When Bad News is Good

• Published in: Advances in Consumer Research Vol. 47; pp. 444 - 445
• Main Authors: Barasz, Kate, Hagerty, Serena
• Format: Conference Proceeding
• Language: English
• Published: Urbana Association for Consumer Research 01.01.2019
• Subjects: Consumer behavior; Decision making
• ISSN: 0098-9258

People generally try to avoid bad news. Most patients would rather not receive a bad diagnosis. Most tenants would rather not receive word that their rent is being raised. Most car owners would rather not find out their vehicle needs expensive repairs. But are there circumstances under which the opposite might be true-when bad news may, paradoxically, be good? Bad news often has the unique effect of guiding subsequent decisions and actions-even when those decisions and actions are themselves undesirable. For instance, a patient who receives objectively bad news ("You have a very severe injury") may be required to undergo surgery, while a patient receiving objectively better news ("You have a moderate injury") may have to choose between surgery and other treatment options. Thus, despite leading to an undesirable outcome, bad news may allow people to evade a difficult decision (i.e., whether or not to have surgery) and create situations in which people feel better with-and even actively prefer-objectively worse news. Thus, this research investigates when and why bad news is good. We compare people's news preferences-the information they hope to receive-when there is versus is not a consequent decision to be made (e.g., whether or not to have surgery). Further, when there is a decision to be made, we compare the affective responses of those who receive worse versus better news, and investigate the mediating role of decisional conflict. In so doing, we demonstrate the conditions under which a systematic preference for objectively bad (versus objectively better) news can emerge. Finally, we explore a potential perverse incentive created by bad news: forgoing opportunities to improve one's circumstances (or move toward "better" news) if such a move introduces a difficult decision. STUDY 1 All participants imagined having a shoulder tendon tear and awaiting news about the exact diagnosis; we simply varied whether participants had a decision to make based on news received. In the decision conditions, surgery was required above a specific threshold length (varied between-subjects as 2cm, 3 cm, or 4cm); below that threshold, participants could opt for surgery but the decision was theirs to make. In the no decision conditions, threshold lengths were discussed but no consequent decision was required. The DV was preferred tendon tear length. Naturally, smaller injuries are objectively better. However, we found that when facing a difficult decision, participants preferred larger, "above-threshold" tears (M=1.67cm, SE=.O5) than participants not facing a decision (M=1.28, SE=.O5; F(l, 542)=28.25, p<.001, q2=.O5). In addition, preferred tear length increased as the size of the threshold increased (M2cm=1.34, SE=.O6; M3cm=1.45, SE=.O6; M4cm=1.63, SE=.O6; F(2, 542)=5.09, p=.006, r|2=.O18). Critically, these main effects were qualified by a significant interaction (F(2, 542)=4.01, p=.O19, p2=.O15): among the decision conditions, there were significant differences in preferred tear length as a function of threshold size (F(2, 272)=6.04, p=.003, r|2=.O4); however, there were nonsignificant differences among the no decision conditions (F(2,270)= 1.15, p=.32, r|2=.008). In other words, bad news was not indiscriminately preferred-it was only when a difficult decision was required and was specific to the threshold above which a decision was eliminated. STUDIES 2A-B Studies 2A and 2B tested another medical scenario-whether or not to undergo an appendectomy-and varied the valence of news people received. In S2A, participants (betweensubjects) either learned they had a 50% chance of appendicitis/50% chance of false alarm ("better news") or that they had a 95% chance of appendicitis/5% chance of false alarm ("worse news"). (Importantly, a pretest confirmed that "worse news" was, indeed, perceived to be worse.) All participants were then told they would have to decide about surgery, and asked how happy, relieved, and anxious (RC) they felt. Paradoxically, participants receiving worse news reported greater positive affect than participants receiving better news (p = .01,d=.36). S2B used the same scenario but a within-subjects design: participants chose which type of news ("better" or "worse") they would rather receive. Most people (74%; vs. chance: p< .001) preferred worse news. By contrast, a pretest among a separate sample revealed that observers believed the opposite would be true: most (74%, 95% CI = [.65, .83]) believed that someone receiving the "worse news" would feel worse. STUDY 3 While worse news may beget worse outcomes, it can deliver greater certainty, which can lower decisional conflict and increase positive affect. Therefore, bad news should be most beneficial when people are especially conflicted, whereas bad news should not confer incremental benefit when an easy