To e or Not to e? Mail or Online Surveys?

Mail surveys are an economical choice compared to phone surveys or personal interviews, but online surveys are priced even lower. Expect to spend 35-50% less for an e-survey than you would for a mail survey. There are even bigger discounts when comparing “e” to phone.

Although online surveys are quicker and less expensive, the results don't always do as good a job at representing the population of interest. They fall short in a few areas:

Sampling Frame. Because you have street addresses for all recipients, it is theoretically possible for anyone in your circulation to be sent a mail survey. Online surveys can only be sent to those for whom you have e-mail addresses. These people might be different (better educated, younger, more affluent, etc.) from those people whose e-mail addresses you don’t have.

Valid addresses. Undeliverable mailed surveys usually don't exceed 2-3% while they can exceed 30% for e-surveys, further reducing the number of readers you are relying on to accurately represent your population.

Non-response bias. When the goal of your survey is to use the responses to represent the entire circulation, the degree to which the population is truly represented depends upon how similar those who do respond to the survey are to those who don't. Thus, the higher the response rate, the more accurate the survey. You can expect considerably higher mail response rates than for online surveys.