Experiments using Human Participants
When performing experiments on human subjects it is important that approved procedures are followed. This is for the benefit of the subject and to limit any liability that may fall on the experimenters or the University when it has encouraged the experimental programme.
Full guidance on what to do before you start your project can be found at Experiments with human participants. Please read the guidelines carefully before planning experiments involving human subjects and discuss with your supervisor if in any doubt. Your supervisor will need to complete your application form for you.
1. If your project involves deception (i.e. misleading the subjects), or anything else that does not have the full and informed consent of the subject at the time the work is conducted, the ignorance of the subject must only persist for some justified time period. Everything must be explained to them afterwards, in a debriefing session. Nobody must remain under a mis-apprehension.
2. The storage methods and erasure procedures for personal data must be documented. If it is essential to contact respondents subsequently (e.g. for prizes), it may be acceptable to request an email address, but this should be optional. If email addresses are collected, your data will then be subject to the terms of the Data Protection Act. Many surveys incorporate demographic data (age, gender, education etc). This should be minimised - you should not collect any demographic data unless it is related to a specific research question.
Not Allowed Without Special Permission
Experiments with human subjects are not allowed without explicit further permission from the Computer Laboratory. If your work will intersect with any human subjects or any of the forms below you must submit an online application for approval form to the Ethics committee.
- experiments on participants who are vulnerable or unable to give informed consent (e.g. children, people with learning disabilities),
- publication of any result that contains personal information not in anonymous form,
- significant financial inducements (excluding reasonable expenses),
- prolonged or repetitive testing,
- discussion of sensitive topics (e.g. drug use, sexual behaviour, political activity, mental health, gender or ethnic status),
- drugs, substances administered, tissue samples, etc.,
- discomfort, stress, anxiety, inherent physical or emotional risk to participants or any third party,
- the co-operation of a 'gatekeeper' to gain initial access to participants (e.g. ethic community groups, nursing homes).
The form and further information can be found at Ethical review of research with human participants.
Your experimental plan will be reviewed by the Ethics Committee and the result will be conveyed to you.
Candidates are strongly recommended to follow an approved evaluation methodology for evaluation work that involves HCI.
For instance, the HCI Lecture Notes (http://www.cl.cam.ac.uk/teaching/0910/HCI/HCI2009.pdf) Section 3.3, "Evaluation techniques", is particularly relevant to Computer Science students planning the evaluation section of their dissertation. The appendix page 24-32 on "Analysis techniques" could also be appropriate for some projects where you wish to carry out an analytical evaluation rather than an empirical one. The suggested supervision exercises on page 36 offer students a chance to have an HCI supervision dedicated to their part II project.