
1
+
y
2
−
y
3
−
y
4
+
y
5
+
y
6
−
y
7
−
y
8
8
)
=
σ
2
8
{\displaystyle \operatorname {Var} ({\hat {\theta }}_{B,1})=\operatorname {Var} ({\frac {y_{1}+y_{2}-y_{3}-y_{4}+y_{5}+y_{6}-y_{7}-y_{8}}{8}})={\frac {\sigma ^{2}}{8}}}
A similar result follows for the remaining weight estimates. Thus, the second experiment gives us 8 times as much precision for the estimate of a single item, despite costing the same number of resources (number of weightings).
Many problems of the design of experiments involve combinatorial designs, as in this example and others.
== Avoiding false positives ==
False positive conclusions, often resulting from the pressure to publish or the author's own confirmation bias, are an inherent hazard in many fields.
Use of double-blind designs can prevent biases potentially leading to false positives in the data collection phase. When a double-blind design is used, participants are randomly assigned to experimental groups but the researcher is unaware of what participants belong to which group. Therefore, the researcher can not affect the participants' response to the intervention.
Experimental designs with undisclosed degrees of freedom are a problem, in that they can lead to conscious or unconscious "p-hacking": trying multiple things until you get the desired result. It typically involves the manipulation – perhaps unconsciously – of the process of statistical analysis and the degrees of freedom until they return a figure below the p<.05 level of statistical significance.
P-hacking can be prevented by preregistering researches, in which researchers have to send their data analysis plan to the journal they wish to publish their paper in before they even start their data collection, so no data manipulation is possible.
Another way to prevent this is taking a double-blind design to the data-analysis phase, making the study triple-blind, where the data are sent to a data-analyst unrelated to the research who scrambles up the data so there is no way to know which participants belong to before they are potentially taken away as outliers.
Clear and complete documentation of the experimental methodology is also important in order to support replication of results.
== Discussion topics when setting up an experimental design ==
An experimental design or randomized clinical trial requires careful consideration of several factors before actually doing the experiment. An experimental design is the laying out of a detailed experimental plan in advance of doing the experiment. Some of the following topics have already been discussed in the principles of experimental design section:
How many factors does the design have, and are the levels of these factors fixed or random?
Are control conditions needed, and what should they be?
Manipulation checks: did the manipulation really work?
What are the background variables?
What is the sample size? How many units must be collected for the experiment to be generalisable and have enough power?
What is the relevance of interactions between factors?
What is the influence of delayed effects of substantive factors on outcomes?
How do response shifts affect self-report measures?
How feasible is repeated administration of the same measurement instruments to the same units at different occasions, with a post-test and follow-up tests?
What about using a proxy pretest?
Are there confounding variables?
Should the client/patient, researcher or even the analyst of the data be blind to conditions?
What is the feasibility of subsequent application of different conditions to the same units?
How many of each control and noise factors should be taken into account?
The independent variable of a study often has many levels or different groups. In a true experiment, researchers can have an experimental group, which is where their intervention testing the hypothesis is implemented, and a control group, which has all the same element as the experimental group, without the interventional element. Thus, when everything else except for one intervention is held constant, researchers can certify with some certainty that this one element is what caused the observed change. In some instances, having a control group is not ethical. This is sometimes solved using two different experimental groups. In some cases, independent variables cannot be manipulated, for example when testing the difference between two groups who have a different disease, or testing the difference between genders (obviously variables that would be hard or unethical to assign participants to). In these cases, a quasi-experimental design may be used.

42
data/en.wikipedia.org/wiki/Design_of_experiments-3.md Normal file
View File

@ -0,0 +1,42 @@
---
title: "Design of experiments"
chunk: 4/5
source: "https://en.wikipedia.org/wiki/Design_of_experiments"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:30.097260+00:00"
instance: "kb-cron"
---
== Causal attributions ==
In the pure experimental design, the independent (predictor) variable is manipulated by the researcher – that is – every participant of the research is chosen randomly from the population, and each participant chosen is assigned randomly to conditions of the independent variable. Only when this is done is it possible to certify with high probability that the reason for the differences in the outcome variables are caused by the different conditions. Therefore, researchers should choose the experimental design over other design types whenever possible. However, the nature of the independent variable does not always allow for manipulation. In those cases, researchers must be aware of not certifying about causal attribution when their design doesn't allow for it. For example, in observational designs, participants are not assigned randomly to conditions, and so if there are differences found in outcome variables between conditions, it is likely that there is something other than the differences between the conditions that causes the differences in outcomes, that is – a third variable. The same goes for studies with correlational design.
