Averages

[InsaneWaya]

Ok I've been going back and forth on if it really is relavent in any way but i want to see what people think. When you log on and view your profile theres always that "Global Average" thing and then under it there is what you have given and what you have gotten. Now can we really consider this a average? I mean think about it, the most popular AMVs will usually get more ops, and usually to get a popular AMV it has to be better than most. So since a majority of ops go to many of the better AMVs can it really be considered a true average? Also, if you see a very bad AMV you usually wouldn't spend the time to give an op, dont lie, I've seen it and done it before, i think if its worse people don't enjoy giving the op as much so the AMVs that are bad/need work get ignored and the average is thrown off more. Im a little curious, what do you personally feel about this? Can it really be considered a true average? I don't know if this has ever been gone over, im just doing this on a whim, sorry if it has been.

[dragontamer5788]

Opinions, like any other voluntary statistic, are inherently biased. Respondents will tend to be the guys who really liked the video (particularly, because you need to make an opinion to put it in your "favorites" list).

Problem is, there isn't much more that can be done, unless you become big brother and force a truly random subset of the AMV community to judge a random set of AMVs. So yeah, it is awfully hard to collect rigorous statistics under these conditions.

Nothing wrong with that, we just have to keep in mind the problems of voluntary surveys and other possible biases. After all, very few surveys actually qualify as a simple random survey. (the assumed survey that most statistics are based on).

Lets put it this way. If a video is really popular and has 100,000 downloads... the number of opinions it has will be probably in the hundreds. So... maybe ~0.5% of the downloaders actually made an opinion. What about the other 99.5% of the downloads? Regardless, I doubt it is even possible to get a rigorous statistic unless peopled are paid to do the reviews...

[CrackTheSky]

It should be pretty obvious that it's not a "true average", since the numbers all range between 7 and 9, and not 4-6 which would indicate a true bell curve. However, of the opinions that have been given, that is the mathematical average. If you compare your scores against the global average, you'll just get numbers relative to what's been given on the .org.

In other words, the bell curve has just been shifted a few numbers to the right, and that's what you compare to.

Logically it shouldn't matter, because the fact that bad videos don't get as many ops should reflect on your own videos as well. As long as the number of positive ops you get is decently larger than the number of negative ops you get, you should be able to compare to the global average pretty accurately (insofar as something this subjective can be "accurate").

[Knowname]

if you see a very bad AMV you usually wouldn't spend the time to give an op

this is why true assholes are a dying breed >.<

[Knowname]

also I attest that 7 IS average, not 5. 5 is fucking fail. we should all strive too attain a 7, it's not average as in ~meh~ what a homely chick. It's average as in a competitive baseline. When talking about a non-competitive space like math (based on facts) 5 is average, but amvs is not fact, amvs is art, art is NOT factual AT ALL. 7 is the baseline we should all strive to be at at the very LEAST. There is a difference between a competitive average and average as a fact. Who teaches this?? Public schools?? lol. No wonder the world is in such chaos. Now I digress

[CrackTheSky]

also I attest that 7 IS average, not 5. 5 is fucking fail. we should all strive too attain a 7, it's not average as in ~meh~ what a homely chick. It's average as in a competitive baseline. When talking about a non-competitive space like math (based on facts) 5 is average, but amvs is not fact, amvs is art, art is NOT factual AT ALL. 7 is the baseline we should all strive to be at at the very LEAST. There is a difference between a competitive average and average as a fact. Who teaches this?? Public schools?? lol. No wonder the world is in such chaos. Now I digress

First of all (and I really don't want to derail this thread), art isn't competitive so if you're going to argue that because art isn't based in facts like math is, you can't automatically equate it with competition. There can be competition within art, but it's not intrinsically competitive, like you seem to be assuming. And I don't think in the sense of leaving ops, AMVs are competitive, so your terminology is confusing. Secondly, based on your logic, art should rate higher than mathematics dictates is "average". Why? I don't understand why we should strive to achieve a global average around 7.

Additionally, I've noticed (and I do this myself) that people, when they leave ops, if they have a negative opinion about a video, don't tend to rate accordingly (at least in a mathematical sense). For example, "low" numbers are 5-7, not 1-4. You rarely see honest opinions that dip into the lower end of the scale. I think this is probably because the global average is skewed to the right so people tend to skew EVERYTHING to the right. Or maybe people are afraid of upsetting the editor in question. Fact is, 5 is the center of the scale. 5 is "average", and people shouldn't sulk to see fives on ops they receive.

[simofc90]

And I don't think in the sense of leaving ops, AMVs are competitive

actually they are if you count the 10% list, that's because generally imho opinions are more useless than usefull

[CorpseGoddess]

Opinions, like any other voluntary statistic, are inherently biased. Respondents will tend to be the guys who really liked the video (particularly, because you need to make an opinion to put it in your "favorites" list).

Problem is, there isn't much more that can be done, unless you become big brother and force a truly random subset of the AMV community to judge a random set of AMVs. So yeah, it is awfully hard to collect rigorous statistics under these conditions.

Nothing wrong with that, we just have to keep in mind the problems of voluntary surveys and other possible biases. After all, very few surveys actually qualify as a simple random survey. (the assumed survey that most statistics are based on).

Lets put it this way. If a video is really popular and has 100,000 downloads... the number of opinions it has will be probably in the hundreds. So... maybe ~0.5% of the downloaders actually made an opinion. What about the other 99.5% of the downloads? Regardless, I doubt it is even possible to get a rigorous statistic unless peopled are paid to do the reviews...

I've been thinking for a while that it would be cool to have a "this week's 10 random vids" on the members' page. Just pull randomly from the entire category---it would give vids that have been around for ages with no downloads a chance to get an audience. Things get lost in the shuffle very quickly here, I think.

Just my two cents.

[Qyot27]

(particularly, because you need to make an opinion to put it in your "favorites" list)

When did that happen?

[mirkosp]

(particularly, because you need to make an opinion to put it in your "favorites" list)

When did that happen?

Never. I just suppose that he never noticed that there's the box to add a video to the favs under the video comments.