Michael

What I'm saying is there is no way to verify the data. Not that the person doing the survey has to personally know them. It's as scientific as any phone poll or from a person standing in a mall. Which, for the most part are ± 5% accurate to people who take phone and mall surveys.

__________________

To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.

To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.

To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.

milwaukeebeers44

If your wife wants a scientifically valid survey, she needs for it to be administered to a sample group representative of the general population (in terms of race, gender, ethnicity, etc). If she is interested in only a particular subset, then she needs to state that specifically in her survey methodology, and exclude results from those that do not fit the criteria.

I have quite a bit of experience myself conducting scientific studies- I have a masters degree from the University of Michigan in epidemiology

Posting the survey on this type of board is, in effect, skewing the sample. . .Look back to the poll results in the main forum from a couple of weeks ago. The majority of the posters here are (presumably) white males between the ages of 20 and 50. This introduces tremendous bias.

If authentication of the results were of utmost importance, there would have been a rather large subject recruitment for the study, and the study design would be something more reliable than a cross-sectional study. She's looking at a rather large exposure, and I really don't understand how she would expect to draw any kind of conclusion from the results she would get (not talking about causality- cross sectional studies can't infer causality- rather, I don't know how any kind of statement other than "futher in-depth study is needed" can be made from this data).

__________________
“You always try to know what you need, try to manage the game given the game situation, ... and I think that's one of the things that Tom does best.” - Bill Belichick


To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.

Fixit

Maybe I should have specified that she's not putting all of her eggs in the PatsFans forum basket, for Pete's sake. Seeing as some of the top minds in the in the Criminology field are A) on her committee, and B) have signed off on her research methods, I might take their word for it.

Guys, seriously, I posted the link here to get some additional responses from people on a football message board, not to have the entire study hinge on it.

Long story short, and I mean this with respect, it's your option to help out or not, but none of you know what the ***** you're talking about with regards to her study.

Thank you for the hyperlink to "epidemiology", by the way, as there's no way I'd know what that was. Not pretentious at all.

milwaukeebeers44

Over three-quarters of the US population has no idea what that word means- most people think I'm a skin doctor.

I was just trying to be helpful. You know, since I've actually been through the process before and consult on study design for a living.

__________________
“You always try to know what you need, try to manage the game given the game situation, ... and I think that's one of the things that Tom does best.” - Bill Belichick


To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.

Fixit

You were absolutely not trying to be helpful; you were taking an opportunity to denigrate one person's study with absolutely no information about it, all to show how much you know.

You have your Master's in epidemiology...good one on you. Honestly, that's great. She's getting her Doctorate in Criminology, guided by one of the world's foremost experts in Criminal Justice and another PhD who specializes in spatial analysis, geographical mapping systems, and all that other stats-related stuff that makes this l'il English major's head hurt. Again, I'm leaning towards their point of view...no offense intended, of course.

I'm willing to let bygones be bygones (if you care) and all that, but don't pretend like you were merely trying to help out. "Helping out" would be taking the damn survey, being honest, and letting the experts sort through everything.

Christ, sometimes I hate 2 cents.

Fixit

Here you go, directly from my wife:

Hi All -

I want to thank you all for taking the time to think about, respond, and take the survey. It is great to see such a response!

As for some of the concerns that have been raised here, I just wanted to take a moment and address them. First, I am fully aware of the methodological restrictions of a non-random sample survey and I am keenly aware that it is not the most desirable data collection model. Fortunately, for the type of research that I am conducting, I am not attempting to make inferences to a general population sample. Instead, I am interested in how an individual responds throughout the survey – not the sample as a whole. However, I have implemented a weighting procedure that will allow me to at least examine how close (or far) I am from the general online demographic (as I am only looking at online users). While I am not a huge fan of weighting processes, I think in this case it will be valuable to address the concept of generalization.

As for the reliability and validity, there are a number of variables within the survey to test for reliability of respondents. Of course, they will probably not alert me to all of the problems, but during pre-test these measures accounted for a significant percentage of the error effects.

Thank you very much for the interest and time that you have taken to talk about and help with this study. Honestly, every conversation and every response is helpful!

Go Pats!

Fixit

Also, I haven't had my coffee yet, so I'm a bit grumpy.