Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
To put this in perspective, a sample size of 3000 even randomized is relatively meaningless

That defeats the entire point of surveys. Do you understand what surveys are?

The sample size was almost 3,000, and it was random. How many more will satisfy you? 4000? 5000?

Hundreds of thousands of respondents is unlikely, and given timeframes, by the time the firm gathers the information the survey would have been rendered moot.

ex pede herculem. From the sample we judge (or better put, estimate) the whole.

The key is a decent sample size of *representative samples.*

Given the nature of the survey and how it was accessed, PriceGrabbers got the second part right.

Given that, generally, samples of 1,000 are well regarded for national surveys, they also got the first part right.

http://www.dfrank.com/accuracy.htm

With a sample of 100 the 90% level of confidence of any observed percentage is within +/- 8.2%. As the sample increases, the precision increases and the range of error gets smaller.

Thus, when we measure a 49% approval rate, say, with a sample of 1,000 people we can be 90% sure if we do the survey again, the results will be within 2.6 points of the 49%, or between 46.4% and 51.6%. We improve our precision quite a bit by going from 1,000 to 2,000 sample size (0.8 points), but only marginally by going from 2,000 to 3,000 (0.3 points). The selection of sample size should depend upon the consequences of coming to a wrong conclusion, and the cost of adding sample size.


Surveys are at best, indicators. Which is exactly what they are meant to be, and this particular one is no different.
 
Last edited:
Does this iPhone 5 purchase prediction factor in the umpteen million Verizon iPhones which appear to be stuck with IOS 4.2.xx for their remaining life? Since the firmware update problem has proven to be unresolvable over the last 4 months, perhaps it is a hardware problem requiring replacement of the VZW iPhone 4 with the new iPhone 5.
 
That defeats the entire point of surveys. Do you understand what surveys are?

The sample size was almost 3,000, and it was random. How many more will satisfy you? 4000? 5000?

You're wasting your time. These people never took or failed statistics and will never understand how it works. I see the same dumb comments on every survey, and trying to convince them is as useless as trying to talk to them about religion.
 
Does anyone know if HSPA+ and LTE require different chips and/or hardware?
Right now, they are two separate chips which is why LTE handsets tend to be bulky and suffer from poor battery performance. That's two big reasons why Apple probably won't include LTE in this year's iPhone.

The third strike against LTE (right now) is rudimentary market adoption. Apart from Verizon's fledgling LTE network and a handful of Scandinavian countries, LTE has not been widely deployed. Most of the world will deploy their LTE networks over the next 12-18 months, going operational in late 2012 or 2013.

The most likely scenario at this point is an HSPA+ capable iPhone. Historically, Apple has released handsets based on widely deployed cellular network technology. HSPA+ would certainly fall under this description. LTE, not so much.
 
Does this iPhone 5 purchase prediction factor in the umpteen million Verizon iPhones which appear to be stuck with IOS 4.2.xx for their remaining life? Since the firmware update problem has proven to be unresolvable over the last 4 months, perhaps it is a hardware problem requiring replacement of the VZW iPhone 4 with the new iPhone 5.
Nah, the CDMA iPhone 4 will get iOS 5, making this a non-issue.

Clearly, Apple has deemed that putting more development resources in adding iOS 4 functionality for CDMA users would hamper development of iOS 5. CDMA iPhone 4 users simply need to exercise some patience.
 
"Survey: 35% of U.S. Consumers Planning to Purchase Next-Generation iPhone"

Bzzt. Wrong.

Either the OP doesn't understand statistics or, more likely, Macrumors is trying yet again to artificially boost pageviews with misleading headlines.

35% of RESPONDENTS, which according to the original article do not appear to be random but self-selected online consumers, indicated they would purchase the next gen iPhone.

To put this in perspective, a sample size of 3000 even randomized is relatively meaningless as it's less than 1/1,000,000th of the US population. Even less meaningful when you consider that the sample is actually nonrandom, and that the particular set of respondents being online could easily have skewed the results skyward.

Bzzzt.... You're partially right. Yes, they should have said "35% of on-line consumers are looking to buy the iPhone 5". That would have been more accurate.

But the rest of what you say is really not important in statistics. I've done tons of surveys over the years and trends don't lie unless you're really off on your audience. What I mean is, survey 3000 people on welfare, or 3000 members of the Millionaire Club and you're going to get skewed information. Assuming they get a wide variety of on-line shoppers, you don't need a huge sampling to see the trend. My guess is it's pretty accurate for people who shop on-line - or that could be easily translated into most American consumers since the profiles are so close.
 
When people talk about inexpensive or free phones in return for contracts...

... they shouldn't forget to include the $9 AT&T 3GS refurbs, or the $49 3GS new phones, or the cheap or free iPhones of all types that have been available outside the USA on contract.

