Springer Nature is making SARS-CoV-2 and COVID-19 research free. View research | View latest news | Sign up for updates

Double or nothing?! Small groups making decisions under risk in “Quiz Taxi”

  • 321 Accesses

  • 1 Citations


This paper investigates the behavior of contestants in the game show “Quiz Taxi” when faced with the decision whether to bet the winnings they have acquired on a final “double or nothing” question. The decision in this natural experiment is made by groups of two or three persons. This setup enables the decision-making process to be studied with regard to group and communication characteristics. The contestants show fairly risk averse behavior. There is also a significant heterogeneity in attitude to risk. In particular, all-female groups are much less likely to go for the risky option. Furthermore, decision-making behavior appears to vary across differently composed groups and prior performance. The study also documents the importance of discussions: The propensity to gamble increases with discussion length, and the correlation between communication content and the final choice is strong, indicating that time and subjective context are important features of decisions made under risk.

This is a preview of subscription content, log in to check access.


  1. 1.

    The game show “Quiz Taxi” is the German version of the English show “Cash Cab” which first aired in 2005. It was sold to over 25 countries and is still running in several of those, e.g., USA and Australia.

  2. 2.

    Since “Quiz Taxi” contestants know about the design of the decision problem and must have subjective probabilities for the possible outcomes, their decisions are made under risk, not under uncertainty. Therefore, the term “risk” is used in the following (see Knight 1921, for the seminal contribution about the difference between risk and uncertainty).

  3. 3.

    “Deal or No Deal” has become very popular among economists. See Blavatskyy and Pogrebna (2008, 2010), De Roos and Sarafidis (2010), and Gee (2007) for further investigations.

  4. 4.

    In the first season, the master question had not been introduced yet and the monetary rewards were lower. In the last season, a new wild card was introduced by which the group composition could change during the cab ride.

  5. 5.

    In the interview mentioned above, the host notes that the reasons for backing out of the cab are very diverse. He speculates that some people are afraid that they might make a fool of themselves, simply do not want to be on TV, or are just accompanied by the wrong persons. In two episodes, people who declined to play were shown. Their reasons for not participating were time constraints.

  6. 6.

    According to the producer, basically all cab rides recorded were aired, especially if the master question was reached. Therefore, an ex-post selection process can be ruled out.

  7. 7.

    Contestants can win back a previously used passersby-wild card by successfully solving an additional task when the cab stops at a red traffic light.

  8. 8.

    The age is estimated by appearance of the contestants if no direct information is revealed during the show. Three broad categories are used: below 30 years of age, 30–50 years of age, and above 50 years of age. Due to a very low number of observations in the last group, the latter two are combined for the empirical analysis.

  9. 9.

    Since the process of recording this data is subject to measurement error, all information obtained from the transcripts is cross-checked. Furthermore, controlling for the person who coded the data does not have any influence on the obtained estimation results.

  10. 10.

    We treat the group decision basically like the decision of an individual. While this abstracts from explicitly modeling the aggregation of the potentially different preferences of the group members, accounting for group characteristics and looking separately at groups that are initially undecided allows for an implicit description of how the groups might deal with diverging opinions.

  11. 11.

    A linear probability model (OLS) or a logit model deliver identical estimation results. The estimation results are available from the authors upon request.

  12. 12.

    The argument for setting the wealth level to zero is based on prospect theory (Kahneman and Tversky 1979) where individuals are assumed to evaluate gains and losses directly and not with regard to their overall wealth (see also Beetsma and Schotman 2001; Fullenkamp et al. 2003). With wealth set to zero, the critical probability is \(p_{i}^{*} = 0.5^{1-\gamma }\) with \(0 < \gamma < 1\).

  13. 13.

    The assumption of low wealth also serves the purpose of being able to estimate a \(\gamma \ge 1\) which is undefined in the case of zero wealth, but not likely given the observed decisions. For \(p_{i} = 0.75\), \(\gamma \) is estimated at approximately 5, but the maximum likelihood procedure does not yield completely stable results for high probabilities of success because expected utility maximization would imply choosing to play the master question in most cases which is incompatible with 67 % of groups not doing so.

  14. 14.

    The explanatory power of this regression is rather low (Pseudo \(R^{2}\) of 0.07). The predicted probability’s mean is 0.76 with a standard deviation of 0.12, a minimum of 0.36, and a maximum of 0.98.

  15. 15.

    Clearly, the comparability of all of these estimates is very limited because they are obtained from very different samples over very different monetary ranges using various methodologies. Therefore, the external validity of estimates of risk aversion parameters is questionable (Rabin 2000).

  16. 16.

