When designing surveys for the web using any product in the Ultimate Survey suite of online survey software, there are several things to consider when designing surveys for the web compared to other mediums.
Here is our list of things that you should think about before designing your web survey:
1. Type of user and their internet ability Is the user an expert or novice internet user? This will effect the speed at which the respondent answers the survey or form. Therefore, this should be taken into account when determining the complexity and length of each page of your online survey.
2. Bandwidth of the user Similar to number 1, this should be used as a factor when considering the length and complexity of your online survey.
3. Number of questions per page While paper form surveys can generally run the gamut in terms of length, computer (and specifically web) based surveys require more thought when it comes to this issue. Generally speaking, we recommend that each page have approximately 5 questions. In our testing, this has generally resulted in the least users that abort the survey and the most thorough answers/time spent on each question. Use pages in your online surveys to separate the questions in to logical sections.
4. General usability When applying a style to your survey, consider the color scheme and font choice that you use. The web survey should be pleasing to the eye and easy for the respondent to read. Poor choice of color design, inconsistent fonts, and small font size can all lead to respondents aborting the survey. In addition, we recommend that you use our survey software's column features when creating radio or checkbox questions with long lists of answers. This will greatly improve the responses that you receive from the survey respondents.
Overall, these are the first steps in creating high quality web based surveys with our survey software. We believe that following a strict and consistent design process will greatly improve the quality of your surveys and the responses that you get from your respondents.
