Posted by bob on April 20, 2000, at 10:51:47
Well, Cam and Soreteh almost blew my cover:
"1.Am I the only person gullible enough to start wondering if Cam really is bob? And also easily amused enough to think it is funnier and funnier if he isn't.[Why hasn't lower case bob posted on this thread? Why did Cam put a pointy nosed winking guy on a post?]"
The reason I did not chime in to the prior discussion is that I will be working with Dr. Bob in conducting this research. We agreed that it would be best for Dr. Bob to "test the waters" before bringing me into the picture. Having reread and thought over the response to the original item posted by Dr. Bob, we agreed it was time to move on to the next stage, starting with this item.
Below are, first of all, a recap and interpretation of what was entered on the first item. Second, I'll be providing some explanation of how we will be proceeding with the work and, more to the point, why do the work in the first place.
The recap:
----------
Minus responses by Dr. Bob and myself, there were a little more than 30 posts that addressed the thread's main topic. Here are some of the more general issues raised:Goal: group dynamics *is* a pretty general term. The goal is to examine how this group has established and maintains a self-help community through the internet. Of obvious concern are the patterns of interaction, the culturally defined values and standards for communication, and other issues that define Babbleland as a "self-help community". There is a fairly extensive literature base on the role that self-help groups play in patient care, some of it including issues such as how the medical profession regards the efficacy (or hinderance) of such groups in contrast to the beliefs of the group members themselves. While I haven't done the journal crawl yet, I doubt there is much--if any--published work on web-based forums such as this. I'll talk about methodology below.
Names/anonymity: it is standard practice in social research to guarantee the anonymity of the participants. All the same, it is also critical in qualitative/ethnographic research to use quotes "from the field" to validate or elucidate specific findings. In such cases and particularly because this *is* a public forum, permission will be sought to cite specific posts ... even if posted anonymously/pseudonymously ... when possible. Again, the focus of the research will be on the group more than it is on individuals, so it is highly unlikely that we will be "tracking the history" of any one person's posts throughout the life of Babble.
On a related note, a concern was raised in the first item about how, with this board's ease of switching identities, it would be possible to track individuals with any certainty. This is closely related to what we might need to face when examining the posts of this "anonymous" person. Again, this is yet another reason to turn away from a unit of analysis that focuses on individuals to one that looks at interactions between individuals. The interaction becomes the unit of analysis, not the individual.
All the same, it was heartening to see so many folks volunteer their real names and/or their willingness to actively participate. A second phase of the investigation may introduce interviews, so such help will be needed.
I would also like to note that of the 30+ responses, only one stated a clearly negative view towards conducting the research. For that person, and for those out there who did not feel free to say the same, I hope the explanations offered in this thread as well as the discussion to follow will calm any concerns and convince you of the value of this work.
Yet another note on the issue of confidentiality -- that of informed consent. Dr. Bob is checking into this with his organization. I'm not bound by any organization, but I am bound by ethical standards for conducting social science research. Once some preliminary analyses have been piloted on the public data, we will have a better idea of what the nature of the full study will be and can submit it to an institutional review board for Human Subjects Review. While granting copyright to Dr. Bob may state who legally "owns" this data, it is not enough for ethical integrity. We will keep you informed on this issue as it develops.
Other general issues:
Several people asked to be informed of the results, and some mentioned the therapeutic value this work might have for them. Absolutely -- you will be informed of the results. In fact, as explained below, your feedback will be needed to provide one measure of validation for the work. "Audience participation" will make for a better study.As one person mentioned, the only compensation we can offer is the knowledge gained from the study. No grant monies are supporting this work, and we're doing it in our "copious" spare time. All the same, we hope that the results will offer some true value not just to the Babbleland community, but to other on-line self-help groups as well as to the professional therapeutic community.
Several thought that work had already been on-going given Dr. Bob's copyright statement. Nope -- this is the first effort. And sorry to disappoint, but I'm not one of his grad students or post-docs in need of data. It's a case of two scientists involved with a fascinating data set who agreed over a beer in SoHo one night that something really needed to be done about it. Why now? For me (since I'll be conducting most of the grunt work) it's a matter of opportunity -- a promising software package that fits the methodology, a broadband connection piped into my home, and the prospect of my summer vacation looming just over the horizon.
