In order for an ABA study to be considered valid it must have all of the following components.
At least 1 subject
An experiment must have at least one subject. This can be either an individual or group. It can also have multiple participants.
Setting
An experiment must have at least one setting or place in which the experiment takes place. although there could be more than one if using a multiple baseline across settings experimental design.
Behavior
An experiment must have at least behavior or dependent variable, although there could be more than one if using a multiple baseline across behaviors experimental design.
Treatment
An experiment must have at least treatment or independent variable, although there could be more than one if comparing the effects of multiple interventions on behavior such as in an alternating treatment design.
Experimental design
An experiment must have a formal method in which it is conducted known as an experimental design. This could be a withdrawal design, a multiple baseline design, alternating treatment design, changing criterion design or a variation of any of these.
System for Measuring
An experiment must also have a system for measuring data and a way to visually analyze the date which is usually in the form of a graph.
References
Cooper, J. O., Heron, T. E., & Heward, W. L. (2007). Applied Behavior Analysis (3rd Edition). Hoboken, NJ: Pearson Education.