TMF data mining

[darvon96]

I was just wondering if Jeff or Myriads have been solicited by anyone to do some statistical analysis and/or data mining with the TMF database as part of some sociological or psychological study. I’ve been involved in data mining of medical records for pharmaceutical and medical research projects in the past and recognize that the information here would be a treasure for some researchers if the data could be organized properly. I don’t know what DB vBulletin is based on, but if it is created under one of the mainstream systems, I’m sure it’s possible.

Also, what would be your (members) reaction if you knew the data was being used for research? Would it change the way you conduct yourself. Would you be more hesitant to express yourself freely? Even if you knew it would be impossible to relate any of the data back to you personally.

 

[EBunbury]

I'm not a moderator here, and I imagine they'll reply as well. As an academic involved in research, I have served on the committees that set up rules for research involving human subjects, which this would be.

I'm sure darvon96 is aware of this already, but some of our members may not know that there are federal laws that require the informed consent of all legally competent human research subjects. While this would be simple to arrange (It could be made a log-in requirement, like the "under-18" law we all had to click on to get here.), the fact that there are many inactive or rarely-active members creates a legal limbo that two federal courts have so far been unwilling to address in similar cases.

Would it change the way members post? For me personally, probably not, but I'm absolutely sure we would lose membership over it, and that some members would behave differently. For that reason, I don't expect the mods to be too interested.

However, in the past, we have had at least one member take a different research approach. She was a graduate student who was seeking data for her degree research, and she posted a message briefly describing her research and giving a website for anyone who wanted to participate. I doubt the mods would have a problem with that.

[I suspect you may get some more colorfully-expressed responses - have to keep an eye on this thread! ;-) ]

🙂 E. Bunbury

 

[darvon96]

I'm sure darvon96 is aware of this already, but some of our members may not know that there are federal laws that require the informed consent of all legally competent human research subjects.

Actually this law only applies in situations where the subject can be identified in some way. Quite a bit of statistical information can be generated from completely de-identified data. How useful de-identified data is depends on the type of research being done. A researcher looking for specific case studies might have a difficult time, but someone doing group trend and behavior analysis could make good use of cumulative information.

 

[TMF Jeff]

So far nobody has approached me about anything like that...

 

[lk70]

I was just wondering if Jeff or Myriads have been solicited by anyone to do some statistical analysis and/or data mining with the TMF database as part of some sociological or psychological study. I’ve been involved in data mining of medical records for pharmaceutical and medical research projects in the past and recognize that the information here would be a treasure for some researchers if the data could be organized properly. I don’t know what DB vBulletin is based on, but if it is created under one of the mainstream systems, I’m sure it’s possible.

Also, what would be your (members) reaction if you knew the data was being used for research? Would it change the way you conduct yourself. Would you be more hesitant to express yourself freely? Even if you knew it would be impossible to relate any of the data back to you personally.

If when I went to register for TMF, I was told that anything I posted could be used in a research study, I wouldn't have registered. Don't care if the data is "deintentified" or not.

 

[TKForce]

Cool

Just let me know when it's going
to happen so I can delete all my posts
and leave the TMF for good. 😀 lol

 

[darvon96]

If when I went to register for TMF, I was told that anything I posted could be used in a research study, I wouldn't have registered. Don't care if the data is "deintentified" or not.

Interesting. What's your reaction to the fact that your medical records may be part of research databases around the world? I'm sure no one would choose to stop seeing a doctor when necessary because of this. For better or worse, this is part of the world we now live in, and it's only going to become more prevalent in the future.

 

[TKForce]

Interesting. What's your reaction to the fact that your medical records may be part of research databases around the world? I'm sure no one would choose to stop seeing a doctor when necessary because of this. For better or worse, this is part of the world we now live in, and it's only going to become more prevalent in the future.

Darn nosey doctors, I'm glad I kiiled all my doctors
and slashed and burned all my med files.

Oh, wait shouldn't have said that if this is going to be in an analysis.
How do I delete my posts again?
HOW?!
SOMEONE HELP BEFORE I GO TO JAIL!!!

darvon96, in case your wondering this is how it changes my posts.
Helpful aren't I? lol

 

[EBunbury]

Actually this law only applies in situations where the subject can be identified in some way. Quite a bit of statistical information can be generated from completely de-identified data. How useful de-identified data is depends on the type of research being done. A researcher looking for specific case studies might have a difficult time, but someone doing group trend and behavior analysis could make good use of cumulative information.

