210 réponses à ce sujet

[Zine2]

Cazlee wrote...

evisneffo wrote...

...Did that poll really happen? Wow. But I'm curious: what would be the use of conducting a test-screening-type poll the day before release? It would obviously have been too late to change anything should the poll have indicated that that was necessary.

There isn't any valid reason to do the test screening a day before release. That's why anyone that believes the author is seriously lacking in common sense.

It actually makes sense given this context - the ending was kept hidden from the team until the last minute (as detailed in the Weekes post). Team did not like ending and tried to prove it by getting poll data.

[Ziggeh]

count_4 wrote...

Just so we're on the same page here: Could you explain what kind of bias you are referring to? 
I'm in no way proficient in this area so I'm just comparing the bias in internet polls with the dev poll here. For me the bias in internet polls comes from the fact that unhappy players are more likely to vote than happy ones a.k.a silent majority.
As the dev poll requires everyone in the audience to vote after experiencing the ending without any prior knowledge, where would any bias come from that makes the poll non-representative (given the audience is diverse enough and not just a hundred 14yo)?

Internet polling is a whole different barrel of bias, but some concerns to look for in this one, or the top of my head, because it's early:

"BioWare leaked the endings to its fans and polled them to gauge their
reaction"

How does one select a random sample when the intitial population are people invested enough to have made themselves known as "fans"? You need to know what that's based on to use the data, but even if it's something fairly wide like previous purchasers, that doesn't account for people who weren't fans previously. It would be exclusive, and therefor wouldn't represent the whole of your population. If it was even wider and by "fans" it means "some people" then it's including people who are irreleveant.

After that it's your standard things like: selection. If the poll was say, conducted by e-mail you'd have yourself a self selection poll again, only the extremes would respond. sample size, leading questions etc

[Zine2]

MikeC99 wrote...

Re BSN poll - what basis do you have to assume that because the poll is based in BSN that it is a biased sample in terms of good or bad ending? I don't see that as proved at all. It's a 'biased' sample only in so far as it would be reasonable to assume that only people who have an interest in Bioware games would bother to register. I just don't see any logic in assuming that BSN = end haters per se. End lovers had the same opportunity it seems to me.

Yep, which is also why I noted how the ME3 Origins Registration requirement is potentially even blurring the line even more. Everyone who plays the game gets an account now.

[SimonTheFrog]

For me the biggest complaint is "Space magic" and related to that incoherence with the universe described in the trilogy.

If that would be fixed i could live with a rail-road ending.

[Zine2]

Ziggeh wrote...
How does one select a random sample when the intitial population are people invested enough to have made themselves known as "fans"? You need to know what that's based on to use the data, but even if it's something fairly wide like previous purchasers, that doesn't account for people who weren't fans previously. It would be exclusive, and therefor wouldn't represent the whole of your population. If it was even wider and by "fans" it means "some people" then it's including people who are irreleveant.

After that it's your standard things like: selection. If the poll was say, conducted by e-mail you'd have yourself a self selection poll again, only the extremes would respond. sample size, leading questions etc

You're stretching here. My turn to drag you from the brink

First of all, we don't actually know the definition of "fan". Metholodogy, as you pointed out, is unknown.

Secondly, and more importantly, any poll regarding a very specific product like ME3 must in fact be limited to either the people who are playing, or will potentially play the game. Complaining about how a poll only covered Mass Effect 3 players and not the whole population - including toddlers and old people who don't play video games - is on the silly side.

You can, in fact, limit yourself to only a segment of the population when doing a survey, and that segment can be "current ME3 players". You just need to specify this segment as part of the study.

[Ziggeh]

Zine2 wrote...
Dude, check the statement before that. I did not say BSN Player Base == Entire Player Base. I said "If X, Y, and Z hold true, THEN that is true".

X, Y and Z cannot hold true. By definition we are most invested as a group than the population in question.

Zine2 wrote...
You're the one making the assumption that internet polling is always beyond random. It often is, I'm pointing out it's not always.

Always is. As I say, it's not simple selection that's wrong with an internet poll. It is impossible for them to remove bias.

Zine2 wrote...
But you cannot prove that it actually has selection bias simply because it's an Internet poll. You have to show that it does in fact significantly diverge from the results of a poll that DOES conform to random sampling.

No, it's inherent. You literally have to select if you want to take part. Whether it's random by accident or not, it's a self selection poll before it even starts.

Zine2 wrote...
Competently, like a major company should?

I'm saying it's not possible. Competent or not.

Modifié par Ziggeh, 18 avril 2012 - 10:03 .

[Zine2]

Ziggeh wrote...

Zine2 wrote...
Dude, check the statement before that. I did not say BSN Player Base == Entire Player Base. I said "If X, Y, and Z hold true, THEN that is true".

X, Y and Z cannot hold true. By definition we are most invested as a group than the population in question.

No, you're totally wrong now. Origins registration is mandatory, your "by definition" argument has been obliterated.

