The best length for your survey is also a tricky issue. On the one hand, it seems that since you have people’s attention — why not make sure that you get all of the information that you might need! However, when surveys get to be too long, participants will simply give up and not respond any further.
Another thing that you will need to consider is whether to use open-ended questions, which means allowing the respondent to provide his/her ideas directly (i.e., “fill in the blank” type questions) or to use a “check the category” response type. Open-ended questions allow for you to get all of the different ideas that come to mind from the respondent; however, in some cases respondents appreciate having choices to click since it’s easier than thinking up their own ideas. It is also easier to analyze the results from the “check the category” questions because they will fall into a limited number of definitive categories.
Constructing the survey
Now that you are constructing the survey, you need to decide if you are going to use a mail survey, an online survey, a phone survey or in-person interviews. There are advantages and disadvantages to each type, and your selection should be driven by the audience you are trying to reach and the medium which will be most effective for them.
Online surveys have become very popular because they can be relatively inexpensive and the results are provided in an easy-to-use electronic format. There are different software packages you can use; two examples: www.surveymonkey.com or www.zoomerang.com. If you are collecting data during a meeting or gathering, you can use an audience interaction tool, i.e., clickers; see www.iclicker.com as an example. One drawback to online surveys is you must have an email address for all in your sample. Using only those in the sample with emails will not give you a representative sample of the population you are targeting.
Once you have the survey questions set and have decided on the medium you will use, it is very important to pretest your survey. Many times this step gets missed only to hear from those developing the survey that “if only they had pretested the survey they would have figured out what questions were not effective.” When pretesting your survey, find people to complete the pretest who have the same background as those who will ultimately be answering your survey. It is common to use phrases or words that have meaning to certain groups of people — so having the appropriate group pre-test your survey will help you determine if you have any wording/phrases that would be confusing for respondents.
An effective tactic which will improve your response rate is to briefly outline in a short paragraph why you are doing the survey. For example, “We are considering: “We are considering offering a new service in our market area, and are collecting responses from both our current and potential customers regarding your opinions. Your input is highly valued and will allow us to better serve our region.”
Administering your survey
To increase your response rate, and thus the value of the data you collect, you will most likely need to follow up — as people are busy and will need a reminder. Depending on the medium you use (mail or email), you can send out your reminder via the same medium. If you are keeping the responses anonymous, you may have to contact people who have already responded. This is a bit awkward, but can be addressed with a statement such as: “If you have already responded to our survey, thank you very much for your valuable input. If you have not had a chance to complete our survey, we would appreciate your input.” If your survey is anonymous, be sure and let your audience know this.
Finally, you will need to gather up the data from your survey. If you used an online mechanism, this will come to you in electronic form. If you performed your survey in some other form, you will need to get the data entered into an electronic form, usually a spreadsheet for ease of analysis.