Home  >>  Announcement

This article appeared in STB-61, May 2001.

The Stata Journal begins publication fourth quarter 2001

by H. Joseph Newton

After ten years of continuous publication, the Stata Technical Bulletin is converting to the Stata Journal. The Stata Journal, Volume 1, Issue 1, will begin publication fourth quarter 2001. You should be getting your first copy around October. This issue, the 61st issue of the Stata Technical Bulletin and the first issue of its eleventh year, will be the last issue of the Bulletin.

How the Journal will differ from the Bulletin:

  • whereas the Bulletin was published bimonthly, the Journal will be quarterly;

  • whereas the Bulletin was printed black-on-white on 8 1/2 by 11 inch paper, stapled, the Journal will be standard journal size (7 1/4 by 9 1/4), bound, and with a color cover;

  • whereas the Bulletin included only unreviewed inserts, the Journal will include reviewed articles together with columns;

  • and whereas the Bulletin was always professional, the Journal will be even more professional, understated, and yes, academic.

I am very excited about this development and, once you see it, I know you will be, too.

As editor of the Journal, I will still be seeking articles from you, but I can now be less restrictive on what kinds of articles are appropriate. Programs with documentation will always be welcome but so will other kinds of articles, especially expository articles.

For those who want to follow along as the Journal develops, you can visit the website http://www.stata-journal.com. That website will become the official website of the new Journal and, among other things, will provide guidelines for authors together with subscription information. All STB subscribers will automatically have their subscriptions rolled over to Stata Journal subscriptions; see an76 for details.

I think the best way to understand the new Journal is to understand what was good and not so good about the Bulletin.


For the past ten years, the STB has served as a means of distributing new commands and Stata updates, both user-written and "official". When the STB began under the editorship of Joseph Hilbe, there was no Internet or, at least, no Internet that was used to distribute software and updates, and, for its first three years, the STB grew, fulfilling this need. The Stata software was, from the beginning, user programmable, and the STB became the primary way users distributed to other users what they had done. Users back then subscribed to the STB with diskettes, and these diskettes contained the updates.

For the next three years the STB continued under the editorship of Sean Becketti, growing and prospering even as the Internet grew and prospered.

For the past five years, the STB has continued under my editorship.

In the last five years, however, the growth of the Internet along with the growth in both the number and the diversity of Stata users has both led me and forced me to gradually introduce changes in the STB. In particular, the Internet, Stata's website, and the Statalist listserver (statalist@hsphsun2.harvard.edu) now allow instant communication among users and, moreover, improvements to Stata software actually allow it to search the Internet for desired statistical capabilities — whether written by StataCorp or by users — and instantly to install what it finds.

The STB is no longer the primary vehicle for distributing user-written programs and, with Stata 6, StataCorp itself stopped distributing official updates through the STB.

Meanwhile, with the growth of Stata, the number of application areas that Stata users are drawn from has also grown. The user base now has wider variation in statistical experience and techniques used.

Over the past five years, STB "inserts" have become less announcements and short articles describing user-written programs and more longer articles describing complicated programs as well as more general articles about how Stata can be used to analyze interesting datasets.

This change is reflected in the fact that, between Volume 5 and Volume 9, the number of pages increased by 51% (238 to 360) while the number of articles remained almost the same (55 to 59).

In short, not only Stata but also the environment in which Stata users operate has changed enormously over the decade, 1991 to 2001, that the STB has been appearing. It is clear that we should take stock and consider the best direction for our journal.

The editors of the Stata Technical Bulletin (STB) met in Boston during the North American User Group Meeting in March to discuss the past and future of the STB. (Two editors were unable to attend: Patrick Royston and Joanne Garrett but contributed to discussions by email.) Also attending was William Gould, president of StataCorp.

We came to the unanimous conclusion that the STB itself needed to change and that the changes needed to be substantial enough that a name change was also warranted.

When the STB began, timeliness was of primary importance. Nowadays, printed matter cannot compete with the Internet in that respect. Printed material, if it is to compete, outscores because it is more considered, more substantial, and more trustworthy. Because we want to focus on this, we came to the conclusion that the Journal should be printed less often (4 times per year rather than 6), allowing us to have more time to have articles reviewed, and so making the articles even more considered, substantial, and trustworthy.

