Table 3.
Proportion of Study 3 participants mentioning each coded response type
| Cognitive operation types | Coded response | Original Loss (n = 93) | Original No‐Loss* (n = 76) | Salience Big‐5 Loss* (n = 106) | Salience Big‐9 Win* (n = 90) | Reverse Order* (n = 105) | #comparison Instruction* (n = 96) | Less numerate (n = 274) | More numerate† (n = 292) |
|---|---|---|---|---|---|---|---|---|---|
| Identification | Probability of win | 63.4% | 75.0%b | 53.8% | 67.8%c | 59.0% | 58.3% | 60.2% | 64.0% |
| Probability of loss | 30.1% | 18.4% | 33.0% | 32.2% | 31.4% | 32.3% | 28.5% | 31.5% | |
| Pie chart | 2.2% | 5.3% | 5.7% | 2.2% | 2.9% | 2.1% | 3.3% | 3.4% | |
| Amount to win | 68.8% | 36.8%a , c | 61.3%a | 68.9% | 65.7% | 68.7% | 54.7% | 70.0%d | |
| Amount to lose | 61.3% | 11.8%a | 71.7%a , c | 62.2%c | 65.7%c | 70.8% | 50.0% | 67.8%d | |
| Avg # of identifications | 2.26 | 1.47a | 2.25c | 2.33c | 2.25c | 2.32 | 1.97 | 2.37d | |
| Comparison | Comparison p(win) and p(lose) | 7.5% | 5.3% | 9.4% | 8.9% | 9.5% | 11.5% | 9.5% | 8.2% |
| Cost of bet | 3.2% | 11.8%c | 3.8%c | 7.8%c | 3.8% | 6.2% | 1.5% | 9.9%d | |
| Comparison $9 vs. 5¢ loss | 9.7% | 0.0% | 14.2%c | 17.8%c | 15.2%c | 16.7%c | 9.1% | 19.5%d | |
| Avg # of comparisons | .31 | .17c | .27c | .34c | .29 | .34c | .20 | .38d | |
| Calculation | Expected value calculation | 2.2% | 1.3% | 0.9%c | 2.2%c | 1.0% | 1.0%c | 0.0% | 2.7%d |
| Other calculation | 17.2% | 18.4%c | 9.4%c | 15.6%c | 12.4%c | 16.7%c | 6.2% | 22.6%d | |
| Avg # of calculations | .19 | .20c | .10c | .18c | .13c | .17c | .06 | .25d | |
| Evaluation | Feeling or evaluation of probability to win | 34.4% | 42.1% | 24.5% | 21.1% | 26.7% | 27.1% | 28.5% | 29.1% |
| Feeling or evaluation of robability to lose | 10.8% | 5.3% | 16.0% | 15.6% | 12.4% | 10.4% | 11.3% | 12.7% | |
| Feelings or evaluation about win | 22.5% | 15.8% | 20.8% | 18.9% | 21.0% | 26.0% | 18.6% | 23.3%d | |
| Feelings or evaluation about loss | 40.9% | 5.3%a | 50.0% | 38.9% | 41.9% | 40.6% | 30.3% | 44.5%d | |
| Avg # of evaluations | 1.09 | .68a | 1.11 | .94 | 1.02 | 1.04 | .89 | 1.10d |
Coded responses for each bet condition were compared separately to the Original Loss condition in logistic regressions controlling for numeracy (in the comparison of Loss vs. No‐Loss, numeracy's interaction with condition was also included because it was the only time the interaction predicted attractiveness); similar linear regressions were conducted for the average number of each operation type.
A significant difference in the indicated condition vs. the Original Loss condition.
A significant interaction between numeracy and the comparison of the Original Loss and No‐Loss conditions.
A significant numeracy difference when controlling for the condition comparison (and the interaction when the two original conditions were compared).
In a separate set of analyses, we included all participants (all conditions) and examined whether numeracy differences existed, controlling for dummy variables of each condition.
A significant numeracy difference.