How we change what others think, feel, believe and do
Hypotheses are are the heart of much social research. Here's some useful notes about them.
A hypothesis is a predictive statement.
A researcher starts with a hypothesis and determines whether it is true or false, and the conditions for truth or falsehood.
Where there are several hypotheses, they are often abbreviated with a capital 'H' and a subscript (often a number).
The primary (or alternative) hypothesis (or just the first in a long list) is often written as H1.
Experiments often seek to prove the primary hypothesis to be true and may be considered to have 'failed' if this is not achieved. In reality, an experiment succeeds when it correctly concludes that the primary hypothesis is true or false. It only fails when its conclusions are incorrect (or that a correct conclusion is incorrectly reached).
The null hypothesis (often written as H0) is the opposite of the primary hypothesis in that it includes all other outcomes.
Note that if H1 is that X happens, then H0 includes both Y happens (which is other stuff) and that nothing happens. 'Not X' means anything but that which is predicted by the primary hypothesis.
For example if you hypothesize that feeding cats dog biscuits will make them purr more, then the null hypothesis would include both that feeding cats dog biscuits makes them purr less or feeding them dog biscuits does not change their purr rate.
Experiments that show that the null hypothesis is true do not necessarily show that the primary hypothesis is always false, only that it is not true in this situation. This is the same principle in law where a person is found not guilty (but not necessarily innocent).
Rather than giving up when their primary hypothesis is not proven, many researchers will simply return to their studies, seeking more specific conditions in which it may be proven true.