Collecting Useful Survey Data

Whether you are polling employee satisfaction, measuring customer satisfaction, or conducting another type of survey, it is critical that you construct questions that will yield the information you need. The following guidelines will help you create an effective survey that provides useful data

• Take time to think about what you hope to learn from your survey and compose questions that will provide this information. Do not ask overly general questions. Be as specific as possible. By themselves, general questions like "How satisfied are you with your job?" or "How satisfied are you with our customer service?" will not tell you anything about what people are satisfied or dissatisfied with (These questions are helpful in measuring "importance"). Create a list of the key metrics of whatever it is you want to measure and write specific survey items that address each of these areas. Once you have composed your survey items, try answering them as if you were a respondent. If the answers you would provide would not be useful, then you need to rewrite the items.
• Ask both quantitative and qualitative questions. Quantitative data makes for nice PowerPoint presentations, but might leave you wondering what is behind the numbers. By including "write-in" questions, you can find out. Even though reading through the comments provided by respondents is time consuming, these comments can often reveal important information not uncovered in the numbers.
• Include demographic questions. It is likely that different groups of people responding to your survey will have different answers to the items. The only way to uncover where these differences lie and pinpoint potential issues is to include the appropriate demographic questions. Think about the different demographic aspects of your respondent population and be sure to include the necessary demographic questions that will allow you to break down the responses into key respondent groups.
• Pilot the survey. If possible, pilot the survey with a small group of respondents. This will help uncover any potential problems with your survey, such as ineffective items or areas in which additional insight is needed. You can also include some extra items at the end of the pilot survey that ask about the survey itself, such as appropriateness of length & content, ease of response, etc.