Games

= Games = toc

**Definition**
Experimental games are social interactions, structured by rules for play that allow researchers to examine behavior and by extension reveal social norms that have implications for economic theory .

**Relevant Characteristics**
Experimental games were first developed and used in laboratory settings in the U.S. (mostly on college campuses) but anthropologists and other researchers have since adapted these games for use in the field. Using the results of lab experimental games, researchers have established some experimental regularities that anthropologists have used as benchmarks for their findings in the field.

 Research participants play anonymously for real money, and efforts are made to minimize deception in experimental design. Common experimental games include: prisoners’ dilemma, public goods games, ultimatum game, dictator game, trust game, gift exchange game, third party punishment game.

 There is no minimum sample size to use this method in the field but a larger sample size will usually allow researchers to address more questions and data patterns.

 Economic games are intended to test attitudes around economic cooperation and reciprocity, such as the presence of in-group favoritism. They have been used in cross-cultural comparison of cooperation. Researchers attempt to control for as many factors as possible in these interactions, but anthropologists have shown that all games are framed (intentionally or unintentionally) in some way and have begun in recent years to look at “framing effects” in experimental games. Such studies  look at the extent to which participants play the game differently when it is associated or explained with terms that refer to certain culturally specific forms of collectives or cooperatives.

**“Method Made Easy”**

 * 1) Select a type of game appropriate to your research topic or design your own game. This step may include adapting a given game to a particular context using ethnographic information.
 * 2) If possible, pre-test the game (most likely this will be among non-participants).
 * 3) Choose a facilitator appropriate to game goals and context, keeping in mind that the facilitator’s relationship with participants will likely influence game outcomes.
 * 4) Recruit participants from the group(s) of interest.
 * 5) Explain rules clearly to all participants, including a way of testing game comprehension. The facilitator may want to do a quick demonstration of the game, or a test round with participants.
 * 6) Play the game. Repeat if necessary. Take notes on player deliberations and results throughout.
 * 7) After the game, interview each player individually, asking them to explain their decisions in the game.

**Advantages**

 * Comparability across subject groups
 * Replicability
 * Advantage of experimental games in the field:
 * <span style="font-family: Arial,sans-serif; vertical-align: baseline;">They are low cost.
 * <span style="font-family: Arial,sans-serif; vertical-align: baseline;">Funds granted by funding agencies have higher purchasing power in many other countries, which helps in analysis of low stakes and high stakes behavior.
 * <span style="font-family: Arial,sans-serif; vertical-align: baseline;">Games provide an alternative to the Likert scale to study trust and other social norms.
 * <span style="font-family: Arial,sans-serif; vertical-align: baseline;">More concrete questions tied to actual behavior can complement the other behavioral data anthropologists are likely to have.

**Limitations**

 * <span style="font-family: Arial,sans-serif; vertical-align: baseline;">Outside of US lab settings, researchers have documented difficulties in explaining the games, which often seem strange and confound research participants.
 * <span style="font-family: Arial,sans-serif; vertical-align: baseline;">“Framing effects” are difficult to control. There will likely be unforeseen framing effects including terms used in explanation, the space in which the game is played, and the identity of game facilitator.
 * <span style="font-family: Arial,sans-serif; vertical-align: baseline;">In their original form, such games were meant to be administered by someone not known to the research subjects; anthropologists usually are well known to their research subjects.
 * <span style="font-family: Arial,sans-serif; vertical-align: baseline;">Research subjects typically approach these games as they do informal social situations; therefore the games’ results may not reveal norms that come into play in other more formal or organized settings.
 * <span style="font-family: Arial,sans-serif; vertical-align: baseline;">Money does not necessarily have a uniform value within or across cultures.

**Analysis**

 * <span style="font-family: Arial,sans-serif;">Detailed fieldnotes can be analyzed and coded as text with the aid of software.
 * <span style="font-family: Arial,sans-serif; vertical-align: baseline;">Participants’ game strategies can be documented and analyzed statistically, including multivariate and bivariate analyses. This may not be the focus of your study if you choose to use games primarily as an elicitation technique.

