ABSTRACT
Science progresses by examining its assumptions–often finding that they are faulty. Although this proposition was originally stated for substantive knowledge, it also applies to measuring in the social-behavioral sciences. In fact, our measuring processes are behavioral processes: Almost every datum is produced by a person–a subject or an observer. Although we assume we know what is going on in the process leading to a datum, we have little established knowledge or theory about the process. Hence we need mini-theories or rationales, one for each type of measuring. That theorizing can–in fact, should–draw upon general substantive theory but should be carried beyond such theory to describe and explain the behavior in the measurement context. The behavioral process in the person making a recorded measurement is behavior generated by the research process and would not have occurred if research were not being done. Even when archival or other records of behavior are used, someone takes the behavioral step of producing measurements from those records.
