Forums / Discussion / General

Hey all, I'm doing a survey as a kind of research, and I'd love it if you all filled it out. Basically, I'm trying to figure out the impacts that avatars and profile pics have on a user's experience. I think it's interesting that on the Internet, we've come to use custom pictures to represent ourselves rather than real images. As such, I'm trying to figure out what extent this difference has on our interactions with each other, as well as impressions.

This is the link for part one, since my free account only allows ten questions a survey:

http://www.surveymonkey.com/s/GXSYJ7T

I'll get part two up when it's done.

Part 2:

http://www.surveymonkey.com/s/GX3WXWD

Apparently can't do this shit on a mobile device, so instead I'll just straight up say it: if I dislike a persons avatar, I am probably going to dislike them. For example if someone has an avatar of my least favorite Pokemon, they are my least favorite person. Of course, there are exceptions to this; some people have shit avatars but nice personalities, or vice versa.

Use Google Docs next time.

I found this survey very interesting, thanks for that, Brawler.
It's actually kinda funny that this came up, because when I was choosing my most recent avatar I was wondering if people would read my posts/see me differently. Glad to see I am not the only one with this mindset.

Are we not allowed to take the survey if we're under 18? That's the lowest age option.

Interesting survey, I did it. Honestly I like sites who allow avatars over those who don't, it makes it a bit more…personal I think. Helps give you a better insite to whoever is posting. At least generally. I also like sites who allow signatures, I think that is an even better way to express since they do not need to be small and can have other info about you you couldn't necessarily fit into an avart like image. But that's better for other sites, I'm not sure they'd work well here.

Oh, shoot, I knew I was forgetting something. Just put 18 to 24, and I'll change that category to 24 or younger in the results.

Bumping to see if I can get any more results.

Suppose I'll do this.

And yes, avatars do say a ton about the type of person you are. Not necessarily your personality, but your choices, influences, stuff you like, etc.

Well, yes, I think that's obvious. But this is about the impact other people's avatars have on a user's experience.

Also, I think I'm going to leave this open for about a week, and then collect the data. I've only got about 25 results so far, and that's obviously not enough for a legitimate conclusion.

Very interesting. I'm not really sure how you'll be recording those more open-ended questions, but I look forward to the results regardless.

I just changed my avatar a few hours ago…

I gladly filled your survey. Avatars have a lot of influence on the voice of the poster. I'm reading the post above in a manly voice.

You're a mod Brawler, but that doesn't justify pointless bumping. So please don't, thank you.

Back from the depths of where we are not?

Well, I don't know Brawler's expertise, but there are qualitative methods where you seek to identify themes that occur with the responses of your sample. Generally, you let respondents say whatever they want, and then you "code" those responses based on themes that your previous literature review had identified. You essentially turn open-ended responses into categorical questions without preventing new themes that you didn't expect from being recorded.

There are even qualitative research software like ATLAS and NVivo that help researchers code these answers (I find Excel and common sense works, but I don't know much about qualitative research.)
 
Problem about online surveys is that you deal with people who like to troll. Leaving respondents space to get "cute" and clever (or so they think) generally gets you several if not a decent percentage of answers that are just bad jokes.

Verbose wrote:

Problem about online surveys is that you deal with people who like to troll. Leaving respondents space to get “cute” and clever (or so they think) generally gets you several if not a decent percentage of answers that are just bad jokes.

I think the biggest thing that an avatar projects about a person is their gender. When I met DPF and Ogreenworld, I thought they were both girls, because their avatars were of female anime characters.

This especially applies because of the disproportionate amount of males on internet culture websites, so we generally assume someone to be male unless we have evidence otherwise, and an avatar is a pretty big piece of evidence.

Personally, the person's user name has more of an impact on me than the avatar.

Gender is one of the few things that I get from seeing a person's avatar. I think this mostly has to do with the fact that nowadays there are so many guys with ponies in their avatars and whatnot.

If you have a sample size of 30, then losing 3 or 4% of them, because people think they're funny, can basically ruin the study (analyses' stability will drop off significantly.) You may not be able to do any meaningful analyses because of those people.

If you have a sample size of 3,000 and 3 or 4% of them decide that they're witty, then you have to go through and read 90-120 answers that aren't usable for each question you asked. Weeding out themes in qualitative responses is hard enough as is. But it's not like the researcher can just skip your answers, because they have to make sure that your answer is assessed for usability
 
Basically, if your researcher is on business, then you're either ruining his/her study or giving him/her a headache by wasting a few hours on something they didn't get any data out of.

Yes please don't do that.

But what if you get a serious response from way out in left field? I assume in general you're not going to want to disregard outliers, especially in this kind of survey. I mean, what do you do with it?

I'm not sure what Brawler's plans are, so you'd have to ask him. He may not want to give out his analysis methodology until the data collection phase has ended though.
 
Generally, if you recognize a theme in the responses you're not expecting based upon your literature review, then that might indicate the need for another category.

