Research integrity and publication ethics [2010 lecture]

非常感谢你，凯瑟琳。

Thank you very much Catherine.

我非常高兴能来到这里，和大家聊聊这个话题。作为《柳叶刀》的编辑，同时也是出版伦理委员会的副主席，这个议题对我来说意义非凡。

I'm really delighted to be here and talk about this topic which is very close to my heart in various guises as The Lancet editor but also as the Vice Chair of the Committee on Publication Ethics.

今天我想跟大家简单介绍一下出版伦理委员会（COPE）是做什么的，我们如何支持编辑们，同时也分享一些想法：作者、研究者、机构、资助方、编辑和出版方，如何能更好地协作。我认为，这个领域我们还有很多工作要做。

And what I'm going to do today is talk to you, tell you a little bit about what COPE is, the Committee on Publication Ethics, what we do, how we help editors, but also give you sort of thoughts on how we all, authors, researchers, institutions, funders, editors, publishers can work better together to in this field because I think that's this is where we need to do a lot more work in.

但在深入之前，我们先来看几个基本定义。

But before I go into this, let's just look at some of the definitions.

这些概念可能大家都很熟悉，但关于研究或出版不端行为的定义，其实五花八门，差异很大。

They might be all very familiar to you, but there are lots and lots of very different definitions of research or publication misconduct, the other side of research integrity out there.

比如，美国负责处理这类问题的国家机构——研究诚信办公室，就给出了一个非常严格且狭窄的定义。

So for example, the Office of Research Integrity, which is The US based national body to look at this, has a very, very strict and narrow definition.

他们只认定三种行为：捏造数据、篡改数据和剽窃，也就是这里标红的FFP。

And that is fabrication, falsification, plagiarism only, FFP, the red, in red here.

而且，后来才补充规定，这些行为必须是故意为之。

And it has to be and that was added only later on intentional.

我不确定。

I don't know.

我觉得这确实很难说，尤其是在数据造假方面，有人会辩解说，‘我真没意识到不该把那些异常值排除在外’之类的。

I think you know that's sometimes very difficult especially in falsification people do argue, you know, oh I that I didn't know that I couldn't could leave these out couldn't leave these outliers out out of my analysis or so.

所以我不确定‘故意’这个说法有多实用，但这就是他们现在的定义。

So I I don't know how helpful this intentional is, but this is what what they define it as.

我们COPE的定义要宽泛得多。

I define or we in COPE as well define it much much much broader.

比如，挂名作者这种行为，我们也视为出版不端。

For example, even gift authorship we regard as publication misconduct.

现在大家通常把这类行为称为‘存疑的研究做法’。

Widely that is known as questionable research practices now.

虽然比造假轻一些，但仍然是科研诚信的另一面。

And it's less serious than fabrication, but it is still the other side of research integrity.

关于这类行为的发生频率，目前几乎没有可靠的数据。

There are very little data out there how often this happens.

这时候我通常会问观众：有谁见过严重的不端行为吗？

At this point, I usually ask the audience, let's see, who here has witnessed serious misconduct?

没人举手。

No one.

牛津真是个绝佳的地方。

Oxford is a fantastic place.

有谁见过可疑的研究行为吗？

Who has witnessed questionable research practices?

有三个人。

Three.

看来我们正在逐步深入这个话题。

Well, we're getting sort of more into this area.

显然这里有个趋势，这些百分比来自已发表的文献，但有人可能会质疑这是否准确。

So it's obviously there is a curve here and what the percentages refer to, this is from published literature, but one can argue whether that's correct.

造假、捏造和剽窃的比例是0.1%到1%，而可疑的研究行为则在10%到50%之间。

So falsification, fabrication, plagiarism is 0.1 to 1% and questionable research practice is 10% to 50%.

我觉得这个比例可能比50%还高，但要看你怎么定义。

I'd argue that's probably higher than the 50%, but it depends how you look.

而科研诚信则处于这个光谱的另一端。

And responsible conduct of research is at the other end here of this spectrum.

当美国的研究人员被问及这个问题时——这是发表在《自然》上的一项研究，调查了两千多名研究人员——他们是否目睹过严重的科研不端行为，我第一个问题就是关于FFP。

So when US researchers were asked and that's in a published study in Nature, over 2,000 researchers were asked whether they had observed and now this is serious research misconduct, my first question to you, so FFP.

有8.7%的人表示，他们亲眼见过或亲身经历过这种情况。

Eight point seven percent said that they had witnessed or had direct experience of this.

这个比例已经相当高了。

So that's pretty high.

那篇论文还估算过，如果按这个数据，ORI每年要处理的不是24起案件，而是两千三百多起。

And in that paper they calculate that instead of the 24 cases that the ORI deals with every year, they would have had to deal with more than 2,300 from that kind of survey.

但从这些数据里，你能感受到，关于这类行为有多普遍，目前的依据其实很不稳定。

But you get a little bit of a feel how shaky the data are out there in terms of how common is it.

其实我觉得，这并不重要。

I don't think it matters actually.

重要的是这种事情确实发生了。

It happens is all that matters.

那会带来什么后果？

What are the consequences?

这是个非常著名的案例，发生在干细胞研究领域，论文被撤回的情况到处都有。

This is a very famous case, case in stem cell research and papers get retracted all over the place.

最近《自然》杂志上还发表过一篇文章或社论，说撤稿数量正在上升。

There was a recent article or editorial even in Nature which argued retractions are going up.

这是好事还是坏事？

Is that a good thing or bad thing?

你怎么看？

What do you think?

这是好事吗？

Is it a good thing?

这是否说明现在的科研不端行为比以前更多了？

Does it show that we're having more research misconduct out there now than before?

我觉得实际上我们在发现这类问题上越来越敏锐，处理起来也越来越到位。

I'd argue actually we're probably getting better in detecting it and we're getting better in acting upon it.

至于后果，主要是对个人的影响。

Now the consequences, they are personal consequences.

会波及职业生涯和声誉。

They affect careers, reputations.

还有机构层面的后果。

They're institutional consequences.

整个机构的声誉都可能因为处理方式不当而受损。

Again, a whole institution's reputation may be tarnished depending on how they deal with it.

