Create a clear goal for your test. For example, you may want to know if people with many friends procrastinate more or less than people with few friends, or if people enjoy watching movies at home more than in the theater. In the article "Everything You Wanted to Know about Questionnaires," for StatPac, David Walonick, Ph.D suggests having a clear goal is the most important aspect of designing a questionnaire.
Create a list of unique questions that relate to your research question. According to Athabasca University's article, "Questionnaire Design," you can analyze the data from multiple choice questions more easily than open-ended questions. Create multiple choice questions in the format of the popular Likert Scale, which people can answer based on a five-point scale ranging from "strongly agree" to "strongly disagree."
Paraphrase each question to create three or four differently-worded versions. Phrase half of these oppositely from the original question. For example, "I often become distracted when I read" would be changed to "I never get distracted when I read." Having differently-worded versions of the same question allows you to know how consistent people's answers are. Someone who answers "strongly agree" for one question, should also answer "strongly agree" or "agree" for a paraphrased equivalent. They should also answer "strongly disagree" or "disagree" for the version of the question you phrased in an opposite manner.
Proofread your questions for clarity. According to the article, "Questionnaire Design," from Athabasca University, unclear questions are a large source of mistakes in questionnaires. You must ensure that your questions are concise and unambiguous, so that everyone taking your test understands them in the same way.
Randomize your questions and put them in a numbered list. Under each question, include answer options in the following order: "strongly disagree," "disagree," "undecided," "agree," "strongly agree." Include a box beside each option that people can check.
Write clear instructions for how to complete the questionnaire above your questions. If your questionnaire contains questions of a personal nature, notify people of this in your instructions. Explain that people can stop doing the test at any time if they do not feel comfortable.