Re: web based health studies

Dennis Roberts (droberts@netvectors.com)
Sun, 27 Sep 1998 05:26:57 -0400

Spike Jones wrote:

> Ian Goddard wrote:
>
> > IAN: True enough, and there's no guarantee that
> > controlled laboratory research is immune from
> > disinformation, i.e., results that were rigged...
> > ...Also, gathering information from a wider population
> > has its advantages too...
>
> ya, huuuuge advantages. there is an unimaginable treasure
> chest of medical wisdom waiting just below the surface,
> waiting for us to find it. i mean some really true breakthroughs.
> i think we will find these thru web based health surveys and
> data mining. all we need, yes, ALL WE NEED friends, is
> some way to allow a large population to download a lot of
> information about their lifestyle, health habits, health problems,
> etc, into a universally accessible database, while simultaneously
> protecting the database from vandalism (possibly by health
> product manufacturers) and guarding the privacy of those who
> contribute to the database.
>
> how, extropians? how how how how do we do it? spike

What is the size of your survey? Do you want anecdotal info from the net at large to comprise the entire database or might multiple databases that used different survey methods to obtain their data be both more useful and be able to give you both privacy and highly valid data just not simultaneously.
Example: Advertise ( through banner ads on the Internet Link Exchange for example) for individuals to become members of a list of potential survey recipients (email addresses) then choose from that list at random. Mail each selected participant a form (an ordinary e-mail doc with an embedded html doc that contains in its html code a validation number perhaps) that they fill out and return by e-mail. No one can submit more than one copy and by stripping out the html doc from the e-mail there is no way to connect the data with the participant. That system is admittedly imperfect since it requires that the participants trust that the survey administrators actually divorce the html doc from the e-mail address. That would however give you a one submission per person database for those folks who did have enough trust. Make the database size significant. That database could be used as a key to help validate a larger ongoing database to which anyone could contribute simply by filling out a form on a website.

--
Dennis Roberts http://netvectors.com/cnc/webguide.htm
 Boost your website with our start-up page!

"Do not be deceived by...some false secondary power, by which, in
weakness,
 we create distinctions, then deem that our puny boundaries
 are things which we percieve, and not which we have made." ---
Wordsworth