ABA Assist

Experimental Control

Definition

The state of affairs in which the experimenter can make the behavior change by manipulating the independent variable, with confounds ruled out. In single-case designs it is shown by prediction, verification and replication: baseline predicts what would happen without treatment, a return to baseline verifies the prediction, and reintroducing treatment replicates the effect.

Example

  • Each time a token system is in place the behavior improves, and each time it is withdrawn the behavior returns. The team has experimental control.

Not experimental control

  • Behavior improves after treatment starts and never returns when treatment is paused. The effect may be real, but control has not been shown.

See also

Written from

  • sidman-1960

Written from Sidman (1960). No textbook glossary consulted.

This entry has not yet been through clinical review. Treat it as a draft.

Something wrong here? Report it.