What's the issue?
Online methods have become increasingly popular in the last few years, challenging traditional data collection methods and raising new methodological issues in media research. In part, online methods may be used to compensate for the difficulties of offline methods, capitalising on the infrastructure provided by the internet (e.g. in recruiting internet users). In part, online methods are used specifically to research online phenomena (e.g. what do people do in chat rooms?). Online methods raise some new challenges regarding access, consent and ethics, especially but not only when researching children. They also permit research on new phenomena (e.g. blogs, profiles, social interaction online, etc).
Online quantitative methods are generally used for researching the demographics and attitudes of internet users (e.g. via an online survey). Online qualitative methods are more suitable for in-depth study of online cultural and social contexts (e.g. virtual ethnography).
Beyond the role of the facilitator of traditional methods, online applications have offered space for the development of new methods for automated data collection, such as logging and metrics of online visits/usage statistics.
Online (as offline), research questions should be addressed differently in diverse modalities of communication (chat rooms, forums, blogs, etc), taking into account the different features and practicalities of each. For example, everyone can read messages in a forum, regardless of who's the author, precisely because they are public and intended to be read by everyone. You can browse through a guest book from a blog without leaving any trace of your presence, but the same cannot be said when you enter a chat room. Real-time communication makes it awkward to observe without interacting, while asynchronous communication makes it possible.
It is not easy to ensure that all ethical imperatives are met and, at the same time, manage to carry out fieldwork without influencing what is being observed or actually making online research possible. Researchers are at present evolving common practice on some of these issues, and many are discussed in a helpful manner on the electronic discussion list of the Association of Internet Researchers.
Online interviews save time and money, but they have to be prepared properly. Researchers have to know how to use appropriate software, how to conduct the interview online (how to ask questions, etc), and be able to follow (written) cues left by interviewees (which means being able to read between the lines), etc.
MSN and other instant messaging programmes can be used as tools for research. Keeping a record of MSN conversations, with the interviewee's permission for this, is a good way of using the internet as a research tool and information resource.
There are no easy answers to the question of authenticity - whether your interviewees online really are who they say they are.
Considerable value may be drawn from online content itself. Besides specific content produced by online users (like web pages or blogs), most online use leaves visible traces (messages in guest books, forums, etc). Content analysis poses some practical problems, as this abundant material can be hard to manage. One option is to draw a sample (e.g. sampling messages posted in a specific time period or in particular forums).
Pitfalls to avoid
Although online methods provide advantages with regard to access to remote populations and automated data collection, which reduce research time, cost and effort, a researcher should think carefully about disadvantages that may affect the quality of the data collected, such as inaccurate sampling frames, irregular response rates, response duplication and participant deception.
For researchers who aim to conduct an online survey, there are clear difficulties in drawing a random representative sample online. In online surveys, the respondents are self-selected and there is a lack of a central registry of web users that would allow the researcher to follow consistent sampling procedures.
When conducting real-time online qualitative interviews or online focus groups, ask respondents to use the typing progress indicator (a small pen, keyboard or a small icon to indicate which person in a conversation is typing) in order to limit the typing to one person at a time. This prevents fast, furious and blurred interaction where we are unable to tell who is replying and who just sending a message.
Questions to consider
Are you using online methods to compensate for the limits of offline methods, or as a matter of convenience? Or because the internet is specifically of interest to the research question? How can you define the population from which you draw your sample? Are you studying individuals' activities online, or online practices or representations (whose relation to offline individuals is less relevant?). Is it important to know, in your research, that the respondents are of the age or gender that they claim? How will you relate what people do offline and online?
Lobe, B. (2008). Integration of Online Research Methods. Information Technology/Social Informatics collection. Faculty of Social Sciences Press.
Wright, K. B. (2005). Researching internet-based populations: Advantages and disadvantages of online survey research, online questionnaire authoring software packages, and web survey services. Journal of Computer-Mediated Communication, 10(3), article 11. Retrieved from http://jcmc.indiana.edu/vol10/issue3/wright.html.
Hine, C. (2000). Virtual Ethnography. London: Sage.
Hine, C. (Ed.). (2005). Virtual Methods: Issues in Social Research on the Internet, Oxford: Berg.
In my research on hip-hop cultural production and consumption, I found that most hip-hop artists (and fans) that I came across (and who were interviewed) were teenagers and young adults (Simões, 2006). Since I wanted to study hip-hop both offline and online, I chose a field research strategy based on multiple methods, namely participant observation, interviewing, gathering data of different natures (visual, audio, textual documents - photographs, video recordings, audio recordings, and other visual documents, like flyers, stickers, posters, etc). Even though online observation followed, basically, the same principles as offline observation, some specific questions may be raised. In this project, the internet was thus both a research object and a research instrument. (José Alberto Simões, Portugal)
In my online interviews with teenagers (aged 15-18), I noticed that it is of particular importance to find a strategy to reduce the ongoing possibility of distractions and interruptions that might prevent an interviewee from being fully engaged in the interview. I practised three tactics to deal with interruptions and disturbance issues. First, it is crucial to provide participants with flexibility in choosing the time suitable for an interview. Second, we have to inform them in advance about the approximate length of the interviews and ask them to suggest the time which suited them best. Next, it is useful to ask the participants to acknowledge the importance of not suspending the interview once started. When the interviewee requests an interruption (break), to accept it is particularly recommended when the interviewee asks for a shorter break (up to 20 minutes). However, if the interviewee decides to take a longer break, there is a high possibility that the interview would remain uncompleted, so it is best to try to keep them in the interview. There were also cases when interviewees did not announce their breaks but just disappeared. In such instances, I would recommend being patient and tactical at the same time to see the positive side of breaks. I looked at them as an opportunity to read the transcription in order to check what had been discussed already, what still needed to be examined and how to continue the interview. It also gave space for reflection by either party. To sum up, being a good online interviewer means being patient. (Bojana Lobe, Slovenia)