这也会让公众关注到某个机构在这一领域是否开展了应有的科研诚信教育和规范流程。

And it shines a spotlight on the processes and the education in this field that happens or doesn't happen at a particular institution.

甚至可能让整个国家成为焦点。

It can throw a spotlight on a whole country.

有人会觉得不公平，但事实就是如此。

Unfairly some may say, but it does.

自然界的案例中，两年前他们曾报道过一个格拉吉尔的奥地利案例，涉及《柳叶刀》撤回了一篇论文，当时在文末写了一句‘奥地利的国家状况出了问题’，这激怒了大量奥地利人，我可以告诉你。

And in nature they have written about a very a Gragier's Austrian case about two years ago which involved The Lancet where we had to retract a paper and they said at the bottom of it, something is rotten in the state of Austria, which upset an awful lot of people in Austria, can tell you that.

但你想想中国。

But just think of China.

关于中国存在的抄袭问题，相关信息非常多。

There's a lot out there on plagiarism in China.

去年一月，一个中国研究团队就有七十篇论文被撤回。

There had to be 70 papers had to be retracted last January from a particular Chinese group.

所以人们往往会用这种方式去看待整个国家。

So you know people start thinking about whole countries in this sort of way.

整个研究领域都可能因此受损。

And a whole field of research may be tarnished.

这就会牵涉到公众信任的问题。

And that goes into the area of public trust.

当然，最近的一个例子是，即使只是被指控存在研究不端行为，后来证明并不属实，也照样会拖累整个领域。

And of course the most recent example of that even when there's only alleged research misconduct which turns out to be not the case, tarnishes a whole field.

这就是气候数据的例子，今年二月《卫报》的头条说英国科学家在气候数据上发现了问题。

This is the example of the climate data where a Guardian headline in February says UK scientists hit climate data flaws.

我们现在知道，其实并没有学术不端行为。

We now know that there was no research misconduct.

不过，缪尔·拉塞尔的调查报告确实指出，他们一贯缺乏应有的开放态度。

However, the conclusions of Muir Russell's investigation did say there had been a consistent pattern of failing to display proper degree of openness.

这些科学家没有意识到，他们不够透明的做法会损害大学声誉，也会危及整个气候科学领域的公信力。

And the scientists failed to appreciate the risk their lack of transparency posed to the university and to the credibility of a whole field of research, The UK climate science field.

所以我主张预防胜于补救，当然不可能总能提前避免，但我们更应该关注如何培养科研诚信和负责任的研究行为，而不是只盯着学术不端。

So I'd argue prevention is better than cure, but you can't always prevent it of course, but we need to think about fostering research integrity and good and responsible research conduct rather than just thinking about misconduct all the time.

那么，每个人在其中该扮演什么角色？

So what is the role of all in this?

研究人员、作者、机构、资助方、编辑和出版方，各自该做些什么？

What is the role of researchers, authors, institutions, funders, editors, publishers?

这些角色之间当然有很多重叠的地方。

There's a whole of overlap of course in these people.

很多研究人员同时也兼任编辑工作。

Many researchers are also part time editors.

很多院系负责人或机构领导本身也是研究人员和作者。

Many heads of departments or institutions are also researchers and authors.

所以我们每个人身上都带着不同的角色，但这一切都得放在这个大背景下来看。

So we have different hats and different roles, but we need to think about it in all this context.

我认为，研究人员和作者的核心就在于诚实、透明和承担责任。

So researchers and authors, I think it comes all down to honesty, transparency and taking responsibility.

听起来很简单，其实没那么容易。

Sounds very easy, it's not.

但只要你坚持这三条原则，研究行为的规范性就能大大提升。

But if you adhere to these three values, then I think that is goes a long way towards having responsible conduct of research.

所以在开展研究之前，所有作者之间必须明确讨论：谁该列在作者名单里，各自承担什么角色。

So that includes before you set out to do a study, have a clear discussion among all the authors who will be an author and what is their role.

我们经常看到作者署名纠纷，往往发生在论文即将发表前，甚至发表之后。

We see so many authorship disputes, you know, right at the end at publication, before just before publication or after publication.

所以研究一开始，团队就应该坐下来，好好讨论清楚这件事。

So at the beginning of a research study, the group should sit together and have a clear discussion on this.

然后每个人都得对自己的研究工作和成果报告负责。

Then everyone should take responsibility for the conduct and the reporting of research.

有人会说，等等。

Hold it, people say then.

我没法对统计分析负责。

I can't take responsibility for the statistical analysis.

我不是统计学家。

I'm not a statistician.

我也没法对某个基因检测负责。

I can't take responsibility for a particular genetic test.

但我认为你至少得承担一部分责任。团队合作本身就意味着彼此信任，至于怎么信任，得你们自己定。

Well I think you should take some degree of responsibility and you should by having a group of people working together, it is inherent that the group trusts each other and it's up to you how you decide how to trust each other.

这个话题我们稍后再深入聊聊，因为这里头确实有很多混淆，连期刊编辑们也常有不同看法。

We can talk about that perhaps afterwards a little bit more because I think this is an area where there's a lot of confusion and where even journal editors don't always agree.

科学期刊经常说，尤其是那些论文作者超过一百人的物理或遗传学论文，认为这种做法根本行不通。

And science journals often say, especially those with you know more than a 100 authors on a paper, in a physics papers or in the genetics papers, say this is unworkable.

但另一方面，所有这些作者都以作者身份分享了这项研究成果的荣誉。

On the other hand, all of these authors take credit as authors for a particular publication.

我始终认为，你不能只享受成果的荣誉，却不承担相应的责任。

And I always argue that you cannot take credit without also taking responsibility.

研究方法和发现必须公开、完整、诚实地报告。

Open, full and honest reporting of your research methods and the findings.

不能偷工减料，也不能故意忽略异常数据。

No cutting corners, no leaving out outliers.

我们在《柳叶刀》要求所有研究，包括随机对照试验，以及其他我们存疑的研究类型，都必须提交研究方案。

We at The Lancet asked for protocols with all research, with all randomized controlled trials, but also with other types of research if we have any doubts on what we get.

有一次我收到一份方案，觉得他们肯定是弄错了。

Once I got a protocol where I thought they must have made a mistake.

他们发给我的是错误的方案。

They sent me the wrong protocol.