== Statistical control ==
It is best that a process be in reasonable statistical control prior to conducting designed experiments. When this is not possible, proper blocking, replication, and randomization allow for the careful conduct of designed experiments.
To control for nuisance variables, researchers institute control checks as additional measures. Investigators should ensure that uncontrolled influences (e.g., source credibility perception) do not skew the findings of the study. A manipulation check is one example of a control check. Manipulation checks allow investigators to isolate the chief variables to strengthen support that these variables are operating as planned.
One of the most important requirements of experimental research designs is the necessity of eliminating the effects of spurious, intervening, and antecedent variables. In the most basic model, cause (X) leads to effect (Y). But there could be a third variable (Z) that influences (Y), and X might not be the true cause at all. Z is said to be a spurious variable and must be controlled for. The same is true for intervening variables (a variable in between the supposed cause (X) and the effect (Y)), and anteceding variables (a variable prior to the supposed cause (X) that is the true cause). When a third variable is involved and has not been controlled for, the relation is said to be a zero order relationship. In most practical applications of experimental research designs there are several causes (X1, X2, X3). In most designs, only one of these causes is manipulated at a time.
== Experimental designs after Fisher ==
Some efficient designs for estimating several main effects were found independently and in near succession by Raj Chandra Bose and K. Kishen in 1940 at the Indian Statistical Institute, but remained little known until the Plackett–Burman designs were published in Biometrika in 1946. About the same time, C. R. Rao introduced the concepts of orthogonal arrays as experimental designs. This concept played a central role in the development of Taguchi methods by Genichi Taguchi, which took place during his visit to Indian Statistical Institute in early 1950s. His methods were successfully applied and adopted by Japanese and Indian industries and subsequently were also embraced by US industry albeit with some reservations.
In 1950, Gertrude Mary Cox and William Gemmell Cochran published the book Experimental Designs, which became the major reference work on the design of experiments for statisticians for years afterwards.
Developments of the theory of linear models have encompassed and surpassed the cases that concerned early writers. Today, the theory rests on advanced topics in linear algebra, algebra and combinatorics.
As with other branches of statistics, experimental design is pursued using both frequentist and Bayesian approaches: In evaluating statistical procedures like experimental designs, frequentist statistics studies the sampling distribution while Bayesian statistics updates a probability distribution on the parameter space.
Some important contributors to the field of experimental designs are C. S. Peirce, R. A. Fisher, F. Yates, R. C. Bose, A. C. Atkinson, R. A. Bailey, D. R. Cox, G. E. P. Box, W. G. Cochran, Walter T. Federer, V. V. Fedorov, A. S. Hedayat, J. Kiefer, O. Kempthorne, J. A. Nelder, Andrej Pázman, Friedrich Pukelsheim, D. Raghavarao, C. R. Rao, Shrikhande S. S., J. N. Srivastava, William J. Studden, G. Taguchi and H. P. Wynn.
The textbooks of D. Montgomery, R. Myers, and G. Box/W. Hunter/J.S. Hunter have reached generations of students and practitioners. Furthermore, there is ongoing discussion of experimental design in the context of model building for models either static or dynamic models, also known as system identification.
== Human participant constraints ==
Laws and ethical considerations preclude some carefully designed
experiments with human subjects. Legal constraints are dependent on
jurisdiction. Constraints may involve
institutional review boards, informed consent
and confidentiality affecting both clinical (medical) trials and
behavioral and social science experiments.
