When to and When Not to Conduct a Web Survey

By: Amanda Barna

Web surveys are often seen as a faster and cheaper alternative to other research mediums such as telephone surveys and mail surveys. While web surveys are an excellent methodology for many research projects, there are times to use them and times not to use them.

When TO do a Web Survey:

  1. You need to include pictures and graphics in the survey instrument

    Most web survey programs include the capabilities to display pictures such as marketing pieces or other graphics which is extremely helpful when collecting qualitative information about a marketing or advertising concept.

  2. To survey a captured audience such as a membership, employees, or customers

    A captured audience already has an established relationship with the organization that is conducting the survey and therefore is more likely to complete the survey than someone without a relationship to the organization. In addition, the sample is readily available and the final data will be representative.

When NOT to do a Web Survey:

  1. If you need to generalize results to a large population or the general public

    While email or other invitations can be sent to a representative sample of a population, you have very little control over who actually completes the survey and therefore may not end up with a statistically representative sample of the population without significant data weighting. In addition, the sample sizes are often too small to draw statistical conclusions about a large population. The number of people with email addresses and access to the internet is on the rise, there are still significant portions of the general public that do not have access or are not comfortable completing a web survey.

  2. Simply to save money

    A common myth about web surveys are that they are a quick and cheap way to collect data. Web surveys, like other methodologies, require a great deal of planning and effort in order to be effective. While a web survey may cost less money than a large telephone survey, if it does not produce data that is valid or reliable or provide the information needed to make key decisions, then money has been wasted rather than saved.