**Method in Context**
<span style="font-family: Arial,sans-serif;">Some special issues arise from the use of real money in experimental games. Researchers should aim to make a distinction between compensation for participation and game winnings, as well as be aware of the different values that can be ascribed to money across and within different communities.

<span style="font-family: Arial,sans-serif;"> Experimental games are traditionally used by economists to develop and test theories about human preferences. These experiments and analyses are informed by game theory, the study of mathematical models used to describe and predict strategic behavior. Work in economics in the 1990s using experimental games, such as the ultimatum game, supported a theory of rational self-interest <span style="font-family: Arial,sans-serif;">. However, cross-cultural ethnographic work later in the decade came to show the existence of altruistic preferences, and “preferences for equality and reciprocity” alongside selfishness. Controlled experiments often demonstrate that people’s behavior differs from that hypothesized based on game theory. These diversions can be interpreted to support the existence of social norms.

<span style="font-family: Arial,sans-serif;"> Experimental games are usually first run in strict laboratory environments, so that the results serve as a comparison for further contextualized study. Ethnographic data can also be coupled with experimental data to assess the external validity of economic games.

<span style="font-family: Arial,sans-serif;"> Relevant quotations on anthropology and economic games:

<span style="font-family: Arial,sans-serif;"> "Broadly, the findings of experimental economics have shown that individual self-interest is not a good assumption for the behaviour of human subjects and that behaviours derive not from fixed personal qualities but from combinations of person, other persons in the game, the game itself and the context. Players are differentially averse to inequality of outcomes; reciprocity is an important motivating force; players care about actions and intentions of other players and not only about the allocation of material payoffs. These findings are much closer to the more social concept of persons and behaviours which are the foundations of anthropology than the self-interested individual of economic stereotypes and therefore offer a promising starting point for interdisciplinary and multidisciplinary research."

<span style="font-family: Arial,sans-serif;"> “Whether frames are explicitly included by the researcher or spontaneously applied by the participants, their influence on patterns of behavior in experimental economic games must be accounted for when developing and testing theories of cooperation. Anthropologists are uniquely situated to turn this methodological weakness of economic games into an analytic strength. By combining standard versions of economic games with games that are explicitly framed to refer to naturally occurring contexts of cooperation, researchers can better understand how cultural values, norms, and institutions influence decisions while also retaining the ability to compare patterns of cooperation across space and time. In turn, ethnographers may gain new insights about the particulars of people and place by incorporating economic games and other field experiments into ongoing research with more traditional ethnographic methodologies.”

<span style="font-family: Arial,sans-serif;"> “A brief and completely open individual discussion with players after the introduction and explanation, and before the actual play, might show many things about what players are thinking when they play. Postgame interviews are very revealing about the ideas at play in the gaming, but they need to be accompanied by more analysis of broader social and cultural contexts for action and an epistemological stance which recognizes varying degrees of reflexivity at work, is alert to ethnocentric assumptions about behaviour and mindful that western subjects are as culturally and historically embedded as subjects in ‘simple’ societies.”

**Online Resources**
<span style="font-family: Arial,sans-serif;">“List of Games in Game Theory.” <span style="font-family: Arial,sans-serif;"> http://en.wikipedia.org/wiki/List_of_games_in_game_theory

<span style="font-family: Arial,sans-serif;"> “Game Theory Icons.” <span style="font-family: Arial,sans-serif;"> https://www.lri.fr/~dragice/gameicons/

**Further Reading**
<span style="font-family: Arial,sans-serif;">Henrich. J., Boyd, R., Bowles, S., Camerer, C., Fehr, E. and Gintis. H., eds. <span style="font-family: Arial,sans-serif;"> 2004. Foundations of human sociality: Economic experiments and ethnographic evidence from fifteen small-scale societies. Oxford: Oxford University Press.

<span style="font-family: Arial,sans-serif;"> Jackson, C. <span style="font-family: Arial,sans-serif;"> 2011. Research with experimental games: questioning practice and interpretation. Progress in Development Studies 11(3): 229-241.

<span style="font-family: Arial,sans-serif;"> Roepstorff, A., Niewöhner, J., Beck, S. <span style="font-family: Arial,sans-serif;"> 2010. Enculturing brains through patterned practices. Neural Networks 23: 1051-1059.