Welcome to the Listening Post!

Good points about the phone survey, but what about the skew inherent in web-based surveys ... how can that be controlled or mitigated?

Well, there's no skew if you are an e-commerce site and have a list of email addresses for all your customers. Most organizations, though, need to improve their house lists and need to describe web research as being representative only of customers on that list. See Representative Web Surveys Require Good Email Lists of Customers for more detail.

Although your data may be correct regarding preference of web over phone, the implications may be a little off. The rate of participation of a phone survey is much higher than web. What your data might really be implying is that people find it easier to decline an email invite than it is to decline a live request for a survey.

For a nice counterpoint article, check out The death of telephone surveys? Not quite yet by Mike Jaxa-Chamiec.

Web surveys are the way to go in many cases. They are far less expensive to administer and with all of the technologies brought to market over the past few years, easy to gain access to on line. 
 
 
 
We agree that general telephone surveys have become an expensive proposition. The DNC list and saturation of cell phones into the marketplace have made that method of collecting market intelligence more and more difficult, and in some cases less reliable.  
 
 
 
However certain types of surveys (or interviews) are best done with an experienced research interviewer, where he or she is able to probe for root cause or more detail from the respondent. And I have seen many web surveys where the response to the question is not appropriate or even on point. They say the devil is in the details and many times, the details may be left out when filling out a web survey. And for business applications, B2B or HR related interviews, sometimes a more professional approach is a one-on-one telephone interview.

All telephone surveryors miss a critical point in understanding WHY someone might dislike a phone survey. The survey organizations don't pay the callers bill, that phone service is intended for the benefit of whomever pays the bill, not random 3rd parties

Calling from India is as cheap as web surveys with much higher response rate. Add to this statistical weights which reflect the household socio-economic parameters, you get a low cost, scientific survey. Check: digipoll.com; epollster.com 
 

I'm on the do not call list.  
 
Whenever I get s survey call, I ask the caller to "hold-on", put the phone in a drawer, and remove it when the loud beeping lets me know the parasite on the other end has gone after a juicier prospect. 
 
The reason I do this has nothing to do with distaste for surveys or researchers. It's because most of the "surveys" are disguised telemarketing. 
 
I make no claim to be representative, but if you want me to answer your survey, start off by identifying your organization "800-service" on the caller ID won't cut it. State why you're conducting the survey, and what your methodology is. Add some industry fostered legislation to that which ONLY exempts from the do not call list pre-registered surveys with proper methodology and reporting, and maybe I'll take your call. 
 
Congratulations, by the way, on your do not call exemption. Parasites.