The number of internet surveys being conducted has increased dramatically in the last 10 years. The increase in the number of Americans who have access to the internet, and the relatively low cost of conducting web surveys have contributed to the proliferation of surveys conducted over the internet. Web surveys can be used to conduct random sample surveys of members of selected populations who have access to the internet and the skills necessary to complete a survey on the Web (e.g., students, members of voluntary associations, etc.). In addition, web surveys can also be used in conjunction with another mode or modes to conduct probability-based surveys of the general public. Finally, many web surveys are conducted using non-probability based methods, such as surveying members of volunteer panels or people who visit a particular website.

Although more surveys are being conducted via the Web, internet surveys have generally not replaced other modes for surveys of the general population. Instead, they are often employed in combination with another mode to accurately represent the public. Surveys of the general population that rely only on the internet can be subject to significant biases resulting from undercoverage and nonresponse. Not everyone in the U.S. has access to the internet and there are significant demographic differences between those who do have access and those who do not. People with lower incomes, less education, living in rural areas or age 65 and older are underrepresented among those who use the internet and those with high-speed internet access (see the Pew Research Center’s Internet and American Life Project for the latest trends). People also vary a great deal in the skills necessary to complete a Web survey.

There also is no systematic way to sample the general population using the internet. Internet surveys of the general public must first contact people by another method, such as through the mail or by phone, and ask them to complete the survey online. There is no national list of email addresses from which people could be sampled, and there is no standard convention for email addresses, as there is for phone numbers, that would allow random sampling.

Because of these limitations, surveyors use two main strategies for surveying the general population using the internet. One strategy is to randomly sample people using another mode (mail, telephone or face-to-face) and ask them to complete the survey on the web. Some of the surveys may allow respondents to complete the survey by a variety of modes. This method is used for one-time surveys and for creating internet panels (such as the GfK KnowledgePanel), where people regularly complete surveys online. Contacting respondents using probability-based sampling via another mode allows surveyors to estimate a margin of error for the survey (see Why probability sampling for more information). The Pew Research Center has used the GfK KnowledgePanel to assess the public’s knowledge of politics and the news (October 2011January 2013, and August 2013) and for a special survey of gay, lesbian, bisexual, and transgender Americans (April 2013).

The Pew Research Center has also conducted internet surveys of random samples of elite populations, where respondents complete the survey online and are sometimes also able to complete the survey by telephone.  For example, see Public Praises Science; Scientists Fault Public, Media.

Another strategy relies on convenience samples of internet users. Researchers use one-time surveys that invite participation from whoever sees the survey invitation or rely on panels of respondents who opt-in or volunteer to participate in the panel. These surveys are subject to the same limitations facing other surveys using nonprobability-based samples: the relationship between the sample and the population is unknown so there is no theoretical basis for computing or reporting a margin of sampling error and thus for knowing how representative the sample is of the population as a whole. (also see the American Association for Public Opinion Research’s (AAPOR) Non-Probability Sampling Task Force Report and the AAPOR report on Opt-In Surveys and Margin of Error). We discuss results comparing responses from an online survey to those obtained using our traditional RDD telephone survey in Online Polling Offer Mixed Results.

 

Related Publications