What can you do to make your instrument more valid?

Question

What will be an ideal response?

Answer 1

When we want to design a research instrument, such as a questionnaire, we need to make sure that we try and maximise its validity. The first thing we can do is to make sure that
we are aware of the literature around the concepts we want to study, and start off with a theoretical framework that will influence the design of the survey. It is a good idea to get
some experts in the area to look at and comment on the instrument. More importantly, you need to pilot the instrument with a group of respondents similar to the sample you want to use. If, for example, you were going to survey 20 year six classrooms, it would be a good idea to pilot the survey in at least three classrooms. The pilot would be used to
look at any problems, such as items which don’t appear to have been understood or items with no variance (everyone gives the same answer). It can be useful to also interview or
conduct focus groups with your pilot pupils or a subset thereof, to get some more detailed information about possible changes to the instrument, and what they feel has or hasn’t
worked. If possible, it is helpful to collect data on factors that we think might be related to our instrument (for example, if we are looking at academic self-concept we could collect pupil grades), so we can test predictive validity.