2. The results of the survey depend heavily on the research sample that actually takes the survey.
3. One of the biggest mistakes researchers often make is not properly pre-testing survey questions.
4. It is important to choose the best possible method to reach your intended audience, based on available resources - this will help determine whether a survey should be conducted in person, over the phone, online, via mail, or some other venue.
5. Programs like Excel and online surveys like SurveyMonkey.com can help you conduct basic statistical analysis. Programs like SPSS and SAS are needed for more advanced quantitative research.
Creating and conducting a survey can be a great way to better understand and document what is happening in your community. Surveys can take many forms and there are many different ways you can design your questions and analyze your findings. With that said, all surveys should follow some basic steps if you are going to maximize your time and efforts in the process.
Let's start with a brief example from our work at the Metamorphosis Project. Several years ago, we were interested in investigating why different residents experience different feelings of belonging in their neighborhood?
What are the factors that contribute to residents' level of belonging within their community?
What is the influence of a person's connection to the neighborhood storyelling network on their feelings of belonging?
These seemed like questions that would lend well to a survey, because we wanted to get results that were quantitative in nature. We decided that we wanted to speak to an equal number of African American and Latino residents who lived in our Los Angeles study areas to find out. Thanks to some grant money for our project, we came to an agreement with an outside firm that would conduct telephone interviews with community members in our study area. After working for a while to develop and refine our questions, we provided the firm with our survey protocol, which included our neighborhood belonging scale, as well as a number of other questions about residents' demographic information, daily life, media use and communication patterns. After the firm acquired several hundred interview responses, they provided us with the raw data, which we then analyzed using different types of statistical techniques. Our analysis was later reported in published academic papers, presentations, and brochures that were widely distributed.