Wikipedia:Scientific standards

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Wikipedia covers an immense number of topics dealing with natural, social, and formal sciences. In all disciplines of science, research and theories are published under a peer-review system, and all credible journals of science are indexed by one or more of the major indexes of scientific journals.

This proposal addresses Wikipedia's reporting and description of such topics. When writing articles or sections involving science, we rely on the reliable sources to describe the scientific evidence accurately while weighting material in a fashion that reflects the weight of published peer reviewed literature in the field.

Reading and editing science-related articles[edit]

Wikipedia strives to be a thorough, comprehensive, and easily understandable encyclopedia. Readers of Wikipedia are both laypersons and scientists, and articles should be written for both types of audience. Care should be taken to accommodate readers that have no expertise. At the same time, technical content should not be removed or misrepresented just because it is difficult to understand.

Since Wikipedia is an encyclopedia anyone can edit, editors need not have academic credentials or affiliations to participate in science-related articles. Editors should, however, adhere to the standard of verifiability expected in academic and science-related work, as put forth in the Wikipedia policies that relate to sourcing, such as neutrality, verifiability and prohibition of original research. Some degree of scientific literacy may be needed to understand sources that describe scientific observations and theories. The goal is for editors of all levels of education and expertise to collaborate in creating quality articles that reach the widest possible audience.

Accurately describing science[edit]

Science is often collaborative with replication and confirmation done by independent teams working to address the same questions. The vast number of independent sources that come to the same conclusion testify to the existence of a general consensus on how much of the physical world functions. This consensus is reached in part through peer scrutiny of research and ideas. Not all positions or opinions in science are considered equal. The peer-review process is intended to test the validity of the presented work. The consensus is built through a complex social process centered around the body of peer-reviewed information. Wikipedia's core neutrality policy prevents it from participating in the review or consensus-building process (Wikipedia neither endorses nor rejects viewpoints); however, when writing about a given area of science it is important that we accurately describe how experts in that field have received each other's work.

Likewise, when a particular opinion is held by a minority of scientists within a particular scientific discipline, we should not describe it as equal to more accepted theories. Wikipedia's co-founder Jimmy Wales stated early in its development that it is important to accurately describe any inequality that exists within the scientific community:

NPOV does not require us to present all these views as if they are equal! This is one of the things that's hardest to remember about NPOV. If a view is the majority view of a broad consensus of scientists, then we say so. If a view is a minority view of some scientists, scientists who are respected by the mainstream that differs with them on this particular matter, then we say so. And if a view is held only by a few people without any traditional training or credentials, and if that view is dismissed by virtually all mainstream scientists, then we can say that, too.[1]

Perspectives on science[edit]

Wikipedia does not endorse a single perspective on the metaphysical basis of science. In the philosophy of science, there is a wide range of perspectives. An empiricist contends that because natural science endeavors to be objective and unbiased, natural science has no point of view; or if it does, it is exactly a neutral point of view, while a dogmatist contends that science is biased because it holds itself as an authority over all other interpretations of life, such as religious, mythical, spiritual, or humanistic explanations. Either way, Wikipedia seeks to have a neutral point of view, describing all viewpoints accurately, reliably, and in their proper proportion.

Proper weighting[edit]

Within any scientific discipline, a body of peer reviewed articles define the range of theories being examined and evidence being found to support or undermine the theories. In nearly every area of science, there are areas of legitimate scientific disagreement as scientists examine and reinterpret evidence in light of one theory or another. As peer reviewed articles in indexed journals are always a reliable, attributed source of vetted viewpoints, NPOV policy allows these competing conclusions to exist on the same page.

The weight given to each view should be proportional to the published articles in high-impact journals on a subject supporting each view according to the number of citations such articles receive. Wikipedia articles should also give extra weight to more recent literature and the citations from the recent literature, when it is evident that the weight of recent academic research has largely shifted away from a view commonly held, say, 50 years ago.

