The articles in Wiley StatsRef were carefully selected from Wiley’s portfolio of online reference works by a team of expert editors: N. Balakrishnan, Ted Colton, Brian Everitt, Walter Piegorsch, Fabrizio Ruggeri and Jef Teugels. Thousands of new cross-references and links to related articles were added to hugely enhance discoverability and ease of navigation. Following launch, Wiley StatsRef: Statistics Reference Online will be updated with new content several times a year.

Statistics Views spoke to four of the current editors, Geert Molenberghs, Fabrizio Ruggeri, Walter Piegorsch and Paolo Brandimarte, to learn a little more about Wiley StatsRef: Statistics Reference Online.

Can you describe Wiley StatsRef?

Geert Molenberghs (GM): It is Wiley’s premier online statistical reference tool, a mega-encyclopaedia of statistics.

Fabrizio Ruggeri (FR): It should become more and more the place where statisticians, scientists from other disciplines, practitioners should find the answers to their statistical quests, either directly or through pointers to adequate literature. Wiley StatsRef has the guarantee of a peer-review process, based on the cooperation of leading statisticians (both as authors and referees).

Walter Piegorsch (WP): A new, major reference work (MRW) by John Wiley & Sons that brings together important concepts in statistics and data sciences in a coordinated, expert-edited, accessible, interconnected fashion.

Paolo Brandimarte (PB): In my view, Wiley StatsRef is the first place where advanced students, practitioners, and researchers should look when they need a clue on some unfamiliar topic. I see it as a sort of Wikipedia, but a more “authoritative” one, since there is a better check of contents and an editors’ plan. A lot of cross-referencing, as well as a good essential bibliography for each topic, are essential features.

Can you explain how you became involved in the StatsRef project?

GM: I was recruited by Kathryn Sharples to keep an eye on biostatistics and survey methodology.

FR: I was one of the EIC of Wiley’s Encyclopedia of Statistics in Quality and Reliability.

WP: I was an editor-in-chief of one of Wiley’s recent MRWs (the Encyclopedia of Environmetrics, 2nd edition) and had also served as a Section Editor on another MRW (the Encyclopedia of Quantitative Risk Analysis and Assessment) and so was very familiar with: (a) how we built these sorts of MRWs; and (b) a nontrivial percentage of the material that the larger collection contained. I was honored when Wiley asked me to use that knowledge to help build Wiley StatsRef.

PB: I have published five books with Wiley (the first one in 2001, and the sixth one is on its way). Then, my editor (Steve Quigley) decided that I could be involved and asked me about it.

What were your main objectives during the process?

GM: The process is still ongoing. I guess it is important to ensure that all material is up-to-date and that new, emerging areas are properly covered.

FR: Provide a structure of the work (a sort of tree-based classification, but with some secondary classification), identify possible overlaps between papers from different works, add links across the the encyclopedias and think about papers in major need of update (new papers are for the future).

Who should read Wiley StatsRef? Anyone who has a question about an issue in data science and statistics!

WP: To bring together the many, sometimes disparate entries already existing in the large body of MRWs in data science that Wiley had amassed, and to find those that: (a) were most modern/up-to-date; and (b) brought the greatest focus to the topic(s) that they addressed. And of course, to place the final collection in a single, accessible location for the user.

PB: I was not involved from the beginning, as I replaced a former editor (Jef Teugels). At present my objective is to include relevant material for finance (currently it is heavily oriented toward actuarial science), as well as data mining/learning/statistics tools for business applications. Business analytics is a huge and hot topic.

What are the key features that make Wiley StatsRef even more useful than Wiley’s existing online reference work portfolio?

GM: The fact that it consolidates the existing sources and further expands them.

WP: (a) Obviously its single, integrated location on Wiley Online Library. Note that the extensive cross-links that have been built into the project exceed their counterparts in each separate encyclopedia. A cross-link from “survey sampling” to “systematic sampling” may not exist in the Encyclopedia of Environmetrics, since the second entry may only be a subcomponent of the first. But with over 6000 entries in Wiley StatsRef, the cross-links are highly extensive and literally “cross” the individual boundaries of the original works. (I think they are the most valuable feature of the product as a whole); and (b) we edited and updated (and are continuing to update) many of the older entries to bring them into 21st century practice.

FR: It should become an “all-you-wanted-to-know-about-statistics” work, enlarging the scope of the original Encyclopedia of Statistical Sciences, adding more specialised papers coming from the fields of the seven other ones.

PB: The mix of general/methodological articles and application specific ones. Readers may appreciate the ideas that are used in another application field, which is not the case in a domain-specific online reference work.

I think that we should enable readers to comment on papers, in order to collect their needs. Readers make our best reviewers, and should help us in addressing missing parts in the general body of knowledge.

I also think that offering dataset repositories and R scripts illustrating applications would be great.

How do you decide on which material to include from Wiley’s existing online reference work portfolio?

WP: The six founding editors looked at every entry from the entire collection of Wiley data science MRWs and decided which ones: (a) were out-of-date and could be discarded; (b) were overlaps (and so took the best one of the lot); and (c) which were critical and important enough to stand as entries in the new, single collection.

GM: Its relevance and breadth for the field. It is also important that the collection is rich enough to facilitate cross-referencing.

PB: Actually, I am more interested in new material in finance and business analytics, but we should leverage what is already there and connects well with new material.

How often will Wiley StatsRef be updated?

WP: I believe it’s an ongoing process, with around 350 new and updated entries to be contracted every year.

Who should read Wiley StatsRef?

GM: Everyone looking for the essence on a topic, presented in a state-of-the-art fashion. In other words, every researcher and user of statistics!

FR: Statisticians, scientists from other disciplines, practitioners.

WP: Anyone who has a question about an issue in data science and statistics! From, “what’s a standard deviation” to “how do I fit a nonparametric regression model in more than one dimension”? Some answers will be short and serve primarily as pointers to other sources (either other entries in Wiley StatsRef, or other sources in Online Library, or elsewhere). Others will give a fair amount of detail while also providing the same pointers for internal and external further reading.

PB: It is important to make the material relevant for advanced students and practitioners. It is essential that papers are written to educate the reader. On the contrary, most journal papers are written to impress and show technical prowess. As has been pointed out by other people, they are written much like a PhD thesis. Wiley StatsRef will strive for usefulness beyond academia.