Abstract
Covariation information can be used to infer whether a causal link plausibly exists between two dichotomous variables, and such judgments of contingency are central to many critical and everyday decisions. However, individuals do not always interpret and integrate covariation information effectively, an issue that may be compounded by limited numeracy skills, and they often resort to the use of heuristics, which can result in inaccurate judgments. This experiment investigated whether presenting covariation information in a composite bar chart increased accuracy of contingency judgments, and whether it can mitigate errors driven by low numeracy skills. Participants completed an online questionnaire, which consisted of an 11-item numeracy scale and three covariation problems that varied in level of difficulty, involving a fictitious fertilizer and its impact on whether a plant bloomed or not. Half received summary covariation information in a composite bar chart, and half in a 2 × 2 matrix that summarized event frequencies. Viewing the composite bar charts increased accuracy of individuals both high and low in numeracy, regardless of problem difficulty, resulted in more consistent judgments that were closer to the normatively correct value, and increased the likelihood of detecting the correct direction of association. Findings are consistent with prior work, suggesting that composite bar charts are an effective way to improve covariation judgment and have potential for use in the domain of health risk communication.
Highlights
The covariation or contingency between two events concerns the degree to which they are associated and may be defined in terms of their co-occurrence—that is, the extent to which one event is likely to occur given the presence or absence of the other event
In both the contingency table (CT) and bar graph (BG) conditions, judgments of contingency were more negative when problems were easier; this trend was more pronounced in the CT condition, judgments in the BG condition appearing more consistent across the three problems
Understanding covariation information in order to determine a potential causal relationship requires the consideration and integration of four pieces of information: the frequency with which the candidate cause and the effect co-occur, the frequency that the candidate cause is present and the effect does not occur, the frequency that the effect is present in the absence of the candidate cause, and the frequency with which the candidate cause and the effect are both absent
Summary
The covariation or contingency between two events concerns the degree to which they are associated and may be defined in terms of their co-occurrence—that is, the extent to which one event is likely to occur given the presence or absence of the other event. Be interested in whether a revision program boosts examination success before deciding whether to enroll, and a patient with a skin condition might wish to ascertain whether a treatment is effective before deciding whether to use it or not. Given that such judgments can have far-reaching consequences, it is important to understand how they are arrived at and the factors that influence them, including how best to present contingency information to aid comprehension and facilitate optimal decisions
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