In Japan alone, iPhone sales jumped from a few hundred thousand up to several million, only after it was offered for free or nearly so.

Definitely a lower price brings many people in, but Android phone sellers aren't the only ones offering deals.
 
Earlier in the year, I said that I will wait until 2012 when I assume an LTE iPhone will come out. My iPhone 4 is awesome and I'm sure it will serve me well into 2012.

The closer we get to the iPhone 5, the more I want a new iPhone. I'm pretty sure when it finally gets announced, I'm going to be upgrading.
 
The set of people who voluntarily choose to complete a survey on a particular website is nowhere near random.

It's a random sampling of a representative sample.

Perfectly fine, unless you don't understand how surveys work.
 
Last edited:
It makes sense. The fact that iOS is managing to stay competitive with Android, while also costing the consumer a good $100-$200 subsidized, means that people REALLY like iOS. If there was ever a lower priced model subsidized to free, I don't think Android could compete. Many people buy Android for the sole fact that it's a free phone.

I know in the US you can't get an iPhone for free, but in many countries you can. Including mine, and I think the iPhone isn't doing quite as good here as in the US. The number of iPhones I'm seeing in the public seems to have decreased since last year and the number of Android handsets I'm seeing has certainly increased
 
It's a random sampling of a representative sample.

Perfectly fine, unless you don't understand how surveys work.

Given how people responded the other day about iTunes pricing with the Amazon app, and this issue with surveys, looks like Mac Rumors should offer some introductory Sales, distribution and marketing classes to its members! LOL!

Oh... almost forgot, legal classes too for all the armchair lawyers in this group! :p
 
"Survey: 35% of U.S. Consumers Planning to Purchase Next-Generation iPhone"

Bzzt. Wrong.

Either the OP doesn't understand statistics or, more likely, Macrumors is trying yet again to artificially boost pageviews with misleading headlines.

35% of RESPONDENTS, which according to the original article do not appear to be random but self-selected online consumers, indicated they would purchase the next gen iPhone.

To put this in perspective, a sample size of 3000 even randomized is relatively meaningless as it's less than 1/1,000,000th of the US population. Even less meaningful when you consider that the sample is actually nonrandom, and that the particular set of respondents being online could easily have skewed the results skyward.

A self-selecting group doesn't necessarily make the results invalid, just casts it in a questionable light. You have to ask yourself if the source of the self-selection in any way influences the outcome of the question.

For example, if I put a survey up on MacRumors or AndroidCentral about buying an iPhone, you can bet the results are going to be heavily skewed one way or the other. However, if I put that same survey up on eBay or the front page of Yahoo! then there's a greater chance that even a self-selecting group will yield meaningful results.

You also have to consider how the survey is promoted and what's bringing in the self-selecting group. If it's inviting people to answer with "Do you plan to buy an iPhone" then you're probably going to get skewed results. If it's a more neutral question along the lines of "Which phone do you plan to buy next?" then it's less suspect.

So, bottom line, I agree with your comment about self-selection, but there are finer points to this that you overlook.

I disagree completely with you on the random selection point, however. You are entirely wrong about that. A truly randomized selection can yield strikingly accurate results even with a very small percentage of the overall pool.
 
HSPA+ IS 4G. It's not LTE, but real world testing shows it to be nearly as fast or even faster than any other network tech out there in the real world today.

Your links were to comparisons with WiMax and 3G. They left out LTE.

ATT's website says their "4G" can go up to about 6 Mbps.

Real life tests with Verizon's LTE average 6-9 Mbps, and people been reported speeds at least a couple of times faster than that.
 
Your links were to comparisons with WiMax and 3G. They left out LTE.

ATT's website says their "4G" can go up to about 6 Mbps.

Real life tests with Verizon's LTE average 6-9 Mbps, and people been reported speeds at least a couple of times faster than that.

Look at the table here (and read the text if you like). In real world testing, LTE isn't showing to be any better than any of the other 4G technologies out there. In the real world there's no meaningful difference between any of the 4G technologies.

EDIT: whoops, forgot the link. http://www.phonearena.com/news/Verizon-LTE-vs-T-Mobile-HSPA-vs-Sprint-WiMAX_id16416
 
Last edited:
Not me. I'm very happy with the Android phone I switched to, which allows me to do things Apple is only planning to copy in their next two OS releases.
 
Of these respondents, 51 percent indicated that they will buy the smartphone within the first year of release, 30 percent will purchase it before the end of 2011, 14 percent will buy it within the first month, and 7 percent will buy it within the first week.

I'm in that 35%, and fall into the 14% (first month) if not 7% (first week) (depends on when payday is ... :) )

Chomping at the bit waiting for the opportunity to finally sink my Blackberry Tour POC once and for all. But, unwilling to spend full price on a phone which is over a year old and due to be updated within a few months.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.