[Air-l] What to do with all this data

[Maximilian C. Forte]

Hello everyone,

This will be quite an imposition, given that this is an awkward and
open-ended message directed mostly to those on the list who, unlike myself,
have some background in quantitative research methods. Lacking this
knowledge is not the only reason this message will be clumsy: compounding
the problem is the lack of any defined research question (!).

Essentially, I am curious about some of the possible uses of the kind of
data I will describe below. For the past five years I have managed the
Caribbean Amerindian Centrelink (www.centrelink.org) and Kacike: The Journal
of Caribbean Amerindian History and Anthropology (www.kacike.org). I think
of both sites, differences taken into account, in kin terms as siblings
(don't get me wrong: I don't also think of one as male and the other as
female, nor do I have pet names for them).

Over these years, the following kind of data have been amassed, without any
deliberate and focused purpose (other than having for the sake of having):

1. countries and cities where visitors are based
2. referring URLs
3. entry and exit pages + most popular pages on the site
4. numbers of unique and repeat visitors, plus hits
5. hourly, daily, weekly and monthly traffic reports
6. key search terms used
7. key search phrases used
8. entries in guestbooks and online questionnaires
9. e-mail feedback received + messages posted via online email forms
10. entries in online polls

Now, having glanced at that, what kinds of questions could be posed with
reference to this data? What kinds of questions can potentially result in
plausible answers? Market-research types of questions seem obvious, though
that might not be bad for developing some sort of "user model", as I hear
people refer to this. Anything else, maybe more exciting?

Having asked that, would anyone like to suggest some of the best methods for
examining this data, with respect to any questions above, and I can always
chase up methods handbooks to learn more.

(Some questions, of course, are easy to answer based on the above, i.e. "on
which days of the week are sites most visited?" or "based on the countries
where visitors are based, which are the primary national markets for these
sites?"--from the basic questions that the statistical site monitoring
packages themselves answer, I have been able to quickly determine that the
most frequent visitors are based in US universities, tending to visit during
mid-week, and especially during the fall and winter semesters. Other
frequent visitors are .mil people and robots...a friendly audience, I am
sure. On a more qualitative level, all that really strikes me is that those
who deliberately provide feedback are either scholars/students, and
supportive activists--those with largely negative or otherwise critical
opinions have largely chosen silence.)

It has taken me months to get over the hesitation in asking so much in one
question, and then I realized I would probably never do anything with all of
this (apart from coming to impressionistic overviews) unless I asked.

Many thanks and very best wishes for the holidays,

Max.

Dr. Maximilian C. Forte
Assistant Professor
Dept. of Anthropology and Sociology
University College of Cape Breton
1250 Grand Lake Road
P.O. Box 5300
Sydney, NS B1P-6L2, Canada
E-mail: max_forte at uccb.ca
Faculty Web page: http://faculty.uccb.ns.ca/mforte/
Office B.273
Telephone: 902-563-1947