A Short Survey

[MPTheMaster]

I'm going to be studying for my Behavioral Statistics course and I thought I'd make it interesting. Please fill out my short survey and I'll give you the results of my scientific analysis of it in a few days.

Age:
Height (in inches):
***Weight (in ounces ( use http://www.calculateme.com/Weight/Po...oOunces.htm)):
Average amount of hours spent watching tv in a day:
Average amount of hours spent playing video games in a day:
If you go to school (otherwise leave blank), Average amount of hours spent studying in a day:
If you work (otherwise leave blank), Average amount of hours spent working in a day:

Thank you for participating!

***if your standard for weight is something other than pounds, just look up on google how to convert it to ounces.

[Theseus1234]

Age: 18
Height in inches: 5' 5.5" (yes 5.5. I'm very picky about that). 65.5 inches
Weight: 2080 ounces
Hours watching TV a day: 1
Hours playing video games per day: 2-3
Hours spent studying: 8 hours in school, as well as 2-3 hours after school

[Cannibalking]

Age:15
Height (in inches):63.6
***Weight: 1760
Average amount of hours spent watching tv in a day: 1 hour
Average amount of hours spent playing video games in a day: 1 hour
If you go to school (otherwise leave blank), Average amount of hours spent studying in a day: 30 minutes - 1 hours
If you work (otherwise leave blank), Average amount of hours spent working in a day:

[Godfrey I of Leuven]

Age:18
Height (in inches):68.50"
Weight: 3068.83 ounces
Average amount of hours spent watching tv in a day: 2 hours
Average amount of hours spent playing video games in a day: 2 hours
If you go to school (otherwise leave blank), Average amount of hours spent studying in a day: 1 hour
If you work (otherwise leave blank), Average amount of hours spent working in a day:/

I don't know if it's relevant or not, but the reason why i weigh 87kg is because I play rugby and work out regularly. I hope you're not looking for a correlation between sitting in the couch and the weight of someone. Anyway, looking forward to the results.

[Elfdude]

Age: 22
Height: 6'1"
Weight: 2304
Average amount of hours spent watching tv in a day: 2 hours
Average amount of hours spent playing video games in a day: 4 hours
Average amount of hours spent studying in a day: 5 hours
Average amount of hours spent working in a day: 5 hours

[eldaran]

Age: 25

Height: 75 Inches (6' 3")

Weight: 2800 ounces (12.5 Stone)

TV: 2-3 hours

Computer: 3-4

Work: Average of 6 per day (12 hour shifts, 4 days on, 4 days off)

[MPTheMaster]

I'm still collecting surveys from other sources so don't worry this isn't dead. Also, I'm writing a program that can do behavioral statistics. As soon as the program is sufficient enough (I'm just coding a rough console version of it right now before I make it into a true window's program), I'll post my analysis of it. That should probably be in about a month or sooner (depends on how much my newly found addiction Dragon Age: Origins takes over me).

[eisenkopf]

Again, I am amazed how willingly people give away information on themselves.

I cannot help but wonder about the validity of the approach used here. It would be better practice (and more efficient) to use one source, e.g. an online tool like SurveyMonkey or your own online questionnaire, of which you spread the link, than this method, which cannot guarantee the participants' anonymity and leaves you the work of manually enter the values (which is highly error prone, too).

And why do you want to code your own statistics software, when there is so much out there to use?

I'm sorry, but to a researcher all of this does not make sense at all.

[MPTheMaster]

You shouldn't be amazed at how willingly people "give away information on themselves." Out of 178 views, only seven people responded including you. Also, I'm not asking for a credit card information or about anything intimate.

How is one source better practice and more valid than multiple sources? A survey is a survey, and the bigger the unbiased sample size the better. I agree this probably isn't very valid and not at all anonymous, but both those things are irrelevant to what I am doing. This is a forum; none of the participants' expect anonymity or they wouldn't have posted. Also, manually entering values is something that comes with giving out certain surveys and also when getting data from other methods. It being "highly error prone" depends on the person doing it. That's also why there is double and triple checking.

I code my own statistics software for more experience as a programmer. Using an already made program would not help me do that.

I'm sorry, but to a researcher who didn't read or remember the first line of my first post, this wouldn't make sense: "I'm going to be studying for my Behavioral Statistics course and I thought I'd make it interesting."

[eisenkopf]

You shouldn't be amazed at how willingly people "give away information on themselves." Out of 178 views, only seven people responded including you. Also, I'm not asking for a credit card information or about anything intimate.

178 views does not mean 178 people. Even if we assume so, a recall of approx. 3.4% is a dream ratio for anything online related. Spammers would be kings of their discipline if they achieved this ratio.

Height and weight are highly intimate information. A lot can be deduced with high probability from these two values.

How is one source better practice and more valid than multiple sources? A survey is a survey, and the bigger the unbiased sample size the better.

One source is better because the uncertainty introduced by the sampling method varies. If you use several methods (sources), you'll have several types of uncertainty to account for.

It is difficult (impossible) to get an unbiased sample, but this forum certainly is highly biased.

I agree this probably isn't very valid and not at all anonymous, but both those things are irrelevant to what I am doing.

It shouldn't be. You should strive for best practice always.

This is a forum; none of the participants' expect anonymity or they wouldn't have posted.

Then why don't we use our real names?

Also, manually entering values is something that comes with giving out certain surveys and also when getting data from other methods. It being "highly error prone" depends on the person doing it. That's also why there is double and triple checking.

