First off I bough an Android Tablet about 2 years ago. POS. Total POS. After 12 months I got no updates. Many of my apps no longer worked, ton of my apps crashed, and the apps look like crap since they were phone apps blown up. The touch screen was unresponsive, cpu slow as hell, tons of lag. Basically it was such a horrible user experience. Just got the iPad Air. I'm in heaven. This is what a tablet suppose to be.
Got my Android phone about 12 months ago. Same story. I was stuck on an old version of Android. Apps crashed constantly. Some apps don't even open. CPU is slow, TONS OF LAG. Build quality is a joke. Basically the same experience as my Android tablet except shrunken down. Got an 5S a week ago. Amazing. I'll never go back to Android. I don't care how much more 'expensive' Apple is. I put expensive in quotes because after factoring in resale value, user experience, time saved, ect it ain't more expensive.
I just sold my old 16GB iPhone 4 for $220. The resale value for iPhones is incredible totally nullifying all the "iPhones are too expensive" crap.
Sample too small, probably taken here on US. Here the iphone is considered as "cool" phone.
I don't see many users downgrading from a FULL HD premium smartphone to a tinynon-HD smartphone like the iphone, it's just not a smart choice.
However, always there are people who do not necessarily make smart decisions.
LOL. Go buy a TV, dude. What's good to have a hi resolution screen that you eyes can't tell the difference? I don't care if the phone screen go 4k because it's not different from retina with human naked eyes.
Articles like these usually attract non-iPhone users so I'd like to hear what it is that makes people switch from Samsung to Apple.
How about: Not wanting to look like an idiot with a tablet-sized phone pasted along side of their faces? Or, not wanting to make the same dumb mistake twice?
I just sold my old 16GB iPhone 4 for $220. The resale value for iPhones is incredible totally nullifying all the "iPhones are too expensive" crap.
I switched to an iPhone 5 a year ago. Was offered $20 for my mint condition HTC touch phone on eBay. There is usually a $100 minimum spread between a newer high-end Android and an older model iPhone on eBay...
If Apple makes a 5" or 6" iPhone, there will be no reason to use a Droid other than it's cheaper or you hate Apple.
Why stop at 5 or 6 inch screen sizes? Go all the way to 7.9" and get the iPad mini with built-in radio. That way no one will be in doubt that you don't mind looking like a dufus when on the phone.
I see posts like this and often wonder what the heck you were using for a phone. My first smartphone was a Droid X by Motorola when it was first released back in 2010. The iPhone wasn't an option as I was on Verizon and who the hell knew if it was ever coming. The Droid X was amazing phone and it never failed me in the two years I had it. When my two years were up I figured I'd give the iPhone a shot as it was my initial choice for a smartphone. It took some getting use to coming from Android, but after playing with the phone I could see why people loved it. I obviously ended up staying with the iPhone but my Android experience was a good one.
The reason your Droid X was "amazing" was because you were most likely using a feature phone (probably a flip phone?) and Droid X was presumably your first touchscreen smartphone. This is equivalent to upgrading from a horse buggy to a cheap Chinese-made motor vehicle (or a Yugo for those who are old enough to remember what a Yugo is). So your experience with Android would obviously be a good one given that you really had nothing to compare it against.
Wrong. My interpretation of the data presented was completely sound. I have a social science doctorate. Any freshman in an intro stats class will know YOU have no idea what you're talking about.
Really? Why don't you then tell us what the margin or error -- say, at the 95% level -- is likely to be with calculations of sample proportion = 16% in the data, when N = 400?
LOL. Go buy a TV, dude. What's good to have a hi resolution screen that you eyes can't tell the difference? I don't care if the phone screen go 4k because it's not different from retina with human naked eyes.
On a tiny 4" screen hell not, but on a bigger smartphone... yes you can notice the difference.
With higher resolution you can see more on your screen and that is a lot better for browsing web pages and enjoy multimedia.
I agree with you that 4K will be too much for a smartphone but for a tablet or TV it will be nice.
On a tiny 4" screen hell not, but on a bigger phone... yes you can notice the difference.
So you’re saying that the human eye can tell the difference between things it physically cannot tell the difference?
Now, Super Hi-Vision has a place on televisions over 42”. I’ve stood 20’ away from one of those 70” Sharp numbers and the pixels are big as hams. It was disgusting and horribly distracting.
I just sold my old 16GB iPhone 4 for $220. The resale value for iPhones is incredible totally nullifying all the "iPhones are too expensive" crap.
