Telephone Surveys
VanAmburg Group Telephone Research Process
Click graphics for larger versions


Telephone surveys have dominated the consumer research market for the past 50 years as the best method to achieve statistically valid results which can be projected into the entire population. While they are now losing ground daily to online surveys, they remain a staple of market research.

For example, all national political polls rely on the phone survey approach for their immediacy and ability to reach a wide range of respondents throughout much of the world.


You create a sample frame prior to beginning your research, which defines who you need to include for your survey to be representative of your audience. You could decide your sample will match the demographics of your community or the nation. Or you may be more interested in the opinions of American car owners, or of female registered Democrats under 35 who voted in the last election.

Your target audience will help determine if you:

  • Can ask anyone at random to participate.
  • Call randomly, but ask screener questions when you call to filter out those who don't meet your criteria.
  • Should purchase a list already filtered to meet your needs.

In consumer research, if you elect to call randomly, there are two general alternatives for sampling:

  • Random number generation: Within target telephone area codes, acquire an accurate list of three-digit phone prefixes, then randomly generate four-digit suffixes, and call the resulting phone numbers. (Such lists can be purchased.)
  • Nth number selection: This is a fancy term for a simple process. If you are surveying randomly in one municipality, use the phone book. Simplistically, if you need 385 complete surveys and there are 385 pages of names, pick a spot on each page such as the top of the middle column. Call down the middle column (residential listings only) until you complete one survey. Then move to the next page and repeat.


Consumer phone interviewing relies on our ability to reach a random sampling of households. The validity of this approach has been decreasing steadily for the past 15 years because of eight technology and lifestyle changes:

  1. Reluctance to participate: Our experience is that the number of calls required to complete one survey has risen from 4 in 1980 to 18 presently, caused by all of the items below, combined with less discretionary time and survey overload and burnout.
  2. Telemarketing and robo-calling: Increased telemarketing prompted creation of federal and state do not call lists, which some consumers wrongly believe forbids phone research.
  3. Availability: The dramatic decrease of decision makers under 65 at home during weekdays restricts calling to a few hours a day in early evening for most studies.
  4. Answering machines: 77% of U.S. households own them and 55% use them to screen calls.
  5. Caller ID: Introduction and use of Caller-ID on incoming calls (40% of U.S. households in 1999).
  6. Unlisted landline phone numbers: Approximately 35% of U.S. households now have unlisted landline phone numbers (affecting Nth number calling, not random digit dialing).
  7. Cellular phones: 7 million people in the U.S. (about 3% of the adult population or 7% of U.S. households) use cell phones as their only phone, thereby removing them from the available pool of listed numbers. For consumers age 18 to 30 this is particularly dominant, with between 10% and 23% having only a cell phone.
  8. Voice Over IP: Use of Internet VOIP for phone continues to grow steadily, and none of these lines are included in local telephone directories. Jean Mercier has estimated that there are 30 million active Skype users.

Additionally, VIOP numbers can be assigned from almost any area code and multiple three-digit prefixes, decreasing the correlation with residence by zip code.


CONTACT US to help make your research the key to your company's future.

Copyright VanAmburg Group, Inc.
Updated November 4, 2008
On a scale of 1 to 5, how useful was this article to you? (Click selected star to vote)
7305 Responses -
  • Comment