A good hypothesis is based on a good question and should be straightforward, specific, and stated in advance.
A simple hypothesis has one predictable variable and one judgment criterion: premature menopause is associated with a higher risk of fragility fractures.
A composite hypothesis has more than one predictable variable: premature menopause, age, and thinness are associated with a greater risk of fragility fracturesor has several judgment criteria: smoking is associated with a risk for cardiovascular disease and lung cancer.
A composite hypothesis cannot be tested with a simple statistical test.
A specific hypothesis is unambiguous with regard to its components and its variables, and to the statistical test that is to be applied.