    The statistical difference of the coefficients is tested by estimating fully interacted linear OLS and Heckman regressions. The linear models do not yield different results than the probit models, but facilitate the test of statistical significance of the interaction terms (Ai and Norton 2003). The significance level is at least 10 %, but 5 % or 1 % in most cases. When the sample is split into three groups, bold font indicates that at least one group’s coefficient is significantly different from the other two, but not necessarily that all coefficients are significantly different from each other.

  17. 17.

    In this sample split only, the Heckman models deliver a somewhat different result: The coefficient of the stake is more negative for female and mixed teams, but not in a statistically significant way (Table 3). Since the correlation between the risky choice and the stake can be confounded by the unobservable subjective probability of answering the master question correctly, this difference in estimation results suggests that the sex composition affects risk attitude as well as the perceived chances of success or a group’s enthusiasm.

  18. 18.

    Less than ten teams mention that they have seen the show before. Their views on the difficulty of the question are diverse. While the order in which the episodes were aired does not necessarily correspond to the order in which they were produced, the two should be almost identical. Seasons are arranged in pairs to avoid too small samples.


  1. Adams, R., & Ferreira, D. (2010). Moderation in groups: Evidence from betting on ice break-ups in Alaska. Review of Economic Studies, 77(3), 882–913.

  2. Ai, C., & Norton, E. C. (2003). Interaction terms in logit and probit models. Economics Letters, 80(1), 123–129.

  3. Andersen, S., Harrison, G. W., Lau, M. I., & Rutstroem, E. E. (2008). Risk aversion in game shows. In G. W. Harrison & J. C. Cox (Eds.), Risk aversion in experiments. Research in experimental economics (pp. 359–404). Bingley: Emerald Publisher. Vol. 12.

  4. Attanasio, O. P., & Weber, G. (1989). Intertemporal substitution, risk aversion and the Euler equation for consumption. Economic Journal, 99(395), 59–73.

  5. Baker, R. J, I. I., Laury, S. K., & Williams, A. W. (2008). Comparing small-group and individual behavior in lottery-choice experiments. Southern Economic Journal, 75(2), 367–382.

  6. Barsky, R. B. (1997). Preference parameters and behavioral heterogeneity: An experimental approach in the Health and Retirement Study. Quarterly Journal of Economics, 112(2), 537–79.

  7. Beetsma, R. M. W. J., & Schotman, P. C. (2001). Measuring risk attitudes in a natural experiment: Data from the television game show “Lingo”. Economic Journal, 111(474), 821–848.

  8. Blavatskyy, P., & Pogrebna, G. (2008). Risk aversion when gains are likely and unlikely: Evidence from a natural experiment with large stakes. Theory and Decision, 64(2), 395–420.

  9. Blavatskyy, P., & Pogrebna, G. (2010). Endowment effects? “Even” with half a million on the table!. Theory and Decision, 68(1), 173–192.

  10. Bliss, R. T., Potter, M. E., & Schwarz, C. (2012). Decision making and risk aversion in the cash cab. Journal of Economic Behavior and Organization, 84(1), 163–173.

  11. Brooks, R., Faff, R., Mulino, D., & Scheelings, R. (2009). Deal or no deal, that is the question: The impact of increasing stakes and framing effects on decision-making under risk. International Review of Finance, 9(1–2), 27–50.

  12. Cohen, A., & Einav, L. (2007). Estimating risk preferences from deductible choice. American Economic Review, 97(3), 745–788.

  13. De Roos, N., & Sarafidis, Y. (2010). Decision making under risk in “Deal or No Deal. Journal of Applied Econometrics, 25(6), 987–1027.

  14. Dohmen, T., Falk, A., Huffman, D., Marklein, F., & Sunde, U. (2009). Biased probability judgment: Evidence of incidence and relationship to economic outcomes from a representative sample. Journal of Economic Behavior and Organization, 72(3), 903–915.

  15. Dohmen, T., Falk, A., Huffman, D., Sunde, U., Schupp, J., & Wagner, G. G. (2011). Individual risk attitudes: Measurement, determinants, and behavioral consequences. Journal of the European Economic Association, 9(3), 522–550.

  16. Donkers, B., Melenberg, B., & Van Soest, A. (2001). Estimating risk attitudes using lotteries: A large sample approach. Journal of Risk and Uncertainty, 22, 165–195.

  17. Fehr-Duda, H., de Gennaro, M., & Schubert, R. (2006). Gender, financial risk, and probability weights. Theory and Decision, 60, 283–313.

  18. Fehr-Duda, H., Epper, T., Bruhin, A., & Schubert, R. (2011). Risk and rationality: The effects of mood and decision rules on probability weighting. Journal of Economic Behavior and Organization, 78(1–2), 14–24.

  19. Fullenkamp, C., Tenorio, R., & Battalio, R. (2003). Assessing individual risk attitudes using field data from lottery games. Review of Economics and Statistics, 85(1), 218–226.