In the field of toxicology, for example, experimentation is performed
on laboratory animals with the goal of defining safe exposure limits
for humans. Balancing
the constraints are views from the medical field. Regarding the randomization of patients,
"... if no one knows which therapy is better, there is no ethical
imperative to use one therapy or another." (p 380) Regarding
experimental design, "...it is clearly not ethical to place subjects
at risk to collect data in a poorly designed study when this situation
can be easily avoided...". (p 393)

54
data/en.wikipedia.org/wiki/Design_of_experiments-4.md Normal file
View File

@ -0,0 +1,54 @@
---
title: "Design of experiments"
chunk: 5/5
source: "https://en.wikipedia.org/wiki/Design_of_experiments"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:30.097260+00:00"
instance: "kb-cron"
---
== See also ==
Adversarial collaboration – Method of research
Bayesian experimental design – Experimental design framework
Block design – Structure in combinatorial mathematics
Box–Behnken design – Experimental designs for response surface methodology
Central composite design – Experimental design in statistical mathematics
Clinical study design – Plan for research in clinical medicine
Computer experiment – Experiment used to study computer simulation
Controlling for a variable – Binning data according to measured values of the variable
Experimetrics (econometrics-related experiments)
Factor analysis – Statistical method
Fractional factorial design – Statistical experimental design approach
Glossary of experimental design
Grey box model – Mathematical data production model with limited structure
Industrial engineering – Branch of engineering which deals with the optimization of complex processes or systems
Instrument effect
Law of large numbers – Averages of repeated trials converge to the expected value
Manipulation checks
Multifactor design of experiments software
One-factor-at-a-time method – Method of designing experiments
Optimal design – Experimental design that is optimal with respect to some statistical criterionPages displaying short descriptions of redirect targets
Plackett–Burman design – Type of experimental design
Probabilistic design – Discipline within engineering design
Protocol (natural sciences) – Procedural method for the design and implementation of an experimentPages displaying short descriptions of redirect targets
Quasi-experimental design – Empirical interventional studyPages displaying short descriptions of redirect targets
Randomized block design – Design of experiments to collect similar contexts togetherPages displaying short descriptions of redirect targets
Randomized controlled trial – Form of scientific experiment
Research design – Overall strategy utilized to carry out research
Robust parameter design
Sample size determination – Statistical considerations on how many observations to make
Supersaturated designs – Type of experimental design
Royal Commission on Animal Magnetism – 1784 French scientific bodies' investigations involving systematic controlled trials
Survey sampling – Statistical selection process
System identification – Statistical methods to build mathematical models of dynamical systems from measured data
Taguchi methods – Statistical methods to improve the quality of manufactured goods
== References ==
=== Sources ===
== External links ==
A chapter from a "NIST/SEMATECH Handbook on Engineering Statistics" at NIST
Box–Behnken designs from a "NIST/SEMATECH Handbook on Engineering Statistics" at NIST

39
data/en.wikipedia.org/wiki/Further_research_is_needed-0.md Normal file
View File

@ -0,0 +1,39 @@
---
title: "Further research is needed"
chunk: 1/1
source: "https://en.wikipedia.org/wiki/Further_research_is_needed"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:31.281573+00:00"
instance: "kb-cron"
---
"Further research is needed" (FRIN), "more research is needed" and other variants of similar phrases are commonly used in research papers. The cliché is so common that it has attracted research, regulation and cultural commentary.
== Meaning ==
Some research journals have banned the phrase "more research is needed" on the grounds that it is redundant; it is almost always true and fits almost any article, and so can be taken as understood.
A 2004 metareview by the Cochrane collaboration of their own systematic medical reviews found that 93% of the reviews studied made indiscriminate FRIN-like statements, reducing their ability to guide future research. The presence of FRIN had no correlation with the strength of the evidence against the medical intervention. Authors who thought a treatment was useless were just as likely to recommend researching it further.
Indeed, authors may recommend "further research" when, given the existing evidence, further research would be extremely unlikely to be approved by an ethics committee.