[by the way, I know of one Nobel Prize that was inspired by a glass of beer ... who knows? ;^]
Before I get into some more specific comments about methodology, there are three last issues from the previous item I'd like to address:
Being "watched" or the Big Brother Effect: Yes, conducting research on an active discussion can have a chilling effect on people's openness. It may also provoke others to wax eloquent. We will focus on archived data as a result. One concern is to examine how the group established itself; another is in how it evolves. Looking at the data up to Dec31.1999 should provide an extensive pool of information without prying into the concerns people have at this moment.
Quantification and demographics: This has also been raised as part of the "registration" thread. If we intended on tracking the participation of individuals, this would be a primary concern. If we were funneling the posts into a priori coded categories which could then be subjected to some hypothetical-deductive statistical analysis, it would again be critical for a multivariate, hierarchical analysis. We won't be doing that sort of analysis. All the same, demographics of the group will certainly be interesting and will be most useful as an aggregated report to help describe the nature of this community. Otherwise, no, we are not intending on matching responses to the medications or diagnoses of the posters. Again, it may be a concern if we do any interviews in a second stage of analysis, but it is not that appropriate for analysis of the group.
Qualifications: You may be able to assume much about the qualifications of Dr. Bob for this work (obviously, the expertise he brings to this study is his background in medicine and clinical psychology, and he certainly has a strong experiential base in managing and engineering this sort of computer-based discussion group, tho I still haven't asked what sort of data analytic expertise he has), but for those who asked questions about what other expertise might be brought to bear. What I bring to the table is extensive training in an applied research branch of psychology called educational psychology. It does not involve clinical psychology (I don't go to schools and treat children with "issues"). It does require a grounding in several branches of basic psychological research. My areas of strength are in developmental, cognitive, social, and cultural psychology. While I have a fairly strong background in statistical analysis and quasiexperimental design, my own preferences run towards qualitative, field-based work. My approach is somewhat ethnographic, my analytic principles based mostly on hermeneutic and interpretive perspectives. The primary methodology I will be employing for the first level of analysis -- making sense of the archived posts -- will be Grounded Theory (a technique developed by sociologists at, coincidentally, the Unviersity of Chicago).
Looking at the clock, I see I'm going to miss my dentist appointment if I blather on much more, so any detailed discussion of methods will have to wait. Let me explain this much about grounded theory, tho:
This approach is not a deductive method formulated on tecting a hypothesis based upon what is already "known" and seeking to extend that knowledge through the application of "scientific" theory. (At least one poster on the previous thread was concerned about the "ecological validity" of such approaches.)
The name of the method comes from the belief that "theory" is evident in how members of cultures and societies interact and create the ground rules for such interactions. The analysis begins with an open mind and an acknowledgement of what positivists call "bias" and interpretivist call "forestructure" -- what the researcher "knows" from the start, so that the reader of the work may judge how this prior knowlegde either informs or misleads the analysis. The work itself is highly inductive and iterative, relying on words and statements from the field to generate more and more abstract categories and category relationships, until a theory grounded in the life of the field can be developed. This theory is then cross-validated in a number of ways. One important one here, given the interest in seeing the results, is called a "member check" -- bringing the nascent theory back to the members of the field for a serious reality check. Another is to look for similar theories in parallel communities. [For example, some of my educational research on discipline and compliance has looked at studies of prison populations for cross-validation.]
Hopefully, MarkH, this addresses your concern (I'm sure others shared as well):
"The quoted "experts" seemed to be trying to fit their favorite psychological and social theories to the phenomenon, rather than building understanding from the evidence."That's exactly the opposite of what Grounded Theory tries to accomplish. =^)
Anyway, if any of y'all can get through that, that should put some fat out to chew. Comments -- posted here or privately through email -- are always welcome.Cheers, and thanks for the support!
Bob
poster:bob
thread:30694
URL: http://www.dr-bob.org/babble/20000420/msgs/30694.html