TEACHING STUDENTS TO STUDY ONLINE COMMUNITIES ETHICALLY

You Have Students Do What?How can we teach students to do research on Internet-based communications technologies in an ethical fashion? At the Georgia Institute of Technology, we offer a graduate class called “Computer Science 6470, Design of Online Communities”. Approximately forty students enroll each time it is offered (every other spring). Course readings discuss topics like the nature of “community,” identity construction, management of deviant behavior, research methods, and data analysis techniques. For their midterm projects, we require students to study an online community. They are asked to use principles from Amy Jo Kim’s book

Community Building on the Web to structure an investigation of design features of the site and how it functions (Kim, 2000).I imagine the reader trembling at the thought-you send students out to study real people online? Forty of them? Aren’t you afraid they will wreak havoc, and you will find yourself in the office of your Dean or IRB Chair with some explaining to do?Fortunately, in five offerings of the class to date, no havoc has resulted. In this paper, I document how we prepare students to understand the complexities of conducting research online in an ethical fashion.

I show how our approach has evolved over the years since the course was first created in 1998, and share lessons learned. Key design decisions include,* One IRB protocol for the entire class.* A notation in the risks section of the consent form that the researchers are students just learning to do research.* Different consent forms for different types of subjects:A web-based consent form for adult subjects.

This requires an IRB waiver of documentation of consent. (Not a waiver of consent-just of its documentation.)Paper-based parental consent and child assent forms for subjects under 18 years old (and other vulnerable populations).* A research method focused on participant observation and interviews.* Requirement that students do most of their interviews on the phone or in person, not online.* A requirement that students openly describe themselves as researchers, and use no identity deception, even if that is common on the site.*

A prohibition from logging otherwise ephemeral conversation without explicit permission.* Careful choice of study sites:All site choices must be approved by the instructor.Sites that cater to children are discouraged, because of the difficulty of recruiting subjects and getting parental consent.Sites on sensitive subject matter are discouraged, unless the student has a strong personal connection to the subject and extra care is taken.Sites where the students are already regular members are discouraged.Successful sites typically make better study sites than failing ones.*

An approach to disguising material in published accounts that depends on the nature of the subject matter.Subject names are changed.Site names are usually disguised, except where this is impossible.If the site can’t be disguised, then more caution is needed in disguising individuals quoted in the paper.Direct quotes of material that is publicly searchable on the Internet are possible in most cases, but not used if the topic of the paper is sensitive.I will discuss each of these in turn. Additionally, I will review some of the challenges we have faced over the years. In this account, I will assume the reader is already at least somewhat familiar with ethical issues that arise in Internet research. Our current consent form is included as an appendix and a full set of our documents appears on our website . Although our consent form was developed as part of the evolution of this class, members of the Electronic Learning Communities laboratory often find ourselves using modified versions of it in many of our studies of Internet-based phenomena