You say this, but lots of people don't end up selling their old phones. Many old phones are simply sitting in drawers, or more commonly, people update their phones because their old stopped working. You can claim that apple's phones are more reliable than android, and it might be true (probably is), but when your phone goes for a swim, or you drop the phone and the screen cracks, reliability doesn't come into the question.
Many people buying new phones can't rely on getting back $200+ for their old iphone, because their old iphone no longer works. It helps the people that have functioning devices (and actually sell them), but I bet they're actually in the minority. Reading on here the comparisons between Android and iOS are pretty amusing. I switched this year from iOS to android after my iphone 4 went for a swim. I now use a Galaxy S3.... I don't know if I'd call the phone "better" than an iphone 5 (the best available at the time), but it was relatively competitive for my needs. Additionally, the S3 cost far less than the iphone. I paid $400 for an unlocked S3, an unlocked iPhone 5 would be what.... $700? In retrospect, I probably should have saved more money and gotten the Nexus, but I wanted to have an SD card slot on my phone.
I don't expect the same quality out of a $400 phone that I'd get out of a $700 one . . . But overall, I have no regrets. I don't really miss the iOS ecosystem, and I like the freedom I get with Android. Sure there's a worry of Malware, but that is so overblown on these boards, as having 1000X more likely to get malware on Android than iOS (or whatever the number) is still meaningless, as the numbers are basically zero to begin with.
Really? Why don't you then tell us what the margin or error -- say, at the 95% level -- is likely to be with calculations of sample proportion = 16% in the data, when N = 400?
Despite what you might think, I don't jump when people like you say jump. I have nothing to prove. What I will say is that margins of error mean nada when there is no data AT ALL about the population that provides context for a statistic. Furthermore, an N of 400 is uninterpretable when the size of the population and characteristics of the sample are undefined. Anyone can play with sample size to increase power and artificially boost significance.
And IIRC, in some other thread you challenged me to jump through hoops to prove my knowledge. I won't. You're disputing MY statement. I'm really leaving it at that, Mr. 11,000 posts, becasue I have a job. That I work at.
Despite what you might think, I don't jump when people like you say jump. I have nothing to prove. What I will say is that margins of error mean nada when there is no data AT ALL about the population that provides context for a statistic. Furthermore, an N of 400 is uninterpretable when the size of the population and characteristics of the sample are undefined. Anyone can play with sample size to increase power and artificially boost significance.
And IIRC, in some other thread you challenged me to jump through hoops to prove my knowledge. I won't. You're disputing MY statement. I'm really leaving it at that, Mr. 11,000 posts, becasue I have a job. That I work at.
You're the one that made a bombastic statement about the data/analysis being crap. I called you out on it, and you respond by telling me you have a PhD, I post too often, and that you have a job. In other words, you completely avoid the question.
The population size in this case is in the many many dozens of millions, as we know. So, whether we pick 10M, or 20M, or 200M, the margin of error will be largely invariant. You should know that.
As to my specific -- and trivially simple -- question to you about the MoE for the "16%" number, it would be calculated as SQRT[p*(1–p)/N], where p = 0.16 and N = 400. That turns to be a rather low MoE of ±1.83% at the 95% level. 16% ± 1.83% suggests a significant shift from the prior proportion of those switching from Android to iPhone. (All this assumes that the sample was truly random; but there's no reason to assume it's not, unless otherwise known).
Of course, you knew all that, right?
When you post next and are called out on it, back it up.
Ummm doesn't one need a bit more information? If one conglomerated OS sells 2-3X of the other OS, how does it make sense to compare percentages for switching?
Say Android had 300 million phone sales last year vs 100 million for iOS (purely theoretical numbers), then 20% of Android to iOS would be 20million users. 10% of iOS to Android would be 30 million users. Or did I miss something and did this company actually normalize for this?
It depends what your purpose is. If you want to be mathematically correct, then your post above is right. If you're more concerned about posting something that sounds more favorable to Apple, it is better the way it is.
I'm one of those Android switchers. I'd been stuck with US Cellular due to cell reception in my area. I've been waiting 6 years to get an iPhone, but had to use Android in the meantime. Android was alright, but had no interface with my Apple ecosystem. Wherever I went I carried my iPod Touch with me. US Cellular started selling the iPhone November 8th and I was waiting in line to ditch my Galaxy S3!
both the 5S and Note3 have battery life to last the whole day. Of course the Note3 does since it has a massive and heavy battery.
