Systematic observation is a calculated form of observation used to either support or disprove a hypothesis. For an observation to be systematic, it must be free of bias and repeatable.
Casual observation is done haphazardly while sitting in park or moving through a public space. Often referred to as "people watching", casual observation entails simply noticing people and mentally recording their behaviors and actions. Systematic observation is much more rigid and is often used in scientific research. Because systematic observations are commonly used as evidence to either support or contradict varying hypotheses, they must be executed in a controlled manner that lends their results credibility.
The credibility of a systematic observation comes from multiple factors. First, only specific behaviors related to the topic being studied are recorded in systematic observations. Further, attempts are made to eliminate bias of any kind that could skew the observation's findings. This is often done by tasking multiple observers with watching for the same behaviors. If they report back with conflicting findings, the procedures are rewritten and the observation is carried out again. Finally, systematic observations must be made in a way that allows others to replicate them. If others cannot recreate the observation, it is often viewed as unreliable and its findings are not seen as scientifically valid.