10 Insanely Useful Tips on Making a Survey
While it may seem that there is a natural order in the process of creating a survey, the practice itself impose more paradoxical conclusions.
If you hear the advice from the experts, which we highly recommend doing, you’ll realize that not everything in the survey game is what it seems.
Here is a list of tips that may sound like insanities, but are actually useful guidelines for obtaining valuable data.
1. Start from the end
The number one piece of advice is the most paradoxical one: don’t start from the beginning.
Before constructing your survey, carefully set your goals and outcomes. In other words, think where you are heading with the survey project, and what it should achieve. With such a direction in mind, you’ll make sure that you gather only relevant data for the analysis stage.
2. Jot down your answers
Postpone curious examination just yet. It is highly advisable to write the answers you need, to ensure your inquiry only contains relevant and suitably formulated questions.
3. Hide your goal
Contrary to popular belief, experts claim that people have a tendency to help.
If you state your description emphasizing your goals, the respondents will likely provide the answers that help achieve that goal. Ultimately, it will leave you with dishonest impressions and compromised data, neither of which is fruitful for analysis.
4. Less is more
Limit your survey to the restricted number of sections and questions. The professionals claim that is the best way for avoiding “respondent fatigue” (occurs when survey participants become tired of the survey task), the phenomenon which could potentially jeopardize the correctness of the feedback.
5. Don’t speak your own language
Be prudent and design your survey using empathy – avoid any language that the respondents would misunderstand, or that might confuse them. Keep it simple and concise.
6. If you lead one way, the respondent will turn the other way
While taking a survey, people often feel frustrated or irritated if they sense that the questions are trying to impose certain answers. It’s not unlikely they will turn the other way, merely to demonstrate an attitude.
7. If you need demographic data, save the questions for the very end.
Gathering demographic information often involves a generous amount of sensitive questions that altogether rise a “stereotype threat” (situational predicament in which people are or feel themselves to be at risk of conforming to stereotypes).
To prevent it from compromising data, ask the questions about gender, race, income, and alike, at the very end.
8. Click “Ctrl + F” and delete all the “ANDs” and “ORs” from your questions
Exclude all the formulations that ask for more than single information at a time
(double-barreled questions). They are confusing and will jeopardize the accurateness of the survey’s results.
9. Matrix questions are the worst tool for obtaining impressions
Although it’s tempting to use them as a tool for measuring impressions, the statistics say that matrix questions are tiring, and eventually make the respondent check boxes randomly.
Instead, try breaking the big boxes into small scalable questions (using emoticons, for example, is a good idea).
10. Don’t pass out your survey to respondents
Actually, the respondent is the last person you’d like to give your survey.
Always make sure everything is working as it should be beforehand, by asking a friend to answer your questions and provide honest feedback. Editing a survey once it’s out in the world could be harder than making a new one.
Ready to create a survey? We can help.