Surveys are done to answer questions about ‘What Is,’ and they can gather a lot of data that may be statistically analyzed and generalized. Survey instruments can be purchased, found in the literature or other sources, or custom-made by the researcher. Custom-designed surveys provide more flexibility to meet the unique information needs of the organization.
Published surveys should provide reliability and validity information to the user, as well as information on the populations that provided the reliability and validity statistics. Custom-designed surveys should be prepared by a professional and pre-tested with a sufficient number of subjects similar to the target population, so that the reliability and internal consistency of the instrument can be assessed. It is also necessary to subject the survey to expert review to make sure it is valid for the purpose and the population. The expert reviewers may be district administrators, university faculty, statisticians, etc. Checking the education literature for similar survey instruments is also advisable to establish validity.
The target population for the survey should be selected randomly, but in education settings, it is often necessary to administer it to classroom groups. It is good practice, and required in some districts, to require parental permission for students to take a survey. Certain research conditions must be met if statistical inferences are to be made from the results.