Nope, if you have to enter lots of data by hand everyone does a lot of errors. Double and triple checking cost time and effort. There is a reason why many scientists favor quick and dirty online questionnaires (cheaper and faster, less observer bias).

I code my own statistics software for more experience as a programmer. Using an already made program would not help me do that.

You could still code modules for, say, R. This would make you acquainted with a powerful software tool at the same time.

I'm sorry, but to a researcher who didn't read or remember the first line of my first post, this wouldn't make sense: "I'm going to be studying for my Behavioral Statistics course and I thought I'd make it interesting."

That is laudable. But the gathering of primary data and the statistic analysis of it is a very difficult and time-consuming effort. My intention was (is) to offer constructive criticism that might make your work easier and more valid.

[MPTheMaster]

178 views does not mean 178 people. Even if we assume so, a recall of approx. 3.4% is a dream ratio for anything online related. Spammers would be kings of their discipline if they achieved this ratio.

You're also forgetting all the people who only saw the thread title and said, "A short survey? Pass."

Height and weight are highly intimate information. A lot can be deduced with high probability from these two values.

What can be deduced with high probability from height and weight? Approximate height and weight are seen all the time in our daily lives.

One source is better because the uncertainty introduced by the sampling method varies. If you use several methods (sources), you'll have several types of uncertainty to account for.

It is difficult (impossible) to get an unbiased sample, but this forum certainly is highly biased.

But some methods cover the uncertainties of others and vica versa. Besides, different methods are used together all the time (like cross-sectional and longitudinal research). I know this forum is highly biased, but as I said before that is irrelevant to what I am doing.

It shouldn't be. You should strive for best practice always.

I completely agree with that for official studies that matter. Alas, this is only me studying concepts in behavioral psychology.

Then why don't we use our real names?

So then you are saying we do have anonymity online? Make up your mind. They either have anonymity or they don't.

Nope, if you have to enter lots of data by hand everyone does a lot of errors. Double and triple checking cost time and effort. There is a reason why many scientists favor quick and dirty online questionnaires (cheaper and faster, less observer bias).

You don't even know how much data I am receiving. I'm hoping for at least 50. Not everyone makes a lot errors; if you go carefully and slowly, not rushing through. Note, I am talking about small data sets in the 50-100 range. For my class, we do them by hand in order to learn. I agree, online questionnaires like you are saying would be better, but again, I am doing this for a class in order to get a better understanding of the mathematics involved.

You could still code modules for, say, R. This would make you acquainted with a powerful software tool at the same time.

But are the modules in C#? Regardless, I am coding for two reasons: to get a better understanding of the math at work and how an entire program works.

That is laudable. But the gathering of primary data and the statistic analysis of it is a very difficult and time-consuming effort. My intention was (is) to offer constructive criticism that might make your work easier and more valid.

I appreciate your constructive criticism and will use it in the future. The way I am doing it right now is to learn; usually learning entails doing work less efficiently in order to have a better understanding of the how and why.

[eisenkopf]

You're also forgetting all the people who only saw the thread title and said, "A short survey? Pass."

True, I forgot about them.

What can be deduced with high probability from height and weight? Approximate height and weight are seen all the time in our daily lives.

A lot. I'm not talking about identification here, I am talking about targeting legal marketing and illegal scams here. Height and weight are two basic variables that potentially tell a lot about habits and health. If it wouldn't be so, why are you asking for them together with information about media usage?

But some methods cover the uncertainties of others and vica versa. Besides, different methods are used together all the time (like cross-sectional and longitudinal research). I know this forum is highly biased, but as I said before that is irrelevant to what I am doing.

Okay.

I completely agree with that for official studies that matter. Alas, this is only me studying concepts in behavioral psychology.

But it matters to you, doesn't it? So it's important.

So then you are saying we do have anonymity online? Make up your mind. They either have anonymity or they don't.

I have made up my mind. There's a lot of room between those two poles. People use nicks here because they want to remain anonymous. It does not mean that they remain so. If you would look at all the things I have written here over the past years, you'll get a fairly detailed picture of me. I still hope it's not enough to identify me.

You don't even know how much data I am receiving. I'm hoping for at least 50. Not everyone makes a lot errors; if you go carefully and slowly, not rushing through. Note, I am talking about small data sets in the 50-100 range. For my class, we do them by hand in order to learn. I agree, online questionnaires like you are saying would be better, but again, I am doing this for a class in order to get a better understanding of the mathematics involved.

Fair enough.

But are the modules in C#? Regardless, I am coding for two reasons: to get a better understanding of the math at work and how an entire program works.

They are in R, but you can link C and C++ for computationally intensive tasks.

I appreciate your constructive criticism and will use it in the future. The way I am doing it right now is to learn; usually learning entails doing work less efficiently in order to have a better understanding of the how and why.

[Veliky Kaiser Theos]

Age: 17
Height: 6 foot 2
Weight: 2,680 ounces
An average of 10 minutes a day watching T.V.
An hour and a half for video games.
Approx. 8 Hours a day schoolwork/studying/homework including weekends.

[Edgar Arakelyan]

Age:15
Height:5 foot 9
Weight:2240 ounces
Avg hours spent watching TV:1hr
Avg hours spent playing video games:0-1 hr
I go to high school; Avg amount of hours spent studying:1-3 hours

[Eskali]

Age: 19
Height (in inches): 74
Weight(in Ounces): 2240
Average amount of hours spent watching tv in a day: 1
Average amount of hours spent playing video games in a day: 5
If you work (otherwise leave blank), Average amount of hours spent working in a day: 4