Research Concepts

Models

Your model describes the way in which you expect that the variables you measure interact. Sometimes the model is also called the theory.

Just the act of selecting variables for data collection constitutes a rudimentary model. For example, a model of how voters choose between candidates would likely include the voter's party affiliation but probably not include the voter's shoe size.

The purpose of a theory is to provide an explicit framework for all of your assumptions about what variables are relevent and how they interact. Since we all bring assumptions to any problem, it is useful to have a theory that attempts to lay out all of our assumptions and their consequences. A danger with any research process is hidden assumptions that influence our observations or conclusions in unexpected ways that introduce bias.

Often a research project will involve independent and dependent variables. Every experiment necessarily involves at least one variable of each type. An "independent variable" usually influences in some way an outcome, or "dependent variable." In the case of voter preferences, the voters' party affiliations might influence how they vote.

Generally the researcher expects that changes in the value of the independent variable are associated with changes in the value of the dependent variable.

For another example, "gender" and "income" are known to be related in most professions, with females having lower income one exception to this is modeling). In this example, "gender" is the independent variable and "income" is the dependent variable.

Similarly, higher GRE scores are often associated with higher gradepoints in graduate school. In this example, gradepoint is the dependent varialbe and the skills measured by the GRE are the independent variable.

In both of these examples, our model posits a specific directional relationship between the independent and dependent variables. Sometimes your research might simply conjecture that there is a relationship without specifying a direction ahead of time. For example you might wonder if there males or females are more likely to exceed the speed limit on the Broken Arrow Expressway.