Studies finding that a treatment has no noticeable effects are sometimes greeted with statements that "more research is needed" by those convinced that the treatment is effective, but the effect has not yet been found. Since even the largest study can never rule out an infinitesimal effect, an effect can only ever be shown to be insignificant, not non-existent. Similarly, Trish Greenhalgh, Professor of Primary Care Health Sciences at the University of Oxford, argues that FRIN is often used as a way in which a "[l]ack of hard evidence to support the original hypothesis gets reframed as evidence that investment efforts need to be redoubled", and a way to avoid upsetting hopes and vested interests. She has also described FRIN as "an indicator that serious scholarly thinking on the topic has ceased", saying that "it is almost never the only logical conclusion that can be drawn from a set of negative, ambiguous, incomplete or contradictory data."
== Addressing the phrase ==
Greenhalgh suggests that, because vague FRIN statements are an argument that "tomorrow's research investments should be pitched into precisely the same patch of long grass as yesterday's", funding should be refused to those making them. She and others argue that more thought and research is needed into methods for determining where more research is needed.
Academic journal editors were banning unqualified FRIN statements as early as 1990, requiring more specific information such as what types of research were needed, and what questions they ought to address. Researchers themselves have strongly recommended that research articles detail what research is needed. This is conventional in some fields. Other commentators suggest that articles would benefit by assessing the likely value of possible further research.
== Example ==
Both the needfulness and needlessness of further research may be overlooked. The blobbogram leading this article is from a systematic review; it shows clinical trials of the use of corticosteroids to hasten lung development in pregnancies where a baby is likely to be born prematurely. Long after there was enough evidence to show that this treatment saved babies' lives, the evidence was not widely known, the treatment was not widely used, and further research was done into the same question. After the review made the evidence better known, the treatment was used more, preventing thousands of pre-term babies from dying of infant respiratory distress syndrome.
However, when the treatment was rolled out in lower- and middle-income countries, early data suggested that more pre-term babies died. It was thought that this could be because of a higher risk of infection, which is more likely to kill a baby in places with poor medical care and more malnourished mothers. The 2017 version of the review therefore said that there was "little need" for further research into the usefulness of the treatment in higher-income countries, but further research was needed on optimal dosage and on how to best treat lower-income and higher-risk mothers.
Further research was done, and found the treatment did actually benefit babies in lower-income countries, too. The December 2020 version of the review stated that the "evidence [that the treatment saves babies] is robust, regardless of resource setting (high, middle or low)" and that further research should focus on "specific understudied subgroups such as multiple pregnancies and other high-risk obstetric groups, and the risks and benefits in the very early or very late preterm periods".
== In culture ==
The idea that research papers always end with some variation of FRIN was described as an "old joke" in a 1999 epidemiology editorial.
FRIN has been advocated as a position politicians should take on under-evidenced claims. Requests for further research on questions relevant to political policy can lead to better-informed decisions, but FRIN statements have also been used in bad faith: for instance, to delay political decisions, or as a justification for ignoring existing research knowledge (as was done by nicotine companies). Policymakers may also not know of existing research; they seldom systematically search databases of research literature, preferring to use Google and ask colleagues for research papers.
FRIN has been advocated as a motto for life, applicable everywhere except research papers; it has been printed on T-shirts, and satirized by the "Collectively Unconscious" blog, which reported that an article in the journal Science had concluded that "no further research is needed, at all, anywhere, ever".
The webcomic xkcd has also used the phrase as a topic, for self-satire, and as a bathetic punchline.
== References ==

77
data/en.wikipedia.org/wiki/HARKing-0.md Normal file
View File

@ -0,0 +1,77 @@
---
title: "HARKing"
chunk: 1/1
source: "https://en.wikipedia.org/wiki/HARKing"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:33.664349+00:00"
instance: "kb-cron"
---
HARKing (hypothesizing after the results are known) is an acronym coined by social psychologist Norbert Kerr that refers to the questionable research practice of "presenting a post hoc hypothesis in the introduction of a research report as if it were an a priori hypothesis". Hence, a key characteristic of HARKing is that post hoc hypothesizing is falsely portrayed as a priori hypothesizing. HARKing may occur when a researcher tests an a priori hypothesis but then omits that hypothesis from their research report after they find out the results of their test. Post hoc analysis or post hoc theorizing then may lead to a post hoc hypothesis.
