Table 1. Comparison between relevant... | Emerald Publishing

Table 1.

Comparison between relevant extant research and the current study

Research	Numeric format manipulation	Summary of findings	Compare ratio vs. (%)	Different Numerators (1 / 20, 5%)	Underlying process	Boundary condition
Kirkpatrick and Epstein (1992)	1 red of 10 vs 10 red of 100	More participants draw from the bowl with the larger number of beans	No	No	No	No
Denes-Raj and Epstein (1994)	Ratio of large (9 red of 100 [9%]) vs small numbers (1 red of 10 [10%])	More people chose the bowl with the larger number of red beans (9 / 100)	No	No	No	No
Yamagishi (1997)	2,414 out of 10,000 vs 24.14 out of 100	Cancer is judged as riskier when labeled as killing 2,414/ 10,000 people vs 24.14 / 100 people	No	No	No	No
Pacini and Epstein (1999)	Optimality at issue (choose between 1 / 10 and 7 / 100 or 9 / 100) vs optimality not at issue (choose between 1 / 10 and 1 / 100)	When optimality is not at issue, people engage in heuristic processing; when optimality is at issue, responses depend on individual differences	No	No	No	No
Monahan et al. (2002)	Ratio (20 / 100) vs percentage (20%)	Higher risk estimates for ratio vs percentage only for physicians working in forensic facilities; no difference for the rest	Yes	No	No	No
Brase (2002)	Simple frequency (1 in 3), single-event probability (0.33), relative frequency (33%) and absolute frequency (90 mil)	Simple frequencies (ratios) clearer and easier to understand than the other formats; single-event probabilities difficult to understand	Yes	Yes	No	No
Tan et al. (2005)	Probability (5%) vs ratio (1 out of 20) of vaccine side effects	No significant difference in likelihood to accept the vaccine	Yes	Yes	No	No
Pinto-Prades et al. (2006)	X out of 100 vs. Y out of 1000	Risk perceived as higher when denominator is 1,000 (vs 100)	No	No	No	No
Keller and Siegrist (2009)	Ratio vs Pictogram vs Paling Perspective Scale (displaying the same frequency – e.g. 1:112)	Pictogram format results in significantly lower risk rating compared with Paling Perspective Scale and the ratio with numerator 1	No	No	No	No
Peters et al. (2011)	Frame (positive vs negative vs combined) by format (ratio vs percentage) – 10% vs 10 / 100, 90% vs 90 / 100	Lower risk perceptions for positive vs negative frame. Less numerate participants perceive lower risk with percentage vs ratio format; no difference for highly numerate	Yes	No	No	No
Slovic et al. (2000)	10% vs 10% of 100 vs 1 / 10 vs. 10 / 100 vs 20% vs 2 / 10 vs 20 / 100 (Study 3)	Patient judged as posing higher risk in ratio vs percentage frame. Judgments of the likelihood of harming someone higher for percentage vs ratio	Yes	Yes	No	No
Raghubir (2008)	S1: same numerator with small/large denominator S2: 90.3 / 100,000 vs. 903 / 1,000,000 S3: 0.903 / 1,000 vs. 90.3 / 1,000,000	Smaller numerators lead to perceptions of higher risk when denominator is not salient. Larger numerators lead to higher perceived risk when the denominator is salient	No	No	No	Yes (denom. salience)
Our study			Yes	Yes	Yes	Yes

Research	Numeric format manipulation	Summary of findings	Compare ratio vs. (%)	Different Numerators (1 / 20, 5%)	Underlying process	Boundary condition
Kirkpatrick and Epstein (1992)	1 red of 10 vs 10 red of 100	More participants draw from the bowl with the larger number of beans	No	No	No	No
Denes-Raj and Epstein (1994)	Ratio of large (9 red of 100 [9%]) vs small numbers (1 red of 10 [10%])	More people chose the bowl with the larger number of red beans (9 / 100)	No	No	No	No
Yamagishi (1997)	2,414 out of 10,000 vs 24.14 out of 100	Cancer is judged as riskier when labeled as killing 2,414/ 10,000 people vs 24.14 / 100 people	No	No	No	No
Pacini and Epstein (1999)	Optimality at issue (choose between 1 / 10 and 7 / 100 or 9 / 100) vs optimality not at issue (choose between 1 / 10 and 1 / 100)	When optimality is not at issue, people engage in heuristic processing; when optimality is at issue, responses depend on individual differences	No	No	No	No
Monahan et al. (2002)	Ratio (20 / 100) vs percentage (20%)	Higher risk estimates for ratio vs percentage only for physicians working in forensic facilities; no difference for the rest	Yes	No	No	No
Brase (2002)	Simple frequency (1 in 3), single-event probability (0.33), relative frequency (33%) and absolute frequency (90 mil)	Simple frequencies (ratios) clearer and easier to understand than the other formats; single-event probabilities difficult to understand	Yes	Yes	No	No
Tan et al. (2005)	Probability (5%) vs ratio (1 out of 20) of vaccine side effects	No significant difference in likelihood to accept the vaccine	Yes	Yes	No	No
Pinto-Prades et al. (2006)	X out of 100 vs. Y out of 1000	Risk perceived as higher when denominator is 1,000 (vs 100)	No	No	No	No
Keller and Siegrist (2009)	Ratio vs Pictogram vs Paling Perspective Scale (displaying the same frequency – e.g. 1:112)	Pictogram format results in significantly lower risk rating compared with Paling Perspective Scale and the ratio with numerator 1	No	No	No	No
Peters et al. (2011)	Frame (positive vs negative vs combined) by format (ratio vs percentage) – 10% vs 10 / 100, 90% vs 90 / 100	Lower risk perceptions for positive vs negative frame. Less numerate participants perceive lower risk with percentage vs ratio format; no difference for highly numerate	Yes	No	No	No
Slovic et al. (2000)	10% vs 10% of 100 vs 1 / 10 vs. 10 / 100 vs 20% vs 2 / 10 vs 20 / 100 (Study 3)	Patient judged as posing higher risk in ratio vs percentage frame. Judgments of the likelihood of harming someone higher for percentage vs ratio	Yes	Yes	No	No
Raghubir (2008)	S1: same numerator with small/large denominator S2: 90.3 / 100,000 vs. 903 / 1,000,000 S3: 0.903 / 1,000 vs. 90.3 / 1,000,000	Smaller numerators lead to perceptions of higher risk when denominator is not salient. Larger numerators lead to higher perceived risk when the denominator is salient	No	No	No	Yes (denom. salience)
Our study			Yes	Yes	Yes	Yes

Source(s): Authors’ own work