Standard Dictator Game Research Articles

Seemingly altruistic behavior and strategic ignorance in a dictator game with potential loss.

Previous studies have indicated that in the standard binary version of the dictator game, people are less likely to behave altruistically when given the opportunity to be strategically ignorant. The present study aims to assess the robustness of individuals' strategic ignorance in the context of the emergence of empathic concern. It is reasonable to hypothesize that traditional standard dictator games may not be an optimal context for the emergence of empathic concern. Accordingly, the purpose of the present study is to use a dictator game with loss potential that facilitates player empathic concern to investigate the emergence of (seemingly) altruistic behavior and strategic ignorance in the context of empathy. The results of a web experiment with 200 Japanese adults showed that our manipulation of loss had no effect on the occurrence of altruistic behavior and strategic ignorance. Moreover, even in situations in which the participant, as a dictator, is aware that the recipient has suffered a loss, they behave as if they have a legitimate reason not to act altruistically by being strategically ignorant. This result demonstrates the robustness of strategic ignorance and indicates that evoking empathic concern may not have enough of an effect to influence altruistic behavior in the laboratory experiment.

The willingness to pay for voice in dictator games

Voice is valuable in relationships, but how much are people willing to pay for it? This paper reports findings about recipients’ willingness to pay for voice in standard dictator games. From our data, we obtain two key findings: first, people underestimate the true value of voice and, second, they are particularly skeptical of the value of voice after an irrevocable allocation decision has already been made (i.e., people underrate the benefits from anticipated messages).

On the Interpretation of Giving in Dictator Games When the Recipient is a Charity

In experimental dictator games, subjects typically give substantial portions of their endowment to their partner, which has been interpreted as evidence that individuals are altruistic. Two influential studies find that when the choice set is expanded to allow subjects to take money away from an anonymous partner, they no longer give money. This suggests that when partners are anonymous, the seemingly altruistic behavior observed in standard dictator games is an artifact of the limited choice set. A vast literature employs dictator games in a different important context where the recipient is not anonymous: charitable giving. In this study, we examine whether generosity vanishes upon allowing subjects to take money when dictators are partnered with a charity instead of an anonymous partner. While we replicate the result that subjects are substantially less generous when the option to take money away from an anonymous partner is added to their choice set, we find that subjects remain generous when the take options are added when the partner is a charity. Regardless of whether subjects are able to take money from the charity or not, subjects give positive amounts just as often, the mean and median donation amounts are similar, and the modal choice is to give the entire endowment to the charity. These results suggest that dictator games remain a useful tool in studies of charitable giving.

Are women more generous than men? A meta-analysis

We perform a meta analysis of gender differences in the standard windfall gains dictator game (DG) by collecting raw data from 53 studies with 117 conditions, giving us 15,016 unique individual observations. We find that women on average give 4 percentage points more than men (Cohen’s d=0.16), and that this difference decreases to 3.1% points (Cohen’s d=0.13) if we exclude studies where dictators can only give all or nothing. The gender difference is larger if the recipient in the DG is a charity, compared to the standard DG with an anonymous individual as the recipient (a 10.9 versus a 2.3% points gender difference). These effect sizes imply that many individual studies on gender differences are underpowered; the median power in our sample of standard DG studies is only 9% to detect the meta-analytic gender difference at the 5% significance level. Moving forward on this topic, sample sizes should thus be substantially larger than what has been the norm in the past.

Dictator game variants with probabilistic (and cost-saving) payoffs: A systematic test

• I test the validity of dictator game variants with probabilistic payoffs. • I include probabilistic decision or player payment, and role uncertainty. • I relate transfers to complementary individual difference measures of prosociality. • Data show that dictator behavior is sensitive to the applied methods. • Results confirm that role uncertainty in particular distorts validity. Driven by methodological concerns, theoretical considerations, and previous evidence, I systematically test the validity of common dictator game variants with probabilistic payoffs. Using a unified experimental framework, I include four approaches and compare them to a standard dictator game: involving fewer receivers than dictators, paying only some players, paying only some decisions, and role uncertainty. I also relate transfers in the dictator game variants to established complementary individual difference measures of prosociality: social value orientation, personal values, a donation to charity, and the Big Five personality factor agreeableness. My data shows that the standard dictator game presents the expected correlations with the complementary measures of prosociality. Involving fewer receivers yields comparably valid results. By contrast, when only some players or decisions are paid or, particularly, when subjects face role uncertainty, the expected associations with complementary prosociality measures are distorted. Under role uncertainty, generosity is also significantly biased upward. I conclude that the validity of dictator game outcomes is highly sensitive to the applied methods. Not all dictator game variants can be recommended for the valid measurement of social preferences.

