100% of anything looks good--the appeal of one hundred percent and the psychology of vaccination

Meng Li M.S., Gretchen Chapman Ph.D

Rutgers University

Poster Presentation, Society of Judgment and Decision Making 28th annual convention, 2007, Long Beach, CA

Abstract

People overweight certainty, even when it is just an illusion. In study 1, participants (N = 470) preferred a vaccine that was 100% effective against viral infections that cause 70% of cancer cases to a vaccine that was 70% effective against infections that cause 100% of cancer cases. Study 2 (N = 129) illustrated the appeal of 100%, even if it does not refer to probability: vaccines with either 100% effectiveness or 100% target range were preferred to other vaccines that were less than 100% effective towards less than 100% target. The preference for 100% effectiveness towards a subset of targets was unaffected by framing the vaccine in a broader target scope. We propose that people overweight 100% in general when they make decisions involving percentage, be it probability, proportion of population, or subset, despite the fact that almost anything can be described as 100% of something.

Paper Summary

People overweight certainty, even if it is an illusion. A vaccine that is half effective is less attractive than a vaccine that is fully effective against half of disease agents (Slovic et. al. 1982). The current research extended this pseudo-certainty finding and revealed a “100%” effect. Using hypothetical vaccines resembling the new HPV (Human Papilloma Virus) vaccine, we investigated whether vaccination intention is influenced by 1) framing of effectiveness 2) salience of other disease risks not protected by the vaccine 3) the appearance of 100%.

Prospect theory predicts that risk elimination within a subset of targets is attractive because a feeling of certainty is involved. In study 1, we asked participants in an internet survey to indicate their intentions to vaccinate towards one of two vaccines. A): 100% effective in preventing virus infections that cause 70% of known cases of a specific type of cancer; B): 70% effective in preventing virus infections that cause all the known cases of a specific type of cancer. Subjects indicated higher vaccination intentions for the first vaccine. This replicated a pseudo-certainty effect— perfect effectiveness for a subset of targets is preferred because it is viewed “certain”, while partial effectiveness for the entire target set is not.

All vaccines protect against a fraction of all disease risks; perfect effectiveness in a small scope is less-than-perfect in a wide scope. Will people realize their feeling of certainty is an illusion if they are reminded of other disease risks? In study 2, we asked subject to rate both vaccines from study 1 on attractiveness (for equivalence, “100% of” was used instead of “all” in vaccine B). We varied sentence structure and scope between subjects (fully crossed). Sentences were structured in either one of two ways: 1): “it is 100% (70%) effective towards virus strains that cause 70% (100%) cases of cancer X”, or 2): “against virus strains that cause 70% (100%) cases of cancer X, it is 100% (70%) effective”. In the narrow scope, only cancer X (target of the vaccines) was mentioned; in the wide scope, an additional cancer Y unprotected by the vaccines was also mentioned. The pseudo-certainty effect was unaffected by sentence structure or by scope. The results excluded the precedence of “100%” to”70%” as an alternative explanation, and demonstrated the robustness of the pseudo-certainty effect.

Study 2 also included 3 vaccines in addition to the two vaccines discussed above. These 3 vaccines were described as 74%, 82%, or 95% effective towards virus strains that cause 95%, 85%, or 74% cases of cancer X, respectively. The net risk reduction for cancer X was held constant among the 5 vaccines. The results indicated that the vaccine “70% effective towards 100% cases” was less appealing than the vaccine “100% effective towards 70% cases”, but more appealing than the three “medium range” vaccines. The preference was not moderated by scope. People seem to overweight “100%”, whether or not it refers to effectiveness or range of target, but they do so more when it refers to effectiveness than when it refers to target range. Subjects were also asked how likely they would be to vaccinate if their favorite vaccine was the only vaccine available among the 5. People who favored vaccines that were either 100% effective or targeting 100% cancer cases indicated higher intentions to vaccinate than people who favored the “medium range” vaccines. This finding further demonstrated the influence of “100%” on vaccination intention.

Overall, the current research demonstrated how vaccination decisions can be influenced by the appearance of “100%”, whether it means certainty, or something else.