Designing Surveys and Questionnaires
The StatPac Inc. website enumerates and explains the different qualities of a good question.
I have learned that a good question: (1) evokes the truth, (2) asks for an answer on only one dimension, (3) can accommodate all possible answers, (4) has mutually exclusive options, (5) produces variability of responses, (6) follows comfortably from the previous question, (7) does not presuppose a certain state of affairs, (8) does not imply a desired answer, (9) does not use emotionally loaded or vaguely defined words, (10) does not use unfamiliar words or abbreviations, (11) is not dependent on responses to previous questions, and (12) does not ask the respondent to order or rank a series of more than five items.
StatPac Inc. further explains that a question must be non-threatening and it should encourage the respondents to produce honest truthful answer by maintaining anonymity to protect their identity and to state policy on confidentiality, especially when the questions contain sensitive issues. Quantifying adjectives in questions, unfamiliar abbreviations as well as vague, complex, and assumptive, anticipative, subjective, and suggestive questions should be avoided.
They should be simple, exclusive, clear, correct or appropriate, and provide meaningful information. There should also be organization, categorization, and smooth transition between items in order to make it easier for respondents to think, contemplate, and analyze their ideas. Multiple choice items with various responses are generally better and more efficient than open-ended questions.
It is important that a question includes all possible responses to prevent respondents develop the feeling of confusion and frustration. Ranking by importance and assigning a percentage to a series of items should be avoided, especially when it exceeds five items, because it is difficult and it makes the answers become less reliable. These guides should be considered before formulating the questionnaire.Sample Essay of EssayTigers.com