Self-selecting into being a dictator: Distributional consequences

We allow for principals to self-select into delegating, or not, the allocation decision to an agent in a modified dictator game. The standard dictator game arises when principal´s choose to make the allocation decision themselves. Dictators thus obtained transfer lower amounts to receivers, relative to when the decision making is passed to an agent under delegation (or in the standard dictator game). Principals choose to be a dictator nearly half of the time. The average amount transferred by individuals who delegate in more than half of the rounds is significantly higher than the quantity transferred by those who choose to delegate in less than half of the rounds. Finally, the distributional consequences of delegating, or not, vary with less inequality obtained when the decision is delegated.

Sharing with a stranger: people are more generous with time than money

ABSTRACT With the present study, we tested whether generosity changes dependent on money or time being shared. During the experiment, participants N = 371 (MAge = 37.5 years, 38.8% female) completed questionnaires measuring social value orientation, moral identity centrality, and honesty-humility. The opportunity cost of time spent on a real effort task was measured with an incentivized method. Then, participants played two versions of a dictator game: either in a standard dictator game, where participants could share payoffs from the real effort task; or in a time dictator game, where participants decided how long they want to work for another participant’s payoff. We tested three hypotheses. (a) Time and money are not equivalent, and participants are more generous with time than with money. (b) Giving time results in higher positive affect than giving money. (c) Participants’ social value orientation, moral identity centrality, and honesty-humility explain the difference between the donations of time and money, and personality traits will have a stronger impact on time decisions than on monetary decisions. We found that approximately 50% of participants were more generous when giving time, this effect was not dependent on the opportunity cost of time. We think that our experiment is the first experiment to unambiguously show this effect. Furthermore, generosity was not related to positive affect and we found no moderating effect of personality traits.

Are elites really less fair-minded?

US power ‘elites’ are substantially less fair-minded than ‘non-elite’ general populations claims a study by Ray Fisman and coauthors [Science 349, 6254 (2015)]. This supposedly explains why US governments, run by people less fair than the citizens they represent, have been uninclined to tackle inequality. We critically replicated the study, because different experimental protocols for ‘elites’ (interactive variants) and ‘non-elites’ (standard dictator games) were used to measure preferences. We find that the protocol potpourri drove the conclusion: under standard protocols, we find no significant class differences, if any opposite ones. Our work highlights the risk of producing false positives when ex post results appear invitingly plausible.

COMMENT ON “GIVING ACCORDING TO GARP: AN EXPERIMENTAL TEST OF THE CONSISTENCY OF PREFERENCES FOR ALTRUISM”

We present clean experimental evidence that a methodological confound was introduced by Andreoni and Miller (2002) that leads to diametrically opposed conclusions regarding comparisons of preferences between categories of fellow human beings distinguished by gender or age. Our study is a warning not to run the Andreoni-Miller tests of the consistency of altruistic preferences on the basis of their ‘interactive’ experimental protocol, and in particular not to perform preference estimations with the purpose of said comparisons. It is more interpretable and controllable to run the traditional, standard dictator game.

Are Women More Generous Than Men? A Meta-Analysis

We perform a meta analysis of gender differences in the dictator game (DG) by collecting raw data from 53 studies with 117 conditions, giving us 15,016 unique individual observations. We find that women on average give 4 percentage points more than men (Cohen's d=0.16), and that this difference decreases to 3.1 percentage points (Cohen's d=0.13) if we exclude studies where dictators can only give all or nothing. The gender difference is larger if the recipient in the DG is a charity, compared to the standard DG with an anonymous individual as the recipient (a 10.9 versus a 2.3 percentage points gender difference). These effect sizes imply that many individual studies on gender differences are underpowered; the median power in our sample of standard DG studies is only 9% to detect the meta-analytic gender difference at the 5% significance level. Moving forward on this topic, sample sizes should thus be substantially larger than what has been the norm in the past.

