One of most common reasons online survey data is confusing or ambiguous is because poorly crafted questions are asked. The wording may be unclear, the measuring system (for example, a Likert scale) may be implemented badly, or the surveyor's bias may have somehow infiltrated the questionnaire. This encourages misleading responses and pollutes your data pool. Bad data can lead to poor decision-making.

Asking the right survey questions can be as much art as science. Below, we'll provide a short list of five question types you should avoid including on your online surveys.

#1 - Imprecise Questions

Many questions lack the precision necessary to yield useful information. For example, suppose you asked your audience whether they were happy with their internet service provider. The question is too vague. Respondents will not know if you're asking about price, outages, speed, or their ISP's customer support. As a result, they'll likely respond to the question they think you're asking.

Be precise. Use specifics. Determine exactly what you would like to know and formulate your questionnaire to ask for the information.

#2 - Out Of Order Questions

If your online survey is longer than a single page, begin with broad, demographic questions to which participants can easily respond. That will reduce your abandonment rate; respondents will be encouraged to continue toward the end. Allow the questions to gradually become more specific. Then, broaden their scope again.

Surveyors often give little attention to the structure with which they design their online questionnaires. If the statements seem to be out of order or lack contextual meaning, respondents will lose interest. That will increase your abandonment rate.

#3 - Multi-Pronged Items

Each item that you include on your questionnaire should focus on a single element. For example, don't ask participants, "Are you satisfied with your ISP's price and reliability?" Price and reliability are unrelated elements. Each deserves its own question. By combining them, you will pollute the value of your data pool because there is no way to determine whether respondents are addressing price or reliability in their answers.

#4 - Leading Questions

The words you choose when designing your survey can play a significant role in the responses you receive. The more objective you can be, the better. For example, suppose you asked your audience, "Don't you agree that your ISP should be less expensive?" The wording suggests that you are hoping for a certain response from participants. It lacks objectivity. The drawback is that many respondents will provide answers that are less than sincere, polluting your data pool in the process.

#5 - Ranked Items

Ordinal questions have a place in some types of online surveys. But, they should be used sparingly. Asking a respondent to rank or rate a list of items creates problems, especially as the list grows. For example, suppose you instructed participants to "place a number between 1 and 10 next to each of the following 10 items, using each number once." The list is large and difficult to rank. Plus, if the items are contextually related to the rest of your questionnaire, participants may begin to rethink their other answers as the rank your list. If you intend to use ordinal questions, limit your list to four or five items.

Online surveys can attract valuable data if your questions are crafted properly. They can help your company measure your customers' level of satisfaction, generate qualified leads, and evaluate your employee training program. They can assist in performing market research and even function as an online job application platform.

Unfortunately, surveyors often become enamored with the technology without devoting enough time to creating objective, specific questions. Avoid that problem by reviewing the list above before creating your next online survey.