Social Value Orientation and Cooperation in Social Dilemmas: A Meta-Analysis Daniel Balliet Singapore Management University Craig Parks and Jeff Joireman Washington State University

Social Value Orientation (SVO) The weights people assign to self and other outcomes in interdependent contexts. –Prosocials (altruists and cooperators) –Proselfs (individualists and competitors)

Purpose of Meta-Analysis Central tendency Variation Moderators –Iterated vs. one-shot –Give-some vs. take-some –Paid vs. non-paid participants –Group size

Iterations vs. One-Shot Dilemmas Partner strategy may reduce SVO- cooperation relationship in iterated dilemmas. –Behavioral Assimilation: Prosocials conform to a proself’s defection during iterated social dilemmas (Kelley & Stahelski, 1970; Kuhlman & Marshello, 1975)

Give-Some vs. Take-Some Two reasons to expect SVO may be more predictive of cooperation in G-S than T-S. –Prospect Theory: Losses (G-S) more important than gains (T-S). Therefore, defection and Cooperation more attractive for proselfs and prosocials, respectively (De Dreu & McCusker, 1997) –Equality norm more salient in take-some dilemmas (Van Dijk & Wilke, 1995), and SVO less predictive when equality norm is salient (de Kwaadsteniet et al., 2006).

Paid vs. Non-Paid Dilemmas Two reasons to expect SVO may be more predictive of cooperation in non-paid dilemmas. –Payment may prime a business frame. This may change prosocials ‘communal’ construal of the dilemma to a ‘business’ construal, thereby reducing their level of cooperation. –Payment may reduce expectations of other’s cooperation.

Group Size All SVO measures include decisions while interacting in a dyad. Does this limit SVO’s ability to predict behavior in groups of 3 or more?

Methods and Analyses Criteria for Selection –Adults –Measure SVO (Ring or TDM) –DV is strict social dilemma Analyses –Correlation as effect size –Prosocial vs. Proself –Mixed-Effects models

Results: Main Effect Prosocial vs. Proselfs –Q(81) = 248, p <.001 –(n = 82) r =.30, 95% CI, LL =.26, UL =.33 –Orwin’s fail safe N = 510 Individualists vs. Competitors –Q(18) = 33.9, p =.013 –(n = 19) r =.27, 95% CI, LL =.19, UL =.34 –Orwin’s Fail Safe N = 110

Results: Moderators Iterated vs. One-Shot –Q (1) =.157, p =.692 –One-Shot: r =.31, 95% CI, LL =.26, UL =.35 –Iterated: r =.29, 95% CI, LL =.24, UL =.34 Give-Some vs. Take-Some –Q (1) = 5.26, p =.022 –Give: r =.29, 95% CI, LL =.25, UL =.34 –Take: r =.22, 95% CI, LL =.17, UL =.27

Results: Moderators Paid vs. Non-Paid Participants –Q(2) = 20.6, p <.001 –Not Paid: r =.39, 95% CI, LL =.33, UL =.45 –Lottery: r =.36, 95% CI, LL =.23, UL =.47 –Paid: r =.23, 95% CI, LL =.20, UL =.26 Group Size –Slope = -.007, Q(1) = 4.75, p =.03

Future Directions Payment X SVO Implications –Is an implicit measure of SVO needed? Research needed on SVO in the context of iterated social dilemmas. –Forgiveness Why is SVO more predictive of G-S games? SVO in strong vs. weak situations. –Do paid/take-some dilemmas have more uniform expectations, construal, and incentives for appropriate behavior?

Questions? Balliet, D., Parks, C., & Joireman, J. (2009). Social value orientation and cooperation in social dilemmas: A meta- analysis. Group Processes and Intergroup Relations, 12,