All your arguments of having a bigger screen are fine and dandy. But the iPadMini Retina would make your Note3 look like a chump. Much larger screen and higher resolution. It's all about preference. For me I want my phone to be a phone. Not a tv in my pocket. Note3 is a compromise product. Too big to be a phone and way too small to be a real tablet. Its a great product if all you can afford is one device. Personally I bought a 5S/iPadAir. The Air is one pound and is a hell lot better of a tablet than the Note3. And my 5S does not break my pants pocket.
Don't lie about the Note3. It LAGS. PERIOD. The only way to stop it from lagging is turning off all the features and rooting the device. But then you risk getting your deviced bricked by Samsung.
Drop test are BS. So are you telling me you rather get in a crash in a plastic car than a metal car? lol.
And by the way the OS you are using on your BRAND NEW Note3 is now officially out of date. I doubt you got updated to KitKat. By next year you will be TWO FULL generations behind on OS. That is pathetic. In contrast the 3 year old iPhone4 gets updated to iOS7. And good luck when you decided to sell your phone. You'll be lucky to get $20
Nope, no lag.
I don't need to carry my smartphone and my table with my Note 3. The Note 3 provide me tablet like experience at a more portable size, something you can't get on the iphone unless you carry a tablet too.
About the IOS7, even the old Galaxy Nexus that I give to my son can do more that IOS7.
IOS7 actually looks like a cheap version of android
Those graphs are terrible to figure out. It's not intuitive to know at a glance what anything is except the first represented data (beginning at 0%). Requires some math ("Let's see, 85% minus 65% equals...").
You're the one that made a bombastic statement about the data/analysis being crap. I called you out on it, and you respond by telling me you have a PhD, I post too often, and that you have a job. In other words, you completely avoid the question.
The population size in this case is in the many many dozens of millions, as we know. So, whether we pick 10M, or 20M, or 200M, the margin of error will be largely invariant. You should know that.
As to my specific -- and trivially simple -- question to you about the MoE for the "16%" number, it would be calculated as SQRT[p*(1–p)/N], where p = 0.16 and N = 400. That turns to be a rather low MoE of ±1.83% at the 95% level. 16% ± 1.83% suggests a significant shift from the prior proportion of those switching from Android to iPhone. (All this assumes that the sample was truly random; but there's no reason to assume it's not, unless otherwise known).
Of course, you knew all that, right?
When you post next and are called out on it, back it up.
I do know all that. Calculating margin of error is one of the most basic of 100-level statistics. You wrote: "All of this assumes the sample was truly random; but there's no reason to assume it's not, unless otherwise known." Wow. Really? Understanding how to calculate a simple statistic using division and square root, yet not knowing how to challenge basic assumptions in absence of any documentation of methods is an error that usually, godawfully, is present in a great deal of bad statistics and the interpretation thereof.
My bombastic statement was a result of reading the article and noting the compete lack of evidence of any random sampling, along with what normally are very inadequate and unreliable data analysis by typical analysts. Anyone reading that article should have been highly suspicious of the stats, especially as regards (as I challenged) the generalizability of the findings.
Your reliance on undocumented assumptions suggests an uncritical eye toward numbers you read. Simply calculating margin of error is not the beginning and end of interpreting these numbers. How generalizable and reliable are the numbers? This is based largely on how well random sampling is conducted.
I do know all that. Calculating margin of error is one of the most basic of 100-level statistics. You wrote: "All of this assumes the sample was truly random; but there's no reason to assume it's not, unless otherwise known." Wow. Really? Understanding how to calculate a simple statistic using division and square root, yet not knowing how to challenge basic assumptions in absence of any documentation of methods is an error that usually, godawfully, is present in a great deal of bad statistics and the interpretation thereof.
My bombastic statement was a result of reading the article and noting the compete lack of evidence of any random sampling, along with what normally are very inadequate and unreliable data analysis by typical analysts. Anyone reading that article should have been highly suspicious of the stats, especially as regards (as I challenged) the generalizability of the findings.
Your reliance on undocumented assumptions suggests an uncritical eye toward numbers you read. Simply calculating margin of error is not the beginning and end of interpreting these numbers. How generalizable and reliable are the numbers? This is based largely on how well random sampling is conducted.
Good. Unless you wish to contradict yourself -- I wouldn't put it past you -- we're in agreement that if the sampling was done right, this is a perfectly valid article and headline.
There is no information there, other than some silly surmise on your part, that suggests otherwise.
It looks like many of the Samsung (and LG) "switchers" are actually upgrading from feature phones.