For example, at one time there were academic papers published supporting a steady state theory of the universe, whereas in recent decades this has been supplanted by a big bang theory. So, an article about the origins of the universe should be based primarily in the recent literature. By taking into account also the citations from the recent literature, older original articles about the Big Bang will not be ignored.

This is an example of where an article on the origins of the universe may be appropriately weighted with an emphasis on studies published in the last twenty years rather than within the entire last 100 years.

Wikipedia editors should avoid edit warring over arguments about whether one POV or the other is the "accepted" or "mainstream" view. When dealing with scientific issues, let the published literature itself shape the weight of the article and respect the contributions of other editors.

Similarly, editors should avoid turning arguments over weight into an excuse for conflict. There is no exacting definition of proper weight that can determine precisely how much evidence for one view or another should be allowed into an article. One cannot reasonably expect editors to compile, or trust the claim of other editors that they have compiled an exact count of articles on all views, multiplied by impact factor of the journals in which they are published, divided by the average age of the articles to determine the appropriate percentage of the article to be allowed to each point of view.

The issue of giving due weight to viewpoints should first be understood to be a guideline, precisely because it cannot be exactly defined. Secondly, the more controversial the article's subject, the less likely there is going to be a unanimity of opinion even among experts. And even if the most experts agree, but a minority of experts have a large public following, clearly a large number of readers may be mostly interested in learning the views of "their experts" and there is nothing wrong with that, and it is in fact an opportunity to provide readers with more evidence from all perspectives.

Wikipedia is not bound by limits of space that limit published encyclopedia articles. Rather than waste time on edit wars over "proper weight", the Project is more rapidly advanced by collaborative editing with a spirit of generosity and openness to seeing as much material presented as possible.

If you feel one POV is being represented by too many citations, the solution is to do your research to see if there are other peer reviewed sources representing your preferred POV that have not been included. Seek to bring balance to the article by adding material, not deleting it. In this way you will help to create a more comprehensive article and more complete bibliography which will be useful to readers who come to Wikipedia looking for information (not truth) and a good start on resources they can study on their own. If a source is misrepresented, delete the misrepresentation or add material from the same source that better represents the source in a balanced fashion. If a resource is redundant, consider if it is possible to add the citation to a section where the findings of similar studies are already reported and at least try to keep the citation...as this shows some respect for the contributing editor's contribution of the citation.

Citations proclaiming "most scientists agree with this view" are never authoritative, as they would have to be backed by a reliable poll of most scientists. Such assertions should not be given undue weight.

Consensus[edit]

Assertions of a scientific consensus require reliable sourcing. It should be acknowledged when there are consensus statements from scientific bodies such as the International Astronomical Union or the International Union of Pure and Applied Chemistry. More commonly scientific consensus takes the form of agreement between multiple independent secondary and tertiary sources, such as specialized texts and textbooks.[2] The existence of consensus is the most common state of established science, established science generally does not have two or more legitimate scientific "sides" within the scientific community. Articles should simply state the consensus view as science. The articles should not elevate minor positions to artificially create counter balance. When sources demonstrate legitimate scientific disagreement then the disagreement will usually be discussed in the literature from both sides of the disagreement. In situations where the proponents of the conflict are relatively equal in proportion, no claim of consensus should be made. The article should describe the subject acknowledging the scientific disagreement.

Pseudoscience[edit]

Main pages: WP:PSCI and Pseudoscience

The statement that something is "backed by science" tends to carry great weight because it implies a tested, evidence-based validity to the idea. Such assertions are often made in marketing a product, idea, or even a belief system, even if it hasn't actually been exposed to the rigors of science.

Pseudoscience has been defined as "claims presented so that they appear [to be] scientific even though they lack supporting evidence and plausibility."[3] Wikipedia is concerned[4] about pseudoscience because it has the potential to make Wikipedia less reliable if not presented properly. Wikipedia does sometimes allow pseudoscience to be presented as a social phenomenon, if it is notable. However, Wikipedia editors must be careful to clearly separate science from pseudoscience in articles and to explain the criteria for distinction. There is broad community support to not misrepresent science by painting pseudoscience as science.