  20. Gee, C. (2007). Risky choice and type-uncertainty in Deal or No Deal?’. Cambridge Working Papers in Economics 758, Faculty of Economics: University of Cambridge November 2007.

  21. Gertner, R. (1993). Game shows and economic behavior: Risk-taking on “Card Sharks”. Quarterly Journal of Economics, 108(2), 507–521.

  22. Hall, R. E. (1988). Intertemporal substitution in consumption. Journal of Political Economy, 96(2), 339–57.

  23. Harrison, G. W., & List, J. A. (2004). Field experiments. Journal of Economic Literature, 42(4), 1009–1055.

  24. Hartley, R., Lanot, G., & Walker, I. (2006). Who really wants to be a millionaire? Estimates of risk aversion from gameshow data. The Warwick Economics Research Paper Series (TWERPS) 747, University of Warwick, Department of Economics.

  25. Hersch, P. L., & McDougall, G. S. (1997). Decision making under uncertainty when the stakes are high: Evidence from a lottery game show. Southern Economic Journal, 64(1), 75–84.

  26. Holt, C. A., & Laury, S. K. (2002). Risk aversion and incentive effects. American Economic Review, 92(5), 1644–1655.

  27. Jianakoplos, N., & Bernasek, A. (1998). Are women more risk averse? Economic Inquiry, 36(4), 620–630.

  28. Johnson, D. K. N., & Gleason, T. R. (2009). Who really wants to be a millionaire? Gender differences in game show contestant behavior under risk. Social Science Quarterly, 90(2), 243–261.

  29. Kahneman, D. (2003). A perspective on judgment and choice: Mapping bounded rationality. American Psychologist, 58(9), 697–720.

  30. Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–291.

  31. Knight, F. H. (1921). Risk, uncertainty, and profit, Library of Economics and Liberty. Retrieved May 17, 2011 from http://www.econlib.org/library/Knight/knRUP.html.

  32. List, J. A. (2006). Friend or foe? A natural experiment of the prisoner’s dilemma. Review of Economics and Statistics, 88(3), 463–471.

  33. Matsen, E., & Strøm, B. (2010). Dominated choices in a simple game with large stakes. Experimental Economics, 13, 99–119.

  34. Metrick, A. (1995). A natural experiment in “Jeopardy!”. American Economic Review, 85(1), 240–253.

  35. Post, T., van den Assem, M. J., Baltussen, G., & Thaler, R. H. (2008). Deal or no deal? Decision making under risk in a large-payoff game show. American Economic Review, 98(1), 38–71.

  36. Rabin, M. (2000). Risk aversion and expected utility theory: A calibration theorem. Econometrica, 68(5), 1281–1292.

  37. Rockenbach, B., Sadrieh, A., & Mathauschek, B. (2007). Teams take the better risks. Journal of Economic Behavior and Organization, 63(3), 412–422.

  38. Stanovich, K. E., & West, R. F. (2000). Individual differences in reasoning: Implications for the rationality debate. Behavioral and Brain Science, 23, 645–665.

Download references


This paper solely reflects the personal views of the authors and not necessarily those of the German Council of Economic Experts or any other institution the authors are affiliated with. We are highly indebted to Tower Productions, nobeo and ProSiebenSat1 for supporting the screening of the archived “Quiz Taxi” episodes, in particular Mark Land, Bernd Berghoff, Markus Schäfer, Rene Steinbusch, Thomas Junggeburth, and Joachim Drees. We thank Jeannette Brosig-Koch, John P. Haisken-DeNew, Christoph M. Schmidt, Thomas K. Bauer, Maarten van Kampen, the anonymous referees, and seminar participants at the 4th RGS Doctoral Conference in Economics 2011, the 2011 Annual Conference of the Scottish Economic Society, the 1st RGS Jamboree 2010, the University of Duisburg-Essen, and the Ruhr-University Bochum for helpful comments and suggestions. We also thank two “Quiz Taxi” contestants for providing us with additional information, Peter Schotman and Philip Hersch for providing us with the estimation codes that they used in their studies, and Michael Klemm for help with the tool that was used to record the data. Financial support by the RGS Econ and the Ruhr-University Bochum is gratefully acknowledged. Marcus Klemm also thanks the Ruhr-University Research School funded by Germany’s Excellence Initiative for further support [DFG GSC 98/1].

Author information

Correspondence to Marcus Klemm.



See Appendix Tables 9, 10, 11.

Table 9 Variable definitions
Table 10 Variable means by group size, sex and age composition, lives left, and season
Table 11 Variable means by initial group opinion, discussion length, and topics discussed

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Keldenich, K., Klemm, M. Double or nothing?! Small groups making decisions under risk in “Quiz Taxi”. Theory Decis 77, 243–274 (2014). https://doi.org/10.1007/s11238-013-9398-8

Download citation


  • Risk attitude
  • Game show
  • Communication
  • Group decision