From the Item Bank

The Professional Testing Blog

Action Design Approaches For Survey Administration

Action design is a set of tools aimed at making it easy for users to take an action or make a behavioral change. In many cases this is an action or change that the user has chosen to take, and yet still needs additional support to carry out, such as exercising, studying, or completing a lengthy survey (Wendel, 2014). Furthermore, while action design includes the goal of good usability, it goes beyond that basic objective by also targeting an application that users find engaging and personally meaningful (Anderson, 2011).

For survey design and administration, action design methods could be used to help influence survey recipients in order to increase their willingness to respond to the survey, and to continue through the entire set of questions. In other words, action design methods can improve both the survey response rate and the completion rate.

There are specific aspects of survey administration where action design techniques can be applied. A few of these are discussed below, along with examples of potential approaches that could be used.

The Cue

In action design the cue is the process of reaching out to potential users, or sending reminders to return to the task. For surveys, the cue would include the initial contact sent to invite people to participate in the study, along with any reminders sent to encourage people to complete an unfinished survey. Behavioral techniques that could be used in this invitational stage might include:

• Frame all messaging to the survey recipient in terms that are personally meaningful to the recipients.
  • Defining the survey on the basis of why the recipient should care (Sinek, 2009) could have a strong effect on survey response and completion rates. (e.g., “The only way we can find out whether our service meets the needs of people like you is to ask you”.)
• Plan to send follow-up contacts to non-responders.
  • Each communication should use a different tone, and a different basis of persuasion, to effectively “target those who failed to respond to earlier contacts” (Dillman, 2000).
• Provide information about social norms (Cialdini, 2006) in terms of the numbers of survey contacts who have already responded, e.g., “Many of our invited participants have already completed the survey.”
  • An alternative approach, when few users have responded at a given point in time, is to use injunctive norms (Martin, Goldstein, Cialdini, 2014). For example, if the information were available, survey recipients could be told that “n% of those contacted have said they would value the results of this survey” or “n% of recipients have agreed that responding to surveys like this is important.”

The Survey

Principles from both action design and survey design can be applied to encourage users to begin the survey and to complete the full survey. Design approaches that target this goal include the following.

• Begin the survey with an item that will be interesting to the respondents, that applies to the majority of them, and that has an easy response format.
  • The first item on a survey is a crucial point in the respondents’ commitment, as he or she was just persuaded to take the survey based on a description of its purpose and value. If the first question does not relate to that purpose, survey recipients may stop participating immediately. Thus, the first item must clearly confirm that the survey will be a meaningful use of the respondents’ time (Dillman, 2000).
• For the overall item order, items that are clearly relevant to the survey purpose should be placed at the beginning, with personal or sensitive questions at the end.
  • This reduces a tendency for users to drop out of a survey before completion, as early items will feel more meaningful to the respondents. And, once respondents have completed the majority of a trustworthy survey, they may be more willing to provide sensitive data as well.
• If instructions are needed within the survey, provide them in a targeted manner.
  • Aim for just-in-time instructions, provided with the item itself the first time the information is needed (e.g., when a new item type is used for the first time).
• Make the task as easy as possible. In some applications, this involves breaking the task into small “winnable” pieces, such as focused survey sections (Wendel, 2014).
  • For longer surveys it may be especially important to break the full survey into smaller “chunks” that can be successfully accomplished in a short period of time. As each section is completed, the respondent can be “rewarded” with an onscreen note of thanks or a reminder about the purpose and value of the survey, and a progress indicator that displays the percentage of the survey that has already been completed.

The behavioral techniques provided here are only some of the action design tools that could be applied within the context of survey design and administration. The larger set of behavioral tools includes useful approaches such as: personalization tied to the user’s self-identity, time-based cues, greater potential rewards, and more.

For more information on similar topics go to Behavioral Design.

References for Action Design

Go to Additional Resources for Action Design for references.

Categorized in: Industry News, Marketing