于是我给作者打了电话，他说：不，不，不，这个就是原始方案。

And I rang up the author and he said no, no, no this was the protocol.

因为它跟我在手头的那篇论文完全对不上。

Because it had sort of nothing to do with the paper that I had in front of me.

他说：是的，这确实是原始方案，但我们后来发现了完全不同的结果，就用另一种方式写了出来。

He said, yeah, this was the protocol, but we found something completely different and they wrote it up in a completely different way.

如果我没看到这份方案，就不会知道他们原本打算做的是完全不同的事，却把次要结果当成了主要结果来报告。

So if I had not seen this protocol, you know, I would have not known that they had actually set out to do something completely different and reported a tertiary outcome measure as if it was the primary.

关于研究局限性的坦诚讨论——作为编辑，我特别欣赏那些在讨论部分如实坦白研究缺陷的作者。

Discussion of Honest discussion of limitations, again as an editor I give a lot of credit to the authors who do go into the who do in the discussion section honestly and openly say, these are the limitations of my research.

别试图掩盖本该提及的问题，否则作为编辑，我肯定会起疑心。

No sort of saying, covering up things that you should have mentioned there because then I, as an editor, certainly I get suspicious.

要公开透明地披露利益冲突以及资助方的角色。

Open and honest reporting of conflict of interest and the role of the funder.

我觉得在这方面，COPE经常看到不少违规行为。

I think that's again, it's an area where we see a lot of transgressions at COPE.

所以一旦发现研究团队存在确凿的不端行为，作者能做的其实很有限，只能接受相应的后果。

So when there's proven misconduct or when misconduct is happening within a research group, there's not much I think authors can do other than facing the consequences.

该撤稿就撤稿，该更正就更正，该怎么做就怎么做。

So agree to a retraction, agree to a correction where it's appropriate.

现在外面还是有很多人搞不清楚这些规定。

There's a lot of confusion still out there.

到底什么时候该撤稿，什么时候该更正？

When is a retraction and when is a correction appropriate?

更正是指研究本身是成立的，只是有些地方需要修正，才能让人正确理解。

Corrections mean that the research in itself stands, but something needs to be corrected for it to be understood.

比如作者署名的问题，有人被漏掉了，可能是无意的，也可能是有意的，这种情况可以更正，因为这并不会让研究本身失效。

So for example, authorship issues, if an author is left out, could be accidentally or not so accidentally, that could be corrected because it doesn't make the research invalid.

或者如果之前没披露利益冲突，现在可以补上。

Or if a conflict of interest has not been reported that could be added.

不过这已经有点模糊了，因为如果利益冲突很严重，审稿人和编辑在知情的情况下，可能会用完全不同的视角来看这项研究。

However, that's already a slightly grey area because if there's a substantial conflict of interest perhaps reviewers and editors would have looked at the research in a different way if they had known about that beforehand.

而当研究本身无效时，就必须撤稿。

And retraction needs to be done when the research is invalid, I.

比如

E.

出现欺诈或数据造假的情况。

If there's fraud or falsification.

还有抄袭，因为这种行为必须从学术文献中清除——你不可能让两篇完全相同的研究同时存在。

But also if there's plagiarism because that's you know needs to be taken out of the literature, you would have had two studies the same.

除此之外，如果研究不是你自己的成果，却重复发表，就会导致两篇看似独立的研究被纳入元分析或系统综述，而实际上它们是重复的，冗余出版也是同样的问题。

Apart from that it's not your own work, you would have also have two studies reporting similar or the same kind of research which will be then taken into a meta analysis or a systematic review as two separate studies where there weren't and the same is for redundant publication.

那么，机构在预防方面应该扮演什么角色？

So the role, what role should institutions take in prevention?

我觉得在这方面，还有很多工作要做。

And I think again here there's a lot to be done still.

作为编辑，当我们遇到问题想联系相关机构时，常常不知道该找谁。

The first thing is often when we write as editors, when we have a case and we want to write to someone who is responsible in this institution, we don't actually know who to write to.

应该明确指定一名负责科研诚信的人员，或者设立一个专门的科研诚信办公室，而且这个部门必须处于较高的管理层级。

So there should be a clearly named responsible person for research integrity or there should be a whole office for research integrity and they should be at a high level.

我一直觉得，如果大学能建立一个集中统一的科研数据存储库，甚至定期接受审计，那会是个非常好的做法，我想知道牛津大学是不是这样做的。

Now I always think it would be a really good idea and I'd be interested to hear whether that happens in Oxford, if there is a central institutional data repository, which may even be check audited at times.

关于科研诚信的教育培训非常重要，必须包含出版伦理的内容。

Education in responsible conduct of research, think is really, really, really important and that should include publication ethics.

这种培训应该覆盖所有人，先从学生开始，接着是研究生，我认为对教职员工尤其关键。

And this should be done for all, for students to start with, but also for postgraduates and I think very important for faculty.

在各个领域，都应该有清晰明确的规范和流程，说明哪些行为是恰当的，哪些是不恰当的。

There should be strong and clear processes and guidelines what is appropriate and what isn't appropriate in various areas.

如果有人违反了这些规定，就必须有相应的后果。

And if these are not followed there should be consequence.

现在很多机构其实都已经制定了相关规范。

I think a lot of institutions now do have guidelines.

英国的大多数机构都有，美国也有相当多的机构设立了这类规范。

Most I would say in The UK and definitely a large number in The US.

你去读这些指南，确实都写得挺好，但真正需要行动的时候，却还是啥也没发生。

And if you read them, they're all really good guidelines, but then, you know, nothing still happens when it's really needed.

每个机构都应该搞一个类似医学界常说的‘死亡与并发症分析会’。

And each institution should have a sort of what we in medicine always call a mortality and morbidity review.

出了问题之后，咱们得从每个案例里吸取教训，避免将来再犯。

When something happened, know, we should learn from each case for the future.

这个情况本来能避免吗？

Could that case have been prevented?

如果能，那到底该怎么避免？

If so, how could it have been prevented?

我们的指南是不是得再严格一点？

And do we need to tighten up our guidelines?

是不是需要制定新的指南？

Do we need new guidelines?

还要针对每个具体案例，让其他人也都了解学习。

And educate others on each particular case.

我们在《柳叶刀》对所有案例都这么做。

We do that at The Lancet with all our cases.