We also knew that we needed to change the emphasis of the articles. As I mentioned, programs with documentation will always be welcome, but we believe that users want and need more expository articles about statistics and using Stata, rather than about Stata as a program.

From there, one thing led to another and, by the end of the meeting, we had more or less designed the Stata Journal. There were and still are many details to be worked out. Nicholas Cox of the University of Durham has agreed to become an editor of the Journal, and he, along with the other Associate Editors of the STB, Kit Baum of Boston College, Joanne M. Garrett of the University of North Carolina, Marcello Pagano of the Harvard School of Public Health, J. Patrick Royston of the UK Medical Research Council Clinical Trials Unit, and Jeroen Weesie of Utrecht University in the Netherlands, will be working with me to design the Journal. I may soon be calling on others to assist.

If you have any thoughts, desires, or comments, please contact us during this design period. You can send email to editors@stata-journal.com, and Nick and I will be sure to read it immediately and to pass along your comments to the other editors as appropriate.

Let me now summarize what we now plan for the Stata Journal.

The Stata Journal

The Stata Journal will potentially include the following types of content:

  • A major focus on expository articles about statistics, especially related to (new) Stata commands, e.g., introductions to cluster analysis using Stata (not necessarily about Stata), frailty modeling using Stata, generalized estimating equations, and so on.

  • Articles about using Stata, e.g,, managing multi-level data in Stata or Stata programming.

  • The Journal will have a number of interesting columns, including a column by Nicholas J. Cox on effective use of the Stata language, both interactive and programming, the best of the FAQs (frequently asked questions) mounted under User Support on the StataCorp website, and so on.

  • There will be book reviews, usually about using Stata or about Stata, but occasionally books in areas deemed important or of interest to Stata users.

  • The Journal will still publish STB-type articles describing new commands by users along with unofficial commands written by StataCorp employees, but we will be asking that such articles contain more expository content.

Articles will be reviewed, which should lead to even better articles than those in the current STB. Naturally, this will be especially important to authors from academia, who are judged on their publication record in reviewed journals, but we intend for all readers to benefit. Reviewers will be attracted from the editorial board, previous authors of the Stata Journal and the STB, and experts from various fields of applied statistics.

The Journal will be published quarterly. We are currently designing the appearance of the Stata Journal, but it will have a glossy, color cover, have the same width and height as the Stata manuals, and be bound. The appearance of the articles will be similar to those in many scholarly journals. We are currently in the process of creating a style sheet for authors based on LaTeX and that will be made available on the http://www.stata-journal.com website (although both ASCII and Word contributions will still be accepted).

Like the STB, the Stata Journal is intended to be for all Stata users, both novice and experienced, with various levels of expertise in statistics, research design, data management, graphics, reporting of results, and of Stata in particular. Each issue should have something of interest to all users. In addition, the Journal will attempt to attract new readers by demonstrating that their research interests are met well by Stata. We aim to disseminate excellent expository material on applied statistics to all researchers and students interested in statistics.

The numerous daily postings on Statalist illustrate very well the readership we have in mind, as those who follow it will appreciate. As with many listservers, the style and content of Statalist discussions have evolved as an expression of members' interests and expertise. Statalist is centered on, but in no sense limited to, Stata users. Members' questions and answers range back and forth through specifics on using Stata to general questions on data management; statistical data analysis and modeling; and what is and is not good practice, statistically, computationally and scientifically. Statalist is widely appreciated, not just as a relatively rapid and effective way of solving Stata problems, but also as a source of wisdom on statistical matters in the widest sense. It is this mix which we seek to emulate, although with more substantial and more durable contributions, in the Stata Journal.

Editorial board and more information

I will be the first Editor of the Stata Journal and Nicholas J. Cox of the University of Durham will be joining me as an Editor. We both can be reached by emailing editors@stata-journal.com.

The current editorial board of the STB will act as the inaugural associate editors of the new journal. They are Kit Baum of Boston College, Joanne M. Garrett of the University of North Carolina, Marcello Pagano of Harvard School of Public Health, J. Patrick Royston of the UK Medical Research Council Clinical Trials Unit, and Jeroen Weesie of Utrecht University in the Netherlands.