== Types ==
Several types of HARKing have been distinguished, including:
THARKing
Transparently hypothesizing after the results are known, rather than the secretive, undisclosed, HARKing that was first proposed by Kerr. In this case, researchers openly declare that they developed their hypotheses after they observed their research results.
CHARKing (or Pure HARKing)
CHARKing or "pure HARKing" refers to the practice of constructing new hypotheses after the results are known and presenting them as a priori hypotheses. CHARKing is often regarded as the prototypical form of HARKing.
RHARKing
RHARKing refers to retrieving old hypotheses from the existing literature after the results are known and presenting them as a priori hypotheses Note that RHARKed hypotheses can be considered to be a priori hypotheses in the sense that they were developed and published prior to knowledge of the current research results.
SHARKing
Suppressing a priori hypotheses after the results of tests of those hypotheses are known.
Active and passive HARKing
Active HARKing occurs when researchers HARK prior to submitting their research report for publication. Passive HARKing occurs when researchers HARK in response to requests by editors and reviewers during the peer review process.
== Prevalence among researchers ==
Concerns about HARKing appear to be increasing in the scientific community, as shown by the increasing number of citations to Kerr's seminal article. A 2017 review of six surveys found that an average of 43% of researchers surveyed (mainly psychologists) self-reported HARKing "at least once". This figure may be an underestimate if researchers are concerned about reporting questionable research practices, do not perceive themselves to be responsible for HARKing that is proposed by editors and reviewers (i.e., passive HARKing), and/or do not recognize their HARKing due to hindsight or confirmation biases.
== Researchers' motivation ==
HARKing appears to be motivated by a desire to publish research in a publication environment that values a priori hypotheses over post hoc hypotheses and contains a publication bias against null results. In order to improve their chances of publishing their results, researchers may secretly suppress any a priori hypotheses that failed to yield significant results, construct or retrieve post hoc hypotheses that account for any unexpected significant results, and then present these new post hoc hypotheses in their research reports as if they are a priori hypotheses.
== Prediction and accommodation ==
HARKing is associated with the debate regarding prediction and accommodation. In the case of prediction, hypotheses are deduced from a priori theory and evidence. In the case of accommodation, hypotheses are induced from the current research results. One view is that HARKing represents a form of accommodation in which researchers induce ad hoc hypotheses from their current results. Another view is that HARKing represents a form of prediction in which researchers deduce hypotheses from a priori theory and evidence after they know their current results.
== Potential costs to science ==
Potential costs of HARKing include:
Translating Type I errors into hard-to-eradicate theory
Propounding theories that cannot (pending replication) pass Popper's disconfirmability test
Disguising post hoc explanations as a priori explanations
Not communicating valuable information about what did not work
Taking unjustified statistical licence
Presenting an inaccurate model of science to students
Encouraging questionable practices in other grey areas
Making us less receptive to serendipitous findings
Encouraging adoption of narrow, context-bound new theory
Encouraging retention of too-broad, disconfirmable old theory
Inhibiting identification of plausible alternative hypotheses
Implicitly violating basic ethical principles
In 2022, Rubin provided a critical analysis of Kerr's 12 costs of HARKing. He concluded that these costs "are either misconceived, misattributed to HARKing, lacking evidence, or that they do not take into account pre- and post-publication peer review and public availability to research materials and data."
== HARKing and the replication crisis ==
Some of the costs of HARKing are thought to have led to the replication crisis in science. Hence, Bishop described HARKing as one of "the four horsemen of the reproducibility apocalypse," with publication bias, low statistical power, and p-hacking being the other three. An alternative view is that it is premature to conclude that HARKing has contributed to the replication crisis.
The preregistration of research hypotheses prior to data collection has been proposed as a method of identifying and deterring HARKing. However, the use of preregistration to prevent HARKing is controversial.
== Ethical concerns ==
Kerr pointed out that "HARKing can entail concealment. The question then becomes whether what is concealed in HARKing can be a useful part of the 'truth' ...or is instead basically uninformative (and may, therefore, be safely ignored at an author's discretion)". Three different positions about the ethics of HARKing depend on whether HARKing conceals "a useful part of the truth".
