Thread "A request to web comic writers and artists" in forum "Networking & Community Projects*"

Posted at 9:07PM, Jul. 17, 2009

Hi there,
My name's Chris Chapman, a student at Edinburgh Napier University and I'm currently working on a dissertation based on web comics. If any of you have the free time to fill in a quick questionnaire I would greatly appreciate it. If you're interested message me with your e-mail and I'll send it over as soon as I can.

Thanks

Chris

Posted at 10:07AM, Jul. 18, 2009

Why not just post the questionnaire here so people can get interested? :]

Posted at 11:07AM, Jul. 18, 2009

Actually, either method makes this survey a convenience sample which would be an extremely poor and flawed data collection methodology for a doctoral dissertation. Determine your population, calculate how many you need to sample in order to get a proper representation of the population, and conduct your survey based upon a random sample.

Posted at 3:07PM, Jul. 18, 2009

Oh, Interesting!
I've been to Edinburgh! I participated in the fringe fest last summer doing spot light for a production with my High School! Go Edinburgh, it was a cool place, even if some locals may not enjoy some of the hype of the fest (from what I've heard).

I'm interested, I'll PQ you!

Posted at 4:07AM, Jul. 21, 2009

There's nothing wrong with using a convenience sample if that's accounted for in the discussion. It may in fact be essential to whatever he's studying. I'm not really sure how else you would readily find a population of webcomic readers without going to a webcomic site. Random sampling assumes he has a list of every possible sample member and can contact them directly, which he doesn't and can't. Even if he did that he would be subject to a low response rate. Might as well do a convenience sample.

You can do a disseration at multiple levels too, doesn't have to be doctoral. :]

Posted at 6:07AM, Jul. 21, 2009

At the risk of getting into a big sampling methodology and terminology argument

At least in the US education system, the term "dissertation" is reserved for the final culminating research paper for doctoral level students. The term "thesis" is used for the Masters and Baccalaureate equivalent. I'm not sure if the terminology is different in other countries though.

While I know very little of his dissertation, performing a convenience sample will only give him a set of people that wanted to participate. Those who wanted to participate may not (will not in most cases) provide a proper representation of the population (kinda like having a study using a town as the population and only asking people in the rich area). If you don't have an accurate representation of the population, you have a biased study. Obviously, having a biased study in a research paper is a bad thing.

Being Drunk Duck's only Doctoral student with two statistics degrees, I kinda have to start questioning this stuff. I know my adviser/dissertation chair would kill me if I said I was going to take a convenience sample (and my doctoral degree is in education, not statistics), so that's why I need to question like this.

Posted at 6:07AM, Jul. 21, 2009

Psh, you guys are such linguistics.

By the way, I did the survey, pretty much what you would think it would say. So, just so you know, he's not a spam bot.
(Don't wanna post some questions without his permission, though.)

Posted at 12:07PM, Jul. 21, 2009

While I know very little of his dissertation, performing a convenience sample will only give him a set of people that wanted to participate.

I've done a master's thesis and I understand statistics and differences between samples. What I'm saying is that even if he somehow figures out the number of his population (unlikely, assuming his 'population' is 'webcomic readers' ), gets a random sample of them and sends them the survey, which would be impossible, he's STILL reliant on the people who feel enough like filling out the survey to respond. While technically that wouldn't be a convenience sample, it would still be very biased in favor of those who cared enough to do it (probably those with strong opinions) which has always been a problem with survey methods. Unless he can somehow round people up involuntarily, make them sit in a room, and all fill out the survey, that bias will always be present to some degree.

It can depend on the subject he's studying as well as the topic of study and many programs will accept studies based on convenience samples if that's properly acknowledged in the paper (so readers may interpret the findings) and if there are few alternatives, especially if it's in a novel or little-researched area (ie a pilot study). It's easy enough to say 'this study does not propose to apply to the entire population.' Since his research is on webcomics, I'm guessing it's for a social science or art and design, which tend to be much more lenient with research sampling than other subject areas in the interest of finding interesting results which may bear further, more rigorous investigation. Education does this very often too I might add. :] It's true that statistics are based on the premise of random sampling, but not all kinds of research need to be based on rigorous statistics in order to be interesting or valid.

Regardless, surely he's got an academic supervisor who knows his subject and methodology so if there were concerns about it they would have been addressed before he ever got to this point. :]

Anyway I've pqed him because I'm happy to fill it out, I've had to wait around hoping for survey responses too. :]

Edit: My dog eats chicken butts, statistics that