Purpose and Ethics of Errordiary

Brain Food

Are we safe sharing our errors online? Isn’t this just asking for trouble?!

We’ve designed Errordiary to make it easy for people to share their errors together. This is to highlight that everyone makes errors and that they quite happen often. We also hope that sharing errors in this way will make talking about them more acceptable and raise debate about how normal errors are.

However, we have found that in designing Errordiary we’ve had to think hard about some challenging ethical issues: what if someone shared an error which led them to being sacked at work? what if someone didn’t know that their errors were being collected on a public archive? what if someone’s error ended up on the front page of a national newspaper? what if serious medical errors were shared online that disturbed patients and their friends and family? what if someone shared an error but then wanted to delete it?

In research we need to maintain high ethical standards so people that take part in our research are informed about the benefits and risks involved. This has come about because a long time ago people did some pretty controversial research with participants. For example, Milgram’s famous experiment caused psychological stress on participants who thought they had electrocuted someone in an adjacent room, and Zimbardo’s prison study involved the guards subjecting other participants to psychological abuse. Nowadays, strict ethical procedures help ensure research is not detrimental to participants and that participants are informed of the risks involved. Errordiary works on the same principles – to get participants to read and understand consent information to inform them of the purpose of the study and the main risks involved.

In our consent information we emphasise that posts should be considered IFP (Immediately and Forever Public) and that people should post on Errordiary responsibly. In fact people should post responsibly on any social media as sharing anything online can have far reaching consequences that people should know about. For example, Paul Chambers found himself in court after tweeting a joke about bombing his local airport. Also, 17 year old Paris Brown found herself in the media spotlight when she was appointed as the UK’s first youth police and crime commissioner – journalists dug up inappropriate tweets she had posted years before (between the ages of 14 and 16) and she resigned under the media pressure. These stories and others all form part of how society is learning to cope with social media, children and adults need to be educated out the risks as we move forward with this technology, which is still a growing and developing area.

In more traditional forms of research there are tight controls on data and what it is used for – the researchers are normally the only people that have access to it. For Errordiary, which is also a public engagement and citizen science endeavor we have intentionally been open about the way we have handled posts and data. By doing this we hope that people engage with the project better, learn more and get more value from it. Even though we are specific about the sorts of things that we intend to do with the data on Errordiary we cannot control for what other people will do with it. Like tweets that are publicly posted they could end up on the national news and the front page of tomorrow’s newspapers! This can happen! Beware!

When you’re using Twitter or posting on Errordiary is it more like talking to friends in a café or being interviewed by a journalist from a national newspaper? The problem is with modern social media is that one can flip from one to the other without any rhyme nor reason, and in the blink of an eye. For this reason some people are cautious and conservative about what they post. For example, one nurse advised another to tweet as if every patient was reading what she said – this made sure that she didn’t say anything she shouldn’t. What simple rules could you adopt to make sure you stay safe and out of trouble with what you post online?

One of the great innovations of Errordiary is that people can tweet errors with the hastag #errordiary, or resilience strategies with #rsdiary, and the post will be included on the website. Do we need to inform people that this is happening? How should we inform them if we decide to do this?

When we started to engage with these issues we found that people generally fall within one of two camps. Camp one believes passionately that this isn’t even an issue: because the posts are already in the public domain then there is no longer confidential and anyone can quote and use that information just as long as it is properly attributed. Camp two believes passionately that even though the data is public people have certain expectations about who is listening and how it is used. In the latter case people still should be informed and given the choice of whether to participate in any particular research endeavor. Some researchers see the easy way out of this conundrum as to just anonymise all the data – then you are not linking any insights or data to any particular person. However, this is seen as disingenuous by some and could breach their intellectual property and copyright in some cases – some people take the view that they have taken time to author and share their posts and so they need to be properly attributed when they are used. See more of this discussion here: http://www.michaelzimmer.org/2010/02/12/is-it-ethical-to-harvest-public-twitter-accounts-without-consent/

As Errordiary has developed from something a few of us have used for fun, to something that has broader engagement with different communities we’ve made the decision that everyone that has their tweet posted on Errordiary, using the hashtag #errordiary or #rsdiary, should be informed. The ideal thing would be to send that person a private message informing them of the tweet if they haven’t already registered with Errordiary. However, this cannot be done. To send a private message on Twitter the two people need to be following each other. The best we can do is to tweet a public message, with a link to further information. From here the person can choose to do nothing, opt-in or opt-out of being included in the project. Importantly, Errordiary only deals with tweets that are already in the public domain so some will see this effort as unnecessary but we think it is good practice anyway. There is a good discussion in this BMJ article about the factors and the sort of judgment call that should be made in this type of research more broadly: Ethical issues in qualitative research on internet communities.

On CHI+MED, which is a research project looking to make medical devices safer and more user-friendly, we’re interested in medical errors and how to reduce them. One of the things we wanted to try and do was to combine this with the everyday Errordiary that everyone uses. However, there was a strong reaction from some of my friends – they didn’t want to see lots of medical errors and they didn’t think others would want to see them either. They thought that people could be worried and disturbed by seeing lots of examples of medical error so we consulted with others and decided that we shouldn’t promote serious medical errors on the normal Errordiary site. This shows that it is important to keep an open dialogue with stakeholders of the site. As the project moves forward we are organizing focus groups, surveys and are opening a discussion forum to explore these issues further, for example:

  • It seems there is a culture of blame in healthcare where the sharing of errors is not encouraged. Is this right?
  • Forward thinking suggests that a more open culture of sharing and learning benefits system safety in the long term. Is this correct?
  • How can we get medical professionals to think about sharing more, and being more open? Does this actually benefit individuals and system safety?
  • How can we reconcile a more open culture in healthcare when the public would rather not know of potential risks and faults?

We also wanted people to stay in control of the information that they shared with us. Following this we have introduced accounts into Errordiary, so people can keep track and control of what they have posted. In addition people can always contact the Errordiary administrators if they are not happy with anything that is posted.

Why should we do it?

In all ethical decisions there is a balance to be had between the benefits and risks involved with the intervention. If there is no potential benefit then there is no decision to be had, because no risk is worth taking. So what is the potential benefit of Errordiary and why should be post?

Errordiary has the potential to engage and inform audiences of human error and resilience research that have not encountered this before. Serendipitously we find that Errordiary is maturing at a time:

  • where universities are encouraging researchers to engage with the general public more through public engagement for greater and broader impact;
  • where citizen science projects are actively involving members of the public in helping with researchers through collecting data and sharing research work;
  • where research councils are encouraging the people they fund to be open about the data they collect, their analysis and their resulting research papers;
  • where issues in patient safety and healthcare need mature debate and guidance on issues related to human error, safety and accident prevention.

Please see Errordiary’s Discovery Zone research area to find out more about the research that this project is enabling.

If you found this interesting, maybe you’d like to take a look at some of the other stories and articles in our Learning Zone?