Morrison Center for Marketing and Data Analytics

Thoughts of the future are commonplace in the daily lives of consumers. Yet, considering the future, with its various uncertainties, can be an intimidating and anxiety-provoking exercise. Yet, no work to date has investigated the ways in which various methods of prospection impact the present-day and anticipated emotions that consumers experience. In this project, we set out to compare different methods of prospection against one another in an attempt to determine which, if any, may benefit consumers. Initial results indicated that not all forms of prospection are created equal: participants who engaged in planning and set intentions indicated that they felt more positive in the current moment than those in the control condition, whereas simulation and prediction afforded no such benefits. We are planning to conduct an additional set of 3 studies to further investigate the emotional outcomes that vary with each prospection type, and to test potential mediators of the relationships we uncover.

Although preventative behaviors are critical for health and well-being, they are also challenging to motivate, particularly among those without an established routine who may be least likely to engage in prevention behaviors. As a result, motivating preventative behaviors, or actions taken today to prevent future adversity, has presented a longstanding behavioral challenge as well as an opportunity for improving welfare. In this research, we aim to demonstrate a key source of heterogeneity in interventions that aim to motivate such preventative behaviors. We will specifically investigate whether calling to mind one’s past adversity in the health or financial wellness domains can encourage preventative behavior, and predict that this messaging intervention will be uniquely effective for those who have been exposed to related adversity. These findings contribute practical insights for social psychological interventions to account for individual-level heterogeneity and increase welfare among those who are most vulnerable to adversity.

Moral rationalization is the process of convincing oneself that an action does not actually violate one’s moral standards (Tsang, 2002). As a result, rationalization fosters an environment conducive to immoral action, and thus is harmful for both individuals and society. The primary purpose of this research is to study the process of rationalization, investigate methods for combating this process, and develop an engaging and interactive intervention that employs these rationalization reduction methods. By interrupting the process that liberates individuals to break with their moral standards, this research serves to reduce immoral behavior and improve the wellbeing of individuals and society.

From deadly viruses that infect our bodies to Siri and Alexa listening to everything we say, concerns over unwanted or unnatural intrusions are common in today’s society. In all of these contexts, a label using a variation of the word “invasive” might be applied (i.e., an invasive disease, an invasion of privacy). Although the normative implications in these specific contexts are fairly clear, where less invasive diseases or technologies are better than the alternative, there are other settings in which a label of “invasive” might come along with different assumptions or values. For example, a patient choosing between treatment options for a cancerous tumor might infer higher rates of success for an invasive surgery vs. a noninvasive approach. In the present research, we seek to explore the assumptions, values, and consequences of practices varying in invasiveness across two key settings with important but distinct implications for consumer choice: consumer privacy and medicine. Specifically, we will investigate how consumers, patients, and doctors define and perceive invasiveness and how potential differences in their interpretations influence judgments and choices.

In just the past four decades, the wealth gap between the richest and poorest families in the United States has grown at an astounding rate: in 1983, families in the top income tier had 28 times the wealth of families in the lowest tier, but by 2016, this grew to 75 times (Pew Research Center, 2020). The consequences of this rapidly rising inequality go beyond the harmful financial consequences for those living in poverty, contributing to imbalances in political power, happiness, and wellbeing (Buttrick & Oishi, 2017; Hing et al., 2019). One of the main tools that a government has to address inequality is redistributive policy, including the use of taxes on the wealthy to provide benefits and services to those in need. Existing evidence suggests that support for redistributive policies is associated with beliefs about whether poverty is driven by dispositional or situational factors. In this research, we investigate (1) whether existing attributions differ for poverty and prosperity, (2) how the framing of redistributive policies conveys narratives around attributions for poverty and prosperity, and (3) how these narratives affect individually held beliefs about the determinants for poverty and prosperity.

Any time you log into Netflix and see “Popular on Netflix Now” or receive a household energy bill comparing your consumption to your neighbors’, someone is trying to influence your decisions by comparing your behavior to a social norm. In past research (Bogard, Delmas, Goldstein, and Vezich, 2020), we demonstrated that this sort of normative feedback is not a monolith, as scholars have typically treated it, but instead contains three separate dimensions: The target (i.e., the specific reference group used as the benchmark), the Valence (i.e., whether one has outperformed or under-performed the Target), and the Distance (i.e., how far from the Target one’s performance is). Whenever someone receives normative feedback, we showed, they are actually receiving information along each of these three dimensions. Through several field and laboratory studies, we documented the independent effect of each dimension on people’s behavioral response to normative feedback.

