Wirelessly posted (Mozilla/5.0 (iPod; U; CPU iPhone OS 4_2_1 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8C148 Safari/6533.18.5)
Makes sense.
YMMV. Frankly, I don't see how it can make sense for a VZW iPhone user to report on a "Last
90 Days" metric when it is literally impossible for him to have owned his iPhone for that long.
You have to wonder about response bias in a survey like this, since the issue is so well known. Do Verizon iPhone users perceive they have fewer dropped calls because they have been led to believe that Verizon has a more reliable network? And do AT&T iPhone users perceive that they have more dropped calls because they have been led to believe that AT&T has a less reliable network?
I'm not saying this data is necessarily inaccurate, but it's a question worth asking in this specific context.
Response bias is likely, but more disconcerting is the Changewave's survey question with a "90 Day" criteria on a product that's barely been available for 45 days.
If Changewave normalized their VZW iPhone survey respondant data, they failed to disclose so, including how. If it wasn't normalized, then some of their report is flat out wrong.
To illustrate, by assuming that they failed to normalize for the temporal disconnect, then since the VZW iPhone4 data is only from a ~45 day period, its reports would be roughly half the value that it should have been for the 90 day period (criteria used for the AT&T customers). Thus, instead of this being a "1.8% vs 4.8%" comparison, it would've been "3.6% vs 4.8%".
And if the average VZW ownership was only 30 days, then it would have been "5.4% vs 4.8%"...yes, in favor of AT&T.
Until Changewave clarifies their analytical method (to assure that this factor was indeed addressed), their "Dropped Calls" comparison claim simply isn't statistically credible.
-hh