Methodology has occupied a peculiar role in the sociological enterprise. There are spokesmen who see little connection between methods, research activities, and the process of theorizing. The sociological enterprise may be said to rest on these elements: theory, methodology, research activity, and the sociological imagination. The interactionist's conception of human behavior assumes that behavior is self-directed and observable at two distinct levels—the symbolic and the interactional. An interactionist assumes that a complete analysis of human conduct will capture the symbolic meanings that emerge over time in interaction. All research methods must provide answers to the problem of causal inference. A method must permit its user to gather data concerning time, order and covariance between variables, while allowing the discarding of rival causal factors. When it is claimed that one variable or process caused another, it must be shown that the causal variable occurs before that it is assumed to cause.