I'm no lawyer, but that's no the interpretation our committee got from lawyers we hired.

🙂 E. Bunbury

 

[darvon96]

I'm no lawyer, but that's no the interpretation our committee got from lawyers we hired.

🙂 E. Bunbury

A few of years ago HIPAA came out with some very stringent rules on data de-identification for electronic medical records to ensure privacy protection. Prior to that, there was a lot of legal jockeying over the subject.

 

[lk70]

Interesting. What's your reaction to the fact that your medical records may be part of research databases around the world? I'm sure no one would choose to stop seeing a doctor when necessary because of this. For better or worse, this is part of the world we now live in, and it's only going to become more prevalent in the future.

Hmm. Interesting. 🙂 This will sound odd as all hell, I'm sure, but somehow I feel like my medical records are less personal and less....inflammatory...than info that would be acquired on this site. To take "data" from this site would mean interpreting an awful lot of narrative, vs looking at any real quantitative data such as age, height, weight, BP, smoker or not, meat eater or not, exerciser or not, etc.

So I don't really care about the medical data.

 

[darvon96]

Hmm. Interesting. 🙂 This will sound odd as all hell, I'm sure, but somehow I feel like my medical records are less personal and less....inflammatory...than info that would be acquired on this site. To take "data" from this site would mean interpreting an awful lot of narrative, vs looking at any real quantitative data such as age, height, weight, BP, smoker or not, meat eater or not, exerciser or not, etc.

So I don't really care about the medical data.

I understand your position exactly lk70. In fact that's why I made the distinction between case study and trend analysis of aggregate data. I was thinking more in terms of data relating to usage habits such as which forums are visited vs. which are posted to, and how that changes over length of membership. Getting anything out of the actual content becomes difficult because it’s so subjective. Anyway I appreciate your comments. I suspect many would share your same view.

 

[TMF Jeff]

I don't know if anyone is concerned about this, but we would never give information about anyone here, to anyone else, for any reason.

 

[TKForce]

I don't know if anyone is concerned about this, but we would never give information about anyone here, to anyone else, for any reason.

It might concern me if I gave this site any real info about me. lol
Shoot, I can't quit cracking jokes.

No, in all seriousness I think that no matter how this analysis is handled,
if it happens in any manner there are enough members here who see this
place as a privet sanctuary and will feel violated by it.
The TMF would most likely lose quite a few members over it.
Not me, I'm paranoid but I love it here. 😀

BTW does anyone else wonder if this thread is the case study? 😉

 

[darvon96]

I don't know if anyone is concerned about this, but we would never give information about anyone here, to anyone else, for any reason.

I have no doubt about that, particularly since the concept for this forum was conceived from your heart.

Medical records are much different than what’s being generated here and the interested parties would also be different. Pharmaceutical companies have very deep pockets and are willing to dig way down for information that could assist in targeting sales or to show product performance, particularly against competitors.

I don’t know of any analogous industry with that kind of cash that would be interested here. Usually educational research has very limited funds unless they are backed by the government or private sector.

Keep in mind though that if the TMF ever does peak the interest of a group of that caliber, you might someday be faced with a very difficult decision.

 

[darvon96]

BTW does anyone else wonder if this thread is the case study? 😉

I’m sure this would be the last thread used for a case study. But when you think about it, there’s nothing stopping some PhD candidate from picking out any one or more of us and tracking our posts. It’s a bit tedious, but if they have the time & motivation, who knows. Obtaining the guts of the database, on the other hand, would interest a whole different kind of animal.

 

[TKForce]

I’m sure this would be the last thread used for a case study. But when you think about it, there’s nothing stopping some PhD candidate from picking out any one or more of us and tracking our posts. It’s a bit tedious, but if they have the time & motive, who knows. Obtaining the guts of the database, on the other hand, would interest a whole different kind of animal.

I just thought you might be studying the reaction your thread has caused. lol 😀

 

[lk70]

Keep in mind though that if the TMF ever does peak the interest of a group of that caliber, you might someday be faced with a very difficult decision.

I sense that Jeff (who I know very very well because I've read at least 50 of his posts), has an ethic that would indicate he wouldn't find it all that difficult a decision. But then, I'm horribly naive about almost everything. Always trusting people and believing in the good. I get screwed every time!

😀