Using an internet forum for Social Research

[Vuk]

I have decided to conduct an Org poll where users will tell their professions and vote on their diet preferences. I will then tabulate this data and arrange it by professions, so can tell which professions like which foods. I will then use this data in my Uni research paper with the following entry under the "Methods" header:

"This data was gathered on an internet forum on which it is possible for a single person to create multiple accounts (and therefore vote multiple times) as well as for bots to register and vote. It is also impossible to know if the information given is accurate or not, or if the people are really who they say they are. Therefore this data is completely useless and should be completely ignored."

What do you think, is it a good idea?

[rasoforos]

What is the point?

[Ibrahim]

I have decided to conduct an Org poll where users will tell their professions and vote on their diet preferences. I will then tabulate this data and arrange it by professions, so can tell which professions like which foods. I will then use this data in my Uni research paper with the following entry under the "Methods" header:

"This data was gathered on an internet forum on which it is possible for a single person to create multiple accounts (and therefore vote multiple times) as well as for bots to register and vote. It is also impossible to know if the information given is accurate or not, or if the people are really who they say they are. Therefore this data is completely useless and should be completely ignored."

What do you think, is it a good idea?

you are the scientist here. do as you please.

[Vuk]

lol, I am just demonstrating how absurd using an internet forum for Social Research is. I was told by my professor that it was a valid option, but I would have to discuss the weakness of that type of research in the methods section. I just thought that it was pretty absurd. :P

[CountArach]

It might also be pointed out that you would be drawing false conclusions from your data - there is no reason to believe that your job determines the food you like.

[LittleGrizzly]

"This data was gathered on an internet forum on which it is possible for a single person to create multiple accounts (and therefore vote multiple times) as well as for bots to register and vote.

If you put it somewhere like the backroom then there is a very limited possibility of that... though a member with multiple accounts could vote twice, anyone who has two active accounts probably uses them for a much better reason then screwing with jobs linking to paticular foods polls....

It is also impossible to know if the information given is accurate or not, or if the people are really who they say they are.

Probably the best criticism of the method, there would be very little reason to lie on the what foods you like, the only reason to lie would be for pure fun... so most would probably give honest answers...

The profession one there is an excuse to lie, make yourself look good, why be a binman when you can be a lawyer, my counter point though is that, we have a fairly mature community so lieing is slightly less likely... alot of people who do work crap jobs are proud that they work so hard... and lastly alot of .org members who have been here a bit will have likely dropped hints or outright told people what thier occupation is, so would not be eager to potentially be branded a liar...

It might also be pointed out that you would be drawing false conclusions from your data - there is no reason to believe that your job determines the food you like.

Going to have to disagree here, atheletes are attracted to healthy food, so would probably come out as having preferences for healthy foods, if we further specify atheletes are likely to eat certain foods depending on thier sport, a wieght lifter and a marathon runner, the weight lifter would have similar diet preferences to other wieght lifters... but probably a little different to the marathon runner... whose diet is much more similar to the other marathon runners...

Then we could get onto people with busy hectic jobs with long hours having preferences for fast easy food... ect.

[CountArach]

It might also be pointed out that you would be drawing false conclusions from your data - there is no reason to believe that your job determines the food you like.

Going to have to disagree here, atheletes are attracted to healthy food, so would probably come out as having preferences for healthy foods, if we further specify atheletes are likely to eat certain foods depending on thier sport, a wieght lifter and a marathon runner, the weight lifter would have similar diet preferences to other wieght lifters... but probably a little different to the marathon runner... whose diet is much more similar to the other marathon runners...

Then we could get onto people with busy hectic jobs with long hours having preferences for fast easy food... ect.

There are always exceptions.

[LittleGrizzly]

There are always exceptions.

Well i think you can find a decent amount of correlation to go between food and jobs, on top of what i already said, a person with a really good job would much more likely eat delicacys and luxurys, a person with a crap job thats hard work will want something cheap and quick...

Its not a huge correlation but more than an exception to the rule i think, someone well read on this type of thing could probably make a rough guess at the type of job someone has by looking at thier diet. the wieghtlifter giving himself away with his muscle foods, the ceo exposing himself via his luxury coorporate lunches including all kinds of delicacys, the factory worker who can't be bothered to do much more than stick something in the microwave or grab a burger on the way home... not a huge correlation but a fairly decent one...

