sexta-feira, 12 de setembro de 2014
Before starting your survey, there are a lot of things to consider. When you’re in the planning stages of your survey, you need to consider what question you’re trying to answer. That will influence the questions in your survey as well as what type of questions the survey will have. Make sure the purpose is pinpointed accurately before formulating your questions. If the report will need to be turned into a graph or chart, that should be considered when forming the survey questions too.
Each topic that you choose should have a rank that lists the most important parts of the question at the top. List each question based on the rank of importance when it comes to answering your base question. For example, if you’re trying to find out if your new product will be well-received, the first questions should ask what people think about that type of product. This will tell you whether the rest of the survey is relevant or not.
When determining the rank of importance in your list of questions, don’t provide too much information about the purpose of the survey. It can create bias that will impact the rest of the questions and skew the survey results.
You’ll also need to figure out your core demographic. Are you marketing to children, men under 30 or women over 50? This will influence the questions you ask, and how you ask them. Children would need a different set of questions than adults.
Internet users can have short attention spans. The survey should be clear and direct. Asking long, essay-like questions can increase the drop out rate for your survey. Survey respondents like neat, orderly questions with page breaks and branching to keep them engaged. Branching will bring the respondents directly to the next appropriate question based on their answers.
There are many types of survey questions from multiple choice to rating scales to comment or essay questions. You’ll need to think about your demographic and the type of answers you need when choosing a question.
Along with short questions, the length of the survey should appeal to people on the Internet who might skim. While you need answers to your questions, a serious, lengthy survey can be a turn off for most respondents. It should take no more than five minutes to complete your survey. That means limiting the length of your questions as well as how many you have.
Once you’ve completed your survey, it’s time to test it with a group of people who can give you intense feedback. It’s a simple way to see if the survey works and iron out any kinks in the timing or responses. You can add a comment section to the testers survey that won’t appear in the live survey. The test group can provide feedback in the comment section to help you tweak your survey questions.
When the test group is finished, review their comments and suggestions and implement any changes before sending out the live survey to any other groups. Don’t forget to remove the comment section. Take your time with this part to ensure that the questions are perfect. This is the last time you’ll be able to change questions. If you change them in the middle of a project, you’ve skewed the results.
Your demographic most certainly has a time when they are online. Single mothers wouldn’t normally be online available for surveys late at night. If you’re asking questions about whether someone wants to work for a boss or themselves, you might get a different response on a Friday compared to a Monday. College students will most likely be unavailable on a Saturday night. These are things to think about when you release your survey, especially if you have specific people to which you’re emailing the survey. Here’s a sneaky tip: Use some of these handy Twitter analytical tools to help you see when your audience is Tweeting the most – and you can send your survey then.
If your survey has complicated questions, is delivered at the wrong time for your demographic or takes longer than five minutes to complete, you won’t get the response that you hope to get. Keep this in mind when preparing and sending out your survey for the best results.