2.1 Why are Research Ethics important?

Activity 5: Thinking about research ethics

What kind of research do you want to do? What might the impact on human subjects be? Think of three ethical issues that might be raised by the research you want to carry out. Share these below and review the contributions of others. Comment on someone else’s if you find something surprising or interesting – or just make your own notes.

Once you’ve finished sharing your ideas and looking at others’ contributions, head on back to section 2.1.

I’d like to research on 2 topics: crowd-source healthcare and artificial intelligence in diagnosis (I wanna focus on the process of gathering verbal information instead of focusing on a specific disease). If I was to pick a third one it would be the creation of public policy based on evidence-based medicine.

The first two topics involve a different approach to classic diagnosis, which could be flawed and lead to wrong diagnosis if they are not conducted properly. Finally public policy is usually implemented “by force” since the State is in charge of its enforcement.

Hi Jorge! Obviously in medical research there can be lots of ways in which ethics can play a part. You don’t say whether you would restrict yourself only to materials that are publicly available (e.g. data harvested from social networks) but obviously this is technically legal. The thing I think it is important to think about is whether or not we have any moral duties to abide by in the way we use open data, especially if we are combining open data sets in ways which might make it easier to individuals to be identified. I think you are right to emphasise the fact that there could even be a moral obligation to do this kind of work with open data if it can potentially relive human suffering - adding a further dimension of moral complexity…

Building on the notion of open data as a potential source of research data that Martin raised in his video, I have been wondering about the ethics of social spaces. Twitter and the blogosphere are potential sources for collecting narrative, opinion and attitudinal data. But, those spaces raise an issue about what is actually public and what is private in Internet spaces, and in particular from an ethical open research perspective.

I have been reading this paper --> http://qrj.sagepub.com/content/8/1/91.full.pdf+html

Wondering whether the team leading the course has some further insights…


I’m currently in educational research, particularly with regards to how students can assist time-strapped lectuers in producing OER, but long term I’d be very interested in studying access to scarce resources from an anthropological and economic perspective. I’ve been interested in how NGOs, community-organised ones in particular, gain, keep and control access to resources of all varieties.

This of course requires a lot of human interaction and if I chose to perform research openly on this topic, individuals would likely be fairly easy to identify even if I focused on protecting their confidentiality.

Hi David! Yes, there are now massive amounts of data out there ‘in public’ which can provide a really interesting source of research data. Arguably material on social networks is a grey area between private and public depending on the privacy settings that are used. But even if someone publishes something openly does that mean we can assume meaningful consent has been given to play with this data?

Hi Thomas! Yes, this is another area where openness can raise issues. Even if data is anonymised it can be possible to identify individuals. This is especially pronounced where different selections of ‘big data’ are brought together and cross-referenced (e.g. through location, age, etc). One of the issues we face in open research is the level of protection we need to offer participants against this; and also we need to think about how far we must imagine the future possibilities of data science as far as we meaningfully can because once the data is open, it’s open!

Our project is researching use and impact of OER. The impact on the human subjects could result in a transformed educational environment for those participating in the research and for their colleagues. Transformation is inevitably difficult, so it is essential to keep the identity of participants confidential, whether they are pro OER or not. Anonymising personal data is relatively easy, but institutions (an experts in certain fields) might be difficult to keep confidential as they might be the only ones of their type in their country. What does one do in that case?

Open data: Can one use interview or survey data collected by others and interpret it differently? How ethical is it to assume a snippet in an interview, conducted by some unknown person in a different context, means what one thinks it does?

I would be interested in researching the impact of introducing learning technologies for people who doesn’t have any formal educational background. As this research would involve people from several backgrounds and demographics, the three ethical issues I foresee are

  1. Revealing the economic background and status of individuals
  2. Though consent would be gained from volunteers, their in-depth understanding about what information they are going to share.
  3. Demographic details would reveal individual identity.

@davidp, I completely agree that the social spaces are potential spaces for collecting data. The users themselves would not be able to classify which part of their content should made available as public or private, though there are options provided by the service providers. In addition, it is bit difficult to verify the authenticity of the content available on the social spaces.