所以但凡我们不得不撤稿，或者遇到严重问题的论文。

So any paper that we have to retract or have serious difficulties with.

目前我们大概同时有七八个这样的案例在处理，给你个概念。

We have about seven or eight ongoing at any given time just to give you an idea.

每个案例结束后，我们都会坐下来，把整个流程重新过一遍。

And each such case we then sit down afterwards and go through the whole process.

我们当时有没有可能更早发现这个问题？

Could we have spotted this earlier?

同行评审专家有没有可能发现？

Could the peer reviewers have spotted it?

我们本可以做些什么来避免这种情况？

What could we have done to avoid it possibly?

从这些案例中，我们已经做出了一些调整。

And we have from these cases made certain changes.

比如，你可能还记得约翰·萨德伯里那个案例，他是位挪威医生，在《柳叶刀》上发表了一篇论文，后来被发现完全是造假的。

For example, you might remember that case with John Sudbury, who was a Norwegian doctor who published a paper in The Lancet, which was then found to be fraudulent, was completely made up.

他录入数据库的那批人里，有一半的出生日期都一模一样，事后一看就很明显。

Half of the people he entered into some database had the same date of birth, so it was very easy to spot afterwards.

但我们其实本可以早点发现，因为他声称这些受试者来自挪威当时根本不存在的一个登记系统。

But we could have spotted this before because he claimed to have these participants from a registry which was at the time not available in Norway.

因此，我们后来制定了新规：只要某个国家有注册研究，最好找一位该国的审稿人，他们很可能提前告诉我们真相，避免论文发表后出问题。

So from that we have then made a rule that anywhere, anytime there is a registered study in a particular country, it's probably helpful to go to a reviewer within that country who could have perhaps told us that before we published the paper.

如果出现学术不端的指控，机构该怎么做？

So what should institutions do if there is alleged misconduct?

我们仍然需要指定一位负责人来处理此事。

Again, we need this named responsible person to deal with it.

而且必须开展一场公正、迅速，最好还是独立的调查。

And then we need a fair, speedy and ideally independent investigation into it.

通常情况下，机构里的负责人要么是涉事人员的密切同事，要么自己就卷入其中，所以让至少一些外部人员参与调查，会非常有帮助。

So often, you know, institutional responsible people are either close colleagues of the people involved or they involve themselves and it's very helpful to have at least some outside people to be part of this investigation.

我说的‘迅速’，是根据我的经验，这类调查通常要花六个月到两年，这根本算不上快。

I'm saying speedy in my experience, most of these investigations take anywhere between six months and two years and that's already not speedy at all.

我觉得六个月应该是一个合理的目标。

I mean, I think six months is probably something to aim for.

六个月以内完成很难，我知道，因为这还得挤在日常工作之外，而且会让人怀疑：调查真的够彻底吗？

Anything under six months is difficult, I know because it's on top of everything and it then raises a question, has this been a thorough investigation?

我们遇到过有人一周内就给出结论，我真要怀疑这种调查到底够不够认真、够不够严谨。

We've had cases where people come back within a week and tell us something and then I seriously question whether that has been a thorough and reasonable investigation into a case.

所有相关方都该得到通知，包括所有期刊编辑和资助方。

And all involved should be informed and that includes all journal editors and also funders.

资助方往往是最难被通知到的一方。

Funders are the least likely to be informed of any outcome.

这些调查和结论最好能公开，这样别人可以引用，也能作为期刊采取行动（比如撤稿）的依据。

These investigations and conclusions should ideally also be publicly available so that they can be cited and that the conclusions can be named in the reasons for any action, for example a journal takes like a retraction.

机构必须有相应的处罚措施。

An institution needs to have appropriate sanctions.

而且我再说一遍，每个案例都应当促使我们重新审视流程和防范措施。

And again, I would say any case should you know, should lead to a review of processes and the safeguards.

那么，编辑们该怎么得到支持呢？

So editors, how can editors be helped?

因为编辑们通常都非常忙。

Because editors are often very, very busy people.

他们本职是全职的研究人员或临床医生，编辑工作只是兼职，而且根本没有接受过如何以编辑身份处理这类问题的培训。

They have this editor's job as a part time on top of their job as a full time researcher or clinician And they have no experience in how to deal with these things from an editor's role.

这时候就轮到COPE出场了——出版伦理委员会。

This is where COPE comes in, the Committee on Publication Ethics.

最初，它只是1997年三四个编辑偶然聚在一起，讨论那些让他们头疼的案例，比如‘你遇到这种情况是怎么处理的？’或者‘你会怎么应对？’

And in the beginning, it was really just a sort of a club of three or four editors getting together in 1997 and start discussing cases that they found difficult like, so what have you done in a case like this or what would you do in a case like this?

就是一个非常随意的互助小组。

Very informal self help group.

但现在，我们已经走得很远了。

We've now come very far.

这三位是COPE的创始人，我称他们为我的三位智者：《柳叶刀》的理查德·霍顿、《英国医学杂志》当时的主编理查德·史密斯，还有《GUT》当时的主编迈克·法灵顿。

These are the three guys who founded COPE so to speak and I have three wise fathers, Richard Horton from The Lancet, Richard Smith who is then editor of the BMJ and Mike Farthing who was then editor of GUT.

第一年成员大概有三四十人，后来会员数量激增，完全出乎我们的意料。

Started off with about 30 to 60 members in the first year and we had this huge increase in membership which took us by surprise.

我可以告诉你，从2004年到2008年，会员数从两千增长了不少，主要原因是一些大型出版商把旗下所有期刊都加入了进来。

I can tell you from 2,000 between 2004 and 2008 and that mainly happened because large publishing companies have signed up all their journals.

比如爱思唯尔，他们把所有期刊都签进来了，一下子我们就多了将近一千八百名新成员。

So Elsevier for example has signed up all their journals and we suddenly have 1,800 more members than we had before they signed up.

