Forum - View topicINTEREST: New Poll: Two-Thirds of Japanese Don't Care About AKB48
|
Goto page 1, 2, 3, 4 Next Note: this is the discussion thread for this article |
| Author | Message | ||
|---|---|---|---|
VORTIA
 SubscriberPosts: 944 |
|
||
|
As much as I hate AKB, this poll is really being read backwards - one-third of THE ENTIRE COUNTRY, man & woman, children & elderly, being interested at all in a single band is an amazing level of popularity!
Last edited by VORTIA on Sun Sep 11, 2016 1:01 am; edited 1 time in total |
|||
|
|||
|
revolutionotaku
Posts: 896 |
|
||
|
Their popularity bubble is about to burst.
|
|||
|
|||
TsubomiKoneko
 Posts: 247 |
|
||
 Uh... okay... that's really not that many people that were polled in the grand scheme of things (way less than 1% of the population of Japan). There's also the fact that AKB48 is going to be more popular among a certain age range/type of people (as most idol groups are likely to be).
The article title is really misleading, for a minute I thought someone literally polled all of Japan in order to see how much of it cared about AKB48 (which would be super weird)... Personally, since the majority of the 1st generation girls have graduated I've lost interest in the group other than checking how their new songs sound. I guess that's the problem with these kinds of groups, unless you're just in it for the music it's likely to not stick with you for long unless you're a fan of more than one girl. If they don't end up getting to participate in the Olympic stuff then that'll be sad for them, but I really don't think it'll be that big of a deal. |
|||
|
|||
Kadmos1
 Posts: 13616 Location: In Phoenix but has an 85308 ZIP |
|
||
|
If they surveyed at least a few million people, then that would be a bit more representative of nearly Japan's nearly 130 million people.
|
|||
|
|||
|
MisterLuck
Posts: 67 |
|
||
Yeah... This poll which consists of 1,376 respondents is stupid. Clearly this poll was created to trivialize people's perspective about AKB48. I guess Qzoo has an underlying hatred to AKB48, perhaps someone was denied a handshake. 
Well here is a fact for this POLLs silliness. Game of thrones season 6 finale was viewed by 8.9 million people in the USA. The USA has a population of 321.42 million, so that means 35/36 of Americans don't care about Game of Thrones... 
Last edited by MisterLuck on Sun Sep 11, 2016 1:25 am; edited 1 time in total |
|||
|
|||
Mr. Oshawott
 Posts: 6773 |
|
||
|
1,376 people is too small of a sample size to gauge fairly the popularity rise or fall of AKB48. The idol group has fans of just about any age, some higher (or lower) than others.
|
|||
|
|||
Top Gun
 Posts: 4792 |
|
||
|
I think some people in here need to read up on scientific polling. No, you do not need to interview millions of people to get a fairly-accurate reading of a population's opinion on some topic.
|
|||
|
|||
|
switchgear1131
Posts: 219 |
|
||
true, but you also need to poll more than 1.5 thousand people in a country of over a 100 million. The polls for many of these things are always so small. I would say at least 10,000 would be needed. Even more so a better indicator of their popularity would be album and ticket sales, what kind of advertisement deals they are getting and the like. If anyone has their thumb on what is popular (or they should to be doing their jobs right) it is the people paying them to advertise their products. |
|||
|
|||
|
FtKaru
Posts: 70 |
|
||
You could just as easily interview another 0.00001% of the population and get a different result that shows the opposite. As stated above, for a country with 130 million people, a sample size of 1376 is incredibly small. You're obviously not going to get results from millions of people, but that number seems quite low for a poll claiming what percent of the general population don't care about something. Maybe my perspective is skewed on this, but 30% still seems abnormally high to me. I was under the impression that AKB48 had faded from relevancy long ago. |
|||
|
|||
|
Top Gun
Posts: 4792 |
|
||
No, here's the thing that everyone gets wrong: you really don't need that many. So long as you choose your respondents by legitimate random sampling, it can be shown mathematically that surveying 1000 people will produce results within a few percentage points of what the overall population would produce. (This is the "margin of error" that's always mentioned with political poll results.) Statistics is a very mature field, and even if you don't have any formal experience with it (I don't really myself), it helps to at least have a basic grasp of how it works. Here's a nice little write-up on sample sizes in layman's terms that I was able to track down. |
|||
|
|||
zrnzle500
 Posts: 3768 |
|
||
|
Yeah this sample size is reasonable if election polling is any indication. While some have more or less than this one, they are usually in that order of magnitude. And given the response rate to polls, finding millions of people to answer polls in a way that gives legitimate random sampling is not plausible.
|
|||
|
|||
relyat08
 Posts: 4125 Location: Northern Virginia |
|
||
The sample size is fine, but yeah, this is being read totally wrong. Almost a third of the country is openly saying that they actually care about this one single group! That is a HUGE percentage of people. If I asked any random hundred people in my town if they cared about Miley Cyrus, I highly doubt it would be a third of them. Superhero movies? Same deal. Or, like MisterLuck pointed out, even Game Of Thrones. What about the Superbowl? Only a little over 100 million tune in to watch that every year, I guess nearly 2/3rds of the US doesn't care about the Super Bowl either.
Last edited by relyat08 on Sun Sep 11, 2016 12:24 am; edited 1 time in total |
|||
|
|||
Tenchi
 Posts: 4536 Location: Ottawa... now I'm an ex-Anglo Montrealer. |
|
||
|
What if a polling company did a poll on the number of Americans/Canadians/British/Australians/etc who care about a currently-popular English-language music act? Somehow I doubt you'd even get close to a third for any musical act you could name.
(I'm not naming names for my example because, to be honest, I'm so out of it when it comes to currently-popular music acts that any group or singer I can name is probably long past their peak popularity.) |
|||
|
|||
|
Utsuro no Hako
Posts: 1052 |
|
||
And you would be wrong. When dealing with large numbers of people, the total population has a negligible effect on the margin of error. A sample size of 1376 gives about a 2.5% margin of error at 95% confidence, and that would be true whether you're talking about Japan, the United States or China (in fact, most of the presidential election polls you see being reported right now have comparable sample sizes). Boosting the sample to 10,000 would only reduce the margin to .9%, which really isn't worth it for how much extra effort would be required. |
|||
|
|||
|
meiam
Posts: 3450 |
|
||
|
You can prove that a small number of people is enough to get a good statistic for the country, but that's using idealize method, often with homogeneous groups. The problem is that a population is compromise of multiple small group each with there own specific statistic, and most recruit method over sample some group and almost ignore others, the only way to fight this is to recruit a very large number of people across multiple means of communication.
For example if you conduct a poll using only landline phone versus a poll using only cellphone you'll get vastly different results simply because different people use different mean of communication. (I forgot where but I once saw a poll that ask people over landline what sex they were and they ended up with a population of well over 50% women because they tend to be the one who pick up the phone and accept to answer survey much more often than men). And as was mentioned, its insane that a third of the country is openly interested in this group, this doesn't throw cold water on there ambition, its throwing fuel to the fire. I doubt even at the top of there popularity the beatle could have claim as much. |
|||
|
|||
 |
All times are GMT - 5 Hours |
|
|
|
Powered by phpBB © 2001, 2005 phpBB Group