The first position is that all HARKing is unethical under all circumstances because it violates a fundamental principle of communicating scientific research honestly and completely. According to this position, HARKing always conceals a useful part of the truth.
A second position is that HARKing falls into a "gray zone" of ethical practice. According to this position, some forms of HARKing are more or less ethical under some circumstances. Hence, only some forms of HARKing conceal a useful part of the truth under some conditions. Consistent with this view, a 2018 survey of 119 USA researchers found that HARKing ("reporting an unexpected result as having been hypothesized from the start") was associated with "ambiguously unethical" research practices more than with "unambiguously unethical" research practices.
A third position is that HARKing is acceptable provided that hypotheses are explicitly deduced from a priori theory and evidence, as explained in a theoretical rationale, and readers have access to the relevant research data and materials. According to this position, HARKing does not prevent readers from making an adequately informed evaluation of the theoretical quality and plausibility of the HARKed hypotheses and the methodological rigor with which the hypotheses have been tested. In this case, HARKing does not conceal a useful part of the truth. Furthermore, researchers may claim that a priori theory and evidence predict their results even if the prediction is deduced after they know their results.
== See also ==
Data dredging
Texas sharpshooter fallacy
== References ==

73
data/en.wikipedia.org/wiki/ICMJE_recommendations-0.md Normal file
View File

@ -0,0 +1,73 @@
---
title: "ICMJE recommendations"
chunk: 1/1
source: "https://en.wikipedia.org/wiki/ICMJE_recommendations"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:34.831046+00:00"
instance: "kb-cron"
---
The ICMJE recommendations (full title, "Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals") are a set of guidelines produced by the International Committee of Medical Journal Editors for standardising the ethics, preparation and formatting of manuscripts submitted to biomedical journals for publication. Compliance with the ICMJE recommendations is required by most leading biomedical journals. Levels of real compliance are subject to debate. As of 9 January 2020, 5570 journals worldwide claim to follow the ICMJE recommendations.
The recommendations were first issued in 1979 under the title "Uniform Requirements for Manuscripts Submitted to Biomedical Journals" (abbreviated URMs and often shortened to "Uniform Requirements"). After a series of revisions, they were given their current name in 2013.
== International Committee of Medical Journal Editors ==
The International Committee of Medical Journal Editors (ICMJE) was originally known as the Vancouver Group, after the location of their first meeting in Vancouver, British Columbia in Canada. As of 2017 members of the ICMJE are:
Annals of Internal Medicine
BMJ
Bulletin of the World Health Organization
Deutsches Ärzteblatt
Ethiopian Journal of Health Sciences
Iranian Journal of Medical Sciences
Journal of the American Medical Association (JAMA)
New England Journal of Medicine
Public Library of Science
Journal of Korean Medical Science
Revista Médica de Chile
The Lancet
The U.S. National Library of Medicine
The New Zealand Medical Journal
The World Association of Medical Editors
Ugeskrift for Læger (Danish Medical Journal)
== Citation style ==
The citation style recommended by the ICMJE Recommendations, which is also known as the Vancouver system, is the style used by the United States National Library of Medicine (NLM), codified in Citing Medicine.
References are numbered consecutively in order of appearance in the text – they are identified by Arabic numerals enclosed in parentheses.
Example of a journal citation:
Leurs R, Church MK, Taglialatela M. H1-antihistamines: inverse agonism, anti-inflammatory actions and cardiac effects. Clin Exp Allergy 2002 Apr;32(4):489–498.
== Manuscripts describing human interventional clinical trials ==
URM includes a mandate for manuscripts describing human interventional trials to register a trial in a clinical trial registry (e.g., ClinicalTrials.gov) and to include the trial registration ID in the abstract of the article. The URM also requires that this registration is done prior enrolling the first participant. A study of five high impact factor journals (founders of ICMJE) showed that only 89% of published articles (articles published during 2010–2011; about trials that completed in 2008) were properly registered prior enrolling the first participant.
A 2016 draft proposal to require that manuscript authors disclose individual patient data relevant to published outcomes within six months of reporting a clinical trial was successfully opposed by Jeffrey M. Drazen, then editor-in-chief of The New England Journal of Medicine. Drazen's claim that "'a new class of research person will emerge,' one who might 'even use the data to try to disprove what the original investigators had posited'" was criticized as protecting intentionally deceptive biomedical researchers.
