In the simplest form, a hypothesis is an assumption that is a tentative explanation of the facts about something. The hypothesis is made in order to test or to examine the assumption's validity or veracity. Hypotheses are inferential in nature, meaning that there is an act of inference that if one thing is true then it follows that a related thing is true. If a fact is true for a sample of a population, then it can be inferred with varying degrees of certainty, that the fact is true for the general population. In some cases, hypothesis is used to infer merely that something needs to be explored so that explanations can be made.
Hypothesis is not theory, despite the incidences of getting to a very fine line of distinction between the two. Theory is generally considered to tested and confirmed to the point where it can be considered to be proved and reliable. Of course, even theory is tested and reviewed and may eventually fail to explain all of the facts about something.
Hypothesis is made when there is not enough evidence to prove that an inference is true. Then hypothesis covers the inference until it can be tested and either validated or goes nowhere. In scientific protocol, the hypothesis is not put forward until there is a test that can be done for the hypothesis. The hypothesis can then serve as a prediction, based on sound inductive or deductive reasoning, that if condition A exists, then condition B exists; that an increase in variable A will cause an increase in variable B, and so on.
In some sociological cases, while testing the hypothesis that change in one causative variable causes change in a dependent variable, other variables may be identified as causative or dependent variables. The hypothesis must then be modified to make new inferences based on the new discovery.
In many sociological investigations, multiple hypothesis apply, especially in the area of large program development and testing, where many causative variables are introduced in the hope that many dependent variables will change for the better; where there are many problems with internal and external validity, and where there are new variables that may be discovered.
In some cases, the hypothesis requires careful identification, definition, and descriptions of the relationships between variables before any inferences can be made.
The most common types of hypothesis are simple versus complex, alternative, null and alternative, cause and effect, and directional or non directional.
A null hypothesis is one which leads to nothing, or to no development of any merit.
An alternative hypothesis is frequently developed in order to be ready for any encounters with intervening, external, or internal threats to validity before the intended hypothesis is tested.
Null and alternative is simply a combination of the two. The goal is for the alternative hypothesis to be null or to go nowhere, and for the intended hypothesis to be proven or verified. The goal is also for one or the other to be null, since if both the main and the alternative hypotheses are valid, then there are problems with all of the hypotheses, and the examination needs to be restructured.
Cause And Effect hypothesis is the most well known, but is much more complex in the social sciences because of the complexity of systems and relationships between components of the systems within which humans operate, and upon which humans act. Cause and effect works with Directional Hypothesis, where the direction of the change or whether the change is quantitatively positive or negative is inferred. In Non Directional Hypothesis, there is no direction specified.
In the social sciences, one causative variable may cause change in many dependent variables. There may be chains and systems of relationship where change in A causes change in B, which causes change in many variables that are dependents of B, one of which may be recursive to any of the points in the chain. This is a common problem with testing hypotheses that involve programs as causative variables, as the program itself is multifaceted, the subjects are multifaceted, and the potential threats to internal and external validity are multifaceted.
This leads us to the Complex Hypothesis, which can go on for pages and pages, especially when all of the facets and complexities of social programs, social movements, social engineering endeavors, and social forecasts are considered in developing unified hypothesis and theory.