[A Terribly Harmful Name]

Ah, statistics statistics! All of them are useless.

[LittleGrizzly]

There are 3 kinds of lie's, Lies, Damned lies... and statistics!

Did you know 43% of all statistics are made up on the spot...

[CountArach]

There are always exceptions.

Well i think you can find a decent amount of correlation to go between food and jobs, on top of what i already said, a person with a really good job would much more likely eat delicacys and luxurys, a person with a crap job thats hard work will want something cheap and quick...

Its not a huge correlation but more than an exception to the rule i think, someone well read on this type of thing could probably make a rough guess at the type of job someone has by looking at thier diet. the wieghtlifter giving himself away with his muscle foods, the ceo exposing himself via his luxury coorporate lunches including all kinds of delicacys, the factory worker who can't be bothered to do much more than stick something in the microwave or grab a burger on the way home... not a huge correlation but a fairly decent one...

Well of course there would be a level of socio-economic bias in the data. At the same time though I would contest that it is a direct result of the job someone does, but rather it is more a factor of the corresponding pay rates. Further, there is no denying that many of those in the lower socio-economic barriers would enjoy lobster and may have had it once - thus proving to be statistical anomalies.

Then we must also look at the inter-relation between race and socio-economic factors in determining food. For example many minority groups play into lower classes and as such there might be a strong leaning towards those food types in determining favourite foods.

This data sample could go on forever... now I am actually interested.

[seireikhaan]

CA- the whole point of statistics is to create generalities. No statistic is perfect, but they can be very helpful, so long as those in need of them use them properly.

[CountArach]

CA- the whole point of statistics is to create generalities. No statistic is perfect, but they can be very helpful, so long as those in need of them use them properly.

Yes I know, but that is the problem I would have with this data - the whole purpose appears to be to draw false conclusions. However, now I am really interested in exploring the similarities between socio-economic class and favourite food type.

[seireikhaan]

Yes I know, but that is the problem I would have with this data - the whole purpose appears to be to draw false conclusions. However, now I am really interested in exploring the similarities between socio-economic class and favourite food type.

Not at all. If that were the point, why do businesses utilize statistics so much? They need to know who to market a product to, where to market it to, where to expand, how they are perceived, etc... Not, of course, that statistics can't be used to draw false conclusions- however, this tends to be much more out of political or cultural convictions, not out of any fault of the statistic. Again, a statistic is just a number- what you draw from it is what's truly important.

[Vuk]

lol, I never said that I would suggest cause and effect. The reason that I am gathering this data is actually unrelated. It is for an economics study, and different professions have been put into different income and social class brackets. This is what is going to compared to diet. I did not mean to get into my project, just suggest how perfectly useless the idea of using an internet forum for social research is. :P

[Cute Wolf]

Well, all you must have is honesty to take the pool.... and I believe there was a lot of honest man here....

[CountArach]

Not at all. If that were the point, why do businesses utilize statistics so much? They need to know who to market a product to, where to market it to, where to expand, how they are perceived, etc... Not, of course, that statistics can't be used to draw false conclusions- however, this tends to be much more out of political or cultural convictions, not out of any fault of the statistic. Again, a statistic is just a number- what you draw from it is what's truly important.

I can't help but feel we are arguing around each other... yet, there are most certainly ways that statistics can be made completely useless and redundant - by way of providing information about a product/person that is misleading then trying to draw an opinion about that. Look at just about any internet poll that links to a damning story for an example of this.

[Vuk]

I can't help but feel we are arguing around each other... yet, there are most certainly ways that statistics can be made completely useless and redundant - by way of providing information about a product/person that is misleading then trying to draw an opinion about that. Look at just about any internet poll that links to a damning story for an example of this.

lol, here is the thing with statistics, they represent something (whether that thing is important or not), simple. You just gotta be smart to tell what they really represent and if it is meaningful or not. It is easy for people to twist statistics around to make em look like something else, but they are always what they are, people just need to be smart enough to tell for themselves what the statistics mean. The fault therefore lies not in the statistics, but in the dishonesty of people who try to use them dishonestly, and the stupidity of people who are not smart enough to tell what they really mean. We just need to use predjudice when looking at them.