== Disclosure of competing interests ==
The ICMJE also developed a uniform format for disclosure of competing interests in journal articles.
== Grey literature ==
The Uniform Requirements were adapted by the Grey Literature International Steering Committee GLISC for the production of scientific and technical reports included in the wider category of grey literature. These GLISC Guidelines for the production of scientific and technical reports are translated to French, German, Italian and Spanish and are available on the GLISC website [1].
== See also ==
Conflicts of interest in academic publishing
EASE Guidelines for Authors and Translators of Scientific Articles
IMRAD
Scientific misconduct
== References ==
== External links ==
About ICMJE
The Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly work in Medical Journals
National Library of Medicine – Uniform Requirements sample references
Journals Following the Uniform Requirements for Manuscripts Archived 10 February 2014 at the Wayback Machine
Use of Uniform Requirements for scientific and technical reports

59
data/en.wikipedia.org/wiki/Steven_N._Goodman-0.md Normal file
View File

@ -0,0 +1,59 @@
---
title: "Steven N. Goodman"
chunk: 1/1
source: "https://en.wikipedia.org/wiki/Steven_N._Goodman"
category: "reference"
tags: "science, encyclopedia"
date_saved: "2026-05-05T03:14:32.428701+00:00"
instance: "kb-cron"
---
Steven N. Goodman (born 1954) is an American epidemiologist and Professor of Epidemiology and Population Health and of Medicine at the Stanford School of Medicine, where he serves as Associate Dean of Clinical and Translational Research. He is co-founder and co-director of the Meta-Research Innovation Center at Stanford (METRICS), a center dedicated to studying and improving the reproducibility and efficiency of biomedical research, alongside John Ioannidis. He is also founder and director of the Stanford Program on Research Rigor and Reproducibility (SPORR).
Goodman has made extensive contributions to the foundations of scientific and statistical inference within the biosciences. In 1999, he coined the term "p-value fallacy" in a pair of landmark papers arguing for the adoption of Bayesian methods in medical research. He is a sibling of journalist and Democracy Now! host Amy Goodman and journalist David Goodman.
== Education and career ==
Goodman received his AB in Biochemistry and Applied Mathematics from Harvard College in 1976, his MD from New York University School of Medicine, and his MHS and PhD in Biostatistics from the Johns Hopkins Bloomberg School of Public Health, completing his doctoral dissertation Evidence and Clinical Trials in 1989 under the supervision of Richard Royall. He completed a residency in pediatrics at St. Louis Children's Hospital, Washington University School of Medicine, and holds board certification in General Pediatrics from the American Board of Pediatrics.
From 1989 to 2011, Goodman served on the faculties of the Johns Hopkins School of Medicine and the Johns Hopkins Bloomberg School of Public Health, where he was co-director of the doctoral program in Epidemiology and Director of the Division of Biostatistics and Bioinformatics in the Department of Oncology (2007–2010). He joined the Stanford School of Medicine faculty in 2011.
Goodman has been a senior statistical editor of the Annals of Internal Medicine since 1987 and served as Editor of Clinical Trials: Journal of the Society for Clinical Trials from 2004 to 2013. He chaired the Methodology Committee of the Patient-Centered Outcomes Research Institute (PCORI) until 2024, where he led their open science and data sharing efforts, and has served as scientific advisor for the national Blue Cross–Blue Shield Technology Assessment Program since 2004. He has served on numerous National Academies committees, including a committee on vaccine safety, a 2012 committee on drug safety (which he chaired), and a 2014 committee on sharing data from clinical trials.
== Research ==
=== Statistical inference and the p-value ===
Goodman's research has centered on the proper measurement, conceptualization, and synthesis of research evidence, with particular emphasis on Bayesian approaches. His early work examined the historical debate between R. A. Fisher's approach to p-values and the Neyman–Pearson hypothesis testing framework, arguing that the combination of the two methods had led to widespread misunderstanding of statistical evidence in medicine.