而且这也不只是局限于我们最初接触的生物医学和科学期刊了，现在我们有了各种类型的期刊，从《考古学报》到《宗教学与科学杂志》，前几天我还看到一份《女性主义政治与舞蹈研究》期刊，挺有意思的——我们也在观察，这些非医学期刊遇到的问题，是不是和生物医学期刊差不多，还是各有不同。

And it also means that we have not only biomedical and science journals with which we started, we now have all sorts of journals from ACTA Archaeological to Zygon Journal of Religion and Science or what did I see the other day, Journal of Feminist Politics and the Dance Journal and it's quite interesting for us to see whether all these journals have very similar problems than biomedical journals or different ones.

到目前为止，除了涉及人体研究的伦理违规这类特殊情况，其他问题其实非常相似，比如作者署名争议、抄袭、一稿多投，几乎都是一样的。

And so far we're finding apart from exceptions such as ethics, unethical research where there are humans involved, that they're very very similar issues such as authorship, plagiarism, duplicate publication, all very very similar.

我们一年开四次会，设立一个论坛，会员编辑可以匿名提交具体案例，大家一起讨论，论坛会给出建议。

We meet four times a year in a forum where our member editors can come and bring cases which are then anonymously discussed and the forum gives advice.

我们自己不做调查，只是对提交上来的案例提供参考意见——当然，编辑们描述的情况也可能有偏差。

So we don't do any investigations ourselves, we give just advice on what is presented to us which you know we might It might be wrong what the editor presents to us.

所以我们总是把问题退回去给编辑，说：根据你提供的信息，我们会这么处理。

So we just always throw it back to the editor and say, from what you present this is what we would do.

编辑们在处理时可以引用COPE的建议，这在我的经验里常常能促成对方回应。

And then editors can cite this cope advice though in their actions, which often helps to elicit responses in my experience.

他们之后会回来告诉我们处理结果。

And they come back to us and tell us the outcomes.

这些案例都会发布在我们的网站上，你可以用关键词搜索，如果你自己遇到类似问题，这些都是很好的参考。

And these cases are published on our website and you can search them by keywords and they're quite a good source if you ever come across a case yourself that you have to deal with.

如果你访问COPE网站，就能看到类似情况下别人是怎么处理的。

If you go to the COPE website you can then see what in a similar case has been done.

大概就是这样。

So that's how it looks.

我们就像围坐在一张桌子旁。

We're sort of sitting around the table.

氛围非常轻松，如果你正好在伦敦，又感兴趣，随时都可以来旁听我们的会议。

It's very informal and if you ever happen to be in London, you're interested, you can also just come and visit that forum.

编辑们通常会带来哪些类型的问题？

What kind of cases do editors bring?

这并不是发病率统计，但根据我们上次的分析，最常见的问题是重复发表或冗余发表，我们当时还试着按问题类型做了分类。

And again, this is not, you know, an incidence rate, but the most common cases are duplicate or redundant publication when we looked last and tried to split them up by problem.

作者身份问题也非常突出，因为这类问题处理起来特别棘手。

Authorship issue is very, very high up there as well because they're difficult to deal with.

我还想特别提一下红色标注的部分：审稿人不当行为和编辑不当行为。

And, but I want to highlight also the, in red there, reviewer misconduct and editorial misconduct.

有时候，问题其实就出在编辑自己身上。

So editors themselves can be the problem.

COPE也非常重视对编辑进行培训，帮助他们以恰当的方式履职。

And COPE very much also likes to educate editors how to behave appropriately as an editor.

我们早在2004年就发布了编辑行为准则，至今仍视其为所有编辑必须遵守的基本规范。

We also have a code of conduct for editors which we launched in 2004 and that we very much regard as a basic standard that all editors should adhere to.

如果编辑违反了行为准则，且期刊自身的投诉机制已经走完，COPE甚至可以介入调查相关投诉。

And COPE can actually investigate complaints against editors if the code is not followed and the journal's own complaint mechanism has been exhausted.

到目前为止，我们大概处理了三四起案例，主要是联系编辑，指出他们在这件事上没有遵守准则，未来必须做出相应改进。

And we have done so probably in about three or four cases so far, Which mainly leads us to educate the editor and go back to the editor and say actually you have not followed the code in this instance and you need to do this and this in future.

实在没办法的时候，我们只能把会员除名，但这并不是我们的目标。

As a last resort, all we can do is throw a member out, but that's not something we're aiming to do.

我们的重点其实是帮助和引导编辑们。

We're much there to educate and help editors.

目前我们正在根据新加入的这些期刊重新审查行为准则，看看它是否适用于所有类型的期刊。

We're currently reviewing the code of conduct in light of all these new journals that I've signed up and to see whether it does apply to all types of journals.

这项工作正在进行中，我们两天后就要开会，仔细讨论这个问题。

And this is work which is just in progress and we're meeting in two days' time to look at that in great detail.

仅从50位来自生物医学不同背景的编辑反馈来看，现有的行为准则基本也适用于其他领域的期刊。

But just from eyeballing all the responses of 50 editors from very different backgrounds in biomedicine, I think the code of conduct as it stands is pretty much applicable also to journals from other fields.

我认为编辑的一项核心职责是，他们往往是最早收到关于研究或出版不端行为举报的人，因此有责任采取行动。

And I think one of the main principles for editors is because they are the first recipients of suspicions very often about studies that might involve misconduct, research misconduct as well as publication misconduct, they have a duty to take action.

过去遇到研究中有点异常的情况，编辑们常常只是直接拒稿了事。

Because what happened before is as soon as something slightly odd happened with research, editors often just rejected the paper.

所以这一点我不太确定，干脆直接拒掉吧。

And so I'm not quite sure about this, let's just reject it.

那接下来会发生什么？

But what happens next?

结果这篇稿子又跑到别的期刊那儿去发表了，级别低一点的期刊。

It gets published just in another journal somewhere down the pecking order of journals.

我觉得我们编辑是有责任的。

And I think that we have a duty as editors.

我们守护的是整个科研记录，而不只是自己这本期刊。

We are guardians of the whole of the research record, not just our particular journal.

这就是COPE所强调的：作为编辑，你有责任认真对待每一个案例，哪怕只是投稿阶段的问题，也必须跟进。

And that's where COPE comes in and says you do have a duty as an editor to take every case serious and take every case forward even when it's just in a submitted paper.

我们还制作了流程图，网站上都能找到，遇到疑问时可以参考，比如重复发表、剽窃这些常见问题，流程图会一步步问你：这种情况发生过吗？

We then developed also flowcharts and they're again a good source on the website to look at if you have doubts yourself in redundant publication, plagiarism, all the common issues where you basically can go down a sort of a chart which says, has this happened?