No, it's inherent. You literally have to select if you want to take part. Whether it's random by accident or not, it's a self selection poll before it even starts.

Again: Not all polls are meant for the whole population.

I'm saying it's not possible. Competent or not.

No, you're simply talking nonesense now.

Modifié par Zine2, 18 avril 2012 - 10:07 .

[Ziggeh]

Zine2 wrote...
First of all, we don't actually know the definition of "fan". Metholodogy, as you pointed out, is unknown.

Sort of my point. Not knowing is the problem.

Zine2 wrote...
Secondly, and more importantly, any poll regarding a very specific product like ME3 must in fact be limited to either the people who are playing, or will potentially play the game. Complaining about how a poll only covered Mass Effect 3 players and not the whole population - including toddlers and old people who don't play video games - is on the silly side.

...

The people who have completed the game are a population, not the population. It just means the group in question.

Zine2 wrote...
You can, in fact, limit yourself to only a segment of the population when doing a survey, and that segment can be "current ME3 players". You just need to specify this segment as part of the study.

Right, and how does one do that before the game has shipped?

[Zine2]

Ziggeh wrote...
Right, and how does one do that before the game has shipped?

Easy. Preorder list

[Cazlee]

Zine2 wrote...

Cazlee wrote...

evisneffo wrote...

...Did that poll really happen? Wow. But I'm curious: what would be the use of conducting a test-screening-type poll the day before release? It would obviously have been too late to change anything should the poll have indicated that that was necessary.

There isn't any valid reason to do the test screening a day before release. That's why anyone that believes the author is seriously lacking in common sense.

It actually makes sense given this context - the ending was kept hidden from the team until the last minute (as detailed in the Weekes post). Team did not like ending and tried to prove it by getting poll data.

The DAY before release?  SPOIL the ending to fans that have yet to play the game? HOW does that make sense?
As I've mentioned in previous posts the author provides no evidence to believe his information, or to justify why Bioware would pull that stunt.  Anyhow, continue defending the merits of an imaginary poll from a misinformed journalist...

[Ziggeh]

Zine2 wrote...
No, you're totally wrong now. Origins registration is mandatory, your "by definition" argument has been obliterated.

And everyone on origin spends time on BSN? The chances of seeing said poll aren't determined by the amount of time spent? And the people who spend more time here aren't more invested than people who don't?

Zine2 wrote...
Again: Not all polls are meant for the whole population.

Heh, no, but always for a population. In the case of internet polls that population will always be "the people who voted in this poll" and nothing else.

Zine2 wrote...
No, you're simply talking nonesense now.

Nope. There is no way to determine our given population prior to release and so no way to accurately sample it.

[Swisspease]

Bioware could possibly have conducted that poll during their launch party held the day before launch. Pure conjecture of course.

Also read my quote by Mark Twain below and you'll get an idea of what I think about statistics.

[Ziggeh]

Zine2 wrote...

Ziggeh wrote...
Right, and how does one do that before the game has shipped?

Easy. Preorder list

And that introduces a bias towards more invested players, eliminating people who purchase games on a whim.

[Banelash]

Wait why is this all about the skewing of internet polling now?

1) this was done by Bioware
2) selected group of people
3) done a day before release

This is a test screening, before everyone else got to play it. Minds fresh without any swaying from either camp. How can you say they are biase?

Edit: Also, isn't BSN forums poll limited to people that actually have the game? Like you have to register the game to this account first? Or can I register 6 accounts to 1 game.

Modifié par Banelash, 18 avril 2012 - 10:14 .

[Zine2]

Cazlee wrote...
The DAY before release?  SPOIL the ending to fans that have yet to play the game? HOW does that make sense?
As I've mentioned in previous posts the author provides no evidence to believe his information, or to justify why Bioware would pull that stunt.  Anyhow, continue defending the merits of an imaginary poll from a misinformed journalist...

Movies do, in fact, show advanced screenings and spoil "fans" many days in advance of release. THat's the whole point of a test screen.

The point is that you do it to a small audience under NDA.

[Psythorn]

I will post this wherever it makes sense - again and again and again:
The hero does NOT need to die to make an ending "artistic". The ending does not even need to be bittersweet to be artistic. I play games for amusement to be entertained and to be distracted from a world that has enough of sadness and bittersweet... I do not get why bittersweet endings are so "hip" in the genre currently. It's OVERDONE !
And in case of ME3 - it's not even done very good. It gives no feel of closure. It's just bitter. It feels forced.
And no it's not that -I'm unable to accept a story to end - even if it is bitter sweet - if it feels right (like Cirsis Core or Assasins Creed Revelations) I would still not complain or might even like it.

[Zine2]

Ziggeh wrote...

Zine2 wrote...

Ziggeh wrote...
Right, and how does one do that before the game has shipped?

Easy. Preorder list

And that introduces a bias towards more invested players, eliminating people who purchase games on a whim.