When pseudoscience may be presented, it must be presented properly. Pseudoscience should always be accompanied by the corresponding scientific view, and should not obfuscate descriptions of relevant scientific evidence or the lack of scientific evidence. All the applicable opinions about pseudoscience should be described in proportion to their prominence, which in science and medicine related discussions is weighted to the majority view among experts in the field of scientific study. In the interest of describing topics fully, articles written about notable pseudoscientific concepts may require a fair amount of detail in the article about the topic, but care should be taken to clearly attribute the views of proponents without implying that such views are actually statements of fact. In contrast, less descriptive detail (or none at all) should be included in articles that are not strictly about a pseudoscience if the putatively relevant pseudoscience in question is of little or no notability.

Not all views on pseudoscientific topics may be science-related (eg. astrology may have epistemological, historical, or cultural views to cover as well), but we shouldn't misrepresent science when mentioning pseudoscience.

Politicization of science[edit]

The politicization of science is the manipulation of science for political gain. It can occur when cultural or political views of scientists or groups of scientists influence their interpretations of scientific findings or their formulation of theories to explain or discount a body of evidence. And it can also occur when government, business, or interest groups exert legal or economic pressure to influence the findings of scientific research or the way it is disseminated, reported or interpreted.

Wikipedia is not a soapbox for pushing political agendas through selective presentation of scientific research. Further, while Wikipedia does describe disputes between notable authorities published in reliable sources, Wikipedia does not engage in disputes.

Bias should not be introduced by politically biased selection of research presented, or by the way in which it is presented or synthesized. The presentation of science in politically-charged Wikipedia articles should be such that the most reliable and best summary of the available observations, data, and theories is presented without editorializing. On the other hand, the best way to counteract a perceived bias (the introduction of material from a reliable peer reviewed article) is not to delete the reference but to (1) ensure that the reference is properly described and (2) add material from other peer reviewed sources of equal or greater merit which offset the viewpoint of the "biasing" article.

Categorization of topics[edit]

Articles on scientific topics should include standard science taxonomy in addition to any relevant folk taxonomy. In choosing category names, or which categories to place articles in, adhere to the classification provided by reliable sources on the topic.

Scientific sources[edit]

The treatment of scientific sources is very important to finding what is established mainstream science. Since wikipedia is open to everyone (and not all editors are familiar with scientific citation) it is important to explicitly describe the value of potential scientific sources. The list below is a description of how modern sources are considered. Historic contexts are not considered in this list, for example a "specialized text" was historically a common and accepted avenue for the presentation of original research. In the present day all original scientific research is expected to be initially presented in a peer-reviewed journal.

  • Peer-reviewed primary publication: In 1968 the Council of Biology Editors Newsletter proposed this definition:

    An acceptable primary scientific publication must be the first disclosure containing sufficient information to enable peers (1) to assess observations, (2) to repeat experiments, and (3) to evaluate intellectual processes; moreover, it must be susceptible to sensory perception, essentially permanent, available to the scientific community without restriction, and available for regular screening by one or more of the major recognized secondary services (e.g., currently, Biological Abstracts, Chemical Abstracts, Index Medicus, Excerpta Medica, Bibliography of Agriculture, etc., in the United States and similar facilities in other countries).[5]

Such a publication indicates the author(s)', usually trained specialists, belief and support for the presented "work". Furthermore an established editor believed the "work" warranted consideration by the scientific community. This decision is made after considering reviews from two or more established researchers in the pertinent scientific field and requesting necessary revisions from the author(s).

