The scientific method of sociology suggests that human behavior is a predictable variable that can be observed and quantified. It suggests that scholars can name a behavior, identify when it will happen, and then observe people to see if the behavior shows up as predicted.
Because this method has been so successful at predicting certain behaviors, sociologists have now become comfortable making general statements, or "laws," regarding human behavior. These laws are generally widely accepted. One such law is: "children who grow up in single parent homes don't do as well as those who grow up in two-parent homes." Another example is: "Girls generally don't do as well as boys on standardized math tests."
Some scholars and researchers believe that these very laws, and the degree to which they are accepted, influence human behavior and make people behave in more predictabile ways. By feeding the brain certain expectations, these laws may in some cases be causing the very behavior that they "predict." For example, girls may be influenced subconciously by the belief that they aren't as good as boys at math. In that case, the prediction or law itself is a significant factor in the very behavior that it "predicts." The problem with teasing out prediction from causality is a major theoretical conversation within the discipline.