Using the construct of 'job satisfaction', the following will get you started with some ideas on the process of designing an instrument.
When constructing an instrument measuring a latent construct such as job satisfaction it is important to conduct a review of the literature in order to explore what instruments may already exist. Part of reviewing these instruments in the literature involves determining the level of construct validity and reliability of the instrument. Does the instrument truly measure what it seeks to measure? Does it perform similarly across different individuals and populations- do the results maintain patterns? Pett, Lackey, & Sullivan (2003) argue that, "When we develop a scale, we are less interested in the scale items than we are in the construct they purport to measure." (p. 16). ...
Provides step by step explanation of how to identify a psychological construct and scale to measure the construct. Includes references.