  • Peer-reviewed review article: Review articles are not primary sources of information. This is usually the first summation of active science put together by well established specialists and their associates. Review articles are often very technical and contain detailed inline citation to primary work. This work is usually the first "artifact" of a scientific consensus.
  • Specialized text: Similar to a review article the specialized text is usually longer and has a slightly lower density of inline references. Also written by a specialist, such "works" are usually reviewed and edited by other specialists. These "works" are often used as texts in higher level courses and as general references by other specialists. While small portions of these texts might deal with controversial science and other portions may quickly become outdated they are usually a fair representation of scientific consensus.
  • Mainstream textbooks: Their intended use is to present a general overview of a field to those not yet trained in the field. Textbooks rarely contain inline citation to primary articles however they may provide references to significant works at the ends of chapters. General textbooks are useful as a starting point for finding general ideas regarding well established mainstream science. Nonetheless, caution is advised because general, introductory texts are not necessarily written by experts in the field at hand. As such they may include erroneous or outdated material, and should be cross-checked against more rigorous sources.
  • Oral and poster presentation: These are often the preliminary and incomplete reports of findings; they are not usually subject to peer review. The more formal and complete peer-reviewed articles intended to reach a wider audience are considered the first official presentations of original research. Unless the poster or a transcript is published, an abstract for these events demonstrates no more than there was an intended presentation on the subject.
  • Patent: By a patent's very nature it does not contain well established science. The science in patents should be treated extremely skeptically because neither the description of, nor the evidence for, a patent must meet the standards of a peer-reviewed journal. Patents must instead meet legal standards that place a higher priority on the patentee's capacity to own a technology than they do on the soundness of the scientific observations. Patents often precede peer-reviewed articles for legal reasons but the peer-reviewed article is still considered the first official presentation of original research to the scientific community.
  • Institute journal: There are a variety of non-peer-reviewed journals published by think tanks, research institutes, and similar organizations. These sources are intended to appear as objective scientific publications. However, the articles from these journals are not the same as peer-reviewed sources. These sources often represent the POV of the commercial, political, or religious "entity(s)" that fund the publication and should be treated as such.
  • Popular media third-party report: Newspaper articles and television reports are generally considered poor sources for scientific fact and information. However, popular media can be expected to reliably quote individuals (though important context may be omitted). There are many assertions that aren't dealt with in mainstream peer-reviewed publications. For example accusations of pseudoscience or censorship can not be made in a peer-reviewed publication. This makes popular media one of the few open channels for such commentary and as a result it should be taken seriously.
  • Scientific third-party report: On the other hand, third-party writing that appears in technical publications are valuable for establishing the context and impact of another work, and can be expected to contain accurate science as well. Articles of this type include "Perspectives" and "News and Views" pieces which appear in scientific journals like Science and Nature, as well as newsmagazines published by scientific societies such as Chemical & Engineering News.
  • Popular media self-documentation: Self-documentation is currently not the expected or approved format for the presentation of original research. In modern science this format should be regarded as little better than self-publication when dealing with original research. As a source of established science these "works" can be considered similar to a textbook if the author is an authority on the subject.
  • Self-publication/press conference: Should not be used as evidence of anything but the authors/presenter's opinion. Such works are considered to be well outside the mainstream scientific community but are appropriate sources for clarifying the opinions of a notable person in a field. Here, notable means that the person is recognized as an expert by his or her peers as demonstrated by more than one peer reviewed journal article on the subject. Self published opinions of anyone below this level of notability should probably be deleted.

Controversies related to science[edit]

A controversy related to science need not imply an actual controversy within science. There may be an established consensus in science and the controversy is really occurring outside science. There are three broad types of situations that may arise in such cases:

  1. Uncontroversial topics; typical standard subjects on which there is broad, undisputed agreement.
  2. Non-scientific controversies; uncontroversial subjects within the scientific community that are nevertheless controversial outside it (e.g. evolution, global warming and ADHD).
  3. Scientific controversies; controversial subjects within the scientific community, e.g. Bohr-Einstein debates.

On scientific topics, Wikipedia weighs published academic discussion in scientific journals more heavily than discussions occurring outside the scientific community.