I’m interested in researching open online learning environments and the participants therein. As such, I envisage an online or virtual ethnographical methodology to be appropriate. Ethical considerations appear to revolve around the level of researcher participation, from lurker-observer to immersed participant, and what constitutes informed consent and the extent to which it is required in these environments.
I’m also interested in researching my topic from an autoethnographical perspective, but I’m somewhat stumped at this stage to identify ethical issues that this might raise.

I collaborate with researchers on a variety of projects, some of which involve human subjects. I also work on projects that use human data that was generated by previous studies (which, presumably, had to pass IRB approval at some stage). Three of the myriad ethical issues that might come up in this kind of research are:

  1. Personal data being shared (or leaked/stolen). Traceability — could you put a name to those genes?

  2. Diagnosis of problems related to the study or not. If you notice a person has a car-crash mutation that will lead to disease, do you tell them? What about a combination of genetic variations that increase a person’s risk of disease by a little bit?

  3. What about the people who will use these data after I do? If I publish a data set and someone down the line figures out how to draw new and interesting conclusions from it…including personally identifying information…that’s a potential problem.

I’m interested in researching participants in open online courses (like this one!) where it is possible to ask questions and interrogate data that are the digital footprints participants may leave in a course. Of course, if the participants are asked to complete a survey, then there is a measure of informed consent. However, as others have pointed out, there is a mass of very interesting data in open and social spaces - in an online course there are the analytics (based on how participants have moved around and used resources etc in the space) and also the records of conversations in forums, comments etc. Is it enough to state during the course enrollment that research will be conducted and even if it is, what are the rules for how to use participant data? What constitutes informed consent?

Thanks for the link to the paper.

This seems to be a question of ‘just because it’s legal, is it ethical?’. We all know by now that our activities in things like Facebook or Amazon are analysed exhaustively and that we would probably be uncomfortable if we realised just how much such companies know about us. Why should MOOCs be different - MOOCs are self evidently a public space? I guess we would feel slightly more hurt if a normally ethical researcher revealed us as unthinking, ill informed or inconsistent in an identifiable way in front of our learning peers than if a faceless giant like Google exploits us.
In the end the researchers’ self-interest (as well as ethics) is likely to guide their approach to informed consent and anonymisation. If people are published against their will, they will stop participating or being open and honest - and that will be the death knell for open discourse in general as well as for the guilty researcher in particular

Thanks Guy. I agree MOOCs(by their nature) are a public space and the discourse around MOOCs has included the ‘learning about learning’ benefits to inform, say, future course design decisions. However there is a difference between using students’ data as a mass to indicate trends and highlighting a particular student or group of students’ behaviour.

I wonder whether people will stop participating, although they may not be entirely open or honest - after all many people know that public websites and services analyse activities and we still continue to use them anyway - they are just too useful and unless they stop being so, on balance it’s a price that seems worth paying.

From the very beginning of a research project, whether conducted openly or not, the ethical use of research data and the conversations surrounding that data needs to be considered carefully and built into the planning of the project. This is especially important if the project starts out as a ‘traditional’ research project and evolves into an open research project. For example, I would like to share research which was originally conducted by email conversations (part of the seeking out sources for information and the conversations which ensued), however before sharing a complete unabridged email conversation I would need to seek permission of all those who participated in the conversation to publish online what was originally a private email. If any person involved in that conversation were to refuse permission for either their name to be included or for them to be quoted anonymously, I would have to respect that preference even if this hindered the open sharing of knowledge. After all, if I published that the source of data was from an archivist at an organisation investigating archives on my behalf, it would be fairly easy for anyone who wanted to know the identity of that person to find this out, even if the email conversation was anonymised. So in seeking to backdate the openness of a research project, it may not be possible to make some data fully open if consent concerning the identities of individuals who have assisted the research is withheld.

My area of research covers reflective practice, mostly in fine art. This area borrows from and seeps into educational theory, psychology and other disciplines.
Often it involves interviewing artists about their personal methods of working and thinking, and so there are issues of revealing and confidentiality there, and sensitive times in art tutorials and crits, etc.
As a practising artist, any research project I do within an institution highly prizes openness in research for public engagement.