For the ones that actually switched from Android, the main factor seems to be that there are more Android (and iOS) users this year to upgrade, compared with Blackberry and feature phone users of which there are less.
Comments
First off I bough an Android Tablet about 2 years ago. POS. Total POS. After 12 months I got no updates. Many of my apps no longer worked, ton of my apps crashed, and the apps look like crap since they were phone apps blown up. The touch screen was unresponsive, cpu slow as hell, tons of lag. Basically it was such a horrible user experience. Just got the iPad Air. I'm in heaven. This is what a tablet suppose to be.
Got my Android phone about 12 months ago. Same story. I was stuck on an old version of Android. Apps crashed constantly. Some apps don't even open. CPU is slow, TONS OF LAG. Build quality is a joke. Basically the same experience as my Android tablet except shrunken down. Got an 5S a week ago. Amazing. I'll never go back to Android. I don't care how much more 'expensive' Apple is. I put expensive in quotes because after factoring in resale value, user experience, time saved, ect it ain't more expensive.
I just sold my old 16GB iPhone 4 for $220. The resale value for iPhones is incredible totally nullifying all the "iPhones are too expensive" crap.
Sample too small, probably taken here on US. Here the iphone is considered as "cool" phone.
I don't see many users downgrading from a FULL HD premium smartphone to a tiny non-HD smartphone like the iphone, it's just not a smart choice.
However, always there are people who do not necessarily make smart decisions.
LOL. Go buy a TV, dude. What's good to have a hi resolution screen that you eyes can't tell the difference? I don't care if the phone screen go 4k because it's not different from retina with human naked eyes.
How about: Not wanting to look like an idiot with a tablet-sized phone pasted along side of their faces? Or, not wanting to make the same dumb mistake twice?
I switched to an iPhone 5 a year ago. Was offered $20 for my mint condition HTC touch phone on eBay. There is usually a $100 minimum spread between a newer high-end Android and an older model iPhone on eBay...
Dogs when new, remain dogs when old.
Why stop at 5 or 6 inch screen sizes? Go all the way to 7.9" and get the iPad mini with built-in radio. That way no one will be in doubt that you don't mind looking like a dufus when on the phone.
I see posts like this and often wonder what the heck you were using for a phone. My first smartphone was a Droid X by Motorola when it was first released back in 2010. The iPhone wasn't an option as I was on Verizon and who the hell knew if it was ever coming. The Droid X was amazing phone and it never failed me in the two years I had it. When my two years were up I figured I'd give the iPhone a shot as it was my initial choice for a smartphone. It took some getting use to coming from Android, but after playing with the phone I could see why people loved it. I obviously ended up staying with the iPhone but my Android experience was a good one.
The reason your Droid X was "amazing" was because you were most likely using a feature phone (probably a flip phone?) and Droid X was presumably your first touchscreen smartphone. This is equivalent to upgrading from a horse buggy to a cheap Chinese-made motor vehicle (or a Yugo for those who are old enough to remember what a Yugo is). So your experience with Android would obviously be a good one given that you really had nothing to compare it against.
@anadtksundaram:
Wrong. My interpretation of the data presented was completely sound. I have a social science doctorate. Any freshman in an intro stats class will know YOU have no idea what you're talking about.
Really? Why don't you then tell us what the margin or error -- say, at the 95% level -- is likely to be with calculations of sample proportion = 16% in the data, when N = 400?
LOL. Go buy a TV, dude. What's good to have a hi resolution screen that you eyes can't tell the difference? I don't care if the phone screen go 4k because it's not different from retina with human naked eyes.
On a tiny 4" screen hell not, but on a bigger smartphone... yes you can notice the difference.
With higher resolution you can see more on your screen and that is a lot better for browsing web pages and enjoy multimedia.
I agree with you that 4K will be too much for a smartphone but for a tablet or TV it will be nice.
Finally, yes I may buy a 4K TV soon.
On a tiny 4" screen hell not, but on a bigger phone... yes you can notice the difference.
So you’re saying that the human eye can tell the difference between things it physically cannot tell the difference?
Now, Super Hi-Vision has a place on televisions over 42”. I’ve stood 20’ away from one of those 70” Sharp numbers and the pixels are big as hams. It was disgusting and horribly distracting.
So why don't you use an iPadMini Retina?
That's much bigger than Note3 and much higher resolution. See how this game can go on forever?
I don't have the need to carry my smartphone and my table with my Note 3. The Note 3 provide me tablet like experience at a more portable size.
I use my tablet at home.