So for example, your literature review says people should say "green" or "red," but several users are saying "blue." Then you'd make another category for "blue" when it comes time to analyze your data.

Now if you're expecting the themes "green" or "red" and respondents are answering "ur momz" or "5," then you can reasonably throw those responses out.

The main thing to look for in qualitative research (in my limited training of it) are themes in responses. Depending on Brawler's needs, he may omit outliers, because it's not a consistent theme in his sample, expected or not (e.g., the theme in question may need to appear in a certain number of responses to be considered a them in the sample and, by proxy, the population of interest.) Or he may include ouitliers, because he's looking for any explanation, because the literature in the area is so weak or inconsistent that he's looking for a base of knowledge.
 
 
But I don't want to hijack his thread anymore than I already have. We can continue this on my wall or via PM if you still have concerns about that.

I did it. On a crappy dumb phone too.
>pic related. It's my crappy dumb phone.

It's for science. Science is always important.

To tell you the truth, I was going to do the survey until I got to the writing part, then I just sort of…

$20 bucks say someone already did.

:I

Free Cake already admitted to doing it.

And I apologize, for A) wasting Teh Brawler's time and B) not taking this as seriously as I should have. It's just that the topic at hand seems so…casual, for lack of better words. It's like the "What candidate wold you rather have a beer with?" question. If the survey was on "What are your opinions and modern politics and the political process before and after taking a U.S. Government class?" which I did take about two weeks ago, I would have been much less joking about things. But I see now that that was not the right approach and apologize for it.

Seriously, I did fill it out.

So, brawler, when can we expect results?

Don't worry about it, you're fine. To be fair, this probably IS a very casual topic, but what makes it interesting is that no one I know of has researched it before. Besides video games, the Internet is the only medium and communication tool that's still developing an academic basis for understanding it, which means a lot of aspects of the theory behind it aren't officially understood. I mean, seriously, it blows my mind that I get to be a part of a website and a business that is LITERALLY the premiere source of Internet analysis and study. And I find this specific topic interesting because it's connected to the psychology, sociology, and communication theory that drives people's actions. It may seem that the question is easy to answer, but until there's real research into it, there's no way to be sure. And even though I don't have a degree (yet), the fact that I can do this and have it possibly be taken seriously simply because of how relatively new the Internet is is exciting to me. In short, I'm helping give birth to an academic field, and it makes me giddy.

Assuming I have enough results to actually come to a conclusion, I'm going to shut down the surveys tomorrow, and write the report over the weekend. Hopefully my lack of experience in writing research papers won't make this a complete mess.

May I ask what your standard or control was? I took the survey last week, and don't remember if you had included any questions pertaining to a non-pseudonym system.

If I'm not too busy I can clean it up for you, just shoot it over to me in an email and I can edit it up.

I feel like a bad person for not knowing what your gender is, Crimson.

It's because past knowledge is conflicting what is right there literally on the screen.

I didn't include a control because I didn't have a hypothesis or a comparable circumstance in the first place. This topic is so vague and unique at this point that it made less sense to treat it like a hypothesis and more as just a source of information on the subject.

I appreciate it, but Verbs has already been helping me. And I'm not so much talking about spelling and grammar as I am logic and does this make any sense whatsoever.

Ah, so it's mostly a compilation and analysis. I was just curious. I didn't know if you were going in with any expectations on the matter or not.

Yeah, most social science/humanities research using qualitative methods are usually exploratory from what I've seen. But that's the cool thing about research on the Internet. It's an entirely different interface when interacting and portraying oneself to others. Unless you can pull from other fields in regards to your background literature, your null hypothesis and alternative hypothesis would be based mostly on simple difference a la omnibus hypotheses (i.e., null=no difference among avatars in any regard; alternative=there is some difference among avatars in some regard.)

BUT WE ALL KNOW THESE THINGS BECAUSE WE'VE ALL BEEN IN SCHOOL FOR WAY TOO LONG.
 
Someone please give me a decently paying job…

I am a girl. Hope that settles some of the little worries in your head.

Thank you for covering that now.
I do feel relieved.

Looks like my avatar projection is just about perfect

I guess I will just break down my thoughts:

Avatars give a large first impression on me. Sombody comes in with a Game Character Avatar, I assume they play video games. They come in as an Anime Character, I assume they watch anime. Etc.

During this initial first impression stage, their avatar affects everything from how I read their voice to. If they change their avatar during this phase, my perception and voice for them changes as well.

However, as time goes on, and they change their avatar regularly enough and as I get to know them better, my mental image of them stabilizes. This stability is rarely ever thrown out of whack, thus even long after somebody like Len changes their avatar and name to Derpy, I still sometimes call him Len.

Same thing applies to drawing, after I learn how to draw somebody as a certain avatar, I will usually not draw them any other way.

However rare, even after some people become stable in my mind they sometimes throw a total curve ball at me out of nowhere and suddenly I am alienated from them…