Except that's not true. Majority of sales are on the first week, including pre-orders, so it's not necessarily biased towards "invested" players. That's just your unsubstantiated assumption.

[Ziggeh]

Banelash wrote...

Wait why is this all about the skewing of internet polling now?

1) this was done by Bioware
2) selected group of people
3) done a day before release

This is a test screening, before everyone else got to play it. Minds fresh without any swaying from either camp. How can you say they are biase?

Bias isn't exclusive to internet polls. It's an issue with all polling.

I'm not even convinced such a poll exists, but giving the guy the benefit of the doubt, the bias begins at step 2. Selecting a representative sample if a hugely tricky business.

[Cazlee]

Zine2 wrote...

Cazlee wrote...
The DAY before release?  SPOIL the ending to fans that have yet to play the game? HOW does that make sense?
As I've mentioned in previous posts the author provides no evidence to believe his information, or to justify why Bioware would pull that stunt.  Anyhow, continue defending the merits of an imaginary poll from a misinformed journalist...

Movies do, in fact, show advanced screenings and spoil "fans" many days in advance of release. THat's the whole point of a test screen.

The point is that you do it to a small audience under NDA.

Lol feedback from  test screenings are used to gauge whether to alter the movie before it is released. There's no point in doing it the day before release (in this case weeks after the game is already gold).

Modifié par Cazlee, 18 avril 2012 - 10:20 .

[Ziggeh]

[quote]Zine2 wrote...

And that introduces a bias towards more invested players, eliminating people who purchase games on a whim. [/quote]

Except that's not true. Majority of sales are on the first week, including pre-orders, so it's not necessarily biased towards "invested" players. That's just your unsubstantiated assumption.[/quote]
Sorry, you might have to explain to me how "people who agree to buy the game before release" includes people who pick games up based on what looks good on a shelf.

How can pre-orders not be biased towards people who are more interested in the game?

[Zine2]

Ziggeh wrote...
And everyone on origin spends time on BSN? The chances of seeing said poll aren't determined by the amount of time spent? And the people who spend more time here aren't more invested than people who don't?

No, I'm pointing out it's blurring the lines. And I'll point out all of the above are merely assumptions and not backed up by actual numbers.

Heh, no, but always for a population. In the case of internet polls that population will always be "the people who voted in this poll" and nothing else.

Again, you are fundamentally misunderstanding statistics and making incorrect sweeping assumptions.

"People who voted in this poll", yes. "Nothing else", not always.

Nope. There is no way to determine our given population prior to release and so no way to accurately sample it.

Again, wrong. It won't be 100% certain, but you can in fact make reasonable assumptions based on picking stuff like from Preorder lists. You're  asking for a ridiculous - indeed nearly impossible - burden of proof.

[Zine2]

Ziggeh wrote...
Sorry, you might have to explain to me how "people who agree to buy the game before release" includes people who pick games up based on what looks good on a shelf.

How can pre-orders not be biased towards people who are more interested in the game?

I'll just apply your own ridiculous standards on you.

Prove that people who pre-order are more invested in the game.

Don't give me assumptions. PROVE it.

[Ziggeh]

Cazlee wrote...

Lol feedback from  test screenings are used to gauge whether to alter the movie before it is released. There's no point in doing it the day before release (in this case weeks after the game is already gold).

They're also done to determine things like target audience and used to decide marketing strategies and release levels.

[Cazlee]

Ziggeh wrote...

Cazlee wrote...

Lol feedback from  test screenings are used to gauge whether to alter the movie before it is released. There's no point in doing it the day before release (in this case weeks after the game is already gold).

They're also done to determine things like target audience and used to decide marketing strategies and release levels.

You don't undertake a major project like that without already knowing who your target market is. Anyhow even if you didn't, the marketing strategy would not be determined one day before release.

[Ziggeh]

Zine2 wrote...

Ziggeh wrote...
And everyone on origin spends time on BSN? The chances of seeing said poll aren't determined by the amount of time spent? And the people who spend more time here aren't more invested than people who don't?

No, I'm pointing out it's blurring the lines. And I'll point out all of the above are merely assumptions and not backed up by actual numbers.

Erm, beg pardon? You need numbers to believe those are true statements? You want statistical proof that not everyone who owns origin uses the bioware social network?

Zine2 wrote...
Again, you are fundamentally misunderstanding statistics and making incorrect sweeping assumptions.

"People who voted in this poll", yes. "Nothing else", not always.

Please explain my fundamental misunderstanding. I'll grant you not always, but it would have to be a very unsual and specific population.

Zine2 wrote...
Again, wrong. It won't be 100% certain, but you can in fact make reasonable assumptions based on picking stuff like from Preorder lists. You're  asking for a ridiculous - indeed nearly impossible - burden of proof.

Jesus H Christ. That's my entire point! It's near impossible. "I'm" not asking for it, I'm saying that for it to be wholey representative of the population in question it has to account for these things, which it can't.