choice exists. To test this, S3 varied both diagnosis (better news or worse news) and disease- appendicitis (like S2) or appendiceal cancer. We replicated S2's results for participants in the appendicitis condition; however, because people confronting cancer strongly prefer action to inaction (Ubel 2012), we predicted (and found) that participants in the appendiceal cancer condition would find the surgery decision relatively easy to make-regardless of the valence of the news-and differences in affective response would diminish (interaction: p=.O48, r|2=.009). We also measured the role of decisional conflict as a mediator and found that decisional conflict varied as a function of news and disease type (interaction: p<.001, r|2=.O3). STUDY 4 What perverse incentives arise from bad news being good? Participants saw Si's shoulder tear scenario; everyone was told they had an "above-threshold" (3.5cm) tear. If interested, they could undergo intensive physical therapy which could reduce the tear size by 1cm, thus making it a "below-threshold" tear. In one condition ("no decision"), this improvement entailed no consequent decision; it would simply improve their injury. In a second condition ("decision"), this improvement also improved their injury, but meant going from definitely needing surgery to having surgery be optional-in other words, moving from having no choice to having a difficult choice. Our DV was interest in undergoing physical therapy (i.e., reducing their injury). People with (versus without) a decision to make were less interested in reducing their injury (p=.004, d=.41). Thus S4 results reveal one possible perverse incentive created by bad news being good: a reluctance to improve a bad situation when objective (but optional) improvement meant introducing a difficult decision. Five studies demonstrated when and why bad news is good. Study 1 varied whether or not a decision had to be made and examined the resulting effect on news preferences: when facing a difficult choice, many people paradoxically preferred objectively worse news-even when worse news begot less preferred (Pretest 1) or objectively worse (Study 1) outcomes. Studies 2A and 2B varied news type (better or worse) and showed that people felt better after receivKate ing worse news. Using mediation and moderation, Study 3 demonstrated the role of decisional conflict: bad news can reduce decisional conflict, which in turn can result in greater positive affect-but only within contexts where the right decision or choice is not obvious. Finally, Study 4 showed that people may be perversely incentivized to stick with bad news if improvement entails introducing a difficult decision. Our findings make several theoretical contributions. First, we contribute to a general understanding of how people respond to difficult, consequential choices. Previous research has shown that consequential choices are emotionally burdensome for any decision-maker (Botti, Orfali, and Iyengar 2009; Kahn and Luce 2003); our results suggest that, under some conditions, such choices may be even more difficult for decision-makers who have received objectively better news. For example, Botti et al. (2009) showed that when facing the gut-wrenching decision of whether to take their premature infants off life support, parents making the choice themselves felt significantly worse than parents for whom the decision was made (i.e., by a doctor). Applying our findings to this paradigm, we might expect that- among the cohort of parents making the decision themselves-those whose newborns had a better prognosis (e.g., very slim chance of survival) may have felt, paradoxically, worse than those whose newborns had the worst prognosis possible (e.g., no chance of survival). We also add to research on decision avoidance. Previous research has documented the many ways in which decision-makers try to dodge decisions already at hand (Anderson 2003)-offloading it onto someone else (Steffel and Williams 2017; Steffel, Williams, and Perrmann-Graham 2016), maintaining the status quo (Samuelson and Zeckhauser 1988), choosing inaction (Ritov and Baron 1992; Tykocinski, Pittman, and Tuttle 1995), or deferring a decision (Dhar 1997; Tversky and Shafir 1992). Adding to this, our findings document another important-and perhaps overlooked-tactic people may use to avoid a decision preemptively: a preference for worse news. Because bad news often eliminates options (Bor et al. 1993), it can result in bypassing difficult choices altogether, which can be beneficial to decision-makers. The fact that bad news can also beget bad outcomes highlights the costs people may be willing bear to avoid a hard choice. Although not the focal element of our investigation, the selfother difference that emerged may be an interesting and fruitful area for future research. In asking observers to make predictions about how news affects decision-makers, we wanted to confirm our conceptualization of news-i.e., that "better news" was, in fact, perceived to be better, and that "worse news" was perceived to be worse. We additionally wanted to demonstrate that the benefit of bad news is