TRORC Guide to Conducting Surveys

A number of towns in the TRORC region have conducted surveys for the purposes of planning. The most common method of surveying residents has been through mailed surveys.
Mailed surveys are not as effective as phone surveys, but they are the least expensive method, requiring a limited amount of manpower to conduct the survey.  A response rate of 25-50% is common. However, the response rate as with the results themselves will differ from town to town.

Things to keep in Mind

  • Know what you want to get from the survey – Be sure that you know your purpose before you start writing questions. Discuss the topics that will be included and be sure they are relevant to your purpose.
  • Know what you want to do with the data you receive – Just because an answer to a question is interesting, does not mean you can do anything with the data. Don’t ask questions that will provide data that does not meet your purpose.
  • Know who you’re surveying: Who will be responding to the survey? Voters? Landowners?
  • Do your research: Don’t try to reinvent the wheel. Look at other surveys. Use them as a starting point as you design your questions.
  • Keep the questions concise: Long winded questions can be confusing and misinterpreted. Be sure your questions are too the point. Less is, as they say, more.
  • Don’t lead the questions: One of the most common mistakes made by survey designers is to phrase the questions in such a fashion that they elicit a response that is expected.
  • Avoid asking for identifying information: People prefer a level of anonymity when they respond to surveys. Don’t ask information that they will be uncomfortable giving you unless it will provide you with data that you must have.
  • Use plain language: Be aware that not all respondents may be able to read exceptionally well or understand complex questions. Do not use big words or confusing concepts.
  • Ask only one question at a time: Avoid “double barrel” questions such as “do you like cats and dogs” as they can cause confusion and poor responses.
  • Keep the answers simple: Use multiple choice responses such as “yes/no” or graded (bad to good) to questions in your survey. Complicated answer methods can be confusing to the respondent.
  • Avoid open ended responses: Unless they are absolutely necessary, avoid questions that have open ended responses. They are difficult and time consuming to tally and they do not usually give you a clear answer to your question. However, you should always include an area in which respondents can add comments (generally at the end of the document) about the questions or the survey itself.
  • Avoid “select all that apply” questions.  Avoid the response option “other”.  Careless responders will overlook the option they should have designated and conveniently mark the option “other.”
  • Put important questions first: You can never be sure that a respondent will completely answer your survey. They may run out of time or feel the effort is too taxing. However, they will commonly answer the first few questions asked.
  • Avoid agreement bias: By framing both positive and negative questions, your respondents will evaluate each question rather than uniformly agreeing or disagreeing to all of the responses.