发生了，没发生。

Yes, no.

如果情况更复杂，就会进入下一步。

And if, then you go down further steps.

所以，简单明确的案例直接按流程图一步步处理就行，我们现在论坛里主要讨论的都是那些没那么清楚的复杂情况。

So very straightforward cases can be easily solved by just going down steps on this flowchart, which means that we now in our forum mainly discuss complex cases where it's not that straightforward.

这些流程图长这个样子。

And this is what these flowcharts look like.

它们也被翻译成了多种语言，主要是有人来找我们，问能不能也出个中文版、韩文版、日文版。

They've also been translated in various languages, mainly when people have approached us and said can we please also have it in Chinese, can we have it in Korean, can we have it in Japanese?

我们总是答应，也会做回译核对，确保准确，但我们会允许这些翻译发生。

And we always say yes, we do check with back translation that it's correct, but we allow that to happen.

在行为准则发布两年后，我们又制定了最佳实践指南，因为意识到编辑们也需要一个努力的方向——虽然不一定每个期刊都适用，尤其是小期刊，但我们听到的是，如果他们能对出版方或其他相关方说：‘我们想这么做，因为这是业界公认的最佳做法’，那会很有帮助。

Two years later after the code of conduct we then developed a best practice guidelines realizing that editors need to also have something to strive towards, perhaps not practical for all journals especially for small journals, but it would be helpful to them, that was what we heard, if they could say something to their publishers or others involved, this is what we would like to do because this is regarded as best practice.

我现在正在重新审阅这份指南，看看是否适用，或者能不能进一步扩展，让它适用于所有类型的期刊，而不仅仅是生物医学类的。

And again, I'm currently reviewing this and see whether it's applicable or whether we can make it, broaden it even further to make it applicable to all types of journals, not just biomedical journals.

COPE 还在开展其他一些工作。

There are other activities that COPE is doing.

我们有一个出版伦理审计工具。

We have a publication ethics audit tool.

会员期刊可以来找我们，想知道他们在最佳实践和行为准则方面的执行情况如何。

So that is a member journal can come to us and say, could we we would like to see how well we are doing in this scheme of best practice and code of conduct.

他们可以使用这个工具，了解自己在这一标准体系中的位置。

And then there's this tool they can actually apply and see where they sit in this spectrum.

我们每季度发布一份通讯，由一名编辑负责，内容大约六到八页，全是出版伦理方面的最新动态。

We have a newsletter, quarterly newsletter with a news editor, just six, eight pages of publication ethics news.

我们还有一个新网站，上面有博客、范例信件和案例，这个网站是2008年上线的。

We have a new website with blogs and sample letters and cases on there, which was developed in 2008.

目前我们正在对网站进行更新。

And we are currently reviewing the website.

预计2011年，我们将会推出一个全新的网站。

So there will be yet another new website coming in 2011 hopefully.

我们每年都会举办研讨会，是非常好的学习资源。

We have annual seminars which are very good educational source.

去年的主题主要是抄袭问题。

Last year's was largely on plagiarism.

每次活动通常都会定一个主题。

They often have a topic.

活动主要在伦敦举办，但从去年开始，我们每年也在华盛顿办一场，时间定在十一月底，这是在美国举办的第二场。

They're held in London, but we also have since last year one held in Washington each year and the Washington one is in at the end of November, which is the second one in The US.

我们还计划在西海岸再办一场美国地区的研讨会。

We're aiming to have a further US based seminar in the West Coast as well.

我们制定了撤稿指南，因为发现编辑们对撤稿这件事总是很紧张。

We've developed retraction guidelines because we found that editors feel very nervous about retractions.

这些指南得到了编辑们的广泛欢迎，毕竟撤稿从来都不是件容易的事。

So these have been very much welcomed by editors because often you know retracting a paper is never an easy thing.

但必须有明确的规范来指导怎么做。

But there need to be clear guidelines how to do this.

比如，必须说明撤稿的原因，任何被撤稿的论文都得清楚地标明状态。

For example, you need to state the reason for retraction, any paper out there needs to be clearly marked as retracted.

有些编辑以为只要把文章从网站上删掉就算撤稿了，但实际上根本不是这么回事。

Some editors have felt that they can just take it down from a website and that equals retraction, which of course it doesn't.

你不能随便把一篇在线文章删掉就当它没发表过，既然发在网上，那就等于已经出版了。

That's just you cannot just take a paper down from somewhere and when it's online it means it is published.

所以，处理这种情况也必须严格遵守撤稿的规范。

So you need to follow the proper retraction guidelines in that as well.

编辑们很担心，问如果我们撤稿，作者不同意，我们会被告吗？

Editors were nervous and said can we be sued by authors if they don't agree with the retraction?

确实，作者并不总是同意撤稿，但作为编辑，你有责任采取行动，即使作者反对，你依然可以撤稿。

Because it does happen, authors do not always agree, but you as editor have, you have a duty to still act and you can definitely retract a paper even when authors don't agree.

我们的律师说，只要按照正确流程操作，编辑根本不会面临被起诉的风险。

And our lawyers tell us that there's no ground tool for, to sue editors when they follow the correct process.

COPE 与科研诚信办公室、欧洲科学基金会、科学编辑理事会、英国科研诚信办公室等机构合作，共同处理出版伦理相关问题。

COPE collaborates with other organizations like the Office of Research Integrity, the European Science Foundation, the Council of Science Editors, The UK Research Integrity Office and others on issues of publication ethics.

这是我们的通讯简报。

And this is our newsletter.

我们的网站长这样。

Our website looks like this.

网站上除了新闻，还有一个博客。

And on the website we also have news, but also this blog.

挺有意思的，每周都会汇总所有关于出版伦理和研究诚信的文章。

It's quite interesting because it picks up on a weekly basis anything written on Publication Ethics or Research Integrity.

从今年八月开始，我们还开了个Facebook页面，方便用那里的用户关注。

We started since August this year to also have a Facebook site for those who are using that.

不过，可能主要是年轻一点的人会去关注Facebook页面。

Although, it's probably sort of the slightly younger generation that gets into the Facebook site.

