In this section you need to think about the "why"s of monitoring and how management choices determine the kind of monitoring program you will design and implement. It is important to recognise the management context so that any program recognises the social and other limitations, and how the program will be interpreted and reported.
The way you approach your problem and how you define the question you will address in your monitoring program will depend upon:
Defining you question (or hypothesis) needs to show clarity and logic, including your assumptions and evidence. You need to leave a paper trail for others to follow; you have think about the ability of someone else to follow your thoughts and intentions if you are not there to explain it. This has the added advantage of making you think clearly about why you are designing the program and what you want out of it. Structured argument can be useful in clarifying the components of your "problem" and makes it easier to go back to redefine your problem if your program did not give the results you anticipated.
Reasons for monitoring
I suspect that most monitoring programs, especially in local government, are started with good intentions but without clear ideas about outcomes. The beliefs that monitoring is a "good thing" and that if we find a problem we can "fix it" are admirable in intent but are recipes for wasting resources and achieving little. A corollary to this is that many programs are based on assumptions/beliefs rather than evidence.
Peter Sandman introduced the concept that: Risk = Hazard + Outrage.
That is, "risk management" needs to recognise both the hazard (a technical concern) and the outrage (a social construct). He argued that when hazard and outrage are not congruent (one is low and the other high) then either a 'real' problem is not addressed or resources are wasted on something inconsequential. A sampling program might have the (covert?) intention, through reporting and education, to bring hazard and outrage into alignment, thereby facilitating further action.
The reason(s) we are proposing to undertake the program will determine the questions we ask and the way we define the "problem". In practice some sampling and monitoring programs will not be 'scientific experiments', being focused on this problem, at this site, under these conditions with no expectations of generality. So long as the researcher knows this and incorporates the consequences into analysis and reporting that can be a valid approach. The concern might remain, however, that a broader view of the 'problem' might have provided an opportunity for a different and more explanatory design.