Working Projects
Although the top 100 business schools emphasize good leadership far more than the prevention of bad leadership, employees in 9.9 million Glassdoor reviews mentioned their leader 70% more often in the negative than the positive section, and negative leader mentions predicted company ratings 1.5 times more strongly. Five experiments across 40 countries (N = 51,433 adults) reveal that people are more willing to sacrifice time and money to avoid a bad leader than to obtain a good leader, across morality and competence, hypothetical and incentivized choice, and ideology and culture.
Revise and Resubmit
Business-size bias in consumer dishonesty
A preliminary study (N = 13,239 businesses) and seven main and seven supplemental studies (N = 10,436 adults) reveal that consumers are less likely to cheat small businesses than big ones. Consumers perceive small businesses as more vulnerable and moral than big businesses, which mediates lower dishonesty toward small businesses. Increasing perceived vulnerability and morality can causally reduce dishonesty toward big businesses.
Revise and Resubmit
Time-as-Money Thinking
We introduce time-as-money thinking, an individual difference capturing consumers’ tendency to view time through a monetary lens, and develop and validate a scale to measure it across 21 studies (N = 9,206 adults). Consumers who view time as money are more likely to pay to outsource tasks, pay premiums to avoid waiting, and choose time-saving offerings, over and above differences in income.
Nine main and three supplemental studies (N = 9,207 adults in the United States and the United Kingdom) reveal a focal self bias. When people compare two versions of themselves, such as their present and past selves or their personal and professional selves, they rate whichever self is the subject of the comparison more favorably, which violates logical symmetry. The bias also appears when people compare other people’s selves, and making a peer focal in self-other comparisons largely eliminates the better-than-average effect.
U.S. Democrat and Republican voters (N = 5,230 adults) could lie anonymously to double their own or another participant’s earnings. For selfish dishonesty, people lied equally whether it harmed their ingroup or outgroup. For coalitional dishonesty, meaning lying without benefiting personally, people lied 9 percentage points more to benefit ingroup than outgroup members.
Eight main and four supplemental experiments (N = 5,272 adults) reveal that creators are perceived as less creative when they integrate AI recommendations at the beginning rather than at the end of their creative work, partly because they are perceived as less effortful and less authentic. Communicating that the creator modified rather than directly implemented AI recommendations attenuates the penalty.
Revise and Resubmit
Recycling rivalry: social identity appeals and recycling
Across two preregistered season-long field experiments at a Norwegian Premier Division football stadium, standalone ingroup identity appeals did not increase recycling (N = 5,712 fans). Combining the identity appeal with a ranking of visiting clubs by recycling rate raised recycling to 52.9%, compared with 15.8% at baseline, and a second experiment (N = 6,598 fans) replicated this against a control (56.7% vs. 34.7%).
Under Review
Happy Growth? GDP-Well-being Relationship Across 160 Countries and Two Decades
Using Gallup World Poll data for 160 countries from 2006 to 2024, we document that 59.9% of countries experienced simultaneous GDP and happiness growth, while 37.5% achieved economic growth without happiness gains. The GDP-happiness correlation ranges from r = .76 in Europe to r = .07 in Africa, and countries with high social conflict exposure convert growth into happiness less effectively.
In Preparation
Greater female variability in wellbeing?
The greater male variability hypothesis proposes that males vary more than females across many traits, but little is known about whether this extends to wellbeing. Using three waves of the Global Flourishing Study (N = 207,092 adults; 23 countries), we find that females are more variable than males in physical health, mental health, financial security, and safety and housing security, and their wellbeing also varies more over time.
Ten studies (N = 6,824 adults) with American and French participants reveal that cancelers underestimate how acceptable others find being canceled on. Consequently, cancelers overestimate how much money to spend on gifts or the time they should spend to make it up to the person they canceled on.
Under Review
Do liars prefer to tell one big lie or multiple small lies?
We examine whether telling one big lie versus multiple small lies for equivalent gains affects individual dishonest behavior. In this registered report, two incentivized experiments (N = 3,600 adults) using anonymous die-rolling tasks test whether reporting outcomes in aggregated, simultaneous, or sequential formats affects how many people lie, how much liars lie, and whether dishonesty escalates across repeated reports.