在座的有谁用Facebook吗？

Who uses Facebook here?

好吧，我收回刚才的话。

Okay, I'll take it back.

我们也在Twitter上更新，欢迎大家关注。

And we also, you can also follow us on Twitter.

在2000年以前，我们主要是个自助小组，但从那以后，逐渐变成了一个有选举产生的官员和理事会的正式组织。

Whereas we were the self help group very much until 2000, we then became sort of a more constitutional body with elected officers and a council.

自2008年起，COPE正式成为一家英国慈善机构。

And since 2008, COPE has become a UK based charity.

我们现在有了一名全职运营经理、一名行政人员，以及一名兼职网页管理员，毕竟随着规模扩大，加上六千五百多种期刊，工作量实在太大了。

We have a full time operations manager now and an administrator and a part time web administrator because with all these expansion and 6,500 journals, we have a lot of work to do.

我目前每周花大约一天时间在COPE上，这已经太多了。

And I'm spending currently about a day, a week on COPE, which is too much.

目前COPE的主席是莉兹·瓦格纳。

Liz Wager is the current chair of COPE.

秘书是来自Cross Medicine的吉妮·巴伯。

The secretary is Ginny Barber from Cross Medicine.

克里斯·格拉夫来自威立·布莱克威尔，目前担任代理司库。

Chris Graff from Blackwell Wiley is acting treasurer.

我们目前正在全力开发一套针对编辑的在线学习模块，一共包含十个主题，涵盖抄袭、数据造假、重复发表等等。

What we're working on very hard at the moment is a e learning module for editors with 10 different modules on plagiarism, fabrication, redundant publication and so on.

我们计划在2011年年初，也就是一月或二月，率先推出第一个模块。

First, hoping to release the first module in early twenty eleven, in January or February 2011.

我们正在讨论是否应该增设在线快速响应机制，毕竟我们每三个月才开一次会，有时候大家需要更及时的建议。

We're discussing whether we should have rapid online responses because every three months we meet only and sometimes people need advice in a more timely fashion.

不过我们也有点担心，眼下可能没足够的时间来落实这件事。

We're just slightly concerned that we wouldn't have enough time actually to do that in the meantime.

但目前我们正在认真考虑，或许可以把这个功能加进去。

But we're in serious discussion to perhaps add that as an option.

我们还在考虑在美国其他地区设立四个站点。

We're also thinking about having four and other locations in The U.

美。

S.

国。

A.

和英。

And The U.

K.

K.

我们还在新加坡办过一场活动，当时正好赶上第二届世界科研诚信大会，反响非常好。

And we did one in Singapore in connection with the Second World Research Integrity Conference that was happening there in July and it was very well received.

我们其实也在为出版商制定一份行为准则，因为出版商已经让旗下所有期刊都签署了承诺，还说：‘我们做得不错吧？’

We're actually also working on a code of conduct for publishers because the publishers have signed up all their journals and they say, oh, haven't we done well?

我们现在也讲究伦理了。

We are now also ethical.

我们已经让所有期刊都签了字。

We've signed up all our journals.

但其实我们还没明确说明出版商该怎么做。

But actually we haven't got anything to say how publishers should behave.

还有很多重要方面，比如编辑独立性，这些都需要在出版商的行为准则里明确规范。

And there's a whole big area in terms of, for example, editorial independence, you know, that needs to be dealt with in a code of conduct for publishers.

我们还特别希望发布一份关于机构、资助方和期刊之间协作的指导文件，这个我们已经开始起草了。

We also would very much like to issue and we kind of started drafting this guidance on cooperation between institutions, funders and journals.

很多问题就出在这里，编辑们本意很好，也想采取行动，但缺乏机构的支持，或者机构根本不作为，又或者拖得太久。

Because this is where a lot of difficulties arise, where editors have all the good intentions and they want to act but they actually haven't got the institutions buy in or the institutions don't do anything or they take far too long.

我觉得我们也非常需要听听来自各个机构的人有什么想法。

And this is something I think which we would very much welcome thoughts also from people in institutions.

我之前提到过新加坡的第二届世界研究诚信大会。

I've already mentioned the Singapore's Second World Research Integrity Conference.

借着这次大会的机会，我们和其他人一起制定了作者和编辑的国际标准。

In context with this conference, we have worked with others to develop international standards for authors and international standards for editors.

作者标准主要反映了编辑们希望作者怎么做、不怎么做。

And the author standards very much what from the point of view what editors expect authors to do or not to do.

这些标准将在2011年3月以《新加坡会议论文集》的形式发布。

And these will be published in March 2011 as the Singapore proceedings.

不过我可以先给你透露一下它们大致是什么样子。

But I give you a little kind of sneak preview of what they look like.

我们强调，编辑要对所发表的每一篇文章负责。

So we're saying editors are accountable and should take responsibility for everything they publish.

是的，这本来是不言而喻的，但现实是，并不是每个人都这么做，也不是每个编辑都认同这一点。

Yes, that goes without saying, but you know, this this isn't isn't followed by everyone or by every editor or not even every editor agrees with that.

编辑应当做出公正、无偏见的决定，不受商业因素影响，并确保同行评审过程公平合理。

Editors should make fair and unbiased decisions independent from commercial considerations and ensure fair and appropriate peer review process.

编辑政策应当做到最大程度的透明。

Editorial policies should be having maximum transparency.

他们应当推动最大程度的透明，以及完整、诚实的成果报告。

They should encourage maximum transparency and complete and honest reporting.

编辑还应通过发布更正和撤稿来维护学术诚信。

And editors should guard integrity by issuing corrections and retractions.

我们还特别谈到编辑如何处理审稿人和编辑自身的不当行为，以及如何应对自身的利益冲突等问题。

And we talk very much also about editors pursuing reviewer and editorial misconduct, dealing with their own conflicts of interest and so on.

这只不过是八页文件的一个简要总结。

This is just a summary of a sort of eight page document.

同样地，我们在新加坡也制定了针对作者的国际标准，要求作者以合乎伦理、负责任的方式报告研究，清晰而诚实地呈现结果——这和我刚才说的差不多。

And similarly, we have this international standards for authors that we've issued there in Singapore where we ask that authors should report everything in an ethical and responsible manner, should present their results clearly and honestly, a little bit what I've already said.

