In the example to the right, it initially seems like religion causes criminal activity (21). But it's important to remember that many research studies do not prove that one factor (like number of churches) causes another (like amount of crime). Instead, they show that the factors are related.

The only studies that can prove that one factor causes another are experiments. In an experiment, the researcher randomly places the research participants in a control group or an experimental group. The research conditions for the control group and the experimental group are exactly alike except for the single factor that the researcher wants to study. Because all other conditions are the same, the researcher is able to conclude that the changes between the control group and the experimental group are caused by the studied factor. In studies other than experiments, the differences between the two groups could have been caused by extraneous factors.

The relationship between churches and crime, for example, is caused by the number of people living in a city. A larger population leads to more churches and more crime.
Do not assume that one factor causes another just because they are related.

Next :Meaningful Relationships

Previous: Statistical Significance vs. Practical Significance