Theories and Models
Economists try to explain social and economic behavior by observing what goes on in the world, noticing some regular patterns that appear to be operative, and trying to explain those patterns. Our observations about the world are the facts. But facts by themselves don't really tell us anything!
Facts are nothing more than disconnected bits of data. It's up to us to link some of them together while discarding the rest. The process of going from specific observations to making generalized explanations about the observed regularities is known as inductive reasoning. It is the same reasoning used by all sciences, including economics.
A hypothesis is a possible explanation for the regularities or patterns we've observed. In science, we focus on hypotheses that can be tested to see if they work. After a hypothesis has been tested, and it seems to provide a reasonable connection between the cause and effect of what we've observed, we can then call the explanation a theory.
Economic theory is a set of related statements about the cause and effect of economic and social phenomena. The way we test economic theories is by making predictions. For example, an economic theory would be "if the price of something goes up, people will buy less of it." We could test that theory by observing whether or not cigarette sales decline after a price increase.
A model is a more formal statement of a theory. It often takes the form of stating the presumed relationship between two or more factors in a precise way. The factors included in a model are called variables because they are things that can change over time. In our earlier example, the price of cigarettes is a variable, and the quantity of cigarettes purchased is another variable. A theory links these two variables together in such a way that explains the causal relationship between them in a testable way.
It's often important to dispense with less relevant aspects of an issue so that we can focus on specific issues. We do this all the time. For example, when trying to determine the cause of an upset stomach, we focus on things like what we ate earlier, the presence of a fever, and so on. We don't usually include factors such as the strength of sunspots or the weather in Brazil. Similarly, when economists try to explain social phenomena, they key in on the factors that seem most relevant. This practice comes from the principle known as Occam's razor, which instructs us to strip away extraneous detail in order to expose the important aspects of a question.
Ceteris paribus: all else equal
When making predictions in order to test our models and theories, we often focus on the relationship between just two variables. In the example given above, we focused on the relationship between the price of cigarettes and how many cigarettes people will buy.
But a whole host of other factors are involved in determining the number of cigarette purchases. For instance, income makes a difference, as does the amount of advertising, styles, the age of the population, the findings of health research, and so on.
In order to focus on the relationship between price and the amount of cigarettes purchased, we need to hold all those other factors constant. To isolate those other factors, we employ the idea of ceteris paribus, which is Latin for all else equal. We can then say that, if all other factors remain constant, an increase in the price of cigarettes will cause a decrease in the number of cigarettes sold.