I just sold my old 16GB iPhone 4 for $220. The resale value for iPhones is incredible totally nullifying all the "iPhones are too expensive" crap.
You say this, but lots of people don't end up selling their old phones. Many old phones are simply sitting in drawers, or more commonly, people update their phones because their old stopped working. You can claim that apple's phones are more reliable than android, and it might be true (probably is), but when your phone goes for a swim, or you drop the phone and the screen cracks, reliability doesn't come into the question.
Many people buying new phones can't rely on getting back $200+ for their old iphone, because their old iphone no longer works. It helps the people that have functioning devices (and actually sell them), but I bet they're actually in the minority. Reading on here the comparisons between Android and iOS are pretty amusing. I switched this year from iOS to android after my iphone 4 went for a swim. I now use a Galaxy S3.... I don't know if I'd call the phone "better" than an iphone 5 (the best available at the time), but it was relatively competitive for my needs. Additionally, the S3 cost far less than the iphone. I paid $400 for an unlocked S3, an unlocked iPhone 5 would be what.... $700? In retrospect, I probably should have saved more money and gotten the Nexus, but I wanted to have an SD card slot on my phone.
I don't expect the same quality out of a $400 phone that I'd get out of a $700 one . . . But overall, I have no regrets. I don't really miss the iOS ecosystem, and I like the freedom I get with Android. Sure there's a worry of Malware, but that is so overblown on these boards, as having 1000X more likely to get malware on Android than iOS (or whatever the number) is still meaningless, as the numbers are basically zero to begin with.
Phil
Really? Why don't you then tell us what the margin or error -- say, at the 95% level -- is likely to be with calculations of sample proportion = 16% in the data, when N = 400?
Despite what you might think, I don't jump when people like you say jump. I have nothing to prove. What I will say is that margins of error mean nada when there is no data AT ALL about the population that provides context for a statistic. Furthermore, an N of 400 is uninterpretable when the size of the population and characteristics of the sample are undefined. Anyone can play with sample size to increase power and artificially boost significance.
And IIRC, in some other thread you challenged me to jump through hoops to prove my knowledge. I won't. You're disputing MY statement. I'm really leaving it at that, Mr. 11,000 posts, becasue I have a job. That I work at.
Despite what you might think, I don't jump when people like you say jump. I have nothing to prove. What I will say is that margins of error mean nada when there is no data AT ALL about the population that provides context for a statistic. Furthermore, an N of 400 is uninterpretable when the size of the population and characteristics of the sample are undefined. Anyone can play with sample size to increase power and artificially boost significance.
And IIRC, in some other thread you challenged me to jump through hoops to prove my knowledge. I won't. You're disputing MY statement. I'm really leaving it at that, Mr. 11,000 posts, becasue I have a job. That I work at.
You're the one that made a bombastic statement about the data/analysis being crap. I called you out on it, and you respond by telling me you have a PhD, I post too often, and that you have a job. In other words, you completely avoid the question.
The population size in this case is in the many many dozens of millions, as we know. So, whether we pick 10M, or 20M, or 200M, the margin of error will be largely invariant. You should know that.
As to my specific -- and trivially simple -- question to you about the MoE for the "16%" number, it would be calculated as SQRT[p*(1–p)/N], where p = 0.16 and N = 400. That turns to be a rather low MoE of ±1.83% at the 95% level. 16% ± 1.83% suggests a significant shift from the prior proportion of those switching from Android to iPhone. (All this assumes that the sample was truly random; but there's no reason to assume it's not, unless otherwise known).
Of course, you knew all that, right?
When you post next and are called out on it, back it up.
Ummm doesn't one need a bit more information? If one conglomerated OS sells 2-3X of the other OS, how does it make sense to compare percentages for switching?
Say Android had 300 million phone sales last year vs 100 million for iOS (purely theoretical numbers), then 20% of Android to iOS would be 20million users. 10% of iOS to Android would be 30 million users. Or did I miss something and did this company actually normalize for this?
It depends what your purpose is. If you want to be mathematically correct, then your post above is right. If you're more concerned about posting something that sounds more favorable to Apple, it is better the way it is.
both the 5S and Note3 have battery life to last the whole day. Of course the Note3 does since it has a massive and heavy battery.
All your arguments of having a bigger screen are fine and dandy. But the iPadMini Retina would make your Note3 look like a chump. Much larger screen and higher resolution. It's all about preference. For me I want my phone to be a phone. Not a tv in my pocket. Note3 is a compromise product. Too big to be a phone and way too small to be a real tablet. Its a great product if all you can afford is one device. Personally I bought a 5S/iPadAir. The Air is one pound and is a hell lot better of a tablet than the Note3. And my 5S does not break my pants pocket.
