The Nature and History of Experimental Control - Boring - 1954 - Article

What does ‘control’ mean?

Control has three meanings:

1. A check: verification – standard of comparison sued to check inferences deduced from an experiment.
2. A restraint: checking in the sense of maintaining constancy – keeping experimental conditions constant and also to alter the independent variable according to precise known predetermination.
3. A guide/directing: producing a precisely predetermined change, a constant and restrained change.

What are Mill’s four methods of experimental inquiry? (Logic, 1843)

1. Method of agreement: if A is always followed by a, then A presumably is the cause of a. Mere agreement does not provide rigorous proof. Inference of causation is not safe when based on just agreement.
2. Method of difference: if A is always followed by a, and no-A is always followed by not-a, then A certainly causes a. this is equal to adding the control observation. This method gives control as a verifying check.
3. Method of residue: if ABC is a known cause of abc, and BC a cause of bc, then A must cause a, even though A can’t be produced without BC nor a without bc.
4. Method of concomitant variations: exists when there is a serie of differences, and in any other pair of concomitances, one concomitance furnishes a comparison or control of the others.

What is control as restraint/guidance?

The meaning of control as restrain/guidance is the common meaning of the term. In science it applies to keeping experimental conditions constant and altering the independent variable according to precise known predetermination.

What is control as a check/standard of comparison?

Control observations, series, and experiments – use of control in its original meaning as a check. It was in the 1870s that the world control started to be used in the sense of a check/standard of comparison regarding where a difference is expected to lie.

• Also around this time the meaning of the world was used in biology by Darwin.
• Wundt was already enough to control a piece of apparatus with an objective check, but the only control he admitted as valid for the human observer was rigorous in psychological observation.
• As experimental psychology developed, the design of its experiments became more elaborate. Speaking approximately, we could say that the formal design first developed in psychophysics, then in reaction experiments, then in memory experiments.
• The first investigation to use the full design (fore-test – practice – after-test) in the experimental group to be compared with (fore-test – nothing – after-test) in the control group was Winch in 1908, and the general use of control groups in these experiments begins then.