Measurement of the conjunction error in social judgment: Answer choice and answer justification

People often judge the probability of two events occurring together to be more probable than the less probable of each of these events occurring separately, thereby demonstrating the conjunction error. “Correct” responses are those in which participants rank a single-element statement of low probability as more probable than a conjunction consisting of a low-probability statement and an additional statement. We demonstrated in two studies that task structure was related to the number of people who chose the single-element statement as more probable. However, relatively few participants provided a rationale for their answer choice based on the conjunction rule. In a third study, responses on the answer choice and answer justification measures converged. In addition to serving as a sensitive dependent measure, participants’ justifications when answering conjunction problems may provide insight into their reasoning.