Here's what our survey guru (former NORC person) said in response to this. Take it for 
whatever value you take from a thirdhand source: 

===================
I checked with Survey Sampling, Inc., about two months ago, and the person there said 
they do not include any cell phone numbers in their samples.  Af far as I know, nobody 
includes cell phone numbers in their samples.  

In all the legitimate phone surveys and polls, everybody is selected from a list of 
phone numbers.  (The only way the interviewer knows your name is that you tell her.)  
The numbers are generated by computers, so many of them do not really exist.  But 
companies like Survey Sampling make their living by guaranteeing that a large 
percentage do exist.  As far as I know, all the exchanges that are used (e.g. 414-382) 
really do exist.  But cell-only exchanges are not used in preparing samples.  For 
example, the area code 917 is used only for cell phones, and nobody who has a 917 area 
code will ever get called for a poll.

This is a bias that will need to be fixed, but as far as I know, it has not been 
fixed.  Breslin is correct.  Of course, most cell phone users also have landline 
phones, so they can be included in polls.
===================

Paul Smith
Alverno College
Milwaukee

---
You are currently subscribed to tips as: [EMAIL PROTECTED]
To unsubscribe send a blank email to [EMAIL PROTECTED]

Reply via email to