Non-scientific controversy[edit]

Topics that are controversial only outside the scientific community are distinguished by the trait that the prevailing sources on the topic do not acknowledge alternative perspectives within the scientific community. In such cases, those offering alternative examples may feel discriminated against and might wish to rectify the situation by presenting their "side" of the story in Wikipedia. Wikipedia is not an appropriate venue for this. The marginalization and ignoring of minority opinions about observable reality that occurs within the academic communities that use the scientific method should be reflected in Wikipedia.

Such topics may nevertheless be notable enough for inclusion in other areas. For example, there may be political, religious, social or cultural movements that have received significant notice outside the scientific laboratory. Articles dedicated to such controversies merit inclusion in Wikipedia, but special care should be made not to mislead the reader into thinking that the controversy has any legitimacy within the scientific community. In articles devoted to subjects which are at the fringes of legitimate research or even clearly outside it, the conventional understanding associated with the novel idea should be discussed and framed as such. Articles should not adopt an "in-universe" treatment of such a topic (a treatment from the fringe or non-scientific perspective); nor should the proponents of such topics be given a venue to promote their proposals. Wikipedia should describe an idea only if it has received notice by third parties, and the descriptions should be dispassionate and acknowledge the marginalization of the idea.

Scientific controversy[edit]

For truly scientific controversial topics, there must be acknowledgment in reliable sources that a controversy exists. This means that both sides of the argument must acknowledge that the other side exists in the venues appropriate to the scientific community. A small number of primary sources, even when published in respected peer-reviewed journals, does not indicate a legitimate controversy if others in the community have not noticed or commented on them. Critical review is the key indicator of significant controversy. When one side of the putative controversy alleges that a controversy exists while the other side (as well as any expected third-party observers) essentially ignores the papers and other allegations of those asserting the controversy, there is no way to source the existence of a real controversy without engaging in non-neutral original research. Therefore, Wikipedia should not report the existence of such a claimed controversy. In particular, any extraordinary claim that purports to revolutionize a field must be acknowledged by professional experts in the field before Wikipedia reports on its existence in articles devoted to that field.

Scientific evidence in non-scientific topics[edit]

Articles which are not about a scientific topic, such as those concerned with religious, spiritual, or philosophical ideas, are properly written plainly without argumentative rejoinders or caveats. Only if the article happens upon a subject which is unequivocally subject to scientific evidence should such evidence be presented. This should be done in a way that does not detract from the primary significance of the subject. For example, an article on the Virgin Mary discusses her role in religious thought of the various relevant traditions, not the scientific likelihood of every event that is associated with her. In the related article on the Virgin birth of Jesus, the scientific issues surrounding this article of faith should be presented dispassionately without any hint of editorial bias toward or against any particular view. Great care must be taken to present fairly the complete range of relevant scientific evidence, including opinions of scientists who may or may not share the religious or philosophical presumptions central to the subject. Wikipedia articles are not battlegrounds. The reader must be left with a knowledge of all significant perspectives on a subject, and information to pursue them further. Wikipedia does not advocate any particular perspective over any other.

See also[edit]

Notes[edit]

  1. ^ "NPOV and 'new physics'". Retrieved 2008-08-13. 
  2. ^ See Wikipedia:RS#Consensus.
  3. ^ Shermer, M. 1997. Why People Believe Weird Things: Pseudoscience, Superstition, and Other Confusions of Our Time. New York: W.H. Freeman. ISBN 9780716730903 OCLC 36260669
  4. ^ See WP:PSCI, part of the core WP:NPOV policy.
  5. ^ Ellsworth B. Cook, "Proposed Definition of A Primary Publication", Council of Biology Editors Newsletter, November 1968, pp. 1-2; reprinted in Herbert Stegemann and Barbara Gastel, "Council Classics", Science Editor, March – April 2009, Vol 32, No 2, pp. 57-8.