Regret the Error

Last fall, Times executive editor Bill Keller instructed the "Siegal Committee" — a group of reporters, editors, copy editors and one photographer — to look into a variety of issues related to increasing reader’s confidence and building the Times’ credibility. Back in November, we printed some excerpts of a memo by Allan M. Siegal (first found by Slate) that outlined some of the key questions the Committee was going to address.

That should provide you with some background before reading the Siegal Committee’s Report, which was just made public today. An accompanying article from a Times writer is also available here. There is an entire section of the report that deals with "Reducing Factual Errors" and at first glance we are pleased by some of the recommendations (most of which are in line with some of the suggestions we made in this post, not that we’re trying to take credit…) We also have to admit that we’re puzzled by one thing. In the Report’s introduction, it states:

This committee has brought together 19 highly experienced news people (11 supervisory  editors, 6 reporters, a copy editor, and a photographer), again under Al Siegal’s chairmanship.

But at the end of the report it lists 20 names for the so-called "Credibility Group." The last is "Fred Andrews, Consultant." Through a few online searches, it appears that Mr. Andrews is, or was, a consultant in "news administration" for The Times. We’re not suggesting a cover-up, but it is puzzling why someone listed as a Committee member at the end of the Report is left out of the introduction.

We’ve included some relevant excerpts from the Report below. We’ll follow up with an analysis post soon. UPDATE: Read our analysis here.

…Just as important, there is an immense amount that we can do
to improve our journalism  and cement our readers’ confidence in us. We
must be yet stricter about anonymous sources. We  must reduce the
garden-variety factual errors that corrode our believability…

…the newspaper needs to create a new set of tools — tools that
give us  flexibility in how we respond to critics; tools that help us
communicate with audiences that we currently make little or no effort
to reach; tools that make the newspaper more transparent and  credible;
and tools that give our journalists assurance that they will be
defended when they are  subjected to unfair attack…

4. Reducing Factual Errors
Recommendations

I.
Newsroom management should issue a strong statement clarifying
fact-checking  responsibilities and reiterating that avoiding error is
everybody’s responsibility. To  resolve any lingering uncertainty, it
should state that:

Writers are accountable for the
accuracy of every fact in their copy. Backfield  editors are
responsible for overall accuracy and fairness and for enforcing
standards. Copy editors should check verifiable error-prone facts as
time allows  and consult with writers about all factual changes.

Writers
should be expected to read edited copies of their stories, as well as
headlines,  captions, graphics and related elements when practicable.
All staff members have a duty  to notify a responsible editor of any
possible errors in copy, before or after publication in  print or on
the Web.

Though our training classes are making inroads, there
remains a surprisingly widespread  belief in many parts of our
operation that fact-checking is someone else’s job.

II.  A newsroom-wide corrections tracking system should be set up
to detect patterns of  errors and take action to avoid repetition. An
editor should be appointed to manage it.

We strongly
believe that the database should be used to identify patterns of errors
and  ways to avoid them, and emphatically not as a way to compile error
counts or rates by  name. The editor in charge would use the data to
analyze trends in errors, to propose  better practices and to inform
staff training.

III. We should take greater advantage of
electronic tools, both for gathering and checking  information and as
part of the production and corrections processes.

Instruction
on using electronic fact-checking tools should be encouraged for all
and made  mandatory for new staff members and newly promoted backfield
editors.

We need to find ways to encourage tapping into
the expertise of our staff. A searchable  system should be set up to
show names and telephone extensions of people with language  skills or
special expertise, whether military knowledge or an interest in opera
or in  firearms or car repair.

We should make it easier
for readers to contact us with complaints about errors. A form  or a
dialogue box on the Web should walk readers through the complaint
process; the  “Tomorrow” house ad that ordinarily appears on the
weather page should be clearer  about how to report errors in the paper.

We
should establish and diligently use an electronic system to distribute
photos and  caption information to copy desks promptly, as soon as
picture editors have made their  selections.

We should work to
develop plagiarism detection software sophisticated enough to fit our
needs and should let the staff know that we would use it not routinely
but to check any  plausible suspicion or accusation.

IV.  Nytimes.com should improve its electronic posting and archiving of corrections.

Corrections
should be posted as promptly as possible, even before they appear in
the paper. A  correction should appear in the text of the online
article, with a note appended to inform  readers of the change.

Nytimes.com
should stop its current practice of keeping outdated and possibly
inaccurate  multiple versions of news reports posted for several days.
The final New York print version,  when it becomes available, should
supersede all others.    

Discussion   

1. Error Tracking.
Last
year we published almost 3,200 corrections. We can do better. Our goal
should be to  eliminate error, beyond acknowledging it and correcting
it.

The proposed database would track the types and causes of
errors that lead to corrections.  The data would come from a mandatory
form filled out by the individual(s) responsible for an  error. It
would include a draft of the proposed correction, with an explanation
of how the mistake  happened and how it could have been avoided.

Raw
counts of an individual’s errors can be simplistic and misleading. We
are greatly  concerned that making them a routine part of evaluations
would breed resentment and undercut  the goal of having everyone
recognize that the system’s purpose is to reduce error. Only masthead
editors, department heads and the editor in charge of overseeing the
error-tracking system should  have access to names in the database.
(This is not an argument against the current and growing  practice of
discussing accuracy in written and oral performance reviews, provided
the approach is  not blindly numeric.)  The editor in charge of the
tracking system would prepare occasional memos (or intranet  postings)
about accuracy — notably including praise for good catches — and would
work with  departments to establish periodic corrections bulletins or
similar efforts at the desk level. This job  could be part of a senior
editor’s duties, or it could be given to a rotation of midlevel
editors, who  would return to their regular duties with a deeper
understanding of our best accuracy practices.   

 Report an error