研究方法必须清晰明确地符合出版要求，并且作者应共同承担责任。

Methods clearly and unambiguously should adhere to publication requirements and they should take collective responsibility.

我很想听听你对这个问题的看法。

I'd be really interested in hearing your thoughts on that.

你觉得我们作为作者，该怎么真正落实这种共同责任呢？

Whether you feel where does how can we grasp this nettle of collective responsibility as authors?

研究论文的作者署名必须准确反映每个人对工作的实际贡献。

And the authorship of research publications should accurately reflect individuals' contributions to the work.

很多期刊会在论文末尾附上作者贡献说明，我强烈建议新编辑和新期刊也采用这种方式。

A lot of journals have contributor statements at the end of a published paper and I very much recommend that to new editors or journals.

但并不是所有期刊都这么做。

But not everybody does.

有些国家有个不成文的规定，不管有没有参与研究，部门主管都必须列在作者名单里。

And there are particular countries where it's just de rigueur that the head of department always has to be an author on each paper whether they have anything to do with it at all or not.

资金来源和相关利益冲突必须如实披露。

Funding sources and relevant conflicts of interest should be disclosed.

那么，编辑们在处理这类问题时，通常会遇到哪些困难呢？

Now what do editors have difficulties with and when they try and resolve these cases?

这非常耗时，尤其是当你已经身兼三份其他工作，编辑只是额外的负担。

It's time consuming, especially if being an editor is your job on the top of three other jobs already.

当你质问作者时，他们往往根本不回你。

You get no reply from authors when you challenge them.

如果你需要进一步联系他们所在的机构，机构负责人也常常置之不理。

When you need to go further and go to the institutions there's no reply often from the head of institutions.

这时候，COPE就能帮上忙了。只要你写信时提到‘这个问题已经提交给出版伦理委员会讨论过’，听起来就像个权威老机构，有分量、有威慑力——虽然我们实际上没什么实权，但这么一说，作者们往往就更愿意配合了。

And this is where COPE does help because if you can say this has been discussed at the Committee on Publication Ethics and you write that in your letter and it sounds like a grand old committee, you know, with a lot of teeth where we don't actually have teeth, but it just sounds like that, you as an editor find you often get a better response.

即使最终开展了调查，结果也可能不够彻底。

Then even if an investigation is done, it might be inadequate.

有时候连相关机构都不存在，那你真是束手无策，根本找不到该找谁。

There may be no institution, then you're stuck, you don't know who to go to.

我们曾经有个案子，最后实在没办法，只好放弃了。

And we had once a case where we finally gave up.

我们努力追查了好几年，但那个作者本身就是一家私人机构的负责人，而他妻子正好是这家机构的经理，说白了，我们已经找不到还能找谁了。

We tried to pursue it for years, but it was like the author was the head of a private institution where his wife was the manager of that institution and you know, it's sort of you've run out of places to go to basically.

我们确实会偶尔要求作者提供原始数据供我们核查，但后来我们渐渐不再这么做了，因为最后堆得满屋子都是数据文件，根本处理不过来。

We do say that we occasionally may ask for authors to provide their original data for us to look at, but we've moved away a little bit from that because we we then end up having, you know, boxes and boxes and boxes full of data.

作为编辑，我们拿到这些数据之后到底该怎么做呢？

And what what as editors do we do with that?

所以我们干脆把问题推回给机构，让他们自己去调查，查清楚了再告诉我们结果。

So we throw that back at the institution and say, you do your investigation and then you let us know what happened.

如果涉嫌学术不端但无法证实，该怎么办？

What to do if alleged misconduct is unproven?

《柳叶刀》曾经有个著名案例，当时我们问作者：你的原始数据在哪？

The Lancet, we always had this very famous case where the author came back to us when we said, where are your data?

他回答说：哎呀真抱歉，数据都被白蚁吃掉了。

Oh, I'm so sorry they've been eaten by white ants or termites.

在那个国家，这种说法听起来还挺合理，可几年后，同样的事发生在《英国医学杂志》身上，他又说了完全一样的话。

In his country that was entirely plausible until he said exactly the same a few years later to the BMJ when something similar happened.

所以那里白蚁实在太多了。

So there were a few too many white ants there.

于是我们决定，在这种情况下，发布一份关注声明，明确说明我们的疑虑，并完整交代事情经过，包括作者的解释。

And we decided that we would, in that case, what we do is issue an expression of concern, stating exactly what our concern is and explaining exactly what happened including the explanation of the author.

那以后对作者我们该怎么处理？

And what do we do with authors in the future?

有些期刊会禁止作者三年、五年，甚至永久禁稿。

Some journals ban authors for three years, five years, forever.

COPE并不建议这么做，因为实际操作起来太难了。

COPE doesn't recommend anything like that because it's very difficult to do that.

但说实话，作为编辑，我们的记性都特别好。

But in practice I have to say as editors we just have very long memories.

那么编辑该如何预防研究或出版不端行为呢？

So how can editors prevent research or publication misconduct?

这更难了，我觉得我们能施加的影响其实很有限。

That's more difficult and we only have indirect influence I'd say.

我们可以推动使用像CONSORT或STROBE这样的报告规范。

We can try and promote reporting standards like CONSORD or STROBE.

正如我之前说的，我们通过要求提交研究方案、伦理审批文件或试验注册信息，来倡导诚实与透明。

We promote honesty and transparency by asking for protocols as I've explained or asking for ethics approval forms or say trial registration.

作者贡献声明能有效促使作者认真思考作者署名的问题。

Contributor statement goes a long way in making at least the authors think about authorship.

他们总不能在最后写上‘某某对这项研究毫无贡献’吧。

They can't very well put at the end such and such had no role at all in any of the study.

我们可以就出版伦理问题撰写社论和评论文章。

We can write editorials and commentaries on issues of publication ethics.

接下来我们就谈到筛查环节了。

Then we get into the area of screening.

现在我们可以在两个方面筛查不端行为：一是用CrossCheck之类的工具检测抄袭或重复发表。

We can now screen for misconduct in two instances and that is for plagiarism or redundant publication using crosscheck or similar software.

我们刚在《柳叶刀》开始这么做，但只针对部分论文，不是全部。

And we've just started doing that at The Lancet for only some papers, not all.