Motivational crowding out effects in charitable giving: Experimental evidence

This paper tests motivational crowding out in the domain of charitable giving. A novelty is that our experiment isolates alternative explanations for the decline of giving, such as strategic considerations of decision-makers. Moreover, preference elicitation allows us to focus on the reaction of donors characterized by different degrees of intrinsic motivation. In the charitable-giving setting subjects donate money to the German “Red Cross” in two consecutive donations. The first dictator game is modified, i.e., donors face with equal probability an ex post reimbursement or a subsequent payment. The second game is a standard dictator game. We find that after an ex post change in the price of giving of the first donation, substantially more donors with a high degree of intrinsic motivation decrease donations than subjects with a low degree of intrinsic motivation and donors who did not experience a price effect. In a replication study we find support for these results for subjects who have previously participated in at least one economic experiment.

Deciding on behalf of others does not mitigate selfishness: An Experiment

We test whether deciding on behalf of a passive third party makes participants less selfish in a subsequent decision on behalf of themselves. We find that, in a standard dictator game and in a modified dictator game that allows for ”moral wiggle room”, the experience of having decided for others does not mitigate selfishness

Gender and altruism in a random sample

We study gender differences in altruism in a large random sample of the Swedish population using a standard dictator game. Beside a baseline treatment we implement a priming treatment where participants are reminded of their gender, and two treatments with known male and female counterpart respectively. We find suggestive evidence that women are more altruistic than men only in the priming treatment. A post-hoc analysis using data on interviewer gender to explore gender context effects indicates that priming affects behavior only in mixed-gender contexts.

Shall we pay all? An experimental test of Random Incentivized Systems

This study seeks to investigate the impact of payment procedures on behaviour in a standard Dictator game. Although the Between-subjects Random Incentivized System (BRIS) is increasingly utilized in experimental economics, its use could be contested from a purely theoretical point of view. In the BRIS, all subjects play the Dictator game, but only a percentage of subjects (e.g., 1 out of 10) are paid. The aim of this article is to test whether the results obtained using this system are consistent with those obtained using a conventional full payment system in which every subject receives payment. An additional, hypothetical treatment is also run. Results show a very similar pattern between the BRIS and the full payment treatment, whereas the hypothetical payment mechanism leads to far fewer egoistic players and far more egalitarians. These findings lend support to the use of BRIS as a valid experimental methodology.

The Ring of Gyges in the Laboratory: The Effect of Tangibility and Earned Money on Giving and Taking

We report the results of a real-donation experiment in which we test the effect on giving to charity of tangible and intangible house and earned money. We also expand the action set, allowing our participants to take from, as well as give to, their charities. In our intangible treatments, the allocation decision is indicated on a standard decision sheet. In our tangible treatments, endowments are received as currency and participants indicate their allocation decisions by anonymously allocating their endowments between own and charity envelopes. We test if in a highly anonymous environment with cash in hand, will participants give to, or take from charity? We also test if participants’ behavior depends on the nature of their endowment: earned or house money? We report three results. First, tangible endowments are found to reduce, but not eliminate, charitable giving. Second, there is significantly more taking in the tangible treatments than the intangible treatments. Finally, we do not observe a consistent earned money effect. With tangible endowments, earned money reduce giving. For intangible endowments, we find that earned money increase giving. This is in contrast to prior studies that found that earning one’s endowment reduced giving in standard dictator games (i.e., those with anonymous other subjects as recipients). Our results suggest that the nature of the recipient, and possibly the action set, are important factors for giving decisions in the laboratory. Implications for experimental design are discussed.

Motivational Crowding Out Effects in Charitable Giving: Experimental Evidence

This paper tests motivational crowding out in the domain of charitable giving. A novelty is that our experiment isolates alternative explanations for the decline of giving such as strategic considerations of decision makers. Moreover, preference elicitation allows us to focus on the reaction of donors characterized by different degrees of intrinsic motivation. In the charitable-giving setting subjects donate money to the German Red Cross in two consecutive stages. The first dictator game is modified, i.e., donors face with equal probability an ex post reimbursement or a subsequent pay. The second game is a standard dictator game where we control for the decline of giving. We find that subjects with a high degree of intrinsic motivation, who received a reimbursement, reduce their donations more than four times as much as equally motivated individuals who did not experience a payment.

