Variables and quantity: what else?

The article deals with topics as the choice of variables, their measurement and statistical analysis, the deductions based on data. All variables can be misleading if used in a misleading way, but the Author contends that the psychology based on the variables has not the possibility to represent selected samples of inner processes and contents. Quantitative analyses based on linear causality and probabilistic inference pose many problems, but some alternative approaches devised to cope with these problems are indicated. An hermeneutic approach aware of the constructivist ground of the scientific knowledge is proposed.