These findings invite a natural question: How do these three dimensions interact? That is, how does the effect of one dimension depend on the specific values of the others? Is negatively valenced feedback always bad, or might slightly negative feedback actually improve performance more than moderately positive feedback? Does it depend on whether performance is being compared against a low-, middle-, or top-performer? The goal of the present investigation is to answer these sorts of questions.

We analyze the impact of a temporary shock to food supply on households' dietary choices. We use hurricane-induced closures of grocery stores, which are typically short term. Results show that store closures influence household purchasing patterns even after the grocery store has reopened. We find a decrease in the nutritional value of household purchases for the six-month period after the store has reopened, despite no change in total expenditures. This finding supports the hypothesis that supply factors play a substantial role in shaping household diets.

How do consumers assess whether to purchase products at a given price? One approach is to consider whether prices equitably reflect input costs. A company could present labor costs in terms of either time or money e.g., “1/2 hour” vs. “$7.50” of labor (for a fixed rate of $15/hour). Prior research shows that time is often substituted for other attributes (e.g., content, quality) and that highlighting time or human effort might increase valuations. Given this, we might expect companies to be better-off when communicating costs framed in terms of time, instead of money. However, results from four pilot experiments seem to suggest the opposite. In these studies, consumers indicated being more likely to purchase a product when costs were framed in terms of money rather than time. Furthermore, participants perceived greater time investment for products framed by financial transparent costs, rather than temporal transparent costs and this time investment was strongly associated with purchase likelihood. Our initial findings seemingly challenge the conventional wisdom from the literature; however, they are not mutually incompatible. Instead, our work may shed light upon how framing costs in terms of money may paradoxically increase perceptions of invested time.

Consumers dislike debt and utilize many strategies to avoid it. We propose one such strategy by considering how consumers respond to information about incomes vs. balances. Income is a flow and represents the rate at which a consumer’s balance grows. Balance is a stock, representing the accumulation of prior inflows (income) and outflows (spending). Though mathematically translatable (either can be derived from the other), we propose consumer spending may be sensitive to either income or balance information. Specifically, we propose income information will cause people to systematically underspend their financial inflows in a strategic attempt to avoid balance deficits. This strategy avoids debt; however, it also leads to accumulation. Whereas consumers who observe balance information may repeatedly spend down their accounts, consumers with only income information will accumulate balance from unspent inflows. Results from a preregistered and incentivized experimental game lend support to these predictions. Across four between-subject conditions, participants underspend their income most frequently—and therefore accumulate more money over time—when presented with information about their income and not their balance. This ongoing research seeks to further our understanding of how and why consumers make spending decisions in the face of different forms of financial information.

In choice contexts, consumers frequently consider past information received from other people. Friends share their experiences from restaurants, critics write movie reviews, and past guests leave online reviews for an AirBnB. Source memory, the ability to link a claim to its original source, is an essential aspect of accurate recall, attitude formation, and subsequent decision making. Across a set of pre-registered experiments (N = 3,308), and a variety of consumer environments, we find that source memory is more accurate for subjective opinions than for objective factual statements, indicating that opinions are more likely to be correctly attributed to their sources than are factual statements.

With the main effect robustly observed, we aim to understand the downstream consequences, boundary conditions, and underlying processes of this effect. One ongoing set of studies focuses on the role of source expertise. Another set of studies manipulates the perceived objectivity of individual claims as well as the extent to which claims provide information about the source, to understand the processes of the observed effect of claim objectivity on source memory. In this research, we replicate and expand upon an important observed effect, aiming to further our understanding of the role of claim objectivity in source memory.

This research first examines whether and under what conditions inferences about the prestige of an award suffer when women or racial minorities win the award (“If she won this award, it cannot have been that important”). And second, how to prevent such biased prestige inferences. We conduct lab experiments to study how award recipients from diverse backgrounds impact perceptions of the award. Our focus lies on scholarly awards in STEM fields, where we vary the stereotypicality of the awards by comparing research excellence awards (stereotypically given to white male, high status) to service awards. We predict that awards will be deemed less prestigious when women and racial minorities win them. Moreover, we predict that the stereotypicality of awards moderates this effect. We hypothesize that the prestige of research excellence awards (stereotypically given to white males) will be derogated to a greater extent with minority recipients (vs majority recipients) than service awards. Lastly, we test what interventions can reduce or eliminate such biased prestige evaluations. We predict that a greater award value (prize money) and more information on the selection process (e.g., the jury) will close the perceived prestige gap between majority and minority award recipients.