Don't lie about the Note3. It LAGS. PERIOD. The only way to stop it from lagging is turning off all the features and rooting the device. But then you risk getting your deviced bricked by Samsung.
Drop test are BS. So are you telling me you rather get in a crash in a plastic car than a metal car? lol.
And by the way the OS you are using on your BRAND NEW Note3 is now officially out of date. I doubt you got updated to KitKat. By next year you will be TWO FULL generations behind on OS. That is pathetic. In contrast the 3 year old iPhone4 gets updated to iOS7. And good luck when you decided to sell your phone. You'll be lucky to get $20
Nope, no lag.
I don't need to carry my smartphone and my table with my Note 3. The Note 3 provide me tablet like experience at a more portable size, something you can't get on the iphone unless you carry a tablet too.
About the IOS7, even the old Galaxy Nexus that I give to my son can do more that IOS7.
IOS7 actually looks like a cheap version of android
See here a small list:
https://www.youtube.com/watch?v=uVTrazT99Ps
Those graphs are terrible to figure out. It's not intuitive to know at a glance what anything is except the first represented data (beginning at 0%). Requires some math ("Let's see, 85% minus 65% equals...").
Why not use the obvious: bar graphs!
All in all, good topic.
You're the one that made a bombastic statement about the data/analysis being crap. I called you out on it, and you respond by telling me you have a PhD, I post too often, and that you have a job. In other words, you completely avoid the question.
The population size in this case is in the many many dozens of millions, as we know. So, whether we pick 10M, or 20M, or 200M, the margin of error will be largely invariant. You should know that.
As to my specific -- and trivially simple -- question to you about the MoE for the "16%" number, it would be calculated as SQRT[p*(1–p)/N], where p = 0.16 and N = 400. That turns to be a rather low MoE of ±1.83% at the 95% level. 16% ± 1.83% suggests a significant shift from the prior proportion of those switching from Android to iPhone. (All this assumes that the sample was truly random; but there's no reason to assume it's not, unless otherwise known).
Of course, you knew all that, right?
When you post next and are called out on it, back it up.
I do know all that. Calculating margin of error is one of the most basic of 100-level statistics. You wrote: "All of this assumes the sample was truly random; but there's no reason to assume it's not, unless otherwise known." Wow. Really? Understanding how to calculate a simple statistic using division and square root, yet not knowing how to challenge basic assumptions in absence of any documentation of methods is an error that usually, godawfully, is present in a great deal of bad statistics and the interpretation thereof.
My bombastic statement was a result of reading the article and noting the compete lack of evidence of any random sampling, along with what normally are very inadequate and unreliable data analysis by typical analysts. Anyone reading that article should have been highly suspicious of the stats, especially as regards (as I challenged) the generalizability of the findings.
Your reliance on undocumented assumptions suggests an uncritical eye toward numbers you read. Simply calculating margin of error is not the beginning and end of interpreting these numbers. How generalizable and reliable are the numbers? This is based largely on how well random sampling is conducted.
I do know all that. Calculating margin of error is one of the most basic of 100-level statistics. You wrote: "All of this assumes the sample was truly random; but there's no reason to assume it's not, unless otherwise known." Wow. Really? Understanding how to calculate a simple statistic using division and square root, yet not knowing how to challenge basic assumptions in absence of any documentation of methods is an error that usually, godawfully, is present in a great deal of bad statistics and the interpretation thereof.
My bombastic statement was a result of reading the article and noting the compete lack of evidence of any random sampling, along with what normally are very inadequate and unreliable data analysis by typical analysts. Anyone reading that article should have been highly suspicious of the stats, especially as regards (as I challenged) the generalizability of the findings.
Your reliance on undocumented assumptions suggests an uncritical eye toward numbers you read. Simply calculating margin of error is not the beginning and end of interpreting these numbers. How generalizable and reliable are the numbers? This is based largely on how well random sampling is conducted.
Good. Unless you wish to contradict yourself -- I wouldn't put it past you -- we're in agreement that if the sampling was done right, this is a perfectly valid article and headline.
There is no information there, other than some silly surmise on your part, that suggests otherwise.
For the ones that actually switched from Android, the main factor seems to be that there are more Android (and iOS) users this year to upgrade, compared with Blackberry and feature phone users of which there are less.