Wikipedia's policy stance against original research is intended to prevent editors from inserting their own opinions into articles. Our articles should be based on reliable sources without implying any conclusions derived from improper synthesis.
Invariably, articles will quite rightly draw from more than one source. So some forms of synthesis are allowed. It can be legitimate for a single compound statement to be supported by more than one source, even in cases where the complete statement is not a rephrasing of information found in a single individual source. Doing so can help editors avoid the appearance of plagiarism while also improving the quality of our articles.
When an article contains a strong statement about the world, it is often desirable to support it with a top tier source from a recognised expert writing in a leading Journal or book published by a reputable university press. Often such sources assume the reader has a basic grounding in the subject being covered, and so will use technical terms without explaining them. In such cases it can be good practice to define the term based on a second source. Similarly, an advanced source might use a common phrase in a context that might confuse the lay reader. To illustrate, consider a case when we want to add to our article on Capital control the following information from advanced source A: "Flight taxes have attracted more opposition than any other form of capital control."
A bright but non expert reader might assume flight tax refers to a tax for travelling by air. So the sentence we add to our article might include a definition from introductory source B making it clear what fight tax means in this context. So the actual statement we add to our article might read:
Source A , + possibly Source B
The definition of flight tax might be sourced to a standard economics text book or alternatively the sentence could even remain cited just to source A as the definition could be viewed as common knowledge.
Sometimes multiple sources provide a fuller picture when taken together, such as when source A points out the reaction to a particular event in one country while source B covers the reaction to the same event in a second country. Sometimes it will be good encyclopaedic writing to combine the information from the two sources into a single sentence.
It is not always original research for an editor to make a judgment that different names used in different sources refer to the same topic. When an editor wrote the article on SM53 trams, some sources called them "SM53", others called them "Høka". They are the same thing. Recognizing that fact, and deciding to use both sources for the article can be a good editing decision.
When abundant source material is available, where the information can be neatly and encyclopaedically summarized in a different form than the way its presented in the sources, then it is not original research to do so, as long as you are not inventing any new information or misrepresenting the source material. Some time back, it was argued that we were not allowed to add Barack Obama to the List of presidents of the United States until someone published such a list with Obama on it; that idea was soundly rejected.
Trivially simple interpretations. These are usually so non-controversial that they are no more original research than routine calculations. To source "Alberta borders on Saskatchewan to the east, British Columbia to the west, the Northwest Territories to the north, and the US state of Montana to the south.", it should be enough to point to a map of Canada where the provinces are marked. Even though combining your vocabulary knowledge of compass directions along with the map of Canada to reach this conclusion is technically a kind of synthesis, this is not what Wikipedia's No Original Research policy was designed to prevent. To summarize: original research is not allowed on Wikipedia, but there is no prohibition against research.[further explanation needed]
At the same time there may be cases when an interpretation may only seem trivial. A notable case,[which?] which involved much debate in Wikipedia, is combining data from various statistical tables. The main caution is that different source may use different criteria in creating tables; they may not always be compatible, so that the combined table may be misleading.[why?]
Any wikipedia policy or guideline takes precedence over any essay. Therefore the editor who wants to combine sources should refrain from doing so, if met with objections. This may mean leaving information out or trying to find a different single source that fully supports the desired compound statement.