On the Validity of Cost-Saving Payment Methods in Dictator-Game Experiments: A Systematic Test

Abstract Driven by methodological concerns, theoretical considerations, and previous evidence, we systematically test the validity of common dictator-game variants with probabilistic incentives. We include four approaches and compare them to a standard dictator game: involving fewer receivers than dictators; paying only some subjects or decisions; role uncertainty at the time of the transfer decision. We also relate the dictator-game variants to established complementary individual difference measures of pro-sociality: social value orientation, personal values, a donation to charity, and Big Five personality factors. Our data show that dictator behavior is quite sensitive to the applied methods. The standard version of the dictator game has the highest validity. Involving fewer receivers than dictators and not paying for all decisions yields comparably valid results. By contrast, when only some subjects are paid or when subjects face uncertainty about their final player role, the expected associations with complementary pro-sociality measures are distorted. Under role uncertainty, generosity is biased upwards. We conclude that the validity of the dictator-game outcomes is highly sensitive to the applied methods.

The Sticky Anchor Hypothesis: Ego Depletion Increases Susceptibility to Situational Cues

AbstractSelf‐control depletion has been linked both to increased selfish behavior and increased susceptibility to situational cues. The present research tested two competing hypotheses about the consequence of depletion by measuring how people allocate rewards between themselves and another person. Seven experiments analyzed behavior in standard dictator games and reverse dictator games, settings in which participants could take money from another person. Across all of these experiments, depleted participants made smaller changes to the initial allocation, thereby sticking closer to the default position (anchor) than non‐depleted participants. These findings provide support for a “sticky anchor hypothesis,” which states that the effects of depletion on behavior are influenced by the proximal situational cues rather than by directly stimulating selfishness per se. Copyright © 2017 John Wiley & Sons, Ltd.

To know or not to know? Looking at payoffs signals selfish behavior, but it does not actually mean so

Costs and benefits of everyday actions are often not known beforehand. In such situations, people can either make a choice “without looking” at the payoffs, or they can “look” and learn the exact payoffs involved before making the actual choice. Recent studies suggest that the mere act of looking at payoffs will be met with distrust by observers: “lookers” are both less trustworthy and perceived to be less trustworthy than “non-lookers”. Here we extend this line of work by changing the measure of pro-sociality: instead of trustworthiness, we consider altruism. Does “looking at payoffs” signal self-regarding preferences? Do observers’ beliefs match decision makers’ actions? Two experiments demonstrate that: (i) the level of altruism among “lookers” is not different from the level of altruism among “non-lookers”, but (ii) “lookers” are perceived to be less altruistic than “non-lookers”. These results hold both when the measure of altruism is the choice whether to help or not in the so-called “envelope game” (Experiment 1) or when the measure of altruism is the donation in a standard Dictator Game (Experiment 2). In sum, these results uncover a perception gap according to which looking at payoffs signals selfish behavior, but it does not actually mean so.

Individual Differences in Good Manners Rather Than Compassion Predict Fair Allocations of Wealth in the Dictator Game.

One of the most common tools for studying pro-sociality is the dictator game, in which allocations to one's partner are often described in terms of altruism. However, the motivations driving these allocations may represent either emotional concern for others (compassion), adherence to social norms regarding fairness (politeness), or both. In this article, we apply personality psychology to the study of behavior in the dictator game, in which we examine the discriminant validity of distinct pro-social constructs from the Big Five and HEXACO models in relation to allocations of wealth. Across four studies (Study 1: N = 192; Study 2: N = 212; Study 3: N = 304; Study 4: N = 90) utilizing both hypothetical and incentivized designs, we found that the politeness-but not compassion-aspect of Big Five Agreeableness, as well as HEXACO Honesty-Humility, uniquely predicted dictator allocations within their respective personality models. These findings contribute to a growing literature indicating that the standard dictator game measures "good manners" or adherence to norms concerning fairness, rather than pure emotional concern or compassionate motives, and have important implications for how this paradigm is used and interpreted in psychological research.