Applying for Ethics Clearance for a Study or Experiment
The purpose of the ethics clearance process is to ensure that any study using human subjects does not violate the rights of the subjects or endanger them or their livelihoods in any way. For example, if a study publishes data which clearly reveals the identity of individuals, this information may be used by those in authority to take disciplinary action against the participant. However, by following a sensible anonymization procedure, this issue can be mitigated or eliminated.
Note: if you plan to use UCT staff or students as part of your study, you need to apply for ethics clearance first. Once you have ethics clearance you must complete additional forms in order to be granted access to staff or students at UCT. The form contains all the relevant contact details.
The instructions on how to complete each form are embedded in the relevant documents. Please note the following:
- All students undertaking a study involving human subjects must complete the Faculty of Science Researcher Ethics Statement (downloadable as a Word Document).
- Computer science studies which involve simple evaluation of system interfaces or usability can be passed through an expedited process; the ethics clearance form explains what kinds of studies can use this route.
- In all cases you need to provide enough detail in your proposal application to convince the committee that there are no ethical issues or that you can address these satisfactorily. Details required include the population (who, how many) you will use for your study, how they will be recruited, activities they will undertake as part of your research, and compensation to be given to participants, if any. If compensation will not be given please give a reason why not
- Please note the default positions on anonymity of applicants and their rights (such as the right to withdraw from the study, or to request that their data is not used). Make sure to address these issues in the relevant section of the form, as well as the letter sent to each participant in your study.
- You must edit the form letter (informed voluntary consent) attached to the end of the research ethics application form in order to tailor it to your study. All the highlighted issues (such as anonymity and right to withdraw) must be edited appropriately. You must also include the correct contact details at the top of the document and specify that this is for the Department of Computer Science. The statement of informed voluntary consent should be submitted within the same document as the ethics application and should not be extracted as a separate file.
- Please pay attention to all parts of the form; if the default position indicated for a section is all you need, it is still best to state this again. Empty forms look unprofessional and suggest the applicant has not read through all the required information. For the expedited process, the form instructions will indicate what sections need to be completed.
- Note the importance of the informed consent and confidentiality parts of the application: if these are not carefully motivated, the application is likely to be sent back for editing.
- Any other ethical issues which may arise must be highlighted in the application.
- Your supervisor must sign off on your application: in doing so, they are endorsing what you plan to do. Do NOT submit the ethics application before they have read and approved it! A physical signature is not required – you and your supervisor should type your names where indicated on the form.
- Please submit all your materials as a single file on this site (scroll down). You may need to create a Submittable account.
- Send an email to email@example.com to notify the CS Ethics subcommittee of your submission.
- Please allow two weeks for the application to be processed. If your application is not expedited, then once it is processed by the department, Faculty of Science Research Ethics Committee Review will take an additional two weeks. If you do not hear from us within two weeks, please email firstname.lastname@example.org. You can also email us if approval is time-sensitive, however please be aware that we may not be able to rush the approval process, and endeavour to submit the application early.
If you follow the guidelines listed above, your application should be processed swiftly. For any questions about the process, please contact email@example.com.
If you have been asked to revise and resubmit your application, please create a new submission, but submit with the same account as before.
Research with Third Party Data (about people)
Research with Third Party Data (about people)
We frequently have honours students looking at third party data sets. The official line is that as long as you have permission from the steward of the data, then you don't need faculty clearance to do the research. In some cases, the data steward may require an ethics clearance certificate, in which case one can be issued that either says clearance is not required, or that says the research has been reviewed and approved. If this is required please state this clearly in your application.
At the same time, not needing faculty clearance doesn't mean that there are no ethical considerations attached to the research. Supervisors should make sure researchers understand wider implications around third party data, and potential ethical issues that can arise even with use of anonymized data. This includes but is not limited to:
- Issues around how the data was gathered in the first place - were the users informed that their data was being collected, and might be used for research purposes? Even if informed, were they aware of implications? e.g. Facebook)
- Potential for group harm - can findings affect a group of people even if the data itself is anonymized at the individual level?
- Potential for identification of individuals based on aggregate data (i.e. name and ID removed, but you can still identify someone based on a search history)
- Is the anonymized data non-anonymous publicly? If so, there's risk of identification (i.e. Twitter), and it may be required to paraphrase or alter 'quotes' to reduce this risk. Of course any alteration of quotes should be acknowledged in the write-up
- One should also always be aware of selection bias and how that will affect the results.