Means literally ‘to throw something into disorder’. In a research context, this term is used when there are two or more explanations for a significant result, usually because a study cannot control one or more extraneous variables.
See Causal pathway, Confounding variable, Contamination, Experimental method