In 1999, Goodman published his most cited work, a two-part series in the Annals of Internal Medicine titled "Toward Evidence-Based Medical Statistics," in which he identified what he called the "P value fallacy"—the widespread misinterpretation of p-values as direct measures of the probability that a hypothesis is true—and proposed the Bayes factor as a more interpretable alternative for quantifying statistical evidence. He further elaborated on these themes in subsequent publications advocating for Bayesian reasoning in clinical research. His 2008 paper "A Dirty Dozen: Twelve P-Value Misconceptions" catalogued twelve common misinterpretations of the p-value and has been widely cited across multiple disciplines.
=== Research reproducibility ===
Goodman has been a leading figure in efforts to define and improve research reproducibility. In 2016, he co-authored a widely cited Science Translational Medicine paper with Daniele Fanelli and John Ioannidis that proposed a standardized conceptual framework distinguishing among "methods reproducibility," "results reproducibility," and "inferential reproducibility." In 2007, he and Sander Greenland published a critique of Ioannidis's influential claim that most published research findings are false, arguing that the conclusion rested on circular reasoning within its Bayesian framework.
He has also contributed to work on reproducible research practices in medical publishing.
=== Clinical research methods ===
Goodman has made contributions to the methods of clinical trials, comparative effectiveness research, and meta-analysis. He co-authored work on methodological standards for comparative effectiveness research with the PCORI, and published on the need for transformational change in randomized clinical trial design for comparative effectiveness. He also contributed to the development of methods for comparative effectiveness research more broadly.
His work on meta-analysis includes early contributions on the role of evidence in meta-analytic synthesis and later work addressing the problem of inconsistent effects in random-effects meta-analysis. He has also written on the concept of "metabias" as a challenge for comparative effectiveness research.
Additional contributions include work on practical improvements to the continual reassessment method for phase I studies, the misuse of statistical power in interpreting clinical trial results, Bayesian approaches to pediatric clinical trials, and the quality of peer review.
=== Research ethics and causal inference ===
Goodman has contributed to the ethics of clinical research, particularly in the context of learning healthcare systems. He co-authored work with Ruth Faden and Nancy Kass proposing an ethics framework that departs from the traditional research-treatment distinction, and contributed to discussions of ethical considerations in drug safety research. He has also co-authored work on causal inference in public health.
== Awards and honors ==
Myrto Lefkopoulou Distinguished Lecturer, Harvard Department of Biostatistics (2000)
Spinoza Chair in Medicine, University of Amsterdam (2016)
Abraham Lilienfeld Award, American College of Epidemiology (2019)
Elected to the National Academy of Medicine (2020)
== Selected publications ==
Goodman, S. N., Royall, R. (1988). "Evidence and scientific research". American Journal of Public Health. 78 (12): 1568–1574. doi:10.2105/ajph.78.12.1568. PMC 1349737. PMID 3189634.{{cite journal}}: CS1 maint: multiple names: authors list (link)
Goodman, S. N. (2007). "Stopping at nothing? Some dilemmas of data monitoring in clinical trials". Annals of Internal Medicine. 146 (12): 882–887. doi:10.7326/0003-4819-146-12-200706190-00010. PMID 17577008. S2CID 21243453.
Goodman, S. N. (2002). "The mammography dilemma: A crisis for evidence-based medicine?". Annals of Internal Medicine. 137 (5): 363–365. doi:10.7326/0003-4819-137-5_part_1-200209030-00015. PMID 12204023. S2CID 40784168.
Goodman, S. (2011). "Confessions of a chagrined trialist". BMJ Quality & Safety. 20 (Suppl_1): i97–i98. doi:10.1136/bmjqs.2010.046623. PMC 3066790. PMID 21450783.
Moher, D., Naudet, F., Cristea, I. A., Miedema, F., Ioannidis, J. P. A., Goodman, S. N. (2018). "Assessing scientists for hiring, promotion, and tenure". PLOS Biology. 16 (3) e2004089. doi:10.1371/journal.pbio.2004089. PMC 5875754. PMID 29596415.{{cite journal}}: CS1 maint: multiple names: authors list (link)
== References ==