People often get stuck in ruts, continuing unfavorable activities when they could easily switch to preferred alternatives. We deem such behaviors—continuing less-preferred activities while passing up clear opportunities for improvement—change failures. Daily life is filled with instances of change failures—envision an individual struggling to complete a task on their phone (e.g., read, shop, email) rather than switching to a nearby computer where they could do it more easily; or, consider someone continuing to watch an unenjoyable TV program, rather than switching to the fun book sitting next to them on the side table. We propose such behavior-change failures are driven, in part, by a novel underlying cause: behavioral entrenchment, a state of increasing task-set accessibility that makes switching feel more difficult. We aim to further examine behavioral entrenchment, improve our understanding of why it occurs, and identify interventions to attenuate it. More generally, this research will shed light on why people get stuck in ruts and provide insights to help manage behavior change.

Segmentation and targeting are cornerstones of marketing strategy. Segmentation refers to the practice of dividing a market into distinct groups, while targeting describes the selection of which of those groups to serve. Often, these groups are defined by demographic characteristics like race or gender. However, recent cultural movements like Black Lives Matter and #MeToo have heightened concerns about social justice, and now segmentation and targeting by race and gender may seem increasingly inappropriate. These examples raise a natural question, given the ubiquity of these strategies: Do consumers today believe such practices are fair? In this research, we find that perceptions of fairness suffer when advertising specifically targets groups based on race or gender, relative to advertising that broadly engages the general public—even when promoted products and services would be beneficial to the targeted segment.

Decisions under uncertainty are ubiquitous, from estimating the travel time to a destination to deciding what city is most attractive. People need to both learn information that will inform their expectations about uncertain outcomes (Oliver and Winer 1987) and to evaluate options with uncertain attributes (Tversky and Kahneman 1992). While there is extensive research on multi-attribute choice problems (Wallen et al. 2008), this has largely focused on attributes that are known or that are risky. By moving to uncertain attributes, researchers must also account for how people learn and incorporate information that is uncertain (Hasher and Zacks 1979; Hertwig 2015). In one project, we test the impact of metrics on estimates of uncertainty (e.g., how the estimate of a distribution of buyer offers on a house changes when estimated in sale price or home equity). The second project tests how people make choices with multiple, correlated uncertain attributes. Correlated attributes, but specifically negatively correlated attributes, tend to defy the predictions of many choice models (Johnson, Meyer, and Ghose 1989) and thus deserve additional investigation into preference formation.

This paper studies the expansion of dollar store chains in the U.S. retail landscape following the Great Recession (2008–2019). This expansion has been accompanied by growing public concern over the impact on other local retailers and food accessibility in small, rural communities. We develop an empirical framework to evaluate the efficiency of the free entry equilibrium and impact of entry regulation on spatial market structure. A dynamic game of entry, exit and investment into spatially differentiated locations is specified, allowing for chain-level economies of density.

Choosing when to claim Social Security Administration (SSA) benefits is a critical determinant of financial health during retirement. Complicating this important decision is the role of heterogeneity; each individual’s decision is affected by individual factors that limit the usefulness of standardized advice. While past work has examined heterogeneity in claiming decisions, there has been little research to systematically test the psychosocial correlates of early or delayed SSA claiming or the effectiveness of theory-driven interventions meant to help with the claiming decision. Given that prior research has typically tested one intervention or individual difference at a time, we systematically examined how several different interventions affect when consumers intend to claim SSA benefits and how such intentions correlate with individual differences. Using a pre-registered design with replications, we analyzed intended claiming age as a function of theory-driven interventions, individual difference measures, and relevant interactions between interventions and individual differences. In so doing, we not only generate better predictive models of which retirees will decide to claim earlier or later, but also provide a methodological roadmap for data-driven theory building for complex consumer decisions.

When confronted with a set of alternatives, how do consumers choose? This project aims to advance our understanding of consumer choice processes in two important particular contexts through the use of eye-tracking. First, we examine the role of attention to specific attributes and how it varies across joint vs. separate evaluation (e.g., Hsee 1996). We will assess the extent to which the effect of attribute comparisons is driven by (a) attention to those attributes, and/or (b) weight assigned to those attributes, conditional on attention. Through online pilot testing, we have already developed a validated set of stimuli for this project. Second, we will examine multi-stage decisions. Recent behavioral work (Spiller & Ariely 2020) has considered how evaluation of a choice between options (e.g., a gift card that can be used at one of two different stores) depends on its average use rather than just its more-preferred use. Even more recently, we have extended these results to situations with objective (rather than subjective) dominance in preferences. Moving forward, measuring visual attention and modeling its effects on choice, we aim to better understand how value is integrated across alternatives when one immediate option opens the door to multiple differentiated subsequent options.