Surveys can be a useful way to get a lot of information from a lot of people quickly. They can be administered by phone, mail, in person, or on the internet. Researchers can choose to use open ended questions or questions with forced choices, such as with the Likert scale. Researchers want to design surveys to maximize response rate and accuracy of responses. As an example, if a researcher wants to create a survey of cyber-bullying, s/he might want to administer this online as those who are experiencing cyber-bullying are already spending time online. If s/he wants to target adolescents, s/he will want to word his or her survey in a way that it is easily understood by adolescents. If s/he thinks adolescents might not want to spend a lot of time on the survey, s/he might choose to use forced-choice responses instead of open ended questions. This will also reduce the time that s/he spends coding and interpreting open ended responses. S/he might anticipate that some adolescents do not take the survey seriously and that some data might not be accurate.
Include reference from : Passer, M. (05/2014). Research Methods, 1st Edition.